Compare commits

..
Author SHA1 Message Date
jschoubben 858b06a94e Take mesh-sdk at c8e3509, which counts an item's text as a placement of a config value (issue 1658)
mesh/repo-check head 858b06a94e69 merged into trunk 12752911b644 — fail: head 858b06a94e69 does not merge cleanly into trunk 12752911b644: in conflict go.m…
mesh/delivery-group group fix/1656-xterm-config rejected: a member's own check failed
mesh/merge-gate group fix/1656-xterm-config: composed again for the group's new heads
mesh/delivery superseded: a newer head of the same pull request
2026-10-21 12:55:08 +02:00
610 changed files with 4326 additions and 109531 deletions
+12 -3
View File
@@ -90,6 +90,15 @@ redis-provisioner-image:
@echo
@docker image inspect $(REDIS_PROVISIONER_IMAGE) --format 'built {{.RepoTags}} {{.Size}} bytes'
# The proxy that turns a route grant into traffic reaching a workload.
PROXY_IMAGE ?= mesh-route-proxy:$(VERSION)
PROXY_DEV_TAG ?= mesh-route-proxy:development
proxy-image:
docker build --build-arg GO_BASE=$(GO_BASE) -f modules/route-proxy/Dockerfile -t $(PROXY_IMAGE) -t $(PROXY_DEV_TAG) .
@echo
@docker image inspect $(PROXY_IMAGE) --format 'built {{.RepoTags}} {{.Size}} bytes'
# The whole check. Raises a database, runs the repository's own check against it -- merge-check.sh,
# the very script `mesh/repo-check` runs -- and takes the database down again, including when the
# check fails, which is why the teardown is not conditional and the script's exit status is the
@@ -108,7 +117,7 @@ redis-provisioner-image:
# as the script runs them: a bus per test, of the release the mesh runs, makes the suite the same in any
# order, and the timeout bounds a hang to a failure with a stack, never a stalled gate.
check: postgres
@sh merge-check.sh ; status=$$? ; docker rm -f -v $(PG_CONTAINER) >/dev/null 2>&1 || true ; exit $$status
@sh merge-check.sh ; status=$$? ; docker rm -f $(PG_CONTAINER) >/dev/null 2>&1 || true ; exit $$status
# The unit level alone (untagged: no bus, no store, no other process), which a developer runs continuously. The tests
# that need a bus or the store carry `//go:build integration` and run in `check`, not here; without a database the
@@ -126,7 +135,7 @@ fmt:
if [ -n "$$unformatted" ] ; then echo "not gofmt'd:" ; echo "$$unformatted" ; exit 1 ; fi
postgres:
@docker rm -f -v $(PG_CONTAINER) >/dev/null 2>&1 || true
@docker rm -f $(PG_CONTAINER) >/dev/null 2>&1 || true
@docker run -d --name $(PG_CONTAINER) -e POSTGRES_PASSWORD=check \
-p 127.0.0.1:$(PG_PORT):5432 $(PG_IMAGE) >/dev/null
@printf 'waiting for postgres'
@@ -138,7 +147,7 @@ postgres:
echo ' — never came up' ; docker logs $(PG_CONTAINER) | tail -20 ; exit 1
postgres-stop:
@docker rm -f -v $(PG_CONTAINER) >/dev/null 2>&1 || true
@docker rm -f $(PG_CONTAINER) >/dev/null 2>&1 || true
clean:
rm -rf build/
+2 -91
View File
@@ -6,11 +6,7 @@ import (
"os"
"path/filepath"
"strings"
"sync"
"sync/atomic"
"time"
"github.com/novox/mesh-controller/internal/builder"
)
// **The build agent says whether it is busy, and being asked drains it** (novox/hq ADR 0335 §1, §5). Its
@@ -77,77 +73,14 @@ func drainedTooLong(now time.Time) error {
"within %s (novox/hq ADR 0335)", began.UTC().Format(time.RFC3339), drainTooLong)
}
// inHand is true while this process has claimed a build and not yet released it. The mark in the temporary directory
// is only believed while it is: a mark with no build of this process behind it is stale (a crash, a fault in an
// earlier path) and is removed, because left alone it hands back every ask for ever (issue 1768).
var inHand atomic.Bool
// markRetry is how long an agent that cannot mark a build in hand takes no work, and markFault when it last failed to.
// Taking nothing is the point: an ask fetched and handed back at once, many times a second, spends all its deliveries.
const markRetry = 30 * time.Second
var (
markMu sync.Mutex
markFault time.Time
markSaid string
)
// markFaulty says the agent could not mark a build in hand and has not yet been able to since: it takes no work. Once
// markRetry has passed it tries the mark itself, with no ask in hand, so that an ask is fetched only when marking works.
func markFaulty(now time.Time) bool {
markMu.Lock()
failed := markFault
markMu.Unlock()
if failed.IsZero() {
return false
}
if now.Sub(failed) < markRetry {
return true
}
f, err := os.OpenFile(inHandMark(), os.O_WRONLY|os.O_CREATE|os.O_EXCL, 0o644)
if err != nil && errors.Is(err, os.ErrExist) && !inHand.Load() {
_ = os.Remove(inHandMark())
f, err = os.OpenFile(inHandMark(), os.O_WRONLY|os.O_CREATE|os.O_EXCL, 0o644)
}
if err != nil {
markFailed(err)
return true
}
_ = f.Close()
_ = os.Remove(inHandMark())
markMu.Lock()
markFault, markSaid = time.Time{}, ""
markMu.Unlock()
fmt.Fprintf(os.Stderr, "the mark of a build in hand can be made again (%s): taking builds\n", inHandMark())
return false
}
// markFailed notes that a build could not be marked in hand, and says it when it is a new reason.
func markFailed(err error) {
markMu.Lock()
defer markMu.Unlock()
markFault = time.Now()
if said := err.Error(); said != markSaid {
markSaid = said
fmt.Fprintf(os.Stderr, "cannot mark a build in hand (%s), so this agent takes no build for %s and tries again: %v\n",
inHandMark(), markRetry, err)
}
}
// claim marks a request in hand, unless the agent is drained: then nothing is in hand, and the request goes
// back to the queue.
func claim() bool {
if !inHand.Load() {
// Nothing of this process is in hand, so a mark already there is stale.
if err := os.Remove(inHandMark()); err == nil {
fmt.Fprintf(os.Stderr, "removed the mark of a build in hand (%s): no build of this agent is in hand\n", inHandMark())
}
}
// Made, never written over: one build at a time, and a mark already there is a build in hand.
f, err := os.OpenFile(inHandMark(), os.O_WRONLY|os.O_CREATE|os.O_EXCL, 0o644)
if err != nil {
// Unmarked, an ask would say idle while this build runs: the build is not taken.
markFailed(err)
fmt.Fprintf(os.Stderr, "cannot mark a build in hand (%s), so it is handed back: %v\n", inHandMark(), err)
return false
}
_, err = f.WriteString(time.Now().UTC().Format(time.RFC3339) + "\n")
@@ -156,7 +89,7 @@ func claim() bool {
}
if err != nil {
_ = os.Remove(inHandMark())
markFailed(err)
fmt.Fprintf(os.Stderr, "cannot mark a build in hand (%s), so it is handed back: %v\n", inHandMark(), err)
return false
}
if drained() {
@@ -164,7 +97,6 @@ func claim() bool {
fmt.Fprintln(os.Stderr, "asked whether busy, so this agent takes no new build: handed back")
return false
}
inHand.Store(true)
return true
}
@@ -175,7 +107,6 @@ func holding(what string) {
// released ends the build in hand.
func released() {
inHand.Store(false)
if err := os.Remove(inHandMark()); err != nil && !errors.Is(err, os.ErrNotExist) {
fmt.Fprintf(os.Stderr, "cannot remove the mark of the build in hand (%s): %v\n", inHandMark(), err)
}
@@ -213,23 +144,3 @@ func busy() (string, error) {
}
return fmt.Sprintf("busy: %s, in hand since %s; taking no new build", what, since), nil
}
var (
roomMu sync.Mutex
roomSaid bool
)
// noRoom says this node is under the floor of free space (novox/hq issue 1834): it takes no work, so another build node
// takes the ask instead of this one taking it and ending it in an error. Said once when it starts and again when it ends.
func noRoom(workspace string) bool {
err := builder.RoomToWork("/", workspace)
roomMu.Lock()
defer roomMu.Unlock()
if err != nil && !roomSaid {
fmt.Fprintf(os.Stderr, "taking no builds: %v\n", err)
} else if err == nil && roomSaid {
fmt.Fprintf(os.Stderr, "there is room again: taking builds\n")
}
roomSaid = err != nil
return err != nil
}
-77
View File
@@ -1,7 +1,6 @@
package main
import (
"github.com/novox/mesh-controller/internal/builder"
"os"
"strings"
"sync"
@@ -161,79 +160,3 @@ func TestAnUnknownCommandFails(t *testing.T) {
t.Fatalf("help failed: %v", err)
}
}
// 1768/AC2: a mark naming no build of this agent (left by a crash or a fault) is stale: the next ask removes it and is
// run, never fetched and handed back in a loop.
func TestAStaleInHandMarkIsRemovedAndTheAskRuns(t *testing.T) {
t.Setenv("TMPDIR", t.TempDir())
inHand.Store(false)
if err := undrained(); err != nil {
t.Fatal(err)
}
if err := os.WriteFile(inHandMark(), []byte("2026-10-21T18:45:06Z\n"), 0o644); err != nil {
t.Fatal(err)
}
if !claim() {
t.Fatal("a stale in-hand mark made the agent hand back an ask")
}
// A live one — this agent's own build in hand — is never taken over.
if claim() {
t.Fatal("a second build was taken while one is in hand")
}
if _, err := busy(); err != nil {
t.Fatalf("the build in hand was lost: %v", err)
}
released()
}
// 1768/AC3: an agent that cannot mark a build in hand for another reason takes no work, instead of fetching asks and
// handing them back; it takes work again once marking works.
func TestAnAgentThatCannotMarkABuildTakesNoWorkUntilItCan(t *testing.T) {
dir := t.TempDir()
t.Setenv("TMPDIR", dir)
inHand.Store(false)
markFault, markSaid = time.Time{}, ""
t.Cleanup(func() { markFault, markSaid = time.Time{}, "" })
if err := undrained(); err != nil {
t.Fatal(err)
}
// The mark's place is a directory that cannot be removed or made over: a directory named as the mark.
if err := os.MkdirAll(inHandMark()+"/x", 0o755); err != nil {
t.Fatal(err)
}
if claim() {
t.Fatal("an agent that cannot mark a build took it")
}
if !markFaulty(time.Now()) {
t.Fatal("an agent that cannot mark a build still takes work")
}
// Still failing after the retry time: still takes nothing, and no ask was needed to find out.
if !markFaulty(time.Now().Add(2 * markRetry)) {
t.Fatal("an agent whose mark still cannot be made takes work")
}
if err := os.RemoveAll(inHandMark()); err != nil {
t.Fatal(err)
}
if markFaulty(time.Now().Add(4 * markRetry)) {
t.Fatal("an agent whose mark can be made again takes no work")
}
if !claim() {
t.Fatal("an agent that can mark takes no build")
}
released()
}
// 1834/AC4: given a node under the floor of free space, when the taking loop asks, then it takes nothing, and takes again
// once there is room.
func TestANodeUnderTheFloorOfFreeSpaceTakesNoWork(t *testing.T) {
was := builder.FreeBytes
t.Cleanup(func() { builder.FreeBytes = was })
builder.FreeBytes = func(string) (uint64, error) { return 2 << 30, nil }
if !noRoom(t.TempDir()) {
t.Error("a node with 2 GiB free took work")
}
builder.FreeBytes = func(string) (uint64, error) { return 200 << 30, nil }
if noRoom(t.TempDir()) {
t.Error("a node with 200 GiB free refused work")
}
}
-14
View File
@@ -36,17 +36,3 @@ func TestTheSiblingsAnAskNamesReachTheSpecAndTheOutcome(t *testing.T) {
t.Fatalf("a request that names none has siblings %+v", plain.Siblings)
}
}
// 1826/AC2: the held values a check request carries reach the spec; a request that carries none leaves it empty.
func TestTheHeldValuesAnAskCarriesReachTheSpec(t *testing.T) {
spec := checkSpecOf(link.BuildRequest{Check: &link.CheckRequest{Owner: "novox", Repo: "mesh-catalog",
HeldConfig: []link.HeldConfig{{Pull: "novox/mesh-catalog#434", Head: "a", Node: "ace", Module: "n8n", Key: "media-gid",
Value: []byte("2000"), Approved: "the operator"}}}})
if len(spec.Held) != 1 || spec.Held[0].Node != "ace" || spec.Held[0].Key != "media-gid" || string(spec.Held[0].Value) != "2000" ||
spec.Held[0].Approved != "the operator" {
t.Fatalf("the spec reads %+v", spec.Held)
}
if plain := checkSpecOf(link.BuildRequest{Check: &link.CheckRequest{Owner: "o", Repo: "r"}}); plain.Held != nil {
t.Fatalf("a request that carries none has %+v", plain.Held)
}
}
+2 -2
View File
@@ -91,7 +91,7 @@ func TestKillEndsTheBuildRunningHereAndNoOther(t *testing.T) {
t.Errorf("kill said %q", said)
}
// Removed at the kill and again once the build had ended: a container made in between is caught.
if len(removed) != 6 || !strings.Contains(removed[0], "label=mesh.build=build-1") || !strings.Contains(removed[3], "label=mesh.build=build-1") {
if len(removed) != 4 || !strings.Contains(removed[0], "label=mesh.build=build-1") || !strings.Contains(removed[2], "label=mesh.build=build-1") {
t.Errorf("removed %v", removed)
}
if c := h.current(); c.Running != nil {
@@ -160,7 +160,7 @@ func TestAHolderStartingRemovesWhatEarlierBuildsLeftHere(t *testing.T) {
}
return "", nil
}
if n := h.removeLeftBehind(); n != 2 || len(removed) != 3 || removed[2] != "docker rm -f -v c1 c2" {
if n := h.removeLeftBehind(); n != 2 || len(removed) != 2 || removed[1] != "docker rm -f c1 c2" {
t.Fatalf("removed %d: %v", n, removed)
}
h.remove = func(context.Context, string, string, ...string) (string, error) {
+1 -11
View File
@@ -64,8 +64,6 @@ is dialled except the broker.
MESH_NPM_TOKEN the token for it, likewise
MESH_NPM_SCOPE the scope it answers for (default: @novox)
MESH_WORKSPACE where to clone and build (default: a temporary directory)
MESH_BUILD_CGROUP_PARENT the slice every container a build or check starts joins, which carries the
machine's limits on builds (novox/hq issue 1798; default: the runtime's own)
It also builds one module and stops, which is how a mesh is raised — before there is a
broker to take work from or a registry to publish into:
@@ -139,10 +137,6 @@ func run() error {
if err != nil {
return err
}
// A build and a check run below the machine's own work (novox/hq issue 1798): the agent itself at the
// lowest priority, which its git and go children inherit, and every container in the build slice.
_ = syscall.Setpriority(syscall.PRIO_PROCESS, 0, 19)
builder.CgroupParent = strings.TrimSpace(os.Getenv("MESH_BUILD_CGROUP_PARENT"))
workspace := os.Getenv("MESH_WORKSPACE")
if workspace == "" {
workspace = os.TempDir() + "/mesh-builder"
@@ -190,7 +184,7 @@ func run() error {
// Paused or drained, it takes nothing new (novox/hq ADR 0219, ADR 0335): a drained agent finishes the build
// in hand and then says idle, and its container is replaced between builds.
machine := link.MachineOverNATSWith(js, on, seat, link.MachineOptions{
Paused: func() bool { return h.Paused() || drained() || markFaulty(time.Now()) || noRoom(workspace) },
Paused: func() bool { return h.Paused() || drained() },
Taking: func() { taking(seat) },
Claim: claim,
})
@@ -505,10 +499,6 @@ func checkSpecOf(request link.BuildRequest) builder.CheckSpec {
Modules: c.Modules, New: c.New, Manifests: c.Manifests, Judge: c.Judge, Base: c.Base,
Toolchain: builder.ToolchainOf(request.Held), Toolchains: builder.ToolchainsOf(request.Held),
Tests: c.Tests, StoreOnly: c.StoreOnly, SiblingsUnknown: c.SiblingsUnknown}
for _, h := range c.HeldConfig {
spec.Held = append(spec.Held, builder.HeldValue{Pull: h.Pull, Head: h.Head, Node: h.Node, Module: h.Module, Key: h.Key,
Value: h.Value, Approved: h.Approved, By: h.By, At: h.At})
}
for _, s := range c.Siblings {
spec.Siblings = append(spec.Siblings, builder.Sibling{Repository: s.Repository, Number: s.Number, Head: s.Head,
Base: s.Base, Dir: s.Dir})
+36 -66
View File
@@ -30,9 +30,7 @@ import (
// closes at once and the process exits, so its service manager restarts it as a candidate (serve,
// in push.go). A renewal the bus does not answer in time is not a loss (novox/hq issue 477): the
// lease is kept, no new act starts once half the key's age passed unrenewed, and the next renewal
// made opens the gate again. A candidate backs off before it stands (novox/hq issue 906, lease_backoff.go):
// after an epoch lost lately, and while its bus does not answer it, in its own process rather than in a
// restart every five seconds.
// made opens the gate again.
// - **A command run at a shell** — `push` in the installer, the lab, a person repairing a mesh whose
// controller is down (issue 201) — acts **under the holder's epoch** when a controller holds the
// lease: it is the same mesh's word, composed and sent under the store's hold of each machine like
@@ -74,9 +72,6 @@ type actor struct {
// lease over in a second rather than in a renewal of five (novox/hq issue 591). A renewal is given
// the lease's own RenewEvery to answer however often it is made, so a slow bus does not lose it.
renewEvery, poll time.Duration
// lossBase, when set, is the backoff after a lost epoch in place of the key's age: a test's (novox/hq
// issue 906).
lossBase time.Duration
}
// theLease is this process's standing to the lease.
@@ -216,73 +211,48 @@ func (a *actor) serveUnderTheLease(ctx context.Context, inv *inventory.Inventory
a.mu.Lock()
a.serving = true
a.mu.Unlock()
say := func(format string, args ...any) { fmt.Printf(format+"\n", args...) }
// **Backing off before standing** (novox/hq issue 906, lease_backoff.go): after an epoch lost lately, before
// this process touches the bus; and while the bus does not answer it, here rather than in a restart.
if wait := a.backoffAfterLosses(ctx, inv, say); wait > 0 {
if err := sleepOrDone(ctx, wait); err != nil {
return nil, err
}
}
var (
js *broker.JetStream
l *lease.Lease
epoch uint64
unwritable error
)
err = standUp(ctx, sleepOrDone, say, func(ctx context.Context) error {
unwritable = nil
if js == nil {
// Its own connection, held as long as the lease, and named so (novox/hq issue 327).
dialled, err := broker.Dial(address, nats.Name(broker.ConnectionName+" lease"))
if err != nil {
return fmt.Errorf("the mesh is on the bus at %s and this control plane cannot reach it to take the "+
"lease: %w", broker.BareAddress(address), err)
}
js = dialled
}
api, err := jetstream.New(js.Conn())
if err != nil {
return err
}
asserting, cancel := context.WithTimeout(ctx, 10*time.Second)
err = broker.EnsureLeaseBucket(asserting, api)
cancel()
if err != nil {
return err
}
l, err = lease.Open(ctx, api, broker.LeaseBucket, lease.Options{Holder: holderOf(instance),
RenewEvery: a.renewEvery, RenewWithin: lease.RenewEvery, Poll: a.poll, Floor: inv.HighestEpoch, Say: say, Health: controllerHealth, Moved: func(was, floor uint64) {
a.mu.Lock()
defer a.mu.Unlock()
a.reset = time.Now()
a.resetSaid = fmt.Sprintf("the lease bucket was at revision %d with epoch %d already issued: it was "+
"raised again from nothing (a bus whose data was replaced), and its revisions were moved past %d so "+
"no machine refuses the next epoch", was, floor, floor)
}})
if err != nil {
return err
}
epoch, err = l.Take(ctx)
if errors.Is(err, lease.ErrUnwritable) {
unwritable = err
return nil
}
// Any other error: whether another controller acts cannot be told, so this one does not act, and tries
// again.
return err
})
// Its own connection, held as long as the lease, and named so (novox/hq issue 327).
js, err := broker.Dial(address, nats.Name(broker.ConnectionName+" lease"))
if err != nil {
if js != nil {
js.Close()
}
return nil, fmt.Errorf("the mesh is on the bus at %s and this control plane cannot reach it to take the "+
"lease: %w", broker.BareAddress(address), err)
}
api, err := jetstream.New(js.Conn())
if err != nil {
js.Close()
return nil, err
}
asserting, cancel := context.WithTimeout(ctx, 10*time.Second)
err = broker.EnsureLeaseBucket(asserting, api)
cancel()
if err != nil {
js.Close()
return nil, err
}
say := func(format string, args ...any) { fmt.Printf(format+"\n", args...) }
l, err := lease.Open(ctx, api, broker.LeaseBucket, lease.Options{Holder: holderOf(instance),
RenewEvery: a.renewEvery, RenewWithin: lease.RenewEvery, Poll: a.poll, Floor: inv.HighestEpoch, Say: say, Health: controllerHealth, Moved: func(was, floor uint64) {
a.mu.Lock()
defer a.mu.Unlock()
a.reset = time.Now()
a.resetSaid = fmt.Sprintf("the lease bucket was at revision %d with epoch %d already issued: it was "+
"raised again from nothing (a bus whose data was replaced), and its revisions were moved past %d so "+
"no machine refuses the next epoch", was, floor, floor)
}})
if err != nil {
js.Close()
return nil, err
}
gone := make(chan struct{})
switch err := unwritable; {
epoch, err := l.Take(ctx)
switch {
case ctx.Err() != nil:
js.Close()
return nil, ctx.Err()
case err != nil && !errors.Is(err, lease.ErrUnwritable):
// Whether another controller acts cannot be told: this one does not act, and exits to try again.
js.Close()
return nil, err
case err != nil:
// Nobody holds it and the bus will not let it be written: unleased, said, tried again (see above).
a.mu.Lock()
+2 -21
View File
@@ -408,21 +408,12 @@ func issueOnAssign(ctx context.Context, open *stores, node, module string) strin
return ""
}
user := broker.Principal{Kind: broker.KindModule, Node: node, Module: module}.Username()
_, reads := m.OwnSecrets["broker"]
if !reads && !runsTheProvisionerClient(m) {
// A module whose bus accounts are its processes' alone has none of its own: only the processes that
// are not issued yet are issued (issue 1226).
return processAccountsOnAssign(ctx, inv, m, node)
}
if !reads {
if _, reads := m.OwnSecrets["broker"]; !reads {
// A module running the provisioner client alone is minted the client's account (issue 917).
user = broker.Principal{Kind: broker.KindProvisioner, Node: node, Module: module}.Username()
}
if _, minted, err := inv.BusUserHash(ctx, user); err != nil {
if _, minted, err := inv.BusUserHash(ctx, user); err != nil || minted {
return ""
} else if minted {
// Its own account stands and is kept; a process it gained is not issued yet (issue 1226).
return processAccountsOnAssign(ctx, inv, m, node)
}
busAddress, err := broker.BusAddress()
if err == nil {
@@ -435,16 +426,6 @@ func issueOnAssign(ctx context.Context, open *stores, node, module string) strin
return fmt.Sprintf("its bus credential is issued and sealed to %s, and arrives with the push", node)
}
// processAccountsOnAssign issues the process accounts of an assigned module not issued yet, and says what it did.
func processAccountsOnAssign(ctx context.Context, inv *inventory.Inventory, m catalogue.Manifest, node string) string {
line, err := issueMissingProcessAccounts(ctx, inv, m, node)
if err != nil {
return fmt.Sprintf("its process accounts are not issued (%v): `module issue %s --node %s` first — "+
"`push %s` refuses to send %s until it is", err, m.Module, node, node, m.Module)
}
return line
}
// refusedMovingTheController refuses assigning or unassigning the controller's module through a verb (review of
// novox/hq ADR 0272): the node it is assigned to is the control-node, and the control-node's operator account is
// the controller's terminal, so whoever moves the module chooses the terminal. At the terminal alone, as every
@@ -31,7 +31,6 @@ func TestALookComesBackForATierTheBusRefusedWhole(t *testing.T) {
t.Fatal(err)
}
bus := false
asksOfEveryModule(t) // the one ask left is the builder's own (ADR 0453 §5): this test is of what a walk does around it
was := askABuild
askABuild = func(_ context.Context, source buildSource, _, _ string) (string, error) {
if !bus {
@@ -96,7 +95,6 @@ func TestALookFailsTheTierTheBusRefusedPastTheBound(t *testing.T) {
if err := inv.SavePlan(ctx, &p); err != nil {
t.Fatal(err)
}
asksOfEveryModule(t) // the one ask left is the builder's own (ADR 0453 §5): this test is of what a walk does around it
was := askABuild
askABuild = func(context.Context, buildSource, string, string) (string, error) {
return "", fmt.Errorf("%w: nats: no responders available for request", inventory.ErrNoChain)
@@ -116,6 +114,11 @@ func TestALookFailsTheTierTheBusRefusedPastTheBound(t *testing.T) {
t.Fatalf("a bus silent past its bound did not fail the plan in words: %s %q", got.State, got.Note)
}
// Rebuilt by hand while the bus still reloads: said as waiting, not as a failure and not with an empty reason.
joined, said, err := joinAPlan(ctx, open, "a")
if err != nil || !joined || !strings.Contains(said, "not asked yet: the bus did not answer") {
t.Fatalf("a rebuild left for the bus is said as %v %q %v", joined, said, err)
}
// And retried by hand: the module is named as waiting, not as asked again.
got, _ = inv.PlanByID(ctx, "plan-bus")
if got.State == inventory.PlanFailed {
@@ -15,15 +15,6 @@ import (
// seconds) is not a failed build. The module is left unasked and the plan building, and the next look asks it again; only a
// module's own fault, or a bus silent for ten minutes, fails the plan.
// asksOfEveryModule makes every module a test names the builder's own, the one module a walk still asks build-agent
// for (ADR 0453 §5): what a test of that ask's waits reads.
func asksOfEveryModule(t *testing.T) {
t.Helper()
was := buildersOwn
buildersOwn = func(string) bool { return true }
t.Cleanup(func() { buildersOwn = was })
}
func aTierOfThree() (*inventory.Plan, map[string]inventory.Entry) {
p := &inventory.Plan{ID: "plan-1", State: inventory.PlanBuilding, Tiers: [][]string{{"a", "b", "c"}},
Modules: map[string]*inventory.PlanModule{}}
@@ -35,7 +26,6 @@ func aTierOfThree() (*inventory.Plan, map[string]inventory.Entry) {
}
func TestABusThatDoesNotAnswerForAMomentDoesNotFailTheTier(t *testing.T) {
asksOfEveryModule(t)
p, byName := aTierOfThree()
bus := false
was := askABuild
@@ -78,7 +68,6 @@ func TestABusThatDoesNotAnswerForAMomentDoesNotFailTheTier(t *testing.T) {
}
func TestABusSilentForTenMinutesFailsTheTierAsBefore(t *testing.T) {
asksOfEveryModule(t)
p, byName := aTierOfThree()
was := askABuild
askABuild = func(context.Context, buildSource, string, string) (string, error) {
@@ -130,7 +119,6 @@ func TestAPlanWithAModuleWaitingForTheBusIsOneTheLookComesBackTo(t *testing.T) {
}
func TestAModulesOwnFaultStillFailsThePlan(t *testing.T) {
asksOfEveryModule(t)
p, byName := aTierOfThree()
delete(byName, "a")
askModule(t.Context(), p, "a", byName)
-23
View File
@@ -1,23 +0,0 @@
package main
import (
"strings"
"testing"
"time"
"github.com/novox/mesh-controller/internal/inventory"
)
// A retry of a plan failed past the bus's bound, whose module the bus still does not answer, names that module as waiting:
// read while the plan still says failed, it was named nowhere, and the retry said "asked" with nothing after it (issue
// 1403, met again once the hand rebuild no longer resumed the plan first).
func TestARetryNamesTheModulesStillWaitingForTheBus(t *testing.T) {
since := time.Now().UTC()
p := inventory.Plan{ID: "plan-bus", State: inventory.PlanFailed, Tiers: [][]string{{"a"}},
Modules: map[string]*inventory.PlanModule{"a": {BusWaitFrom: &since}}}
waiting := busWaitingOnRetry(p)
_, said := retryWords(p, []string{"a"}, nil, waiting)
if !strings.Contains(said, "a not asked yet: the bus did not answer") || strings.Contains(said, "asked ;") {
t.Fatalf("a retry left for the bus is said as %q", said)
}
}
-4
View File
@@ -645,10 +645,6 @@ func doesWords(act conditions.Action) string {
return "the delivery is stopped"
case act.Verb == broker.ControllerSeat+".plans" && act.Arguments["retry"] != "":
return "the failed builds are asked again"
case act.Verb == broker.ControllerSeat+".push" && act.Arguments["node"] != "":
return "the held builds are sent to " + act.Arguments["node"]
case act.Verb == broker.ControllerSeat+".upgrade" && act.Arguments["release-backlog"] != "":
return "the builds held after the failed delivery are released"
case strings.HasSuffix(act.Verb, ".restart"):
return "its service is restarted on " + act.Machine
}
@@ -68,7 +68,6 @@ func TestAPlanGoesOnWithoutAMachineAwayWhenItsTierIsSent(t *testing.T) {
var mu sync.Mutex
var asks []string
var sends [][]string
asksOfEveryModule(t) // the one ask left is the builder's own (ADR 0453 §5): this test is of what a walk does around it
wasAsk := askABuild
askABuild = func(_ context.Context, source buildSource, _, _ string) (string, error) {
mu.Lock()
-5
View File
@@ -232,8 +232,6 @@ func (f following) hearMerge(ctx context.Context, m link.SourceMoved, now time.T
fmt.Print(nothingReadsIt(repository, m))
return nil
}
// A module the merge adds is registered from the builder's first done key of it, which the builder's write wakes
// (novox/hq ADR 0501 §6): nothing is asked here.
// **Its delivery group's word, before the plans are held** (ADR 0306 decision 2): a slow answer holds up no walk.
own := ownPath(touches)
waits, holding, err := groupWaitHolding(ctx, inv, m, deliverySeatHeld(entries))
@@ -800,9 +798,6 @@ type keptMerge struct {
// FirstBuild says the merge is a first merge into a repository the mesh builds nothing from, kept to ask the first
// build of its modules once mesh-delivery's word on it is kept (novox/hq ADR 0432, issue 1146).
FirstBuild bool `json:"first_build,omitempty"`
// Held says which first builds of the merge were refused for a word the running build-agent does not read: the merge
// waits for a build-agent that reads them, and is never failed (novox/hq ADR 0482 §5, issue 1173).
Held []heldFirstBuild `json:"held,omitempty"`
}
// announcedOf is a kept merge as the forge announced it, and what it moved; empty where the record says none.
+296 -113
View File
@@ -28,6 +28,112 @@ import (
// cheapest next step. That is how novox/hq ADR 0001 records `hal/sdk` reaching 34,636:
// nothing in it was wrong, and no one edit was the one that should have been a new file.
// buildCommand builds a module from its source and records what came out.
//
// **Run where there is a container runtime**, which is why it is a command rather than something
// the control plane does on its own: building needs to run things on a machine, and what the
// control plane may send a machine is bounded by the declaration language. This is the shape the
// builder module will take when it is given work over the broker; today a person runs it, and the
// mesh records the result the same way either way.
// buildOn rebuilds every module the mesh holds that stands on the named module's artifacts — the
// rebuild a changed base needs, which nothing else asks for: their sources did not move, and
// "behind" does not see a base that did (novox/hq 04-ISSUES/131). Bases first among them too.
func buildOn(ctx context.Context, base string, wait time.Duration) error {
open, err := openStores(ctx)
if err != nil {
return err
}
defer open.Close()
held, err := open.inventory.Catalogued(ctx)
if err != nil {
return err
}
against, err := open.inventory.BuiltAgainst(ctx)
if err != nil {
return err
}
var on []inventory.Entry
for _, e := range held {
if standsOnModule(e, base, against) {
on = append(on, e)
}
}
if len(on) == 0 {
fmt.Printf("nothing the mesh holds stands on %s\n", base)
return nil
}
on = orderByBases(on, against)
fmt.Printf("%d module(s) stand on %s:\n", len(on), base)
var failed []string
for _, e := range on {
fmt.Printf("--- %s\n", e.Manifest.Module)
source := buildSource{Repository: e.Source.Repository, Seat: e.Source.Seat}
if err := buildOne(ctx, source, e.Source.Path, e.Source.Ref, wait); err != nil {
fmt.Printf(" %v\n", err)
failed = append(failed, e.Manifest.Module)
}
}
if len(failed) > 0 {
return fmt.Errorf("%d of %d could not be built: %s", len(failed), len(on), strings.Join(failed, ", "))
}
fmt.Printf("\n%d module(s) rebuilt on %s. `push --behind` sends them on\n", len(on), base)
return nil
}
func buildCommand(ctx context.Context, args []string) error {
set := flag.NewFlagSet("build", flag.ContinueOnError)
ref := set.String("ref", "", "the branch, tag or commit to build")
// A module is a repository and a path within it (novox/hq ADR 0069). Empty is the repository's
// root, which is the ordinary case and why this is a flag rather than a second argument.
path := set.String("path", "", "the module's directory inside the repository")
wait := set.Duration("wait", 10*time.Minute, "how long to wait for a builder to answer")
dryRun := set.Bool("dry-run", false, "build and print the manifest, recording nothing")
// Every module whose source has moved, rather than one named repository.
//
// **The mirror of `push --behind`, and the same argument** (novox/hq ADR 0010): the mesh
// already knows which modules are behind their source, so making a person read that list and
// retype each repository is asking them to be the loop. Naming a repository and asking which
// ones need building are different requests, so they are not combined.
behind := set.Bool("behind", false, "every module the mesh holds older than its source has")
on := set.String("on", "", "rebuild every module that stands on this module's artifacts — the rebuild a changed base needs")
// A repository on the mesh's own forge, named by its path there (novox/hq ADR 0111). Without it
// the repository is external, cloned exactly as given — see source.go.
self := set.Bool("self", false, "the repository is a path on the forge holding the git seat")
positionals, err := parseAround(set, args)
if err != nil {
return err
}
if *on != "" {
if len(positionals) != 0 || *behind || *self {
return errors.New("build --on <module> names a base and nothing else")
}
return buildOn(ctx, *on, *wait)
}
if *behind {
if len(positionals) != 0 || *self {
return errors.New("build <repository> or build --behind, not both: one names a " +
"repository and the other asks which need building")
}
return buildBehind(ctx, *wait)
}
if len(positionals) != 1 {
return errors.New("build <repository> [--self] [--path P] [--ref R] [--wait D] [--dry-run] | build --behind | build --on <module>")
}
source := buildSource{Repository: positionals[0]}
if *self {
if err := onASeat(source.Repository); err != nil {
return err
}
source.Seat = gitSeat
}
if *dryRun {
return buildAndShow(ctx, source, *path, *ref, *wait)
}
return buildOne(ctx, source, *path, *ref, *wait)
}
// buildFrom turns what a builder said into what the mesh keeps.
//
// **By digest and path, never by where it was pushed** (novox/hq 04-ISSUES/102). The builder
@@ -249,6 +355,86 @@ func builderCommand(ctx context.Context, args []string) error {
return issueOnTheNewBus(ctx, inv, m, node, address)
}
// buildBehind builds every module the mesh holds older than its source has.
//
// **This is the loop novox/hq ADR 0010 replaced a pipeline with, closed.** The mesh already
// records where each module came from and what its source last had; until this, a person read
// that list and retyped each repository — which is a person being the loop, and the thing a
// pipeline was doing before it was taken away.
//
// Each is built and recorded on its own. **One failing does not stop the others**, for the same
// reason one broken module no longer blocks a machine's whole declaration: a mesh where one bad
// repository holds back nine good ones is a mesh where nobody dares add the tenth.
func buildBehind(ctx context.Context, wait time.Duration) error {
open, err := openStores(ctx)
if err != nil {
return err
}
defer open.Close()
inv := open.inventory
held, err := inv.Catalogued(ctx)
if err != nil {
return err
}
var stale []inventory.Entry
for _, e := range held {
if !e.Source.Current() {
stale = append(stale, e)
}
}
if len(stale) == 0 {
// Said rather than doing nothing quietly: "nothing needed building" and "this did not
// run" must never look the same.
fmt.Println("every module the mesh holds is what its source last had")
return nil
}
fmt.Printf("%d module(s) behind their source:\n", len(stale))
for _, e := range stale {
fmt.Printf(" %s %s < %s\n",
e.Manifest.Module, short(e.Source.BuiltFrom), short(e.Source.Head))
}
fmt.Println()
// Bases first: a module built before the module it stands on is built against the old one
// and reports success (novox/hq 04-ISSUES/131).
against, err := inv.BuiltAgainst(ctx)
if err != nil {
return err
}
stale = orderByBases(stale, against)
var failed []string
for _, e := range stale {
fmt.Printf("--- %s\n", e.Manifest.Module)
// Its own recorded ref, not its head commit: a module tracking a branch should be built
// from that branch, and pinning to the commit the mesh happened to notice would quietly
// turn a tracked branch into a pin.
source := buildSource{Repository: e.Source.Repository, Seat: e.Source.Seat}
if err := buildOne(ctx, source, e.Source.Path, e.Source.Ref, wait); err != nil {
fmt.Printf(" %v\n", err)
failed = append(failed, e.Manifest.Module)
}
}
if len(failed) > 0 {
return fmt.Errorf("%d of %d could not be built: %s",
len(failed), len(stale), strings.Join(failed, ", "))
}
fmt.Printf("\n%d module(s) built. `push --behind` sends them to the machines running them\n",
len(stale))
return nil
}
// buildOne asks a build machine for one repository and records everything that came back.
//
// Separated from the command so `--behind` can walk a list without a second path to the same act.
func buildOne(ctx context.Context, source buildSource, path, ref string, wait time.Duration) error {
_, err := buildOneAsked(ctx, source, path, ref, wait, false, "build")
return err
}
// recordAsked keeps a build request once it is asked (novox/hq issue 325), with who asked it, in a store
// opened for the purpose: the asks reach here from processes that hold none.
func recordAsked(ctx context.Context, r inventory.BuildRequest) {
@@ -325,6 +511,7 @@ func buildOneAsked(ctx context.Context, source buildSource, path, ref string, wa
Seats: seatBases(ctx),
DryRun: dryRun,
}
request.Reuse = reuseAsked(ctx, dryRun)
fmt.Printf("asked for %s", source)
if source.Seat != "" {
fmt.Printf(" (%s)", repository)
@@ -492,6 +679,13 @@ func takeIn(ctx context.Context, inv *inventory.Inventory, result link.BuildResu
recorded.Ref = followedBranch(was.Ref)
}
}
if err := namesNoInstallation(manifest); err != nil {
return manifest, kept, fmt.Errorf("%s built %s (%s), and the mesh does not register it: %w",
result.On, result.Repository, short(result.Commit), err)
}
// **Only from the repository the catalogue builds the module from** (novox/hq ADR 0266): else the trunk
// below is the trunk of whatever repository was built, which may be one an agent made — and a module named
// `sudo` from it would be what the next push sends. Another repository is the operator's, at the terminal.
trunk := result.Trunk
if was, err := inv.SourceOf(ctx, manifest.Module); err == nil && followedBranch(was.Ref) != "" {
trunk = followedBranch(was.Ref)
@@ -499,111 +693,55 @@ func takeIn(ctx context.Context, inv *inventory.Inventory, result link.BuildResu
if trunk == "" {
trunk = "main"
}
j := takenBuild{kept: kept, manifest: manifest, recorded: recorded, ask: byBuild(result.ID), path: result.Path,
trunk: trunk, by: result.On, what: result.Repository, commit: result.Commit}
// The checks only an outcome of build-agent needs: a key is a trunk build by construction (ADR 0501 §2), in no
// scratch namespace, and carries no list of its own.
j.published = func() error {
// **Only a commit on the trunk is published** (novox/hq ADR 0238): a commit off its repository's
// default branch — a pull request's head, a feature branch built by hand, a `rebuild` or `replay
// --register` of one — is for checking, and is never a module's version; nothing could then send it.
// The branch the module already follows is its trunk — never the branch a build was asked at, or a
// build of a feature branch by name would make that branch its trunk.
follows := ""
if was, err := inv.SourceOf(ctx, manifest.Module); err == nil {
follows = followedBranch(was.Ref)
}
if err := publishable(result, follows); err != nil {
return err
}
// **Never a build in the scratch namespace** (novox/hq ADR 0239 §9, ADR 0350 §2): what a check builds is
// registered only as the walk's copy in the module's own namespace.
if scratch := inScratch(result.Made); scratch != "" {
return fmt.Errorf("%s is in the scratch namespace, which only a walk's copy leaves (ADR 0350)", scratch)
}
return nil
}
j.carried = func() error {
// **The newest trunk build's list of predefined keys is published** (novox/hq ADR 0348 §7): a build of the
// catalogue that may be registered keeps its list when its commit is newer than the one kept.
carried, err := carriedList(result)
if err != nil {
return err
}
if carried != nil {
if err := publishPredefined(ctx, inv, result, carried); err != nil {
fmt.Printf("%s: the list of predefined keys it carried is not published: %v\n", result.ID, err)
}
}
return nil
}
manifest, err = registerJudged(ctx, inv, j)
return manifest, kept, err
}
// takenBuild is a build the controller judges and registers, whatever told it of the build: build-agent's outcome (the
// builder's own module's, until R4g, ADR 0453 §5) or the builder's done key (ADR 0501 §1). **One judging for both**:
// the rules of who may register, the gate a build failed, the predefined keys and the rule that an unchanged build
// source moves nothing are the same whichever part built it (ADR 0501 §1, §3).
type takenBuild struct {
kept inventory.Build
manifest catalogue.Manifest
recorded inventory.Source
ask registrationAsk
path string
trunk string
// by and what name the build in what is said: who built it, and what was built ("ace", "novox/mesh-catalog").
by, what, commit string
// published refuses a build that may not become a module's version (off the trunk, in the scratch namespace);
// nil for a key, which is a trunk fingerprint's by construction (ADR 0501 §2).
published func() error
// carried publishes a list of predefined keys the build carried; nil when it carried none.
carried func() error
}
// registerJudged judges a recorded build and registers its module, or says why not.
func registerJudged(ctx context.Context, inv *inventory.Inventory, j takenBuild) (catalogue.Manifest, error) {
manifest, kept := j.manifest, j.kept
notRegistered := func(err error) (catalogue.Manifest, error) {
return manifest, fmt.Errorf("%s built %s (%s), recorded and not registered: %w", j.by, manifest.Module,
short(j.commit), err)
}
if err := namesNoInstallation(manifest); err != nil {
return manifest, fmt.Errorf("%s built %s (%s), and the mesh does not register it: %w",
j.by, j.what, short(j.commit), err)
}
// **Only from the repository the catalogue builds the module from** (novox/hq ADR 0266): else the trunk
// below is the trunk of whatever repository was built, which may be one an agent made — and a module named
// `sudo` from it would be what the next push sends. Another repository is the operator's, at the terminal.
repoID, err := mayRegisterFrom(ctx, inv, manifest.Module, j.recorded, j.ask, j.path, j.trunk)
repoID, err := mayRegisterFrom(ctx, inv, manifest.Module, recorded, result.ID, result.Path, trunk)
if err != nil {
return notRegistered(err)
return manifest, kept, fmt.Errorf("%s built %s (%s), recorded and not registered: %w", result.On,
manifest.Module, short(result.Commit), err)
}
if j.published != nil {
if err := j.published(); err != nil {
return notRegistered(err)
}
// **Only a commit on the trunk is published** (novox/hq ADR 0238): a commit off its repository's
// default branch — a pull request's head, a feature branch built by hand, a `rebuild` or `replay
// --register` of one — is for checking, and is never a module's version; nothing could then send it.
// The branch the module already follows is its trunk — never the branch a build was asked at, or a
// build of a feature branch by name would make that branch its trunk.
follows := ""
if was, err := inv.SourceOf(ctx, manifest.Module); err == nil {
follows = followedBranch(was.Ref)
}
if err := publishable(result, follows); err != nil {
return manifest, kept, fmt.Errorf("%s built %s (%s), recorded and not registered: %w", result.On,
manifest.Module, short(result.Commit), err)
}
// **Never a build in the scratch namespace** (novox/hq ADR 0239 §9, ADR 0350 §2): what a check builds is
// registered only as the walk's copy in the module's own namespace.
if scratch := inScratch(result.Made); scratch != "" {
return manifest, kept, fmt.Errorf("%s built %s (%s), recorded and not registered: %s is in the scratch "+
"namespace, which only a walk's copy leaves (ADR 0350)", result.On, manifest.Module, short(result.Commit), scratch)
}
// **A build that failed its gate is never registered again** (novox/hq ADR 0236): an outcome heard
// twice, or replayed, would otherwise make the build a rollback put back what the module is again,
// and the next push would send it.
if failed, err := inv.GateFailed(ctx, kept.ID); err != nil {
return manifest, err
return manifest, kept, err
} else if failed {
return manifest, fmt.Errorf("%s built %s (%s), which failed its gate on its first machine and was put "+
"back: it is recorded and not registered again — a newer build is", j.by, manifest.Module,
short(j.commit))
return manifest, kept, fmt.Errorf("%s built %s (%s), which failed its gate on its first machine and was put "+
"back: it is recorded and not registered again — a newer build is", result.On, manifest.Module,
short(result.Commit))
}
// Then the module is judged against the list the mesh publishes, as every manifest read again is: one naming a
// key that list does not hold, or of another type, is refused here and never enters the catalogue.
if j.carried != nil {
if err := j.carried(); err != nil {
return manifest, err
// **The newest trunk build's list of predefined keys is published** (novox/hq ADR 0348 §7): a build of the
// catalogue that may be registered keeps its list when its commit is newer than the one kept. Then the module
// is judged against the list the mesh publishes, as every manifest read again is: one naming a key that list
// does not hold, or of another type, is refused here and never enters the catalogue.
if carried, err := carriedList(result); err != nil {
return manifest, kept, err
} else if carried != nil {
if err := publishPredefined(ctx, inv, result, carried); err != nil {
fmt.Printf("%s: the list of predefined keys it carried is not published: %v\n", result.ID, err)
}
}
if manifest.NamesPredefined() {
if _, err := catalogue.ParseManifest(kept.Manifest); err != nil {
return notRegistered(fmt.Errorf("judged against the list of predefined keys the mesh publishes, %w", err))
return manifest, kept, fmt.Errorf("%s built %s (%s), recorded and not registered: judged against the "+
"list of predefined keys the mesh publishes, %w", result.On, manifest.Module, short(result.Commit), err)
}
}
// **A build whose source is unchanged is never a move** (novox/hq issue 280): a rebuild made from
@@ -613,34 +751,32 @@ func registerJudged(ctx context.Context, inv *inventory.Inventory, j takenBuild)
if kept.SourceFingerprint != "" {
stands, raw, err := inv.StandingBuild(ctx, manifest.Module, kept.ID)
if err != nil {
return manifest, err
return manifest, kept, err
}
if stands != "" && stands != kept.ID && len(raw) > 0 {
if was, err := catalogue.ParseManifest(raw); err == nil && was.Module == manifest.Module {
if same, err := catalogue.ParseManifest(raw); err == nil && same.Module == manifest.Module {
fmt.Printf("%s at %s was made from the source %s was: registered with its artifacts, no move\n",
manifest.Module, short(j.commit), stands)
manifest = was
manifest.Module, short(result.Commit), stands)
manifest = same
}
}
}
if err := inv.RegisterModule(ctx, manifest, j.recorded); err != nil {
if err := inv.RegisterModule(ctx, manifest, recorded); err != nil {
if errors.Is(err, inventory.ErrSuperseded) {
return notRegistered(err)
return manifest, kept, fmt.Errorf("%s built %s (%s), recorded and not registered: %w",
result.On, manifest.Module, short(result.Commit), err)
}
return manifest, err
return manifest, kept, err
}
// A bus process the new definition gains gets its account now, on each machine assigned the module, so the
// next push composes a real account (issue 1226).
issueGainedProcessAccounts(ctx, inv, manifest)
// Which repository it is registered from, by the forge's own id (novox/hq ADR 0266).
if err := inv.SetSourceIdentity(ctx, manifest.Module, repoID); err != nil {
return manifest, err
return manifest, kept, err
}
// The keep set just moved, and new bytes just landed (novox/hq ADR 0189). Asked here rather
// than on a timer of its own: this is the only moment either is true. Never fatal — the build
// worked and the module is registered.
collect(ctx, inv)
return manifest, nil
return manifest, kept, nil
}
// inScratch is the first artifact a build made into the scratch namespace; empty when none is.
@@ -681,17 +817,64 @@ func publishable(result link.BuildResult, follows string) error {
return nil
}
// buildAndShow builds and prints the manifest without recording anything.
func buildAndShow(ctx context.Context, source buildSource, path, ref string, wait time.Duration) error {
repository, err := cloneFrom(ctx, source)
if err != nil {
return err
}
if err := verbMayAsk(ctx, source, repository, path); err != nil {
return err
}
ident, err := openIdentity(ctx)
if err != nil {
return err
}
defer ident.Close()
server, err := connectLink(ctx, nil, nil, nil)
if err != nil {
return err
}
defer server.Close()
ask, err := askOverOn(buildSeatHeld(ctx))
if err != nil {
return err
}
defer ask.Close()
result, err := ask.Submit(ctx, link.BuildRequest{
ID: link.NewBuildID(time.Now()),
Repository: repository, Path: path, Ref: ref,
Held: heldBy(ctx), Seats: seatBases(ctx),
DryRun: true,
}, wait)
if err != nil {
return err
}
if result.Failed != "" {
return fmt.Errorf("%s could not build %s:\n%s", result.On, result.Repository, result.Failed)
}
manifest, err := catalogue.ParseManifest(result.Manifest)
if err != nil {
return fmt.Errorf("%s built %s and what came back is not a manifest: %w",
result.On, result.Repository, err)
}
body, err := json.MarshalIndent(manifest, "", " ")
if err != nil {
return err
}
fmt.Println(string(body))
return nil
}
// answers is what the three questions came back with, read once.
type answers struct {
// notAnnounced are the owners some module reads the state of that have announced nothing yet (novox/hq ADR 0495,
// issue 1790): read from `reads` as before until they do; notAnnouncedWhy why that could not be read.
notAnnounced []string
notAnnouncedWhy string
wrong []inventory.Doing
nodes []inventory.Node
quiet []inventory.Node
behind map[string][]string
sources map[string]inventory.Source
wrong []inventory.Doing
nodes []inventory.Node
quiet []inventory.Node
behind map[string][]string
sources map[string]inventory.Source
// waiting is every machine not running what the mesh would send it.
waiting []inventory.Machine
// recordedWaits is, per machine, every move of a recorded module that waits there for a person (novox/hq ADR
+59 -59
View File
@@ -373,7 +373,22 @@ type protectionRule struct {
// branch, else the first glob rule, in the forge's order, that covers it (gitea's first matching rule). Never a
// later rule that happens to be stronger: the forge does not read it.
func judgedRule(rules []protectionRule, branch string) forgeFacts {
switch r := appliedRule(rules, branch); {
var applied *protectionRule
for i := range rules {
if rules[i].Rule == branch {
applied = &rules[i]
break
}
}
if applied == nil {
for i := range rules {
if ruleCovers(rules[i].Rule, branch) {
applied = &rules[i]
break
}
}
}
switch r := applied; {
case r == nil:
return forgeFacts{Why: fmt.Sprintf("no protection rule covers %s", branch)}
case r.Push:
@@ -386,22 +401,6 @@ func judgedRule(rules []protectionRule, branch string) forgeFacts {
return forgeFacts{Guarded: true}
}
// appliedRule is the one rule the forge applies to a branch: the rule named for it, else the first glob rule, in the
// forge's order, that covers it; nil when none does.
func appliedRule(rules []protectionRule, branch string) *protectionRule {
for i := range rules {
if rules[i].Rule == branch {
return &rules[i]
}
}
for i := range rules {
if ruleCovers(rules[i].Rule, branch) {
return &rules[i]
}
}
return nil
}
// ruleCovers says a protection rule's name — a branch, or a glob of them — covers a branch, as the forge reads it.
func ruleCovers(rule, branch string) bool {
if rule == branch {
@@ -470,39 +469,8 @@ func (f *sourceForms) onTheForge(repository, seat string) (owner, name string, o
// repository and path; or, for the build of a merge, when mesh-delivery announced it on the operator's Approve of
// that head (novox/hq ADR 0420 §7 for a moved module, ADR 0432 for a new one), the forge's checks still holding. path is the module's directory as built; branch the trunk it is registered from.
func mayRegisterFrom(ctx context.Context, inv *inventory.Inventory, module string, built inventory.Source,
ask registrationAsk, path, branch string) (int64, error) {
return mayRegisterIn(ctx, inv, newSourceForms(ctx, inv), module, built, ask, path, branch)
}
// registrationAsk is what a registration's warrant is read from (novox/hq ADR 0501 §3): the build an outcome answers,
// whose kept request says whether it was asked at the terminal or for a merge (build-agent's outcome, the builder's own
// module until R4g); or, for a take-in from the builder's done key, the merge the indexer's trunk value names and
// whether the operator asked the registration at the controller's terminal (ADR 0501 §4). A key asks no build, so it
// keeps no request.
type registrationAsk struct {
// Build is the build whose kept request is read; empty for a take-in from a key.
Build string
// Merge is the commit of the merge a key's module is registered from (the indexer's trunk value); empty for an
// outcome, whose request names it.
Merge string
// Terminal is a key registration the operator asked at the controller's terminal for this repository and path.
Terminal bool
}
// byBuild is the warrant of a build's outcome: read from the request kept under its id.
func byBuild(id string) registrationAsk { return registrationAsk{Build: id} }
// String is the ask as the registration's words name it.
func (a registrationAsk) String() string {
switch {
case a.Build != "":
return a.Build
case a.Terminal:
return "the registration asked at the terminal"
case a.Merge != "":
return fmt.Sprintf("the merge %.8s", a.Merge)
}
return "a key"
buildID, path, branch string) (int64, error) {
return mayRegisterIn(ctx, inv, newSourceForms(ctx, inv), module, built, buildID, path, branch)
}
// catalogueReader is the catalogue's manifests, as the router's records are found in them.
@@ -522,19 +490,15 @@ type registrations interface {
// mayRegisterIn is mayRegisterFrom on what it reads of the mesh handed in.
func mayRegisterIn(ctx context.Context, inv registrations, forms *sourceForms, module string, built inventory.Source,
ask registrationAsk, path, branch string) (int64, error) {
buildID := ask.String()
buildID, path, branch string) (int64, error) {
was, err := inv.SourceOf(ctx, module)
isNew := errors.Is(err, inventory.ErrNoSuchModule)
if err != nil && !isNew {
return 0, err
}
terminal := ask.Terminal
if ask.Build != "" {
var err error
if terminal, err = askedHereFor(ctx, inv, forms, ask.Build, built, path); err != nil {
return 0, err
}
terminal, err := askedHereFor(ctx, inv, forms, buildID, built, path)
if err != nil {
return 0, err
}
refuse := func(why string) (int64, error) {
return 0, fmt.Errorf("%w: %s. A module is registered from such a source only by a build asked at the "+
@@ -578,7 +542,7 @@ func mayRegisterIn(ctx context.Context, inv registrations, forms *sourceForms, m
// The forge's protection and the repository's identity are checked below as for every registration.
movedWhy := ""
if why != "" && !terminal {
accepted, err := approvedMove(ctx, inv, forms, ask, built)
accepted, err := approvedMove(ctx, inv, forms, buildID, built)
switch {
case err == nil:
movedWhy, why = accepted, ""
@@ -698,6 +662,37 @@ func sourceWords(repository, seat string) string {
return buildSource{Repository: repository, Seat: seat}.String()
}
// firstBuildWarrant is what the controller checked before it asks the first build of the modules a merge adds to a
// repository nothing is built from: the merge was made on the operator's Approve of its head (approvedAnnounced). It
// travels in the context of that ask alone, in this process, so no verb carries one.
type firstBuildWarrant struct {
repository string // owner/name on the git seat
paths map[string]bool
why string
}
type firstBuildWarrantKey struct{}
// withFirstBuildWarrant is ctx carrying the warrant for asking the first builds of paths in repository.
func withFirstBuildWarrant(ctx context.Context, repository string, paths []string, why string) context.Context {
w := &firstBuildWarrant{repository: repository, paths: map[string]bool{}, why: why}
for _, p := range paths {
w.paths[strings.Trim(p, "/")] = true
}
return context.WithValue(ctx, firstBuildWarrantKey{}, w)
}
func firstBuildWarrantIn(ctx context.Context) *firstBuildWarrant {
w, _ := ctx.Value(firstBuildWarrantKey{}).(*firstBuildWarrant)
return w
}
// covers says the warrant is for exactly this ask: the repository on the git seat, and a module path the merge holds.
func (w *firstBuildWarrant) covers(source buildSource, path string) bool {
return w != nil && source.Seat == gitSeat && strings.EqualFold(strings.Trim(source.Repository, "/"), w.repository) &&
w.paths[strings.Trim(path, "/")]
}
// verbMayAsk refuses, for a call through a verb, a build of a repository the catalogue builds no module from
// (novox/hq ADR 0266): the build node would run what an agent wrote, and its outcome could never be
// registered anyway. url is the repository as it is cloned. At the terminal anything may be asked.
@@ -705,6 +700,11 @@ func verbMayAsk(ctx context.Context, source buildSource, url, path string) error
if startedAtTheTerminal() {
return nil
}
// **Or the first build of a module a merge adds on the operator's Approve** (novox/hq ADR 0432, issue 1146): the
// controller's own ask, carrying the warrant it checked, for that repository and the module paths the merge holds.
if firstBuildWarrantIn(ctx).covers(source, path) {
return nil
}
open, err := openStores(ctx)
if err != nil {
return err
@@ -1,139 +0,0 @@
//go:build integration
package main
import (
"context"
"sync"
"testing"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/testbus"
)
// builderOnTheBus is the builder's and the indexer's buckets on a test's own bus, which the controller's connection for
// the moment (servingBus) reaches: what a take-in from a key reads (R4e, issue 1851).
type builderOnTheBus struct {
js *broker.JetStream
builds, modules, manifests jetstream.KeyValue
}
func aBuilderOnTheBus(t *testing.T) builderOnTheBus {
t.Helper()
js, err := broker.Dial(testbus.URL(t))
if err != nil {
t.Fatal(err)
}
t.Cleanup(js.Close)
api, err := jetstream.New(js.Conn())
if err != nil {
t.Fatal(err)
}
ctx := t.Context()
b := builderOnTheBus{js: js}
for _, c := range []struct {
into *jetstream.KeyValue
bucket string
history uint8
}{
{&b.builds, sdkdelivery.BuilderBuildsBucket, sdkdelivery.BuilderBuildsHistory},
{&b.modules, sdkdelivery.ModulesBucket, sdkdelivery.ModulesHistory},
{&b.manifests, sdkdelivery.ManifestsBucket, 1},
} {
kv, err := api.CreateKeyValue(ctx, jetstream.KeyValueConfig{Bucket: c.bucket, History: c.history})
if err != nil {
t.Fatal(err)
}
*c.into = kv
}
servingBus.Store(js)
t.Cleanup(func() { servingBus.Store(nil) })
withTrunkList(t, true)
// The indexer on this bus is read, not the "not held" a mesh with no bus reads (aMeshOnTheStore).
wasIndexed := indexedModule
indexedModule = theIndexedModule
t.Cleanup(func() { indexedModule = wasIndexed })
return b
}
// indexed writes the indexer's trunk value of module at commit, naming fp, and its manifest: a module of one archive.
func (b builderOnTheBus) indexed(t *testing.T, module, version, repository, path, commit, fp string) string {
t.Helper()
keys, trunk := aKeyFor(module, version, fp, repository, path, commit)
raw := keys.manifests[trunk.Manifest]
value, err := sdkdelivery.EncodeManifest(sdkdelivery.Manifest{Module: module, JSON: raw})
if err != nil {
t.Fatal(err)
}
if _, err := b.manifests.Put(context.Background(), trunk.Manifest, value); err != nil {
t.Fatal(err)
}
trunk.FP = "src1:" + fp[len("src2:"):]
encoded, err := sdkdelivery.EncodeModule(trunk)
if err != nil {
t.Fatal(err)
}
if _, err := b.modules.Put(context.Background(), module, encoded); err != nil {
t.Fatal(err)
}
return trunk.Manifest
}
// built writes the builder's key of module at fp: done, its one archive named with a reference, or in another state.
func (b builderOnTheBus) built(t *testing.T, module, version, fp, manifest, state string) {
t.Helper()
keys, _ := aKeyFor(module, version, fp, "", "", "")
v := keys.keys[module].Build
v.Manifest = manifest
if state != sdkdelivery.BuildDone {
v = sdkdelivery.BuilderBuild{State: state, By: "ace"}
if state == sdkdelivery.BuildFailed {
v.Error = "digests differ: ace sha256:1, g14 sha256:2"
}
}
value, err := sdkdelivery.EncodeBuilderBuild(v)
if err != nil {
t.Fatal(err)
}
key, err := sdkdelivery.BuilderBuildKey(module, fp)
if err != nil {
t.Fatal(err)
}
if _, err := b.builds.Put(context.Background(), key, value); err != nil {
t.Fatal(err)
}
}
// noBuildAsked fails the test if anything asks build-agent for a build (ADR 0501 §6, ADR 0503 §6).
func noBuildAsked(t *testing.T) {
t.Helper()
was := askABuild
askABuild = func(_ context.Context, _ buildSource, path, _ string) (string, error) {
t.Errorf("a build was asked of build-agent, at %q", path)
return "", context.Canceled
}
t.Cleanup(func() { askABuild = was })
}
// asksWithPaths records every build a merge asks for, as repository, path and ref. The merge window's goroutine and
// the test's own may ask at once, so the record is written under a lock (issue 1543).
func asksWithPaths(t *testing.T) *[][3]string {
t.Helper()
asksOfEveryModule(t) // the one ask left is the builder's own (ADR 0453 §5): these tests are of what a walk does around it
var asked [][3]string
var mu sync.Mutex
was := askABuild
askABuild = func(_ context.Context, source buildSource, path, ref string) (string, error) {
mu.Lock()
defer mu.Unlock()
asked = append(asked, [3]string{source.Repository, path, ref})
return "b-" + path, nil
}
t.Cleanup(func() { askABuild = was })
return &asked
}
// theIndexedModule is the look at the indexer as the controller makes it, kept before any test replaces it.
var theIndexedModule = indexedModule
@@ -1,9 +0,0 @@
package main
import "context"
// A test's builder says it writes references, as R4e's does, unless the test says otherwise (issue 1886): no test reads
// the bus for it.
func init() {
builderSaysReferences = func(context.Context) (bool, error) { return true, nil }
}
-243
View File
@@ -1,243 +0,0 @@
package main
import (
"context"
"errors"
"strings"
"testing"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/novox/mesh-controller/internal/builderkeys"
"github.com/novox/mesh-controller/internal/inventory"
)
// theBuilderSays sets what the running builder says for one test.
func theBuilderSays(t *testing.T, says bool, err error) {
t.Helper()
was := builderSaysReferences
builderSaysReferences = func(context.Context) (bool, error) { return says, err }
t.Cleanup(func() { builderSaysReferences = was })
}
// askingBuildAgent records the modules build-agent is asked for, by path.
func askingBuildAgent(t *testing.T) *[]string {
t.Helper()
var asked []string
was := askABuild
askABuild = func(_ context.Context, _ buildSource, path, _ string) (string, error) {
asked = append(asked, path)
return "build-9", nil
}
t.Cleanup(func() { askABuild = was })
return &asked
}
// 1886/AC2: while no running builder says it writes references, a walk asks build-agent for a module of its tier, as
// before R4e, and takes nothing from the builder's key.
func TestAWalkAsksBuildAgentWhileTheBuilderDoesNotSayItWritesReferences(t *testing.T) {
theBuilderSays(t, false, nil)
asked := askingBuildAgent(t)
p, byName := aWalkOfOne("sudo")
askModule(t.Context(), p, "sudo", byName)
s := p.Modules["sudo"]
if s.Key || s.Build != "build-9" || s.State != "asked" || len(*asked) != 1 {
t.Fatalf("the module is not asked of build-agent: %+v, asked %v", s, *asked)
}
}
// 1886/AC3: once a running builder says it writes references, a walk takes the module from the builder's key (ADR
// 0498 §8) and asks build-agent for nothing.
func TestAWalkTakesTheKeyOnceTheBuilderSaysItWritesReferences(t *testing.T) {
theBuilderSays(t, true, nil)
asked := askingBuildAgent(t)
p, byName := aWalkOfOne("sudo")
askModule(t.Context(), p, "sudo", byName)
if s := p.Modules["sudo"]; !s.Key || s.Build != "" || len(*asked) != 0 {
t.Fatalf("the module is not taken from its key: %+v, asked %v", s, *asked)
}
}
// 1886/AC2: a say that cannot be read leaves the module unasked and says why, so the next look asks it: neither path is
// guessed.
func TestASayNotReadLeavesTheModuleUnasked(t *testing.T) {
theBuilderSays(t, false, errors.New("the bus did not answer"))
asked := askingBuildAgent(t)
p, byName := aWalkOfOne("sudo")
askModule(t.Context(), p, "sudo", byName)
s := p.Modules["sudo"]
if s.State != "" || s.Key || len(*asked) != 0 || !strings.Contains(s.Why, "the bus did not answer") {
t.Fatalf("a say not read: %+v, asked %v", s, *asked)
}
if got := unaskedOf(*p); len(got) != 1 {
t.Fatalf("the module is not left for the next look to ask: %v", got)
}
}
// 1886/AC2: the one function every ask goes through lets build-agent build any module while the builder does not say
// it writes references, and refuses it again once it does.
func TestBuildAgentIsAskedForAnyModuleWhileTheBuilderDoesNotSay(t *testing.T) {
was := askedModuleOf
t.Cleanup(func() { askedModuleOf = was })
askedModuleOf = func(_ context.Context, _ buildSource, path string) (string, error) {
return strings.TrimPrefix(path, "modules/"), nil
}
theBuilderSays(t, false, nil)
if err := onlyTheBuildersOwn(t.Context(), buildSource{Repository: "novox/mesh-catalog", Seat: "git"}, "modules/sudo"); err != nil {
t.Fatalf("an ask while the builder does not say is refused: %v", err)
}
theBuilderSays(t, true, nil)
if err := onlyTheBuildersOwn(t.Context(), buildSource{Repository: "novox/mesh-catalog", Seat: "git"}, "modules/sudo"); !errors.Is(err, errNotTheBuildersOwn) {
t.Fatalf("an ask once the builder says is not refused: %v", err)
}
}
// 1886/AC4: an open walk that waits on a key the running builder cannot reference (done without a reference, as R4e's
// own walk did in plan-1792707174627079192) is judged again at the next look: the module is handed back unasked, and
// that look's ask goes to build-agent. With the builder saying it writes references, the same walk keeps waiting.
func TestAnOpenWalkWaitingOnAnUnreferencedKeyIsAskedOfBuildAgent(t *testing.T) {
t.Cleanup(func() {
keyTakes.Lock()
keyTakes.running, keyTakes.waits, keyTakes.taken = map[string]bool{}, map[string]string{}, map[string]bool{}
keyTakes.Unlock()
})
fp := "src2:" + strings.Repeat("e", 64)
unreferenced := builderkeys.Key{Module: "sudo", Fingerprint: fp, Found: true, Revision: 1,
Build: sdkdelivery.BuilderBuild{State: sdkdelivery.BuildDone, Manifest: strings.Repeat("b", 64),
Artifacts: []sdkdelivery.BuiltArtifact{{Name: "tools", Kind: "bundle", Digest: "sha256:1"}}}}
walk := func() (*inventory.Plan, map[string]inventory.Entry, planKeysOf) {
theBuilderSays(t, true, nil)
p, byName := aWalkOfOne("sudo")
askModule(t.Context(), p, "sudo", byName)
keys := planKeysOf{values: map[string]sdkdelivery.Module{"sudo@c0ffee0123456789": {FP2: fp}},
keys: map[string]builderkeys.Key{"sudo": unreferenced}}
return p, byName, keys
}
take := func(string, sdkdelivery.Module, string) { t.Fatal("an unreferenced key was taken in") }
records := recordsOf{builds: map[string]inventory.Build{}}
// Said: it waits for the reference, as ADR 0498 §8 has it.
p, _, keys := walk()
if _, err := settleKeys(t.Context(), records, keys, p, p.Tiers[0], take); err != nil {
t.Fatal(err)
}
if s := p.Modules["sudo"]; s.State != "asked" || !s.Key || !strings.Contains(s.Why, "waits for the reference") {
t.Fatalf("with the builder saying, the walk does not wait for the reference: %+v", s)
}
// Not said: handed back, and the next look asks build-agent.
p, byName, keys := walk()
theBuilderSays(t, false, nil)
asked := askingBuildAgent(t)
changed, err := settleKeys(t.Context(), records, keys, p, p.Tiers[0], take)
if err != nil || !changed {
t.Fatalf("the walk was not judged again: %v %v", changed, err)
}
s := p.Modules["sudo"]
if s.Key || s.State != "" || p.State != inventory.PlanBuilding || !strings.Contains(s.Why, "does not say") {
t.Fatalf("the module is not handed back to be asked of build-agent: %+v %s", s, p.State)
}
for _, m := range unaskedOf(*p) {
askModule(t.Context(), p, m, byName)
}
if s := p.Modules["sudo"]; s.Key || s.Build != "build-9" || len(*asked) != 1 {
t.Fatalf("the next look did not ask build-agent: %+v, asked %v", s, *asked)
}
}
// 1886/AC4: a key the running builder failed while it does not say it writes references fails no walk: the module is
// asked of build-agent, whose build is the one delivered.
func TestAFailedKeyOfABuilderThatDoesNotSayFailsNoWalk(t *testing.T) {
theBuilderSays(t, true, nil)
p, byName := aWalkOfOne("sudo")
askModule(t.Context(), p, "sudo", byName)
_ = byName
fp := "src2:" + strings.Repeat("f", 64)
theBuilderSays(t, false, nil)
keys := planKeysOf{values: map[string]sdkdelivery.Module{"sudo@c0ffee0123456789": {FP2: fp}},
keys: map[string]builderkeys.Key{"sudo": {Module: "sudo", Fingerprint: fp, Found: true,
Build: sdkdelivery.BuilderBuild{State: sdkdelivery.BuildFailed, Error: "not stored: tools: no reference"}}}}
if _, err := settleKeys(t.Context(), recordsOf{}, keys, p, p.Tiers[0], func(string, sdkdelivery.Module, string) {}); err != nil {
t.Fatal(err)
}
if s := p.Modules["sudo"]; p.State != inventory.PlanBuilding || s.State != "" || s.Key {
t.Fatalf("a failed key of a builder that does not say failed the walk: %s %q %+v", p.State, p.Note, s)
}
}
// 1886/AC6 (ADR 0522 §6): the parts that write references, the builder and the build modules, are build-agent's own until
// R4g: asked of build-agent and never taken from keys they write themselves, whatever the builder says.
func TestTheReferenceWritersAreAskedOfBuildAgentWhateverTheBuilderSays(t *testing.T) {
theBuilderSays(t, true, nil)
asked := askingBuildAgent(t)
for _, m := range []string{"mesh-build-go", "mesh-build-node", "mesh-build-image", "mesh-delivery-builder"} {
p, byName := aWalkOfOne(m)
askModule(t.Context(), p, m, byName)
if s := p.Modules[m]; s.Key || s.Build != "build-9" {
t.Errorf("%s is not asked of build-agent: %+v", m, s)
}
}
if len(*asked) != 4 {
t.Fatalf("build-agent was asked %v", *asked)
}
}
// 1886 review round 1 (ADR 0522 §2): the guard of the one ask function judges by the say askModule read, never a second
// read: a second read that fails, or that says otherwise, fails no plan.
func TestTheAskGuardUsesTheSayTheWalkRead(t *testing.T) {
was := askedModuleOf
t.Cleanup(func() { askedModuleOf = was })
askedModuleOf = func(_ context.Context, _ buildSource, path string) (string, error) {
return strings.TrimPrefix(path, "modules/"), nil
}
reads := 0
wasSays := builderSaysReferences
t.Cleanup(func() { builderSaysReferences = wasSays })
for _, second := range []struct {
name string
says bool
err error
}{{"a second read that fails", false, errors.New("the bus did not answer")}, {"a say that turned true", true, nil}} {
reads = 0
builderSaysReferences = func(context.Context) (bool, error) {
reads++
if reads == 1 {
return false, nil
}
return second.says, second.err
}
wasAsk := askABuild
askABuild = func(ctx context.Context, source buildSource, path, _ string) (string, error) {
if err := onlyTheBuildersOwn(ctx, source, path); err != nil {
return "", err
}
return "build-9", nil
}
p, byName := aWalkOfOne("sudo")
askModule(t.Context(), p, "sudo", byName)
askABuild = wasAsk
if s := p.Modules["sudo"]; p.State == inventory.PlanFailed || s.Build != "build-9" || s.Key {
t.Errorf("%s: the walk %s, the module %+v", second.name, p.State, s)
}
}
}
// 1886 merge check hang: the look must judge an ask and a settle by one say. A settle that read the say from another
// source than the ask (the bus's bucket, against the walk's read) handed a module back, the next ask took it from its key
// again, and the look's loop never ended: walk_from_keys_integration_test hung.
func TestTheSettleJudgesByTheSameSayAsTheAsk(t *testing.T) {
theBuilderSays(t, true, nil)
p, _ := aWalkOfOne("sudo")
p.Modules["sudo"] = &inventory.PlanModule{}
askModule(t.Context(), p, "sudo", map[string]inventory.Entry{"sudo": {Source: inventory.Source{Repository: "novox/mesh-catalog", Path: "modules/sudo", Ref: "main"}}})
fp := "src2:" + strings.Repeat("1", 64)
keys := planKeysOf{values: map[string]sdkdelivery.Module{"sudo@c0ffee0123456789": {FP2: fp}},
keys: map[string]builderkeys.Key{"sudo": {Module: "sudo", Fingerprint: fp, Found: true,
Build: sdkdelivery.BuilderBuild{State: sdkdelivery.BuildBuilding, By: "shanks"}}}}
if _, err := settleKeys(t.Context(), recordsOf{}, keys, p, p.Tiers[0], func(string, sdkdelivery.Module, string) {}); err != nil {
t.Fatal(err)
}
if s := p.Modules["sudo"]; !s.Key || s.State != "asked" {
t.Fatalf("the settle judged by another say than the ask's: %+v", s)
}
}
-112
View File
@@ -1,112 +0,0 @@
package main
import (
"context"
"errors"
"strings"
"testing"
)
// A12 (ADR 0453 §5, ADR 0498 §7; issue 1851): the one function every ask goes through refuses any module but the
// builder's own, naming the builder's key it is taken from instead; the builder's own module passes the refusal.
func TestBuildAgentIsAskedForTheBuildersOwnModuleAlone(t *testing.T) {
was := askedModuleOf
t.Cleanup(func() { askedModuleOf = was })
askedModuleOf = func(_ context.Context, _ buildSource, path string) (string, error) {
return strings.TrimPrefix(path, "modules/"), nil
}
_, err := askABuild(t.Context(), buildSource{Repository: "novox/mesh-catalog", Seat: "git"}, "modules/sudo", "main")
if !errors.Is(err, errNotTheBuildersOwn) || !strings.Contains(err.Error(), "sudo is taken from the builder's done key") ||
!strings.Contains(err.Error(), "mesh-delivery-builder_builds") {
t.Fatalf("an ask of another module: %v", err)
}
if err := onlyTheBuildersOwn(t.Context(), buildSource{Repository: "novox/mesh-delivery", Seat: "git"},
"modules/mesh-delivery-builder"); err != nil {
t.Fatalf("the builder's own module is refused: %v", err)
}
askedModuleOf = func(context.Context, buildSource, string) (string, error) { return "", errors.New("no store") }
if err := onlyTheBuildersOwn(t.Context(), buildSource{Repository: "novox/x"}, ""); !errors.Is(err, errNotTheBuildersOwn) {
t.Fatalf("an ask whose module cannot be read went on: %v", err)
}
}
// A12: a walk of the builder's own module and one other asks build-agent for the first alone; the other is taken from
// its key.
func TestAWalkAsksBuildAgentForTheBuildersOwnModuleAlone(t *testing.T) {
var asked []string
was := askABuild
askABuild = func(_ context.Context, _ buildSource, path, _ string) (string, error) {
asked = append(asked, path)
return "build-1", nil
}
t.Cleanup(func() { askABuild = was })
p, byName := aWalkOfOne("sudo")
p.Tiers = [][]string{{"mesh-delivery-builder", "sudo"}}
byName["mesh-delivery-builder"] = byName["sudo"]
e := byName["mesh-delivery-builder"]
e.Source.Repository, e.Source.Path = "novox/mesh-delivery", "modules/mesh-delivery-builder"
byName["mesh-delivery-builder"] = e
for _, m := range p.Tiers[0] {
askModule(t.Context(), p, m, byName)
}
if len(asked) != 1 || asked[0] != "modules/mesh-delivery-builder" {
t.Fatalf("build-agent was asked for %v, want the builder's own module alone", asked)
}
if s := p.Modules["mesh-delivery-builder"]; s.Key || s.Build != "build-1" {
t.Fatalf("the builder's own module is not asked: %+v", s)
}
if s := p.Modules["sudo"]; !s.Key || s.Build != "" {
t.Fatalf("the other module is not taken from its key: %+v", s)
}
}
// ADR 0520 §1 (issue 1851): until R4g build-agent and mesh-tools, the toolchain image its builds stand on, are build-agent's
// own, as the builder's module is: the builder writes no key of them, so a walk asks build-agent for them and reads no key;
// mesh-tools-go (its key from mesh-build-image) and any other module of the same tier are still taken from their keys.
func TestAWalkAsksBuildAgentForItselfAndMeshToolsAndReadsNoKeyOfThem(t *testing.T) {
own := []string{"build-agent", "mesh-tools"}
keyed := []string{"mesh-tools-go", "sudo"}
wasModule := askedModuleOf
askedModuleOf = func(_ context.Context, _ buildSource, path string) (string, error) {
return strings.TrimPrefix(path, "modules/"), nil
}
t.Cleanup(func() { askedModuleOf = wasModule })
for _, m := range own {
if err := onlyTheBuildersOwn(t.Context(), buildSource{Repository: "novox/x", Seat: "git"}, "modules/"+m); err != nil {
t.Fatalf("an ask of %s is refused: %v", m, err)
}
}
if err := onlyTheBuildersOwn(t.Context(), buildSource{Repository: "novox/x", Seat: "git"}, "modules/mesh-tools-go"); !errors.Is(err, errNotTheBuildersOwn) {
t.Fatalf("an ask of mesh-tools-go went on: %v", err)
}
var asked []string
was := askABuild
askABuild = func(_ context.Context, _ buildSource, path, _ string) (string, error) {
asked = append(asked, path)
return "build-" + strings.TrimPrefix(path, "modules/"), nil
}
t.Cleanup(func() { askABuild = was })
p, byName := aWalkOfOne("sudo")
p.Tiers = [][]string{append(append([]string{}, own...), keyed...)}
for _, m := range append(append([]string{}, own...), keyed...) {
e := byName["sudo"]
e.Source.Path = "modules/" + m
byName[m] = e
}
for _, m := range p.Tiers[0] {
askModule(t.Context(), p, m, byName)
}
if strings.Join(asked, " ") != "modules/build-agent modules/mesh-tools" {
t.Fatalf("build-agent was asked for %v", asked)
}
for _, m := range own {
if s := p.Modules[m]; s.Key || s.Build != "build-"+m || s.State != "asked" {
t.Fatalf("%s is not asked of build-agent, or is read from a key: %+v", m, s)
}
}
for _, m := range keyed {
if s := p.Modules[m]; !s.Key || s.Build != "" {
t.Fatalf("%s is not taken from its key: %+v", m, s)
}
}
}
-7
View File
@@ -32,13 +32,6 @@ func assertBusObjects(ctx context.Context, inv *inventory.Inventory, r broker.Ra
if err := broker.Raise(r, names); err != nil {
return nil, err
}
// And each node's bucket for the files of its declaration that travel by digest (novox/hq ADR 0486 §4), so a node finds its own
// whenever the bus is up.
if files, ok := r.(interface{ EnsureDeclarationFiles(nodes []string) error }); ok {
if err := files.EnsureDeclarationFiles(names); err != nil {
return nil, err
}
}
// The work queues of the mesh's own roles (novox/hq ADR 0121). The queue before the holder,
// deliberately: work queues until somebody arrives to do it, so assigning a build machine a week
// after something started asking for builds flushes the backlog instead of having lost it.
+7 -3
View File
@@ -241,9 +241,10 @@ func mergeAskPrefix(number int, head string) string {
return fmt.Sprintf("merge-%d-%s-", number, head)
}
// askersOfMerges are the askers a merge's warrant is read under: the teller alone (ADR 0437 §6), since R4e stopped taking
// mesh-delivery's (ADR 0450 §8; issue 1318).
func askersOfMerges() []string { return []string{sdkdelivery.TellerModule} }
// askersOfMerges are the askers a merge's warrant is read under: the teller, and mesh-delivery for a merge before release 3e
// (ADR 0437 §6). Release 4's controller work stops accepting mesh-delivery (ADR 0450 §8); a merge asked under it earlier is
// still found here until then.
func askersOfMerges() []string { return []string{sdkdelivery.TellerModule, theDelivery} }
// caughtUpWarrant judges a chosen merge ask of the router as the warrant of a merge: the ask is the operator's, the warrant
// answers that very ask as it was published, and it is the operator's Approve proved by the channel it came through (ADR 0437
@@ -390,6 +391,9 @@ var findMergeWarrant = func(ctx context.Context, inv catalogueReader, owner, rep
if asker == sdkdelivery.TellerModule && !strings.HasPrefix(id, sdkdelivery.AskKindMerge+"-") {
continue
}
if asker == theDelivery && !named {
continue
}
ask, w, err := closedRecordOf(ctx, inv, asker, id)
if err != nil {
return "", false, "", err
+34 -6
View File
@@ -69,12 +69,12 @@ func TestAForgeThatDoesNotAnswerIsAnErrorNotAnEmptyTrunk(t *testing.T) {
func TestACaughtUpMergeIsAcceptedOnTheSameWarrantAsAnAnnouncedOne(t *testing.T) {
ok := approvedMerge(t)
if _, err := caughtUpWarrant(sdkdelivery.TellerModule, &ok.ask, &ok.warrant); err != nil {
t.Errorf("the operator's Approve under the teller was refused: %v", err)
}
// 1318 (ADR 0450 §8): mesh-delivery's warrant of before the gate is no longer taken.
if _, err := caughtUpWarrant(theDelivery, &ok.ask, &ok.warrant); err == nil {
t.Error("a warrant under mesh-delivery was accepted")
for _, asker := range []string{sdkdelivery.TellerModule, theDelivery} {
w := ok.warrant
w.Asker = asker
if _, err := caughtUpWarrant(asker, &ok.ask, &w); err != nil {
t.Errorf("the operator's Approve under %s was refused: %v", asker, err)
}
}
if _, err := caughtUpWarrant("keycloak", &ok.ask, &ok.warrant); err == nil {
t.Error("a warrant of an asker the controller takes none from was accepted")
@@ -229,6 +229,34 @@ func TestAGroupsWarrantIsFoundByTheAsksWordsForThePullRequestAndItsHead(t *testi
}
}
// Keycloak's merge was asked as mesh-delivery.merge-378-b9104454f9a5-1, before release 3e: the id names the pull request and
// the first twelve digits of its head, and the warrant is found by them under mesh-delivery.
func TestAWarrantAskedUnderMeshDeliveryIsFoundByThePullRequestAndItsHead(t *testing.T) {
ok := approvedMerge(t)
head := "b9104454f9a5e0c4d3b2a1f0e9d8c7b6a5948372"
askID := "merge-378-b9104454f9a5-1"
ask, w := ok.ask, ok.warrant
ask.ID, w.Ask, w.Asker = askID, askID, theDelivery
w.AskDigest = ask.Digest()
withRouter(t, map[string]map[string]aRouterRecord{theDelivery: {
"merge-377-aaaaaaaaaaaa-1": {},
askID: {ask, w},
}})
why, found, reason, err := findMergeWarrant(t.Context(), nil, "novox", "mesh-catalog", 378, head)
if err != nil || !found || !strings.Contains(why, askID) {
t.Fatalf("the warrant was not found: %q %v %q %v", why, found, reason, err)
}
// Another head names another ask, and none is found.
if _, found, _, _ := findMergeWarrant(t.Context(), nil, "novox", "mesh-catalog", 378, "ffffffffffff0000"); found {
t.Fatal("another head found a warrant")
}
// A router that cannot be read is an error, never "no warrant".
routerAskIDs = func(context.Context, catalogueReader, string) ([]string, error) { return nil, errors.New("no answer") }
if _, found, _, err := findMergeWarrant(t.Context(), nil, "novox", "mesh-catalog", 378, head); found || err == nil {
t.Fatalf("an unreadable router read as no warrant: %v %v", found, err)
}
}
// A trunk shorter than a page, or with none, is read without a fault: the pages end when the forge's do.
func TestAShortTrunkIsReadWholeAndAnEmptyOneIsNoMerges(t *testing.T) {
was := trunkCommitsPage
+44 -2
View File
@@ -51,13 +51,55 @@ func attachTheChain(ctx context.Context, inv *inventory.Inventory, server *link.
fmt.Printf("chain: the builds are not laid out under their modules yet, and none is read until they are: %v\n", err)
go layBuildsOutSoon(ctx, inv)
}
// The requests the move of delivery 1 found never handed over are not asked again: the controller asks no build
// (novox/hq ADR 0501; issue 1851, A10), and the builder builds every fingerprint the indexer writes.
// And the requests the move of delivery 1 found never handed over, asked again once (askAgain), once it is serving.
go askAgainSoon(ctx, inv)
// And what delivery 1 recorded of what each machine was sent, copied onto the keys of the assignments (CopySentBuilds).
go copySentBuildsSoon(ctx, inv)
return nil
}
// askAgain asks once more for each build request the move found never handed over (novox/hq issue 981 note 7 Q5): the
// requests a merge could not ask for, which a person's hand `build` would otherwise have to repeat. **At most once**: a
// request is claimed before it is asked and settled after, so a crash between loses the ask, which is where it stood.
func askAgain(ctx context.Context, inv *inventory.Inventory) {
pending, err := inv.RequestsToAskAgain(ctx)
if err != nil {
fmt.Printf("chain: the requests never handed over cannot be read: %v\n", err)
return
}
for _, r := range pending {
if claimed, err := inv.ClaimToAskAgain(ctx, r.ID); err != nil || !claimed {
if err != nil {
fmt.Printf("chain: %s cannot be claimed to be asked again: %v\n", r.ID, err)
}
continue
}
id, err := askABuild(ctx, buildSource{Repository: r.Repository, Seat: r.Seat}, r.Path, r.Ref)
if err != nil {
if serr := inv.AskedAgainAs(ctx, r.ID, "not asked again: "+err.Error()); serr != nil {
fmt.Printf("chain: %s was not asked again, and that could not be kept: %v\n", r.ID, serr)
}
fmt.Printf("chain: %s was never handed over and cannot be asked again: %v\n", r.ID, err)
continue
}
recordBuildRequest(ctx, inv, inventory.BuildRequest{ID: id, Repository: r.Repository, Seat: r.Seat, Path: r.Path,
Ref: r.Ref, Commit: r.Commit, For: r.For})
if err := inv.AskedAgainAs(ctx, r.ID, id); err != nil {
fmt.Printf("chain: %s was asked again as %s, and that could not be kept: %v\n", r.ID, id, err)
}
fmt.Printf("chain: %s, never handed over, is asked again as %s\n", r.ID, id)
}
}
// askAgainSoon waits for the controller to be serving, which an ask needs, and then asks again.
func askAgainSoon(ctx context.Context, inv *inventory.Inventory) {
select {
case <-ctx.Done():
case <-time.After(time.Minute):
askAgain(ctx, inv)
}
}
// kindSentBuildsNotCopied is the condition of the one-time copy of node.sent_builds onto the assignments' keys that has not
// succeeded: until it has, a module an assignment pulled in may be missing from what a machine is read to run.
const kindSentBuildsNotCopied = "sent-builds-not-copied"
+25 -1
View File
@@ -5,6 +5,7 @@ import (
"errors"
"fmt"
"os"
"strconv"
"github.com/novox/mesh-controller/internal/builder"
"github.com/novox/mesh-controller/internal/inventory"
@@ -52,7 +53,7 @@ func forgetChecksOfWalk(ctx context.Context, inv *inventory.Inventory, p invento
if p.WalkSaid != inventory.WalkEnded || p.Delivery == nil {
return
}
for _, pull := range pullsOfWalk(&p) {
for _, pull := range reuseAskOf(&p).Pulls {
said, err := forgetChecks(ctx, inv, builder.PullRef{Owner: pull.Owner, Repo: pull.Repo, Number: pull.Number})
if err != nil {
fmt.Fprintf(os.Stderr, "%s: its check builds stay until the week's sweep: %v\n", p.ID, err)
@@ -61,3 +62,26 @@ func forgetChecksOfWalk(ctx context.Context, inv *inventory.Inventory, p invento
fmt.Printf("%s: %s\n", p.ID, said)
}
}
// forgetChecksCommand is `forget-checks <owner> <repo> <number>`: the verb the delivery's owner asks when the forge
// says a pull request closed unmerged (novox/hq ADR 0350 §5).
func forgetChecksCommand(ctx context.Context, args []string) error {
if len(args) != 3 {
return errors.New("forget-checks <owner> <repo> <number>")
}
number, err := strconv.Atoi(args[2])
if err != nil || number <= 0 {
return fmt.Errorf("%q is not a pull request's number", args[2])
}
open, err := openStores(ctx)
if err != nil {
return err
}
defer open.Close()
said, err := forgetChecks(ctx, open.inventory, builder.PullRef{Owner: args[0], Repo: args[1], Number: number})
if err != nil {
return err
}
fmt.Println(said)
return nil
}
@@ -14,12 +14,23 @@ import (
"github.com/novox/mesh-controller/internal/link"
)
func TestAWalkNamesThePullRequestsOfItsMergesOnce(t *testing.T) {
func TestOnlyAWalksAskTakesAChecksBuildAndADryRunNever(t *testing.T) {
p := &inventory.Plan{Delivery: &inventory.PlanDelivery{Merges: []inventory.PlanMerge{
{Repository: "novox/shop", Number: 7}, {Repository: "http://forge.invalid/novox/web.git", Number: 3},
{Repository: "novox/shop", Number: 7}, {Repository: "novox/shop"}}}}
if want := []link.PullRef{{Owner: "novox", Repo: "shop", Number: 7}, {Owner: "novox", Repo: "web", Number: 3}}; !reflect.DeepEqual(pullsOfWalk(p), want) {
t.Fatalf("the walk names %+v", pullsOfWalk(p))
asked := reuseAskOf(p)
if want := []link.PullRef{{Owner: "novox", Repo: "shop", Number: 7}, {Owner: "novox", Repo: "web", Number: 3}}; !reflect.DeepEqual(asked.Pulls, want) {
t.Fatalf("the walk names %+v", asked.Pulls)
}
walk := withReuse(t.Context(), asked)
if reuseAsked(walk, false) != asked {
t.Fatal("a walk's ask does not carry its leave")
}
if reuseAsked(walk, true) != nil {
t.Fatal("a dry run carries a leave to take a check's build")
}
if reuseAsked(t.Context(), false) != nil {
t.Fatal("an ask that is not a walk's carries a leave")
}
}
+59
View File
@@ -104,6 +104,44 @@ func TestAWalksBuildSaysWhetherItReusedOrBuilt(t *testing.T) {
// build as it judges one the walk built. A scratch reference is refused. Needs the store (MESH_TEST_POSTGRES), and
// is named in testdata/store-tests.
// A walk's leave names the builds its own pull requests' verdicts made at the head each merge merged — never an
// earlier head's — read once per walk, and none when they cannot be read: the builder takes nothing a verdict does
// not name (novox/hq ADR 0350 §2, issue 703).
func TestAWalksLeaveNamesTheBuildsItsPullRequestsVerdictsMade(t *testing.T) {
was := checkedBuildsOf
forgetWalkVerdicts()
t.Cleanup(func() { checkedBuildsOf = was; forgetWalkVerdicts() })
var asked [][]link.PullRef
pull := link.PullRef{Owner: "novox", Repo: "shop", Number: 7}
atHead := link.CheckedBuild{Pull: pull, Check: "build-2", Head: "bbbbbbbbbbbb", Built: link.CheckBuilt{Module: "app", Key: "r1:2"}}
earlier := link.CheckedBuild{Pull: pull, Check: "build-1", Head: "aaaaaaaaaaaa", Built: link.CheckBuilt{Module: "app", Key: "r1:1"}}
checkedBuildsOf = func(_ context.Context, pulls []link.PullRef) ([]link.CheckedBuild, error) {
asked = append(asked, pulls)
return []link.CheckedBuild{atHead, earlier}, nil
}
pulls := []link.PullRef{pull}
heads := map[link.PullRef]string{pull: "bbbbbbbbbbbb0123456789"}
for range 3 {
if got := reuseWithVerdicts(t.Context(), "plan-1", &link.ReuseAsk{Pulls: pulls}, heads); !reflect.DeepEqual(got.Builds,
[]link.CheckedBuild{atHead}) {
t.Fatalf("the leave names %+v, not only the merged head's build", got.Builds)
}
}
if !reflect.DeepEqual(asked, [][]link.PullRef{pulls}) {
t.Fatalf("the verdicts were read %d times for one walk: %v", len(asked), asked)
}
// A merge whose head was not said takes nothing.
if got := reuseWithVerdicts(t.Context(), "plan-1", &link.ReuseAsk{Pulls: pulls}, nil); got.Builds != nil {
t.Fatalf("a merge of no said head names %+v", got.Builds)
}
checkedBuildsOf = func(context.Context, []link.PullRef) ([]link.CheckedBuild, error) {
return nil, errors.New("no bus")
}
if got := reuseWithVerdicts(t.Context(), "plan-2", &link.ReuseAsk{Pulls: pulls}, heads); got.Builds != nil {
t.Fatalf("verdicts that could not be read named %+v", got.Builds)
}
}
// A check is handed only the builds earlier checks of its own pull request at its own head made: another pull
// request's build, or an earlier head's, is never kept (novox/hq issue 703). None when they cannot be read.
func TestACheckIsHandedOnlyItsOwnHeadsBuilds(t *testing.T) {
@@ -132,3 +170,24 @@ func TestACheckIsHandedOnlyItsOwnHeadsBuilds(t *testing.T) {
t.Fatalf("verdicts that could not be read named %+v", got)
}
}
func forgetWalkVerdicts() {
walkVerdicts.Lock()
defer walkVerdicts.Unlock()
walkVerdicts.read = map[string]verdictsRead{}
}
// A walk's merged heads are the forge's word for each merge; a merge announced without its head, or a pull request
// merged at two heads in one walk, has none, and its walk takes no check's build (novox/hq issue 703).
func TestAWalksMergedHeadsAreTheForgesWord(t *testing.T) {
merge := func(number int, head string) inventory.PlanMerge {
return inventory.PlanMerge{Repository: "novox/shop", Number: number, Head: head}
}
p := &inventory.Plan{Delivery: &inventory.PlanDelivery{Merges: []inventory.PlanMerge{
merge(7, "aaaaaaaaaaaa1"), merge(8, ""), merge(9, "bbbbbbbbbbbb"), merge(9, "cccccccccccc")}}}
heads := mergedHeadsOf(p)
shop := func(n int) link.PullRef { return link.PullRef{Owner: "novox", Repo: "shop", Number: n} }
if heads[shop(7)] != "aaaaaaaaaaaa1" || heads[shop(8)] != "" || heads[shop(9)] != "" {
t.Fatalf("merged heads %v", heads)
}
}
+165
View File
@@ -0,0 +1,165 @@
package main
import (
"context"
"fmt"
"regexp"
"sort"
"strings"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// **A delivery group's member is checked beside the other members' heads** (novox/hq ADR 0382, issue 867).
//
// A mesh-host change and a mesh-controller change that need each other could not both pass their own merge checks:
// each was read beside the other's repository at the commit the mesh runs, which holds neither change. So the
// controller asks mesh-delivery's `groups` once for every pull request's check it asks (§1), and names, in the ask,
// the group's other open pull requests into a base of the same name, in a repository the build seat reads beside the
// check — mesh-lab never — each with its whole head. The build seat reads those repositories at the sibling's merge (§2).
// A pull request for which mesh-delivery answered and named no sibling is asked exactly as before (§7).
// wholeCommit is a commit as the forge says a head: forty hexadecimal characters.
var wholeCommit = regexp.MustCompile(`^[0-9a-f]{40}$`)
// besideRepositories are the repositories (lower case, owner/repo) a check reads beside its own, by the directory each
// is found under — the core's, as besideOf clones them, and mesh-sdk beside the controller; never mesh-lab, which is
// always read at its trunk.
func besideRepositories(entries []inventory.Entry) map[string]string {
read := map[string]string{}
for _, e := range entries {
dir, core := coreModules[e.Manifest.Module]
if !core || e.Provided || e.Source.Repository == "" {
continue
}
read[nameOf(e.Source.Repository)] = dir
if dir == "mesh-controller" && e.Source.Seat != "" {
read[nameOf(siblingOf(e.Source.Repository, "mesh-sdk"))] = "mesh-sdk"
}
}
delete(read, "")
return read
}
// nameOf is a recorded repository as owner/repo in lower case, empty when it names no owner.
func nameOf(repository string) string {
owner, repo := sourceOwner(repository), sourceName(repository)
if owner == "" || repo == "" {
return ""
}
return strings.ToLower(owner + "/" + repo)
}
// siblingsFor is the pull request's siblings, as mesh-delivery's groups say them (ADR 0382 §1): the other members of
// the group its head branch names that are open, unmerged and into a base of its own base's name, in a repository the
// check reads beside it (beside: the directories the ask already names). The second answer is why they could not be
// learned, empty when they could: the group's members carry no head or base — a mesh-delivery from before the rule.
func siblingsFor(p link.PullUpdated, entries []inventory.Entry, beside map[string]link.CheckedOut,
lines []groupsLine) ([]link.CheckSibling, string) {
own := strings.ToLower(p.Owner + "/" + p.Repo)
var group *groupsLine
if p.Head == "" {
// A recheck (mesh-delivery's `recheck` through `delivery check`) carries no head branch name: the group is the one
// whose members hold this very pull request, by repository and number.
learnable := false
for i := range lines {
for _, m := range lines[i].Details {
learnable = true
if p.Number != 0 && m.Number == p.Number && strings.ToLower(m.Repository) == own {
group = &lines[i]
}
}
}
if group == nil {
for _, l := range lines {
if len(l.Details) == 0 && len(l.Members) > 0 && !learnable {
return nil, "mesh-delivery named the groups' members without their pull requests (a mesh-delivery from before novox/hq ADR 0382): " +
"the group of a pull request announced without its head branch cannot be found"
}
}
return nil, "" // a lone pull request
}
} else {
for i := range lines {
if lines[i].ID == p.Head {
group = &lines[i]
}
}
if group == nil {
return nil, ""
}
}
if len(group.Details) == 0 && len(group.Members) > 0 {
return nil, "mesh-delivery named the group's members without their heads and bases (a mesh-delivery from before novox/hq ADR 0382)"
}
read := besideRepositories(entries)
var siblings []link.CheckSibling
for _, m := range group.Details {
repository := strings.ToLower(m.Repository)
if repository == own {
continue
}
if !m.Open {
continue
}
if m.Head == "" || m.Base == "" {
return nil, "mesh-delivery named " + m.ID + " without its head or base (a mesh-delivery from before novox/hq ADR 0382)"
}
dir, readBeside := read[repository]
if m.Base != p.Base || !readBeside || dir == "mesh-lab" {
continue
}
if _, there := beside[dir]; !there {
continue
}
head := strings.ToLower(m.Head)
if !wholeCommit.MatchString(head) {
return nil, fmt.Sprintf("mesh-delivery named %s with a head that is not a whole commit (%q)", m.ID, m.Head)
}
siblings = append(siblings, link.CheckSibling{Repository: m.Repository, Number: m.Number, Head: head, Base: m.Base, Dir: dir})
}
sort.Slice(siblings, func(i, j int) bool { return siblings[i].Dir < siblings[j].Dir })
return siblings, ""
}
// withSiblings asks mesh-delivery's `groups` once and says in the request the pull request's siblings, or why they
// could not be learned (ADR 0382 §1): a check never waits on mesh-delivery past groupAskWithin and never fails for it,
// so a fix to mesh-delivery is still checked without it.
func withSiblings(ctx context.Context, request *link.BuildRequest, p link.PullUpdated, entries []inventory.Entry) {
lines, err := groupsAsked(ctx, deliverySeatHeld(entries))
siblingsFrom(request, p, entries, lines, err)
}
// siblingsFrom is withSiblings from mesh-delivery's answer already had.
func siblingsFrom(request *link.BuildRequest, p link.PullUpdated, entries []inventory.Entry, lines []groupsLine, err error) {
if request.Check == nil {
return
}
if err != nil {
request.Check.SiblingsUnknown = fmt.Sprintf("mesh-delivery did not say the group's pull requests (%v): "+
"the check reads no siblings", err)
return
}
request.Check.Siblings, request.Check.SiblingsUnknown = siblingsFor(p, entries, request.Check.Beside, lines)
}
// composedMember says the pull request is a member of a delivery group whose gate is the composed check's (novox/hq
// ADR 0403 clauses 1 and 7): the group has two members or more and mesh-delivery says so on its line. The group is
// found as siblingsFor finds it, by its head branch name, or, for a recheck that carries none, by the pull request's
// repository and number — not by siblings, which name only what the check reads beside it.
func composedMember(p link.PullUpdated, lines []groupsLine) bool {
own := strings.ToLower(p.Owner + "/" + p.Repo)
for _, l := range lines {
found := p.Head != "" && l.ID == p.Head
for _, m := range l.Details {
if p.Head == "" && p.Number != 0 && m.Number == p.Number && strings.ToLower(m.Repository) == own {
found = true
}
}
if found {
return l.ComposedGate && max(len(l.Details), len(l.Members)) >= 2
}
}
return false
}
+248
View File
@@ -0,0 +1,248 @@
package main
import (
"context"
"encoding/json"
"errors"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// **A delivery group's member is checked beside the other members' heads** (novox/hq ADR 0382 §1, §7, issue 867): the
// controller asks mesh-delivery's `groups` once for every check it asks, and names in the ask the group's other open
// pull requests into a base of the same name, in a repository the build seat reads beside it — mesh-lab never.
const (
headHost = "1111111111111111111111111111111111111111"
headController = "2222222222222222222222222222222222222222"
headSDK = "3333333333333333333333333333333333333333"
headLab = "4444444444444444444444444444444444444444"
headHq = "5555555555555555555555555555555555555555"
)
func theCoreOnTheGitSeat() []inventory.Entry {
onSeat := func(module, repository string) inventory.Entry {
e := fromRepo(module, repository, "")
e.Source.Seat = "git"
return e
}
nats := onSeat("nats", "novox/mesh-catalog")
nats.Source.Path = "modules/nats"
held := catalogue.Manifest{Module: "mesh-delivery", Claims: []catalogue.Claim{{Name: catalogue.DeliverySeat, Scope: catalogue.ScopeMesh}}}
return []inventory.Entry{onSeat("mesh-controller", "novox/mesh-controller"), onSeat("mesh-host", "novox/mesh-host"),
onSeat("node-tools", "novox/mesh-tools"), nats, {Manifest: held, On: []string{"anchor"}}}
}
func besideTheCore() map[string]link.CheckedOut {
return map[string]link.CheckedOut{"mesh-controller": {Repository: "c"}, "mesh-host": {Repository: "h"},
"mesh-tools": {Repository: "t"}, "mesh-catalog": {Repository: "k"}, "mesh-sdk": {Repository: "s"}, "mesh-lab": {Repository: "l"}}
}
func member(repository string, number int, head, base string, open bool) memberLine {
return memberLine{ID: repository + "@" + head[:12], Repository: repository, State: "checked", Number: number, Head: head,
Base: base, Open: open}
}
func group(id string, members ...memberLine) groupsLine {
line := groupsLine{ID: id, Details: members}
for _, m := range members {
line.Members = append(line.Members, m.ID+" ("+m.State+")")
}
return line
}
func TestASiblingIsTheGroupsOtherOpenPullRequestIntoABaseOfTheSameName(t *testing.T) {
entries := theCoreOnTheGitSeat()
pull := link.PullUpdated{Owner: "novox", Repo: "mesh-host", Number: 114, Base: "main", Head: "feat/x", Commit: headHost}
for _, c := range []struct {
what string
lines []groupsLine
want []string
unknown string
}{
{"the other member, open, same base", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "main", true))},
[]string{"mesh-controller=novox/mesh-controller#346@" + headController + " onto main"}, ""},
{"a member into a base of another name is not one", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "release", true))},
nil, ""},
{"a member not open (merged or closed) is not one", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "main", false))},
nil, ""},
{"mesh-lab is never one", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-lab", 5, headLab, "main", true))},
nil, ""},
{"a repository the check reads nothing of is not one", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/hq", 9, headHq, "main", true))},
nil, ""},
{"mesh-sdk, read beside the controller, is one", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-sdk", 7, headSDK, "main", true))},
[]string{"mesh-sdk=novox/mesh-sdk#7@" + headSDK + " onto main"}, ""},
{"another group is none of its business", []groupsLine{group("feat/y",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "main", true))},
nil, ""},
{"a group of its own alone has none", []groupsLine{group("feat/x", member("novox/mesh-host", 114, headHost, "main", true))},
nil, ""},
{"an older mesh-delivery names no heads", []groupsLine{{ID: "feat/x",
Members: []string{"novox/mesh-host@111111111111 (checked)", "novox/mesh-controller@222222222222 (checked)"}}},
nil, "without their heads and bases"},
{"a member without its base", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "", true))},
nil, "without its head or base"},
{"a head that is no commit", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController[:12]+"zzzzzzzzzzzzzzzzzzzzzzzzzzzz", "main", true))},
nil, "not a whole commit"},
} {
got, unknown := siblingsFor(pull, entries, besideTheCore(), c.lines)
var said []string
for _, s := range got {
said = append(said, s.Dir+"="+s.Repository+"#"+itoa(s.Number)+"@"+s.Head+" onto "+s.Base)
}
if strings.Join(said, ";") != strings.Join(c.want, ";") {
t.Errorf("%s: siblings %v, wanted %v", c.what, said, c.want)
}
if (c.unknown == "") != (unknown == "") || !strings.Contains(unknown, c.unknown) {
t.Errorf("%s: could not learn %q, wanted %q", c.what, unknown, c.unknown)
}
}
// A pull request into a release branch has siblings into the release branch of the same name (ADR 0358 §1).
release := pull
release.Base = "feat/57"
lines := []groupsLine{group("feat/x", member("novox/mesh-host", 114, headHost, "feat/57", true),
member("novox/mesh-controller", 346, headController, "feat/57", true), member("novox/mesh-tools", 8, headLab, "main", true))}
if got, _ := siblingsFor(release, entries, besideTheCore(), lines); len(got) != 1 || got[0].Dir != "mesh-controller" {
t.Errorf("into a release branch: %+v", got)
}
// A recheck announces the pull request without its head branch: its group is the one holding it (repository, number).
pull.Head = ""
got, unknown := siblingsFor(pull, entries, besideTheCore(), lines)
if len(got) != 1 || got[0].Dir != "mesh-tools" || unknown != "" {
t.Errorf("a recheck of a group member: %+v %q", got, unknown)
}
// A pull request in no group, rechecked, is alone: no sibling, and nothing said.
pull.Number = 999
if got, unknown := siblingsFor(pull, entries, besideTheCore(), lines); got != nil || unknown != "" {
t.Errorf("a lone pull request: %+v %q", got, unknown)
}
if got, unknown := siblingsFor(pull, entries, besideTheCore(), nil); got != nil || unknown != "" {
t.Errorf("no groups at all: %+v %q", got, unknown)
}
// A mesh-delivery that names members without their pull requests cannot find the group of a recheck: it says so.
older := []groupsLine{{ID: "feat/x", Members: []string{"novox/mesh-host@111111111111 (checked)"}}}
pull.Number = 114
if got, unknown := siblingsFor(pull, entries, besideTheCore(), older); got != nil || unknown == "" {
t.Errorf("an older mesh-delivery: %+v %q", got, unknown)
}
}
func itoa(n int) string { b, _ := json.Marshal(n); return string(b) }
// Every check asks mesh-delivery's `groups` once; the ask names the sibling's repository, head and base.
func TestEveryCheckAsksMeshDeliveryOnceAndNamesItsSiblings(t *testing.T) {
entries := theCoreOnTheGitSeat()
asked := 0
was := groupsAsked
t.Cleanup(func() { groupsAsked = was })
groupsAsked = func(_ context.Context, held bool) ([]groupsLine, error) {
asked++
if !held {
t.Error("the seat is held, yet it was asked as not held")
}
return []groupsLine{group("feat/x", member("novox/mesh-host", 114, headHost, "main", true),
member("novox/mesh-controller", 346, headController, "main", true))}, nil
}
request := link.BuildRequest{Check: &link.CheckRequest{Owner: "novox", Repo: "mesh-host", Base: "main", Beside: besideTheCore()}}
withSiblings(t.Context(), &request, link.PullUpdated{Owner: "novox", Repo: "mesh-host", Base: "main", Head: "feat/x"}, entries)
if asked != 1 {
t.Fatalf("mesh-delivery was asked %d times for one check", asked)
}
want := link.CheckSibling{Repository: "novox/mesh-controller", Number: 346, Head: headController, Base: "main", Dir: "mesh-controller"}
if got := request.Check.Siblings; len(got) != 1 || got[0] != want || request.Check.SiblingsUnknown != "" {
t.Fatalf("the ask names siblings %+v, unknown %q", got, request.Check.SiblingsUnknown)
}
body, _ := json.Marshal(request.Check)
if !strings.Contains(string(body), `"siblings":[{"repository":"novox/mesh-controller","number":346,"head":"`+headController+
`","base":"main","dir":"mesh-controller"}]`) {
t.Fatalf("on the wire: %s", body)
}
}
// When mesh-delivery does not answer in time (or is not held, or answered without the fields), the check reads no
// sibling and its request says it could not learn them, so a fix to mesh-delivery is still checked without it.
func TestWhenMeshDeliveryDoesNotAnswerTheCheckReadsNoSiblingAndSaysSo(t *testing.T) {
was := groupsAsked
t.Cleanup(func() { groupsAsked = was })
for _, c := range []struct {
what string
err error
held bool
}{{"it did not answer", context.DeadlineExceeded, true}, {"nothing serves it", errors.New("nothing serves mesh-delivery.groups"), true},
{"no holder on record", errors.New("no holder of mesh-delivery is on record to say it"), false}} {
entries := theCoreOnTheGitSeat()
if !c.held {
entries = entries[:4]
}
groupsAsked = func(_ context.Context, held bool) ([]groupsLine, error) {
if held != c.held {
t.Errorf("%s: held %v", c.what, held)
}
return nil, c.err
}
request := link.BuildRequest{Check: &link.CheckRequest{Owner: "novox", Repo: "mesh-host", Base: "main", Beside: besideTheCore()}}
withSiblings(t.Context(), &request, link.PullUpdated{Owner: "novox", Repo: "mesh-host", Base: "main", Head: "feat/x"}, entries)
if len(request.Check.Siblings) != 0 || !strings.Contains(request.Check.SiblingsUnknown, c.err.Error()) ||
!strings.Contains(request.Check.SiblingsUnknown, "reads no siblings") {
t.Errorf("%s: siblings %+v, unknown %q", c.what, request.Check.Siblings, request.Check.SiblingsUnknown)
}
}
}
// **A pull request for which mesh-delivery answered and named no sibling is asked exactly as before** (ADR 0382 §7):
// byte for byte the request it was without the rule.
func TestAPullRequestWithNoSiblingIsAskedByteForByteAsBefore(t *testing.T) {
entries := theCoreOnTheGitSeat()
was := groupsAsked
t.Cleanup(func() { groupsAsked = was })
for _, answer := range [][]groupsLine{nil, {group("feat/other", member("novox/mesh-controller", 346, headController, "main", true))},
{group("feat/x", member("novox/mesh-host", 114, headHost, "main", true), member("novox/hq", 3, headHq, "main", true))}} {
groupsAsked = func(context.Context, bool) ([]groupsLine, error) { return answer, nil }
build := func() link.BuildRequest {
return link.BuildRequest{ID: "build-1", Repository: "r", Ref: headHost, Check: &link.CheckRequest{Owner: "novox",
Repo: "mesh-host", Number: 114, Base: "main", Beside: besideTheCore(), Modules: []string{"mesh-host"}, Judge: "validator"}}
}
before, _ := json.Marshal(build())
after := build()
withSiblings(t.Context(), &after, link.PullUpdated{Owner: "novox", Repo: "mesh-host", Base: "main", Head: "feat/x"}, entries)
got, _ := json.Marshal(after)
if string(before) != string(got) {
t.Fatalf("the request changed:\nbefore %s\nafter %s", before, got)
}
}
}
// What the build seat's verdict says of its siblings reaches the forge's holder in the controller's `checked`.
func TestTheVerdictCarriesTheSiblingsAndWhereTheJudgeCameFrom(t *testing.T) {
read := []link.CheckSibling{{Repository: "novox/mesh-controller", Number: 346, Head: headController, Base: "main", BaseCommit: headHost}}
c := checkedOf(link.BuildResult{ID: "b", Ref: headHost, Checked: &link.CheckRequest{Owner: "novox", Repo: "mesh-host", Number: 114},
Check: &link.CheckOutcome{Verdict: "pass", Summary: "ok", Siblings: read, JudgeFrom: "sibling novox/mesh-controller#346 merge abc",
Validator: "the change's own validator"}})
if len(c.Siblings) != 1 || c.Siblings[0] != read[0] || c.JudgeFrom == "" || c.Validator == "" {
t.Fatalf("the verdict says %+v", c)
}
body, _ := json.Marshal(c)
for _, want := range []string{`"siblings":[{"repository":"novox/mesh-controller","number":346,"head":"` + headController +
`","base":"main","base_commit":"` + headHost + `"}]`, `"judge_from":"sibling novox/mesh-controller#346 merge abc"`} {
if !strings.Contains(string(body), want) {
t.Fatalf("the verdict lacks %s:\n%s", want, body)
}
}
unknown := checkedOf(link.BuildResult{ID: "b", Check: &link.CheckOutcome{Verdict: "pass", SiblingsUnknown: "mesh-delivery did not answer"}})
if body, _ := json.Marshal(unknown); !strings.Contains(string(body), `"siblings_unknown":"mesh-delivery did not answer"`) ||
strings.Contains(string(body), `"siblings":`) {
t.Fatalf("the verdict that could not learn them says %s", body)
}
}
+58 -105
View File
@@ -4,7 +4,6 @@ import (
"context"
"encoding/json"
"fmt"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"io"
"os"
"path"
@@ -162,77 +161,56 @@ func (f following) pullUpdated(ctx context.Context, p link.PullUpdated, w io.Wri
return "", err
}
scope := pullScope(p, entries, read, edges)
// **The merge gate's checks stop** (novox/hq ADR 0503 §3, §5; issue 1299): the checker judges every repository whose
// trunk requires `mesh/check`. A repository whose trunk requires `mesh/repo-check` (hq), whatever else it requires
// (ADR 0513 §4), has its own merge-check.sh asked of build-agent, and said with that layer alone; any other is asked
// nothing and said nothing. What the forge's protection requires is read, never listed (lead decision L11).
// Mesh-config is still told what the head declares (ADR 0502), whatever judges it.
this := pullAtOf(p)
if this.Truncated {
this.Paths, this.Truncated = truncatedPaths(entries, p.Owner, p.Repo, scope.Modules), false
}
if scope.gated() || holdsModules(entries, p.Owner, p.Repo) {
tellPullDeclares(ctx, w, []pullAt{this})
}
required, err := requiredStatusesOf(ctx, p.Owner, p.Repo, p.Base)
switch {
case err != nil:
fmt.Fprintf(w, "%s/%s#%d (%.8s): what %s's protection requires cannot be read, so nothing is asked and nothing is "+
"said: %v\n", p.Owner, p.Repo, p.Number, p.Commit, p.Base, err)
return "", nil
case !asksRepoCheck(required):
fmt.Fprintf(w, "%s/%s#%d (%.8s): %s does not require %s, and %s; nothing is asked\n", p.Owner, p.Repo, p.Number,
p.Commit, p.Base, sdkdelivery.StatusContextRepoCheck, checkerWords(required))
// **A cycle of edges fails the check, naming every module and edge** (novox/hq ADR 0419 clause 6), a lone pull
// request's as a group's composed check is: no order exists for what it changes.
alone := orderMember{ID: p.Owner + "/" + p.Repo + "@" + twelve(p.Commit), Repository: p.Owner + "/" + p.Repo, Base: p.Base,
Head: p.Commit, Number: p.Number, Paths: p.Paths, PathsTruncated: p.PathsTruncated, Removed: p.Removed, BaseSHA: p.BaseSHA}
if failing, cyclic, err := cycleVerdict(ctx, "", []orderMember{alone},
map[string]orderReach{alone.ID: {Moved: scope.Modules}}, edges, entries); err != nil {
return "", err
} else if cyclic {
failing.Commit, failing.Trunk = p.Commit, twelve(p.BaseSHA)
fmt.Fprintf(w, "%s/%s#%d (%.8s): %s\n", p.Owner, p.Repo, p.Number, p.Commit, failing.Summary)
sayChecked(ctx, w, failing)
return "", nil
}
return f.repoCheckAlone(ctx, w, p, scope, entries)
}
// asksRepoCheck says what a trunk's protection has the controller ask: its repository's own merge-check.sh whenever the
// trunk requires `mesh/repo-check`, whatever else it requires (ADR 0513 §4: a trunk requiring both has the checker judge
// `mesh/check` and the repository check asked beside it), and nothing otherwise.
func asksRepoCheck(required []string) bool {
return slices.Contains(required, sdkdelivery.StatusContextRepoCheck)
}
// checkerWords says who judges a trunk that requires no `mesh/repo-check`.
func checkerWords(required []string) string {
if slices.Contains(required, sdkdelivery.StatusContextCheck) {
return "the checker judges its " + sdkdelivery.StatusContextCheck
}
return "does not require " + sdkdelivery.StatusContextCheck + " either, so nothing is said"
}
// repoCheckAlone asks build-agent for a repository's own merge-check.sh alone, for a pull request of a repository whose
// trunk requires `mesh/repo-check`, whether or not it also requires `mesh/check` (ADR 0503 §5, ADR 0513 §4): no gate, no module built, its verdict said with the
// repo-check layer alone.
func (f following) repoCheckAlone(ctx context.Context, w io.Writer, p link.PullUpdated, scope checkScope,
entries []inventory.Entry) (string, error) {
inv := f.open.inventory
// **A member of a group whose gate is the composed check's runs its repository's check alone** (novox/hq ADR 0403
// clause 1): mesh-delivery is asked once, and says so on the group's line. Told nothing, the pull request is
// checked alone, gate included (clause 6).
lines, groupsErr := groupsAsked(ctx, deliverySeatHeld(entries))
composed := groupsErr == nil && composedMember(p, lines)
// The change plan of the commit at hand (ADR 0238): what a merge of it would build and send, posted
// with the verdict whatever the verdict is.
plan := changePlanOf(p.Owner+"/"+p.Repo, p.Base, p.Commit, scope.Reach, entries, func(module string) (inventory.Upgrade, bool) {
u, err := inv.UpgradeOf(ctx, module)
return u, err == nil
})
fmt.Fprint(w, planText(plan))
direct := directVerdict(p, plan)
if composed {
scope.Modules, scope.Dependents, scope.New, scope.Manifests = nil, nil, nil, nil
composedGate(&direct)
}
switch {
case !scope.Mesh:
direct.RepoCheck = &link.CheckLayer{Verdict: "warning", Summary: "the repository is not the mesh's: nothing run"}
fmt.Fprintf(w, "%s/%s#%d (%.8s): the repository is not the mesh's: said, nothing run\n", p.Owner, p.Repo, p.Number,
p.Commit)
case !scope.gated() && !scope.Mesh:
fmt.Fprintf(w, "%s/%s#%d (%.8s): %s, and the repository is not the mesh's: said, nothing run\n",
p.Owner, p.Repo, p.Number, p.Commit, noModuleTouched)
sayChecked(ctx, w, direct)
return "", nil
case p.MergeCheckSaid && !p.MergeCheck:
case !scope.gated() && p.MergeCheckSaid && !p.MergeCheck:
direct.RepoCheck = &link.CheckLayer{Verdict: "warning", Summary: noMergeCheck}
fmt.Fprintf(w, "%s/%s#%d (%.8s): %s\n", p.Owner, p.Repo, p.Number, p.Commit, noMergeCheck)
fmt.Fprintf(w, "%s/%s#%d (%.8s): %s; %s\n", p.Owner, p.Repo, p.Number, p.Commit, noModuleTouched, noMergeCheck)
sayChecked(ctx, w, direct)
return "", nil
}
request, err := checkRequestFor(ctx, f.open, p, checkScope{Mesh: true, From: scope.From}, entries)
request, err := checkRequestFor(ctx, f.open, p, scope, entries)
if err != nil {
return "", err
}
request.Check.Plan = &plan
request.Check.Composed = composed
// **The group's other open pull requests are read beside it** (novox/hq ADR 0382 §1).
siblingsFrom(&request, p, entries, lines, groupsErr)
seat := buildSeatHeld(ctx)
ask, err := askOverOn(seat)
if err != nil {
@@ -242,42 +220,23 @@ func (f following) repoCheckAlone(ctx context.Context, w io.Writer, p link.PullU
if err := ask.Ask(ctx, request); err != nil {
return "", err
}
fmt.Fprintf(w, "%s/%s#%d (%.8s): asked %s to run its own merge-check.sh alone, as %s\n", p.Owner, p.Repo, p.Number,
p.Commit, seat, request.ID)
// **A newer request replaces the one still waiting, and a newer key stops the one running** (novox/hq ADR 0406 §3, §4,
// issue 974): the check queue keeps one request a pull request, and a running check watches its key. Nothing is asked
// of the forge here (issue 726).
what := "its own merge-check.sh alone: " + noModuleTouched
if composed {
what = "its own merge-check.sh alone: the gate is its group's composed check's"
}
if scope.gated() {
what = fmt.Sprintf("the gate over %s (a merge would build %d module(s) in %d tier(s))",
strings.Join(append(append([]string{}, scope.Modules...), prefixedAll("new:", scope.New)...), ", "),
scope.Width, scope.Tiers)
}
fmt.Fprintf(w, "%s/%s#%d (%.8s): asked %s to check it before it merges — %s — as %s\n", p.Owner, p.Repo, p.Number,
p.Commit, seat, what, request.ID)
return request.ID, nil
}
// requiredStatusesOf is what the forge's protection of a repository's branch requires before a merge, as the forge
// picks the rule it applies (the one named for the branch, else the first glob covering it). A variable so a test needs
// no forge.
var requiredStatusesOf = func(ctx context.Context, owner, repo, branch string) ([]string, error) {
var rules struct {
Rules []protectionRule `json:"rules"`
}
js, err := aBus()
if err != nil {
return nil, err
}
defer js.Close()
raw, _ := json.Marshal(map[string]any{"owner": owner, "repo": repo})
answer, err := link.Ask(ctx, link.OverNATS{Conn: js.Conn()}, "gitea", "gitea_branch_protection_get", raw, 30*time.Second)
if err != nil {
return nil, err
}
if answer.Error != "" {
return nil, fmt.Errorf("gitea.gitea_branch_protection_get: %s", answer.Error)
}
if err := json.Unmarshal(answer.Result, &rules); err != nil {
return nil, err
}
r := appliedRule(rules.Rules, branch)
if r == nil {
// empty-on-error: not an error; no rule covers the branch, so it requires nothing, which the caller says
return nil, nil
}
return r.RequiredStatuses, nil
}
// composedWords is what a group member's own verdict says of its gate layer (novox/hq ADR 0403 clause 1).
const composedWords = "judged by its group's composed check"
@@ -406,6 +365,15 @@ func trunkOf(s inventory.Source) string {
return "main"
}
// sourceName is the name of a recorded repository, a path on the git seat or a URL: novox/mesh-host → mesh-host.
func sourceName(repository string) string {
parts := strings.Split(strings.Trim(strings.TrimSuffix(repository, ".git"), "/"), "/")
if len(parts) < 2 {
return ""
}
return parts[len(parts)-1]
}
// namedForTheCheck are the builds earlier checks of the same pull request at the same head named, read once for the
// check (novox/hq issue 703): asked again because only the trunk moved, the check keeps what they built of a key
// still the same. Another pull request's build, or one of an earlier head, is never kept. None when they cannot be
@@ -626,13 +594,11 @@ func checkedOf(result link.BuildResult) link.Checked {
// A delivery group's composed check (novox/hq ADR 0239): every head it judged, this one first.
if g := result.Checked; g.Group != "" {
c.Group = g.Group
// Every member is named with its trunk commit (novox/hq issue 1106): the seat's word, else the one the forge
// announced with the head, so the forge's merge tool never meets a member whose verdict names none.
c.Members = append(c.Members, link.CheckedMember{Owner: g.Owner, Repo: g.Repo, Number: g.Number,
Commit: result.Ref, Trunk: firstTrunk(c.Trunk, g.Trunk)})
Commit: result.Ref, Trunk: c.Trunk})
for _, m := range g.Members {
c.Members = append(c.Members, link.CheckedMember{Owner: m.Owner, Repo: m.Repo, Number: m.Number,
Commit: m.Ref, Trunk: firstTrunk(memberTrunk(result.Check, m), m.Trunk)})
Commit: m.Ref, Trunk: memberTrunk(result.Check, m)})
}
}
}
@@ -661,16 +627,6 @@ func checkedOf(result link.BuildResult) link.Checked {
return c
}
// firstTrunk is the first trunk commit named, as twelve characters; empty when none is.
func firstTrunk(named ...string) string {
for _, n := range named {
if n != "" {
return twelve(n)
}
}
return ""
}
// memberTrunk is the trunk commit the build seat merged one of a group's heads into (novox/hq ADR 0323 §6), empty
// when it said none.
func memberTrunk(outcome *link.CheckOutcome, m link.GroupMember) string {
@@ -695,9 +651,6 @@ func prefixedAll(prefix string, items []string) []string {
// sayChecked says a merge check's verdict on the bus, where the forge's holder hears it.
func sayChecked(ctx context.Context, w io.Writer, c link.Checked) {
// **No merge-gate layer from R4e, for any repository** (ADR 0503 §3, §5; issue 1299): the forge's holder writes
// `mesh/repo-check` from the repo-check layer alone, and the checker owes `mesh/check`.
c.Gate = nil
repo := "none"
if c.RepoCheck != nil {
repo = strings.ToUpper(c.RepoCheck.Verdict) + " — " + c.RepoCheck.Summary
@@ -706,8 +659,8 @@ func sayChecked(ctx context.Context, w io.Writer, c link.Checked) {
if c.Trunk != "" {
judged = " merged into trunk " + c.Trunk
}
fmt.Fprintf(w, "%s: %s/%s#%d at %.8s%s checked on %s: repository %s\n", c.ID, c.Owner, c.Repo, c.Number,
c.Commit, judged, orSomewhere(c.On), repo)
fmt.Fprintf(w, "%s: %s/%s#%d at %.8s%s checked on %s: gate %s — %s; repository %s\n", c.ID, c.Owner, c.Repo, c.Number,
c.Commit, judged, orSomewhere(c.On), strings.ToUpper(c.Verdict), c.Summary, repo)
if checkEvents == nil {
return
}
+97
View File
@@ -0,0 +1,97 @@
//go:build integration
package main
import (
"context"
"encoding/json"
"fmt"
"testing"
"github.com/nats-io/nats.go"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// Acceptance 5 of milestone 2 (novox/hq ADR 0406 §3, issue 974), live, with the store and a real bus: two merges into one
// repository within a minute, with two open pull requests, leave each pull request with exactly one queued check, by the
// stream's rule. Each merge moves the trunk, and mesh-delivery asks the controller to check both open pull requests again
// against it (`delivery check`, which is the controller's own path for a head announced); nothing moves a request, nothing
// asks the forge, and the queue is read back whole.
func TestTwoMergesWithinAMinuteLeaveEachOpenPullRequestOneQueuedCheck(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
js, _ := aBuildQueue(t)
servingBus.Store(js)
t.Cleanup(func() { servingBus.Store(nil) })
holdTheDeliverySeat(t, open)
was := askDeliveryOwner
t.Cleanup(func() { askDeliveryOwner = was })
askDeliveryOwner = func(_ context.Context, _ *nats.Conn, verb string, _ map[string]any) (json.RawMessage, error) {
if verb != "groups" {
t.Fatalf("asked the owner %s", verb)
}
return json.RawMessage(`[]`), nil
}
// The repository the two pull requests are into, as a module the mesh builds from it.
if err := open.inventory.RegisterModule(ctx, catalogue.Manifest{Module: "mesh-controller", Version: "1"},
inventory.Source{Repository: "novox/mesh-controller", Ref: "main", BuiltFrom: "c0", Head: "c0"}); err != nil {
t.Fatal(err)
}
heads := map[int]string{7: "7777777777777777777777777777777777777777", 8: "8888888888888888888888888888888888888888"}
check := func(number int, trunk string) {
t.Helper()
p := orderMember{ID: fmt.Sprintf("novox/mesh-controller@%.12s", heads[number]), Repository: "novox/mesh-controller",
Base: "main", Head: heads[number], Number: number, Paths: []string{"main.go"}, ModuleDirsSaid: true,
BaseSHA: trunk, Ready: number == 8}.pull()
p.MergeCheck, p.MergeCheckSaid = true, true
if err := (following{open: open, oneShot: true}).PullUpdated(ctx, p); err != nil {
t.Fatal(err)
}
}
// Announced against the trunk as it stands, then asked again after each of two merges.
for _, trunk := range []string{"1111111111111111111111111111111111111111", "2222222222222222222222222222222222222222",
"3333333333333333333333333333333333333333"} {
check(7, trunk)
check(8, trunk)
}
q, err := link.ReadQueue(ctx, js, link.TheBuildMachine)
if err != nil {
t.Fatal(err)
}
seen := map[int]string{}
for _, a := range q.Asks {
c := a.Request.Check
if c == nil {
t.Fatalf("%s is no check", a.ID)
}
if _, twice := seen[c.Number]; twice {
t.Errorf("pull request %d has two queued checks", c.Number)
}
seen[c.Number] = c.Trunk
if a.State != link.AskWaiting {
t.Errorf("%s is %s", a.ID, a.State)
}
if want := "3333333333333333333333333333333333333333"; c.Trunk != want {
t.Errorf("pull request %d is queued against %.8s; want the newest trunk %.8s", c.Number, c.Trunk, want)
}
}
if len(seen) != 2 {
t.Fatalf("queued checks for %v; want one for each of the two open pull requests", seen)
}
// By the streams' own count: the ready one on its stream, the other on its.
for stream, want := range map[string]uint64{broker.ChecksReadyStream: 1, broker.ChecksOtherStream: 1} {
info, err := js.Context().StreamInfo(stream)
if err != nil {
t.Fatal(err)
}
if info.State.Msgs != want {
t.Errorf("%s holds %d request(s); want %d", stream, info.State.Msgs, want)
}
}
}
@@ -1,109 +0,0 @@
//go:build integration
package main
import (
"context"
"encoding/json"
"errors"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// C1 (ADR 0503 §3, §5; issue 1299, lead decision L11), with the store and a real build queue: a pull request of a
// repository whose trunk requires mesh/check asks build-agent nothing and says nothing; one whose trunk requires
// mesh/repo-check alone (hq) has its own merge-check.sh asked, and every verdict said names no gate layer; a protection
// that cannot be read asks and says nothing.
func TestThePullRequestChecksAreAskedOnlyForARepositoryTheCheckerDoesNotJudge(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
js, _ := aBuildQueue(t)
servingBus.Store(js)
t.Cleanup(func() { servingBus.Store(nil) })
said := &saidOnTheBus{}
checkEvents = said
t.Cleanup(func() { checkEvents = nil })
if err := open.inventory.RegisterModule(ctx, catalogue.Manifest{Module: "mesh-controller", Version: "1"},
inventory.Source{Repository: "novox/mesh-controller", Ref: "main", BuiltFrom: "c0", Head: "c0"}); err != nil {
t.Fatal(err)
}
// hq is cloned as a module built from it is, from its URL: no forge is held in this test.
if err := open.inventory.RegisterModule(ctx, catalogue.Manifest{Module: "hq-docs", Version: "1"},
inventory.Source{Repository: "http://forge.invalid/novox/hq.git", Ref: "main", Path: "docs", BuiltFrom: "c0",
Head: "c0"}); err != nil {
t.Fatal(err)
}
required := map[string][]string{"novox/mesh-controller": {"mesh/check"}, "novox/hq": {"mesh/repo-check"}}
was := requiredStatusesOf
requiredStatusesOf = func(_ context.Context, owner, repo, _ string) ([]string, error) {
if r, ok := required[owner+"/"+repo]; ok {
return r, nil
}
return nil, errors.New("the forge did not answer")
}
t.Cleanup(func() { requiredStatusesOf = was })
queued := func() int {
q, err := link.ReadQueue(ctx, js, link.TheBuildMachine)
if err != nil {
t.Fatal(err)
}
return len(q.Asks)
}
pull := func(repo string, number int, mergeCheck bool) link.PullUpdated {
p := link.PullUpdated{Owner: "novox", Repo: repo, Number: number, Base: "main", Commit: "abcdefabcdefabcdefabcdefabcdefabcdefabcd",
BaseSHA: "1111111111111111111111111111111111111111", Paths: []string{"main.go"}, ModuleDirsSaid: true}
p.MergeCheck, p.MergeCheckSaid = mergeCheck, true
return p
}
f := following{open: open, oneShot: true}
if err := f.PullUpdated(ctx, pull("mesh-controller", 7, true)); err != nil {
t.Fatal(err)
}
if n := queued(); n != 0 || len(said.events) != 0 {
t.Fatalf("0503 §3: a repository the checker judges asked %d check(s) and said %v", n, said.events)
}
if err := f.PullUpdated(ctx, pull("unread", 3, true)); err != nil {
t.Fatal(err)
}
if n := queued(); n != 0 || len(said.events) != 0 {
t.Fatalf("L11: a protection that cannot be read asked %d check(s) and said %v", n, said.events)
}
if err := f.PullUpdated(ctx, pull("hq", 9, true)); err != nil {
t.Fatal(err)
}
q, err := link.ReadQueue(ctx, js, link.TheBuildMachine)
if err != nil {
t.Fatal(err)
}
if len(q.Asks) != 1 || q.Asks[0].Request.Check == nil || len(q.Asks[0].Request.Check.Modules) != 0 ||
q.Asks[0].Request.Check.Repo != "hq" {
t.Fatalf("0503 §5: hq's repository check is not the one ask: %+v", q.Asks)
}
// A head with no merge-check.sh is said at once, with the repo-check layer alone.
if err := f.PullUpdated(ctx, pull("hq", 10, false)); err != nil {
t.Fatal(err)
}
if len(said.events) != 1 {
t.Fatalf("hq's head with no merge-check.sh was said %d time(s)", len(said.events))
}
var c link.Checked
if err := json.Unmarshal(said.bodies[0], &c); err != nil {
t.Fatal(err)
}
if c.Gate != nil || c.RepoCheck == nil || c.RepoCheck.Verdict != "warning" {
t.Fatalf("0503 §5: the verdict said names a gate layer, or no repo-check layer: %+v", c)
}
// A trunk requiring both (hq, ADR 0513 §4): the checker judges mesh/check, and the repository check is still asked.
required["novox/hq"] = []string{"mesh/check", "mesh/repo-check"}
if err := f.PullUpdated(ctx, pull("hq", 11, true)); err != nil {
t.Fatal(err)
}
if q, err := link.ReadQueue(ctx, js, link.TheBuildMachine); err != nil || len(q.Asks) != 2 ||
q.Asks[1].Request.Check == nil || q.Asks[1].Request.Check.Repo != "hq" {
t.Fatalf("0513 §4: a trunk requiring both had no repository check asked: %+v %v", q.Asks, err)
}
}
-43
View File
@@ -250,31 +250,6 @@ func TestAVerdictNamesTheTrunkCommitItJudged(t *testing.T) {
}
}
// **A group's composed verdict names every member's trunk commit** (novox/hq issue 1106): a member the build seat named no
// trunk for (an SDK member whose base the seat did not merge, or a seat from before the rule) is named by the trunk the forge
// announced with its head, and the primary the same, so the merge tool never reads "names no trunk commit" on a member
// whose trunk did not move.
func TestAGroupVerdictNamesEveryMembersTrunkTheSDKMemberToo(t *testing.T) {
group := &link.CheckRequest{Owner: "novox", Repo: "mesh-catalog", Number: 7, Base: "main", Group: "feat/x",
Trunk: "dddddddddddd0000",
Members: []link.GroupMember{
{Owner: "novox", Repo: "mesh-sdk", Number: 8, Ref: "cccccccccccc0000", Base: "main", Trunk: "eeeeeeeeeeee0000"},
{Owner: "novox", Repo: "mesh-host", Number: 9, Ref: "bbbbbbbbbbbb0000", Base: "main", Trunk: "ffffffffffff0000"}}}
c := checkedOf(link.BuildResult{ID: "g", Ref: "aaaaaaaaaaaa0000", Checked: group, Check: &link.CheckOutcome{
Verdict: "pass", Head: "aaaaaaaaaaaa", Gate: &link.CheckLayer{Verdict: "pass"},
Members: []link.CheckedMember{{Owner: "novox", Repo: "mesh-host", Commit: "bbbbbbbbbbbb", Trunk: "222222222222"}}}})
want := []string{"dddddddddddd", "eeeeeeeeeeee", "222222222222"}
if len(c.Members) != 3 {
t.Fatalf("the composed check names its members %+v", c.Members)
}
for i, m := range c.Members {
if m.Trunk != want[i] {
t.Fatalf("member %d (%s/%s) is named with trunk %q, want %q (the seat's word first, else the announced one): %+v",
i, m.Owner, m.Repo, m.Trunk, want[i], c.Members)
}
}
}
// **The merge made for a check is never registered** (novox/hq ADR 0323 §2): a build of it — a commit on no branch
// of the forge — is refused, naming the trunk. The rule of ADR 0238 §1 stands unchanged; this holds it for the
// merge commit a check makes.
@@ -316,21 +291,3 @@ func TestARecheckedHeadsDirectVerdictNamesItsTrunk(t *testing.T) {
// A check stopped by its pull request's key moving on says no verdict (issue 974, review 2): an error said for it would name the
// live head, and the newer request is the one that judges. A check that failed for any other reason still says its error.
// C2 (ADR 0513 §4, ADR 0503 §5; issue 1299): a trunk requiring mesh/repo-check has its repository check asked, whether or
// not it also requires mesh/check (hq requires both); a trunk requiring mesh/check alone, or neither, is asked nothing.
func TestARepositoryCheckIsAskedWheneverTheTrunkRequiresIt(t *testing.T) {
for _, c := range []struct {
required []string
asks bool
}{
{[]string{"mesh/check", "mesh/repo-check"}, true},
{[]string{"mesh/repo-check"}, true},
{[]string{"mesh/check"}, false},
{nil, false},
} {
if got := asksRepoCheck(c.required); got != c.asks {
t.Fatalf("a trunk requiring %v asks its repository check: %v, want %v", c.required, got, c.asks)
}
}
}
-6
View File
@@ -233,12 +233,6 @@ func prepareSend(ctx context.Context, open *stores, node string) ([]string, erro
if !slices.Contains(in.BusMinted, "provisioner."+node) {
continue
}
default:
// A bus process's account nobody issued still gets the stand-in, until the composition leaves
// the module off (issue 1753): the warning names the verb (issue 1226).
if line := processAccountWarning(m, node, name, in.BusMinted); line != "" {
sayWarning(ctx, line)
}
}
if _, err := inv.SecretForModule(ctx, node, m.Module, name); err == nil {
did = true
+2 -8
View File
@@ -200,7 +200,6 @@ func composeQuestion(ctx context.Context, open *stores, node string, mode compos
}
sayFound(ctx, d)
out := sendableOf(d)
out.fileReferences = readsFileReferences(in.Input, node)
if !withOrder {
out.Sequence, out.Epoch, out.Generation = 0, 0, 0
}
@@ -239,11 +238,7 @@ func markUnresolvable(err error) error {
// still names anything foreseen is refused, said, and never sent.
func composeSend(ctx context.Context, open *stores, node string) (sendable, error) {
declared, _, err := composeSendValidated(ctx, open, node)
if err != nil {
return declared, err
}
// Never a declaration naming an image the store does not hold (novox/hq issue 1742): that module is left out.
return withoutMissingImages(ctx, open, node, declared), nil
return declared, err
}
// composeSendValidated is composeSend and what the node-engine's validator refuses of the declaration it composed, for the
@@ -286,7 +281,6 @@ func composeSendValidated(ctx context.Context, open *stores, node string) (senda
if len(left) == 0 && len(in.wouldMake) == 0 {
sayFound(ctx, d)
out := sendableOf(d)
out.fileReferences = readsFileReferences(in.Input, node)
// The sender stamps the order it allots (order.stamp); a composition for a send carries none of its own.
out.Sequence, out.Epoch, out.Generation = 0, 0, 0
here, _ := nodeOf(in.Input, node)
@@ -372,7 +366,7 @@ func sayFound(ctx context.Context, d compose.Declared) {
// sendableOf is a composed declaration as the controller's sends know it.
func sendableOf(d compose.Declared) sendable {
out := sendable{Resources: compactKept(d.Resources), Sequence: d.Sequence, Epoch: d.Epoch, Generation: d.Generation,
out := sendable{Resources: d.Resources, Sequence: d.Sequence, Epoch: d.Epoch, Generation: d.Generation,
ReadsModules: d.ReadsModules, moduleOf: d.ModuleOf, buildIDs: d.BuildIDs, modules: d.Modules,
localNetwork: d.LocalNetwork, Received: d.Received, Mesh: d.Mesh, BusUsers: d.BusUsers,
LeftOut: d.LeftOut, leftOutWhy: d.LeftOutWhy, withheld: d.Withheld, unbound: d.Unbound, foreseen: d.Foreseen,
@@ -0,0 +1,68 @@
//go:build integration
package main
import (
"context"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// A member of a group whose gate is the composed check's is asked of the build seat with no module: its repository's
// check alone (novox/hq ADR 0403 clause 1). With no word from mesh-delivery, or one that does not say the gate is
// composed, it is checked alone with its gate (clause 6).
func askedForMember(t *testing.T, answer []groupsLine, answerErr error) link.CheckRequest {
t.Helper()
open := aMesh(t)
ctx := t.Context()
js, _ := aBuildQueue(t)
servingBus.Store(js)
t.Cleanup(func() { servingBus.Store(nil) })
if err := open.inventory.RegisterModule(ctx, catalogue.Manifest{Module: "mesh-controller", Version: "1"},
inventory.Source{Repository: "novox/mesh-controller", Ref: "main", BuiltFrom: "c0", Head: "c0"}); err != nil {
t.Fatal(err)
}
was := groupsAsked
groupsAsked = func(context.Context, bool) ([]groupsLine, error) { return answer, answerErr }
t.Cleanup(func() { groupsAsked = was })
p := link.PullUpdated{Owner: "novox", Repo: "mesh-controller", Number: 346, Base: "main", Head: "feat/x", Commit: headController,
Paths: []string{"main.go"}, ModuleDirsSaid: true, MergeCheck: true, MergeCheckSaid: true}
if err := (following{open: open, oneShot: true}).PullUpdated(ctx, p); err != nil {
t.Fatal(err)
}
q, err := link.ReadQueue(ctx, js, link.TheBuildMachine)
if err != nil || len(q.Asks) != 1 {
t.Fatalf("asks %d: %v", len(q.Asks), err)
}
return *q.Asks[0].Request.Check
}
func TestAComposedMembersCheckIsItsRepositoryChecksAlone(t *testing.T) {
line := group("feat/x", member("novox/mesh-controller", 346, headController, "main", true), member("novox/mesh-catalog", 350, headHost, "main", true))
line.ComposedGate = true
c := askedForMember(t, []groupsLine{line}, nil)
if len(c.Modules) != 0 || len(c.New) != 0 || len(c.Dependents) != 0 || !c.Composed {
t.Fatalf("the member's check is asked as %+v; want no module and composed", c)
}
}
func TestAMemberOfAGroupOfAnOlderMeshDeliveryKeepsItsOwnGate(t *testing.T) {
line := group("feat/x", member("novox/mesh-controller", 346, headController, "main", true), member("novox/mesh-catalog", 350, headHost, "main", true))
c := askedForMember(t, []groupsLine{line}, nil)
if !strings.Contains(strings.Join(c.Modules, ","), "mesh-controller") || c.Composed {
t.Fatalf("an older mesh-delivery's member is asked as %+v; want its own gate", c)
}
}
func TestAMemberWhoseGroupCannotBeLearnedIsCheckedAloneWithItsGate(t *testing.T) {
c := askedForMember(t, nil, context.DeadlineExceeded)
if !strings.Contains(strings.Join(c.Modules, ","), "mesh-controller") || c.Composed ||
!strings.Contains(c.SiblingsUnknown, "reads no siblings") {
t.Fatalf("the member is asked as %+v; want its gate and why its siblings are unknown", c)
}
}
+80 -5
View File
@@ -1,21 +1,50 @@
package main
import (
"context"
"encoding/json"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// A delivery group's gate is the composed check's, and its members' own checks run their repository's check alone
// (novox/hq ADR 0403 clauses 1 and 7). Pure: no store, no bus.
// The heads of a group's members, by repository.
const (
headHost = "1111111111111111111111111111111111111111"
headController = "2222222222222222222222222222222222222222"
)
// A member of a group whose line says composed_gate is a composed member; one whose group says nothing of it (an older
// mesh-delivery), one of a group of a single pull request and a lone pull request are not.
func TestAMemberOfAComposedGroupIsKnownByTheGroupsLine(t *testing.T) {
two := func(composed bool) groupsLine {
l := group("feat/x", member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "main", true))
l.ComposedGate = composed
return l
}
p := link.PullUpdated{Owner: "novox", Repo: "mesh-host", Number: 114, Base: "main", Head: "feat/x", Commit: headHost}
recheck := link.PullUpdated{Owner: "novox", Repo: "mesh-host", Number: 114, Base: "main", Commit: headHost}
for name, c := range map[string]struct {
p link.PullUpdated
lines []groupsLine
want bool
}{
"composed": {p, []groupsLine{two(true)}, true},
"a recheck, by its number": {recheck, []groupsLine{two(true)}, true},
"an older mesh-delivery": {p, []groupsLine{two(false)}, false},
"another group": {p, []groupsLine{group("feat/y", member("novox/mesh-host", 1, headHost, "main", true))}, false},
"no group": {p, nil, false},
} {
if got := composedMember(c.p, c.lines); got != c.want {
t.Errorf("%s: composed %v, want %v", name, got, c.want)
}
}
alone := group("feat/x", member("novox/mesh-host", 114, headHost, "main", true))
alone.ComposedGate = true
if composedMember(p, []groupsLine{alone}) {
t.Error("a group of a single pull request is composed")
}
}
// A composed member's verdict never says its gate passed: the gate layer, and the verdict that is the gate's, read
// "composed"; a failure or an error stays what it was.
@@ -59,3 +88,49 @@ func TestTheComposedVerdictCarriesEveryMember(t *testing.T) {
}
}
}
// Members whose manifests order their modules in a cycle fail the composed check, naming every module and edge, and no
// ask is made of the build seat.
func TestAGroupInACycleFailsItsComposedCheckWithoutABuild(t *testing.T) {
members := []orderMember{
{ID: "novox/x@" + headHost[:12], Repository: "novox/x", Base: "main", Head: headHost, Number: 7, Paths: []string{"module.json"}},
{ID: "novox/y@" + headController[:12], Repository: "novox/y", Base: "main", Head: headController, Number: 9, Paths: []string{"module.json"}},
}
reach := map[string]orderReach{members[0].ID: {Moved: []string{"mx"}}, members[1].ID: {Moved: []string{"my"}}}
was, wasParent := manifestAt, parentOf
t.Cleanup(func() { manifestAt, parentOf = was, wasParent })
parentOf = func(context.Context, string, string, string) (string, error) { return "", nil }
manifestAt = func(_ context.Context, _, repo, _, ref string) (catalogue.Manifest, bool, error) {
if ref == "main" {
return catalogue.Manifest{}, false, nil // what the pull requests merge into holds no such manifest
}
if repo == "x" {
return catalogue.Manifest{Module: "mx", DefinesSeats: []catalogue.SeatDeclaration{{Name: "s", Serves: []catalogue.Verb{{Name: "a"}}}},
Invokes: []string{"seat:s.b"}}, true, nil
}
return catalogue.Manifest{Module: "my", DefinesSeats: []catalogue.SeatDeclaration{{Name: "s", Serves: []catalogue.Verb{{Name: "b"}}}},
Invokes: []string{"seat:s.a"}}, true, nil
}
c, cyclic, err := cycleVerdict(context.Background(), "feat/cycle", members, reach, []inventory.Edge(nil), nil)
if err != nil || !cyclic {
t.Fatalf("no cycle found: %v", err)
}
if c.Verdict != "fail" || c.Group != "feat/cycle" || len(c.Members) != 2 || c.Gate == nil || c.Gate.Verdict != "fail" {
t.Fatalf("the verdict is %+v", c)
}
for _, w := range []string{"mx", "my", "s.a", "s.b"} {
if !strings.Contains(c.Summary, w) {
t.Errorf("the cycle does not name %s: %s", w, c.Summary)
}
}
// Members with no edge between them are no cycle.
manifestsReadMu.Lock()
clear(manifestsRead)
manifestsReadMu.Unlock()
manifestAt = func(context.Context, string, string, string, string) (catalogue.Manifest, bool, error) {
return catalogue.Manifest{}, false, nil
}
if _, cyclic, err := cycleVerdict(context.Background(), "feat/cycle", members, reach, nil, nil); err != nil || cyclic {
t.Fatalf("a cycle where there is none: %v %v", cyclic, err)
}
}
@@ -1,65 +0,0 @@
package main
import (
"io"
"os"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
)
// Issue 1638 (novox/hq ADR 0480 clause 7), taken in with mesh-sdk's container `groups`: the controller sends a container's
// groups only to a node-engine that reports the capability container-groups, and holds the module on any other, naming it.
const n8nWithAGroup = `{"module":"n8n","version":"1",
"config":{"media-gid":{"type":"number","minimum":1,"required":true,"description":"the group that owns the media library"}},
"resources":[{"id":"server","type":"container","name":"n8n",
"image":"n8nio/n8n@sha256:4846eb2f4b874ab04cde7fc1e249d2ddaec66e9aea64439beb2972cfea88e3c0",
"groups":["${config:media-gid}"]}]}`
// printedBy runs reportUnhostable and returns what it printed.
func printedBy(t *testing.T, node string, plan catalogue.Resolution) string {
t.Helper()
stdout := os.Stdout
r, w, err := os.Pipe()
if err != nil {
t.Fatal(err)
}
os.Stdout = w
reportUnhostable(node, plan)
w.Close()
os.Stdout = stdout
out, _ := io.ReadAll(r)
return string(out)
}
// 1638/AC1: a node that reports the capability gets the module; one that does not has it held, the capability named.
func TestAModuleWithGroupsIsHeldOnANodeThatDoesNotReportTheCapability(t *testing.T) {
m, err := catalogue.ParseManifest([]byte(n8nWithAGroup))
if err != nil {
t.Fatal(err)
}
shelf := catalogue.Shelf{"n8n": m}
reads := catalogue.Node{Name: "ace", Capabilities: map[string]bool{"container-runtime": true, "container-groups": true}}
old := catalogue.Node{Name: "g14", Capabilities: map[string]bool{"container-runtime": true}}
got, err := catalogue.Resolve(shelf, []string{"n8n"}, reads, catalogue.World{})
if err != nil || len(got.Unhostable) != 0 || len(got.Modules) != 1 {
t.Fatalf("a node that reads groups: %v, held %+v", err, got.Unhostable)
}
if out := printedBy(t, "ace", got); out != "" {
t.Errorf("a node that reads groups was told: %q", out)
}
got, err = catalogue.Resolve(shelf, []string{"n8n"}, old, catalogue.World{})
if err != nil {
t.Fatal(err)
}
if len(got.Modules) != 0 || len(got.Unhostable) != 1 {
t.Fatalf("n8n was sent to a node that does not report container-groups: hosted %d, held %+v", len(got.Modules), got.Unhostable)
}
if out := printedBy(t, "g14", got); !strings.Contains(out, "n8n") || !strings.Contains(out, "container-groups") {
t.Errorf("the hold does not name the module and the capability: %q", out)
}
}
@@ -1,110 +0,0 @@
//go:build integration
package main
import (
"encoding/json"
"fmt"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdecl "git.novox.be/novox/mesh-sdk/go/declaration"
"git.novox.be/novox/mesh-sdk/go/overlay"
"git.novox.be/novox/mesh-sdk/go/compose"
)
// 1696/AC2, ADR 0486 §5: a node is sent a file by digest only when its profile reports the capability. A node whose profile lacks
// it, and a node the mesh has no profile of, are sent every content inline; a node that reports it is sent the reference. The test
// goes through composeQuestion and composeSend, the two places readsFileReferences is read (compose_sdk.go).
func TestAFileTravelsByDigestOnlyToANodeThatReportsTheCapability(t *testing.T) {
open := aMesh(t) // anchor and laptop report no file-references
ctx := t.Context()
big := strings.Repeat("0123456789abcdef", 8<<10) // 128 KiB
register(t, open, catalogue.Manifest{Module: "bigfile", Version: "1",
Resources: []map[string]any{{"id": "blob", "type": "file", "path": "/etc/bigfile/blob", "mode": "0644", "content": big}}})
// reader reports the capability; bare has never reported a profile at all.
for i, n := range []struct {
name string
profile map[string]any
}{
{"reader", map[string]any{"capabilities": []any{
map[string]any{"name": "container-runtime", "present": true},
map[string]any{"name": "systemd", "present": true},
map[string]any{"name": sdkdecl.FileReferencesCapability, "present": true}}}},
{"bare", nil},
} {
record, err := open.inventory.AddNode(ctx, n.name)
if err != nil {
t.Fatal(err)
}
if err := open.inventory.SetPlace(ctx, n.name, n.name+".example:51820", "here", false, fmt.Sprintf("10.77.0.%d", i+10)); err != nil {
t.Fatal(err)
}
if n.profile != nil {
if err := open.inventory.RecordProfile(ctx, record.ID, n.profile); err != nil {
t.Fatal(err)
}
}
if err := open.inventory.RecordSealingKey(ctx, record.ID, aPublicKey(t)); err != nil {
t.Fatal(err)
}
if err := open.inventory.RecordOverlayKey(ctx, record.ID, aPublicKey(t)); err != nil {
t.Fatal(err)
}
if _, err := open.inventory.Assign(ctx, n.name, overlay.Name); err != nil {
t.Fatal(err)
}
}
for _, node := range []string{"laptop", "reader", "bare"} {
if _, err := assign(ctx, open, node, "bigfile"); err != nil {
t.Fatalf("assign %s: %v", node, err)
}
}
byDigest := map[string]bool{"laptop": false, "bare": false, "reader": true}
for node, want := range byDigest {
asked, _, err := composeQuestion(ctx, open, node, compose.Reading, false)
if err != nil {
t.Fatalf("%s: %v", node, err)
}
sent, err := composeSend(ctx, open, node)
if err != nil {
t.Fatalf("%s: %v", node, err)
}
for what, s := range map[string]sendable{"question": asked, "send": sent} {
if s.fileReferences != want {
t.Fatalf("%s, %s: fileReferences %v, want %v", node, what, s.fileReferences, want)
}
body, err := s.Body()
if err != nil {
t.Fatal(err)
}
var got struct {
Resources []map[string]any `json:"resources"`
}
if err := json.Unmarshal(body, &got); err != nil {
t.Fatal(err)
}
var found bool
for _, r := range got.Resources {
if r["path"] != "/etc/bigfile/blob" {
continue
}
found = true
inline := r["content"] == big
if want && (inline || r["digest"] == nil || r["size"] == nil) {
t.Fatalf("%s, %s: a node that reads references is sent the content inline", node, what)
}
if !want && (!inline || r["digest"] != nil) {
t.Fatalf("%s, %s: a node that does not report the capability is sent a reference", node, what)
}
}
if !found {
t.Fatalf("%s, %s: the file is not in the declaration", node, what)
}
}
}
}
-253
View File
@@ -1,253 +0,0 @@
package main
import (
"context"
"encoding/json"
"errors"
"fmt"
"sort"
"sync"
"git.novox.be/novox/mesh-sdk/go/compose"
sdkdecl "git.novox.be/novox/mesh-sdk/go/declaration"
"github.com/nats-io/nats.go"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/conditions"
)
// **A declaration's large files travel by their digest to a node that can read one** (novox/hq ADR 0486 §1, §5, issue 1696).
//
// The declaration to a node that reports sdkdecl.FileReferencesCapability carries, in place of the content of every
// file resource of more than sdkdecl.FileReferenceThreshold bytes that may travel so (File.TravelsByDigest: plain content or
// bytes, not sealed, not filled with secrets, not written into another file), its sha256 digest and size. The content is put
// into the bucket of the bus under the hex of the digest before the declaration is sent. A node that does not report the
// capability is sent every content inline, as before: an older node-engine refuses a field it does not know, whole.
// readsFileReferences says the node reports the capability of reading a file given by digest (ADR 0486 §5). A node the mesh has
// no profile of reads none.
func readsFileReferences(in compose.Input, node string) bool {
n, ok := nodeOf(in, node)
return ok && n.Capabilities[sdkdecl.FileReferencesCapability]
}
// referencing is the resources as a node that reads references is sent them, and the content of each file that went by
// digest, by the name of its object (the hex of its digest). The resources it is given are not changed. A file that does not
// travel by digest is passed as it is.
func referencing(resources []map[string]any) ([]map[string]any, map[string][]byte, error) {
out := make([]map[string]any, len(resources))
var objects map[string][]byte
for i, r := range resources {
out[i] = r
if kind, _ := r["type"].(string); kind != string(sdkdecl.TypeFile) {
continue
}
// Cheap first: only a file with a content or bytes of the size is worth reading whole.
content, _ := r["content"].(string)
bytesField, _ := r["bytes"].(string)
if len(content) <= sdkdecl.FileReferenceThreshold && len(bytesField) <= sdkdecl.FileReferenceThreshold {
continue
}
raw, err := json.Marshal(r)
if err != nil {
return nil, nil, fmt.Errorf("resource %v: %w", r["id"], err)
}
var f sdkdecl.File
if err := json.Unmarshal(raw, &f); err != nil {
return nil, nil, fmt.Errorf("resource %v is not a file the controller can read: %w", r["id"], err)
}
if !f.TravelsByDigest() {
continue
}
ref, body, err := f.AsReference()
if err != nil {
return nil, nil, err
}
name, err := sdkdecl.ObjectName(ref.Digest)
if err != nil {
return nil, nil, err
}
with := make(map[string]any, len(r)+2)
for k, v := range r {
with[k] = v
}
delete(with, "content")
delete(with, "bytes")
with["digest"], with["size"] = ref.Digest, ref.Size
out[i] = with
if objects == nil {
objects = map[string][]byte{}
}
objects[name] = body
}
return out, objects, nil
}
// namedObjects are the objects of the bucket a body names: the object name of every file resource given by reference,
// with its size. It reads a declaration as the node does, so it also reads the latest one the bus holds.
func namedObjects(body []byte) (map[string]int64, error) {
var envelope struct {
Resources []struct {
Type string `json:"type"`
Digest string `json:"digest"`
Size int64 `json:"size"`
} `json:"resources"`
}
if err := json.Unmarshal(body, &envelope); err != nil {
return nil, err
}
var out map[string]int64
for _, r := range envelope.Resources {
if r.Type != string(sdkdecl.TypeFile) || r.Digest == "" {
continue
}
name, err := sdkdecl.ObjectName(r.Digest)
if err != nil {
return nil, err
}
if out == nil {
out = map[string]int64{}
}
out[name] = r.Size
}
return out, nil
}
// objectStore is what putting the objects needs of a node's bucket.
type objectStore interface {
// Has says whether an object of that name is there, complete, and whether it carries the sweep's mark of being named by no
// node's latest declaration.
Has(ctx context.Context, name string) (have, marked bool, err error)
// Put writes the object under that name, replacing one of the same name (and so erasing its mark).
Put(ctx context.Context, name string, content []byte) error
}
// putObjects makes sure every object of the declaration is in the bucket before the declaration that names it is sent: a node
// that reads the declaration finds each object. An object already there (the same content, the same name) is not written again,
// **unless the sweep marked it unnamed**: then it is written again, so the mark that would have it removed is gone with the
// old object (issue 1696, review: a mark read just before a send must not take an object the send relies on). Refuses on the
// first it cannot put, naming it, so the declaration is not sent.
func putObjects(ctx context.Context, store objectStore, objects map[string][]byte) error {
names := make([]string, 0, len(objects))
for name := range objects {
names = append(names, name)
}
sort.Strings(names)
for _, name := range names {
have, marked, err := store.Has(ctx, name)
if err != nil {
return fmt.Errorf("whether the object %s is in the bucket: %w", name[:12], err)
}
if have && !marked {
continue
}
if err := store.Put(ctx, name, objects[name]); err != nil {
return fmt.Errorf("putting the object %s (%d bytes) in the bucket: %w", name[:12], len(objects[name]), err)
}
}
return nil
}
// busObjects is a node's bucket on the controller's own connection.
type busObjects struct{ store jetstream.ObjectStore }
// openBucket opens the bucket of a node's files, making it first if it is not there (it is asserted for every node at the
// controller's start; a node adopted since is covered here).
func openBucket(ctx context.Context, conn *nats.Conn, node string) (busObjects, error) {
js, err := jetstream.New(conn)
if err != nil {
return busObjects{}, err
}
bucket, err := broker.DeclarationFilesBucketOf(node)
if err != nil {
return busObjects{}, fmt.Errorf("the object bucket of node %q: %w", node, err)
}
store, err := js.ObjectStore(ctx, bucket)
if errors.Is(err, jetstream.ErrBucketNotFound) {
if err = broker.EnsureDeclarationFiles(ctx, js, node); err == nil {
store, err = js.ObjectStore(ctx, bucket)
}
}
if err != nil {
return busObjects{}, fmt.Errorf("the object bucket %s: %w", bucket, err)
}
return busObjects{store}, nil
}
func (b busObjects) Has(ctx context.Context, name string) (bool, bool, error) {
info, err := b.store.GetInfo(ctx, name)
switch {
case errors.Is(err, jetstream.ErrObjectNotFound):
return false, false, nil
case err != nil:
return false, false, err
}
_, marked := info.Metadata[declarationUnnamed]
return !info.Deleted, marked, nil
}
func (b busObjects) Put(ctx context.Context, name string, content []byte) error {
_, err := b.store.PutBytes(ctx, name, content)
return err
}
// filesLocks hold a node's bucket for one act at a time in this process: a send puts its objects and publishes the declaration that
// names them under the node's lock, and the sweep judges the node's bucket under the same one, so the sweep never removes an
// object between a send's put and its declaration. Another process (a person's push) is covered by the put erasing a mark and by
// the sweep looking at an object again just before it removes it.
var filesLocks sync.Map
func lockFilesOf(node string) func() {
l, _ := filesLocks.LoadOrStore(node, &sync.Mutex{})
mu := l.(*sync.Mutex)
mu.Lock()
return mu.Unlock
}
// kindFilesNotPut is a node whose files that travel by digest could not be put in its bucket, send after send.
const kindFilesNotPut = "declaration-files-not-put"
// filesNotPutAfter is how many sends in a row may fail to put before the condition is raised: the first send to the machine holding
// the bus fails for want of the grant that very send delivers, and says so, and is not yet a fault.
const filesNotPutAfter = 2
var (
filesFailMu sync.Mutex
filesFailed = map[string]int{}
)
func filesNotPutObservation(node, summary string) conditions.Observation {
return conditions.Observation{Scope: conditions.ScopeMachine, ID: node, Kind: kindFilesNotPut, Token: kindFilesNotPut,
Machine: node, Severity: conditions.Warning, Source: "sends", Summary: summary}
}
// notePut records how putting a send's objects went: a failure counts, the second in a row raises the condition, a success
// clears the count and the condition.
func notePut(ctx context.Context, node string, put error) {
filesFailMu.Lock()
if put == nil {
delete(filesFailed, node)
} else {
filesFailed[node]++
}
n := filesFailed[node]
filesFailMu.Unlock()
keeper := conditionsFrom
if keeper == nil {
return
}
key := filesNotPutObservation(node, "").Key()
switch {
case put == nil:
if _, err := keeper.Clear(ctx, key, node+" was sent its files by digest"); err != nil {
fmt.Printf("the condition %s could not be cleared: %v\n", key, err)
}
case n >= filesNotPutAfter:
summary := fmt.Sprintf("%s's files that travel by digest could not be put in its bucket %d sends in a row, so it is sent them "+
"inline: %v", node, n, put)
if _, err := keeper.Observe(ctx, filesNotPutObservation(node, summary)); err != nil {
fmt.Printf("the condition %s could not be raised: %v\n", key, err)
}
}
}
@@ -1,246 +0,0 @@
//go:build integration
package main
import (
"bytes"
"context"
"encoding/json"
"strings"
"testing"
"time"
sdkdecl "git.novox.be/novox/mesh-sdk/go/declaration"
"github.com/nats-io/nats.go"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/link"
"github.com/novox/mesh-controller/internal/testbus"
)
type sigOf struct{}
func (sigOf) Sign(context.Context, []byte) ([]byte, error) { return make([]byte, 64), nil }
// 1696/AC7 (ADR 0486 §1): on a real bus with its default largest message of 1 MiB, a declaration whose files add up to more than
// 3 MiB is sent as references, its objects are in the bucket, each fetched back whole and verified against the digest and size the
// declaration names; the same declaration inline is refused by the controller, naming size and limit. A changed byte in an object
// fails the verification.
func TestADeclarationOfMoreThanThreeMiBIsSentByReferenceOnARealBus(t *testing.T) { // 1696/AC7
s := testbus.Start(t)
conn, err := nats.Connect(s.ClientURL())
if err != nil {
t.Fatal(err)
}
defer conn.Close()
nc, err := conn.JetStream()
if err != nil {
t.Fatal(err)
}
if _, err := nc.AddStream(&nats.StreamConfig{Name: "NODES", Subjects: []string{"mesh.node.*.declare"},
MaxMsgsPerSubject: 1}); err != nil {
t.Fatal(err)
}
ctx, cancel := context.WithTimeout(context.Background(), 60*time.Second)
defer cancel()
api, err := jetstream.New(conn)
if err != nil {
t.Fatal(err)
}
if err := broker.EnsureDeclarationFiles(ctx, api, "novox"); err != nil {
t.Fatal(err)
}
// Run twice: an assertion of an existing bucket is no error and no change.
if err := broker.EnsureDeclarationFiles(ctx, api, "novox"); err != nil {
t.Fatalf("asserting the bucket again: %v", err)
}
var resources []map[string]any
for i := 0; i < 3; i++ {
resources = append(resources, aFileResource("big-"+string(rune('a'+i)), contentOf("big"+string(rune('a'+i)), 1200<<10)))
}
resources = append(resources, aFileResource("small", "tiny\n"))
inline := sendable{Resources: resources}
byDigest := sendable{fileReferences: true, Resources: resources}
bus := link.OverNATS{Conn: conn, JS: nc}
body, err := inline.Body()
if err != nil {
t.Fatal(err)
}
if err := link.Declare(ctx, bus, sigOf{}, "novox", body, 5*time.Second); err == nil ||
!strings.Contains(err.Error(), "1048576") || strings.Contains(err.Error(), "maximum payload exceeded") {
t.Fatalf("the inline declaration of more than 3 MiB: %v", err)
}
bucket, err := openBucket(ctx, conn, "novox")
if err != nil {
t.Fatal(err)
}
objects, err := byDigest.objects()
if err != nil || len(objects) != 3 {
t.Fatalf("objects: %d %v", len(objects), err)
}
if err := putObjects(ctx, bucket, objects); err != nil {
t.Fatal(err)
}
if err := putObjects(ctx, bucket, objects); err != nil {
t.Fatalf("putting them again: %v", err)
}
refBody, err := byDigest.Body()
if err != nil {
t.Fatal(err)
}
if err := link.Declare(ctx, bus, sigOf{}, "novox", refBody, 5*time.Second); err != nil {
t.Fatalf("the declaration by reference: %v", err)
}
// What the stream holds for the node is the reference form, and it names exactly the three objects.
latest, err := latestDeclarations(ctx, conn, []string{"novox", "ace"})
if err != nil || len(latest) != 1 {
t.Fatalf("the latest declarations: %v %v", len(latest), err)
}
named, err := namedByLatest("novox", latest["novox"])
if err != nil || len(named) != 3 {
t.Fatalf("what the latest declaration names: %v %v", named, err)
}
novoxBucket, err := broker.DeclarationFilesBucketOf("novox")
if err != nil {
t.Fatal(err)
}
store, err := api.ObjectStore(ctx, novoxBucket)
if err != nil {
t.Fatal(err)
}
var envelope struct {
Resources []sdkdecl.File `json:"resources"`
}
if err := json.Unmarshal(refBody, &envelope); err != nil {
t.Fatal(err)
}
for _, f := range envelope.Resources {
if !f.IsReference() {
continue
}
name, _ := sdkdecl.ObjectName(f.Digest)
got, err := store.GetBytes(ctx, name)
if err != nil {
t.Fatalf("the object of %s: %v", f.ID, err)
}
if err := f.Verify(got); err != nil {
t.Errorf("the object does not verify: %v", err)
}
got[len(got)/2] ^= 1
if err := f.Verify(got); err == nil {
t.Errorf("a changed byte in the object of %s passed the verification", f.ID)
}
}
// The sweep against the real bucket: all named, nothing marked; a fourth object named by none is marked, kept a day less a
// minute, and removed at the day.
extra := bytes.Repeat([]byte("x"), 100)
extraName := strings.Repeat("e", 64)
if err := putObjects(ctx, bucket, map[string][]byte{extraName: extra}); err != nil {
t.Fatal(err)
}
b := busSweptBucket{store}
t0 := time.Now()
if r, err := sweepDeclarationFiles(ctx, b, named, t0); err != nil || r.Marked != 1 || r.Removed != 0 || r.Kept != 3 {
t.Fatalf("first sweep: %+v %v", r, err)
}
if r, err := sweepDeclarationFiles(ctx, b, named, t0.Add(24*time.Hour-time.Minute)); err != nil || r.Removed != 0 {
t.Fatalf("a minute short of a day: %+v %v", r, err)
}
if r, err := sweepDeclarationFiles(ctx, b, named, t0.Add(24*time.Hour+time.Second)); err != nil || r.Removed != 1 {
t.Fatalf("a day after: %+v %v", r, err)
}
if infos, err := store.List(ctx); err != nil || len(infos) != 3 {
t.Fatalf("the bucket holds %d objects (%v), want the 3 the declaration names", len(infos), err)
}
}
// 1696/AC8 on a real bucket: the sweep removes an object only if it still carries the mark it judged it by. An object marked
// unnamed, then put again by a send (which erases the mark), is kept when the sweep deletes it by the old mark.
func TestTheSweepsDeleteLeavesAnObjectASendPutAgainAfterItWasMarked(t *testing.T) { // 1696/AC8
s := testbus.Start(t)
conn, err := nats.Connect(s.ClientURL())
if err != nil {
t.Fatal(err)
}
defer conn.Close()
ctx, cancel := context.WithTimeout(context.Background(), 60*time.Second)
defer cancel()
bucket, err := openBucket(ctx, conn, "novox")
if err != nil {
t.Fatal(err)
}
name := strings.Repeat("a", 64)
content := map[string][]byte{name: bytes.Repeat([]byte("y"), 100)}
if err := putObjects(ctx, bucket, content); err != nil {
t.Fatal(err)
}
b := busSweptBucket{bucket.store}
judged := time.Now().Add(-25 * time.Hour).Truncate(time.Second)
if err := b.MarkUnnamed(ctx, name, judged); err != nil {
t.Fatal(err)
}
if _, marked, err := bucket.Has(ctx, name); err != nil || !marked {
t.Fatalf("the object was not marked: %v %v", marked, err)
}
// The sweep has read the mark; a send now puts the object again, which erases it.
if err := putObjects(ctx, bucket, content); err != nil {
t.Fatal(err)
}
if have, marked, err := bucket.Has(ctx, name); err != nil || !have || marked {
t.Fatalf("a send did not put the marked object again: have %v marked %v %v", have, marked, err)
}
gone, err := b.Delete(ctx, name, judged)
if err != nil || gone {
t.Fatalf("the sweep deleted by an old mark: gone %v, %v", gone, err)
}
if have, _, err := bucket.Has(ctx, name); err != nil || !have {
t.Fatalf("the object a send relies on is gone: %v %v", have, err)
}
// And the mark still on it is honoured: marked again and deleted by that mark, it goes.
if err := b.MarkUnnamed(ctx, name, judged); err != nil {
t.Fatal(err)
}
if gone, err := b.Delete(ctx, name, judged); err != nil || !gone {
t.Fatalf("an object still carrying the mark it was judged by is kept: gone %v, %v", gone, err)
}
}
// 1696/AC9 on a real bus: the sweep of a node's bucket takes the node's lock first, so it waits for a send that has put its objects
// and not yet published the declaration that names them.
func TestTheSweepOfANodesBucketHoldsTheNodesLock(t *testing.T) { // 1696/AC9
s := testbus.Start(t)
conn, err := nats.Connect(s.ClientURL())
if err != nil {
t.Fatal(err)
}
defer conn.Close()
ctx, cancel := context.WithTimeout(context.Background(), 60*time.Second)
defer cancel()
const node = "sweep-lock-node"
if _, err := openBucket(ctx, conn, node); err != nil { // the bucket exists, or the sweep would return before the lock
t.Fatal(err)
}
release := lockFilesOf(node) // a send is between its put and its publication
done := make(chan struct{})
go func() {
_, _ = sweepNode(ctx, conn, node)
close(done)
}()
select {
case <-done:
t.Fatal("the sweep judged the node's bucket while a send held it")
case <-time.After(300 * time.Millisecond):
}
release()
select {
case <-done:
case <-time.After(20 * time.Second):
t.Fatal("the sweep did not go on once the send let go")
}
}
@@ -1,534 +0,0 @@
package main
import (
"context"
"encoding/base64"
"encoding/hex"
"encoding/json"
"errors"
"fmt"
"strings"
"testing"
"time"
sdkdecl "git.novox.be/novox/mesh-sdk/go/declaration"
"github.com/novox/mesh-controller/internal/conditions"
"github.com/novox/mesh-controller/internal/link"
)
// contentOf is a deterministic content of n bytes that does not compress to nothing and holds no character JSON escapes.
func contentOf(seed string, n int) string {
var b strings.Builder
for i := 0; b.Len() < n; i++ {
fmt.Fprintf(&b, "%s-%07d-%x;", seed, i, i*2654435761%1000003)
}
return b.String()[:n]
}
func aFileResource(id, content string) map[string]any {
return map[string]any{"id": id, "type": "file", "path": "/etc/" + id, "mode": "0644", "content": content}
}
// controlNodeShape is a declaration shaped like the control node's on 2026-10-21 (issue 1689), the sizes measured then: a bus users
// file of 252,678 bytes, the vault's export of 138,475 bytes (compact), the tool runtime 21 KB, the packet filter 19 KB, the launcher
// 14 KB, and about 600 small resources for the rest (about 286 KB). `more` scales the count of small ones: 1 is the shape, 1.4 forty
// percent more of them.
func controlNodeShape(more float64) []map[string]any {
res := []map[string]any{
aFileResource("nats.bus-users", contentOf("users", 252678)),
aFileResource("mesh-vault.kept", contentOf("kept", 138475)),
aFileResource("node-tools.runtime", contentOf("runtime", 21<<10)),
aFileResource("nftables.filtering", contentOf("filter", 19<<10)),
aFileResource("mesh-host.launcher", contentOf("launcher", 14<<10)),
}
for i := 0; i < int(600*more); i++ {
res = append(res, aFileResource(fmt.Sprintf("module-%03d.config", i), contentOf(fmt.Sprintf("small%d", i), 380)))
}
return res
}
func wireSizeOf(t *testing.T, s sendable) int {
t.Helper()
body, err := s.Body()
if err != nil {
t.Fatal(err)
}
// The signature the controller's key makes is 64 bytes.
msg, err := link.Envelope(body, make([]byte, 64))
if err != nil {
t.Fatal(err)
}
return len(msg)
}
// 1696/AC1: for a node that reports the capability, the two big files of a control-node-shaped declaration carry their digest and size
// and no content, and the declaration is far under the bus's 1 MiB.
func TestAFileOver64KiBTravelsByDigestToANodeThatReadsOne(t *testing.T) { // 1696/AC1
users, kept := contentOf("users", 252678), contentOf("kept", 138475)
s := sendable{fileReferences: true, Resources: []map[string]any{
aFileResource("nats.bus-users", users), aFileResource("mesh-vault.kept", kept),
aFileResource("small", "tiny\n"),
}}
body, err := s.Body()
if err != nil {
t.Fatal(err)
}
var got struct {
Resources []map[string]any `json:"resources"`
}
if err := json.Unmarshal(body, &got); err != nil {
t.Fatal(err)
}
for i, want := range []string{users, kept} {
r := got.Resources[i]
digest := sdkdecl.DigestOf([]byte(want))
if _, has := r["content"]; has || r["digest"] != digest || r["size"] != float64(len(want)) {
t.Errorf("%v goes with content %v, digest %v, size %v; want no content, %s and %d", r["id"], r["content"] != nil,
r["digest"], r["size"], digest, len(want))
}
if r["path"] == nil || r["mode"] != "0644" {
t.Errorf("%v lost its path or mode: %v", r["id"], r)
}
}
if got.Resources[2]["content"] != "tiny\n" || got.Resources[2]["digest"] != nil {
t.Errorf("a small file is not inline: %v", got.Resources[2])
}
if strings.Contains(string(body), users[:200]) || strings.Contains(string(body), kept[:200]) {
t.Error("the content of a file that went by digest is in the body")
}
if n := wireSizeOf(t, s); n > 100<<10 {
t.Errorf("the declaration is %d bytes on the wire, want under 100 KB", n)
}
// The objects to put are the contents, each under the hex of its digest, and a reference verifies against its object.
objects, err := s.objects()
if err != nil || len(objects) != 2 {
t.Fatalf("objects: %v %v", objects, err)
}
for _, want := range []string{users, kept} {
name, _ := sdkdecl.ObjectName(sdkdecl.DigestOf([]byte(want)))
if string(objects[name]) != want {
t.Errorf("the object %s does not hold the content", name[:12])
}
}
named, err := namedObjects(body)
if err != nil || len(named) != 2 {
t.Fatalf("the objects the body names: %v %v", named, err)
}
for name := range objects {
if _, ok := named[name]; !ok {
t.Errorf("the body does not name the object %s it was sent with", name[:12])
}
if _, err := hex.DecodeString(name); err != nil || len(name) != 64 {
t.Errorf("the object name %q is not 64 hex characters", name)
}
}
}
// 1696/AC2: a node that does not report the capability is sent every content inline, as before, and has no object put.
func TestANodeWithoutTheCapabilityIsSentEveryContentInline(t *testing.T) { // 1696/AC2
users := contentOf("users", 252678)
s := sendable{Resources: []map[string]any{aFileResource("nats.bus-users", users)}}
body, err := s.Body()
if err != nil {
t.Fatal(err)
}
var got struct {
Resources []map[string]any `json:"resources"`
}
if err := json.Unmarshal(body, &got); err != nil {
t.Fatal(err)
}
if got.Resources[0]["content"] != users || got.Resources[0]["digest"] != nil || got.Resources[0]["size"] != nil {
t.Fatalf("a node that reads no references is not sent the content inline: %v", got.Resources[0]["digest"])
}
if objects, err := s.objects(); err != nil || len(objects) != 0 {
t.Fatalf("objects for a node that reads none: %v %v", len(objects), err)
}
}
// 1696/AC3: only a file that may travel by digest does: one of exactly 64 KiB, a sealed one, one filled with secrets, one written
// into another file, and a directory stay inline, whatever the node reads.
func TestOnlyAFileThatMayTravelByDigestDoes(t *testing.T) { // 1696/AC3
exact := contentOf("exact", 64<<10)
big := contentOf("big", 64<<10+1)
sealed := aFileResource("sealed", big)
sealed["sealed"], sealed["content"] = "box", ""
sealed["content"] = big // a sealed file's content is none; the body must pass it as it is, whatever it holds
secret := aFileResource("secret", big)
secret["secrets"] = map[string]any{"token": "x"}
into := aFileResource("into", big)
into["into"] = "/etc/other"
s := sendable{fileReferences: true, Resources: []map[string]any{
aFileResource("exact", exact), sealed, secret, into,
{"id": "dir", "type": "directory", "path": "/etc/d", "content": big},
aFileResource("big", big),
}}
body, err := s.Body()
if err != nil {
t.Fatal(err)
}
var got struct {
Resources []map[string]any `json:"resources"`
}
if err := json.Unmarshal(body, &got); err != nil {
t.Fatal(err)
}
for i, id := range []string{"exact", "sealed", "secret", "into", "dir"} {
if got.Resources[i]["id"] != id || got.Resources[i]["digest"] != nil || got.Resources[i]["content"] == nil {
t.Errorf("%s went by digest or lost its content: %v", id, got.Resources[i]["digest"])
}
}
if big := got.Resources[5]; big["id"] != "big" || big["digest"] == nil || big["content"] != nil {
t.Errorf("a plain file of 64 KiB and a byte did not go by digest: %v", big["digest"])
}
}
// A file given as bytes goes by digest on its decoded length, with the decoded bytes in the object.
func TestAFileOfBytesTravelsByDigestOnItsDecodedLength(t *testing.T) {
raw := []byte(contentOf("png", 70<<10))
r := map[string]any{"id": "icon", "type": "file", "path": "/usr/share/icon.png", "bytes": stdBase64(raw)}
s := sendable{fileReferences: true, Resources: []map[string]any{r}}
objects, err := s.objects()
if err != nil || len(objects) != 1 {
t.Fatalf("%v %v", len(objects), err)
}
name, _ := sdkdecl.ObjectName(sdkdecl.DigestOf(raw))
if string(objects[name]) != string(raw) {
t.Fatal("the object is not the decoded bytes")
}
if r["bytes"] == nil {
t.Fatal("the composed resource was changed in place")
}
}
type memObjects struct {
have map[string][]byte
marked map[string]bool
puts []string
fail map[string]error
}
func (m *memObjects) Has(_ context.Context, name string) (bool, bool, error) {
_, ok := m.have[name]
return ok, m.marked[name], nil
}
func (m *memObjects) Put(_ context.Context, name string, c []byte) error {
if err := m.fail[name]; err != nil {
return err
}
m.puts = append(m.puts, name)
m.have[name] = c
delete(m.marked, name)
return nil
}
// 1696/AC4: an object already in the bucket is not put again, one absent is, and a put that fails stops the send, naming the object.
func TestObjectsAreInTheBucketBeforeTheDeclarationIsSent(t *testing.T) { // 1696/AC4
objects := map[string][]byte{strings.Repeat("a", 64): []byte("one"), strings.Repeat("b", 64): []byte("two")}
b := &memObjects{have: map[string][]byte{strings.Repeat("a", 64): []byte("one")}}
if err := putObjects(t.Context(), b, objects); err != nil || len(b.puts) != 1 || b.puts[0] != strings.Repeat("b", 64) {
t.Fatalf("put %v (%v), want the one absent", b.puts, err)
}
// An object the sweep marked unnamed is put again, which erases the mark: a send relies on it.
b = &memObjects{have: map[string][]byte{strings.Repeat("a", 64): []byte("one"), strings.Repeat("b", 64): []byte("two")},
marked: map[string]bool{strings.Repeat("a", 64): true}}
if err := putObjects(t.Context(), b, objects); err != nil || len(b.puts) != 1 || b.puts[0] != strings.Repeat("a", 64) || b.marked[strings.Repeat("a", 64)] {
t.Fatalf("a marked object: put %v (%v) marked %v, want it put again and the mark gone", b.puts, err, b.marked)
}
b = &memObjects{have: map[string][]byte{}, fail: map[string]error{strings.Repeat("a", 64): errors.New("permissions violation")}}
err := putObjects(t.Context(), b, objects)
if err == nil || !strings.Contains(err.Error(), "aaaaaaaaaaaa") || !strings.Contains(err.Error(), "permissions violation") {
t.Fatalf("a failed put: %v", err)
}
}
// fallbackDelivery refuses the first declaration it is sent with the bucket's failure, as a bus that has not yet been given the
// controller's grant does, and takes the second.
type fallbackDelivery struct {
bodies [][]byte
first error
}
func (f *fallbackDelivery) grant(context.Context, []readyNode) error { return nil }
func (f *fallbackDelivery) declare(_ context.Context, _ readyNode, body []byte) (string, error) {
f.bodies = append(f.bodies, body)
if len(f.bodies) == 1 && f.first != nil {
return "", fmt.Errorf("x: %w: %w", errFilesNotPut, f.first)
}
return "digest", nil
}
// 1696/AC5: when the bucket cannot be written yet (the grant travels in the declaration of the machine holding the bus), that send goes
// with every content inline and is made; a send that fails for another reason is not retried inline.
func TestASendWhoseFilesCannotBePutGoesInlineOnce(t *testing.T) { // 1696/AC5
users := contentOf("users", 100<<10)
s := readyNode{node: "novox", declared: sendable{fileReferences: true, Resources: []map[string]any{aFileResource("nats.bus-users", users)}}}
f := &fallbackDelivery{first: errors.New("timeout")}
digests, err := deliver(t.Context(), f, "novox", []readyNode{s})
if err != nil || digests["novox"] != "digest" || len(f.bodies) != 2 {
t.Fatalf("%v %v %d", digests, err, len(f.bodies))
}
if strings.Contains(string(f.bodies[0]), users[:100]) || !strings.Contains(string(f.bodies[1]), users[:100]) {
t.Error("the first body should name the file by digest and the second carry it inline")
}
other := &fallbackDelivery{}
other.first = nil
failing := &failingDelivery{err: errors.New("the bus is down")}
if _, err := deliver(t.Context(), failing, "novox", []readyNode{s}); err == nil || failing.calls != 1 {
t.Fatalf("a failure of another kind: %v after %d send(s)", err, failing.calls)
}
}
type failingDelivery struct {
err error
calls int
}
func (f *failingDelivery) grant(context.Context, []readyNode) error { return nil }
func (f *failingDelivery) declare(context.Context, readyNode, []byte) (string, error) {
f.calls++
return "", f.err
}
// A bucket for the sweep.
type sweepBucketFake struct {
objects map[string]time.Time // zero: named at the last look
deleted []string
}
func (b *sweepBucketFake) Objects(context.Context) ([]bucketObject, error) {
var out []bucketObject
for n, at := range b.objects {
out = append(out, bucketObject{Name: n, UnnamedSince: at})
}
return out, nil
}
func (b *sweepBucketFake) MarkUnnamed(_ context.Context, n string, at time.Time) error {
b.objects[n] = at
return nil
}
func (b *sweepBucketFake) ClearUnnamed(_ context.Context, n string) error {
b.objects[n] = time.Time{}
return nil
}
func (b *sweepBucketFake) Delete(_ context.Context, n string, markedAt time.Time) (bool, error) {
if at, ok := b.objects[n]; !ok || !at.Equal(markedAt) {
return false, nil // put again since it was judged: its mark is gone
}
delete(b.objects, n)
b.deleted = append(b.deleted, n)
return true, nil
}
// 1696/AC6: an object a latest declaration still names is not removed, however long; one unnamed is marked at the first look and removed
// only when the mark is 24 hours old; one named again loses its mark.
func TestAnObjectIsRemovedADayAfterNoLatestDeclarationNamesIt(t *testing.T) { // 1696/AC6
named, unnamed, back := strings.Repeat("1", 64), strings.Repeat("2", 64), strings.Repeat("3", 64)
t0 := time.Date(2026, 10, 21, 12, 0, 0, 0, time.UTC)
b := &sweepBucketFake{objects: map[string]time.Time{named: t0.Add(-100 * time.Hour), unnamed: {}, back: t0.Add(-5 * time.Hour)}}
named3 := map[string]bool{named: true, back: true}
r, err := sweepDeclarationFiles(t.Context(), b, named3, t0)
if err != nil || r.Marked != 1 || r.Cleared != 2 || r.Removed != 0 || len(b.deleted) != 0 {
t.Fatalf("first look: %+v %v %v", r, err, b.deleted)
}
if !b.objects[named].IsZero() || !b.objects[back].IsZero() || !b.objects[unnamed].Equal(t0) {
t.Fatalf("marks after the first look: %v", b.objects)
}
// Still unnamed 23 hours 59 minutes later: kept. A day after the mark: removed. The named ones stay for ever.
if r, err = sweepDeclarationFiles(t.Context(), b, map[string]bool{named: true, back: true}, t0.Add(24*time.Hour-time.Minute)); err != nil ||
r.Removed != 0 || len(b.deleted) != 0 {
t.Fatalf("a minute short of a day: %+v %v %v", r, err, b.deleted)
}
if r, err = sweepDeclarationFiles(t.Context(), b, map[string]bool{named: true, back: true}, t0.Add(24*time.Hour)); err != nil ||
r.Removed != 1 || len(b.deleted) != 1 || b.deleted[0] != unnamed {
t.Fatalf("a day after: %+v %v %v", r, err, b.deleted)
}
if _, ok := b.objects[named]; !ok {
t.Fatal("an object a latest declaration names was removed")
}
// An object named again before the day is up loses its mark and is not removed.
b = &sweepBucketFake{objects: map[string]time.Time{unnamed: t0}}
if _, err := sweepDeclarationFiles(t.Context(), b, map[string]bool{unnamed: true}, t0.Add(23*time.Hour)); err != nil || !b.objects[unnamed].IsZero() {
t.Fatalf("named again: %v %v", b.objects, err)
}
}
// What a node's latest declaration names is read from the signed envelope as stored on the bus; an unreadable one is an error and
// never "names none", which would mark an object a node needs.
func TestTheObjectsALatestDeclarationNamesAreReadFromTheBus(t *testing.T) {
s := sendable{fileReferences: true, Resources: []map[string]any{aFileResource("a", contentOf("a", 70<<10))}}
body, _ := s.Body()
msg, _ := link.Envelope(body, []byte("sig"))
named, err := namedByLatest("novox", msg)
name, _ := sdkdecl.ObjectName(sdkdecl.DigestOf([]byte(contentOf("a", 70<<10))))
if err != nil || len(named) != 1 || !named[name] {
t.Fatalf("%v %v", named, err)
}
if none, err := namedByLatest("ace", []byte(`{"declaration":"e30=","signature":""}`)); err != nil || len(none) != 0 {
t.Fatalf("a declaration naming none: %v %v", none, err)
}
if _, err := namedByLatest("novox", []byte("not json")); err == nil {
t.Fatal("an unreadable latest declaration was read as naming none")
}
}
// 1698/AC2 (ADR 0486 §9): a declaration shaped like the control node's, for a node that reads references, stays under half of the
// bus's largest message on the wire; the inline form of the same does not fit at all; and the same shape with 40 percent more resources
// fails the test's own limit, so a change that eats the room fails in a pull request and not on the live mesh.
func TestTheControlNodesDeclarationInTheReferenceFormStaysUnderHalfTheBusLimit(t *testing.T) { // 1698/AC2
const limit = 1048576
shape := sendable{fileReferences: true, Resources: controlNodeShape(1)}
n := wireSizeOf(t, shape)
t.Logf("the control node's shape in the reference form: %d bytes on the wire, %.1f%% of %d", n, 100*float64(n)/limit, limit)
if n > limit/2 {
t.Errorf("the reference form is %d bytes on the wire, over half (%d) of the bus's largest message", n, limit/2)
}
if inline := wireSizeOf(t, sendable{Resources: controlNodeShape(1)}); inline < limit*3/4 {
t.Errorf("the inline form is %d bytes, which is not the shape that was over the limit", inline)
}
grown := wireSizeOf(t, sendable{fileReferences: true, Resources: controlNodeShape(1.4)})
if grown <= limit/2 {
t.Errorf("the shape with 40%% more resources is %d bytes, which the test would let pass", grown)
}
}
func stdBase64(b []byte) string { return base64.StdEncoding.EncodeToString(b) }
// A sweep that read an object's mark before a send put it again must not remove it: the object, now without the mark, is left.
func TestASweepDoesNotRemoveAnObjectASendPutAgainAfterItWasJudged(t *testing.T) { // 1696/AC8
n := strings.Repeat("4", 64)
t0 := time.Date(2026, 10, 21, 12, 0, 0, 0, time.UTC)
b := &sweepBucketFake{objects: map[string]time.Time{n: t0}}
judged := bucketObject{Name: n, UnnamedSince: t0}
b.objects[n] = time.Time{} // the send put it again: no mark
if gone, err := b.Delete(t.Context(), judged.Name, judged.UnnamedSince); err != nil || gone || len(b.deleted) != 0 {
t.Fatalf("an object put again was removed: %v %v %v", gone, err, b.deleted)
}
}
// One node's bucket is held by one act at a time in this process: the sweep waits for a send that has put its objects and not yet
// published the declaration that names them (1696/AC9).
func TestTheSweepOfANodesBucketWaitsForASendThatHasPutItsObjects(t *testing.T) { // 1696/AC9
unlock := lockFilesOf("lock-test-node")
got := make(chan struct{})
go func() {
defer lockFilesOf("lock-test-node")()
close(got)
}()
select {
case <-got:
t.Fatal("the sweep took the node's bucket while a send held it")
case <-time.After(50 * time.Millisecond):
}
other := lockFilesOf("lock-test-other-node") // another node's bucket is not held
other()
unlock()
select {
case <-got:
case <-time.After(2 * time.Second):
t.Fatal("the sweep did not take the node's bucket once the send was done")
}
}
// A put that fails send after send raises a condition; the first failure does not (the grant it waits for travels in that very
// declaration), and a put that works clears it (1696/AC10).
func TestAPutThatFailsSendAfterSendRaisesAConditionAndClearsWhenItWorks(t *testing.T) { // 1696/AC10
store := conditions.NewInMemory()
k := conditions.NewKeeper(t.Context(), conditions.Options{Store: store, History: store,
Teller: &conditions.Told{}, Now: func() time.Time { return time.Date(2026, 10, 21, 14, 0, 0, 0, time.UTC) }})
t.Cleanup(func() { k.Close(t.Context()) })
was := conditionsFrom
conditionsFrom = k
t.Cleanup(func() { conditionsFrom = was })
key := filesNotPutObservation("cond-node", "").Key()
open := func() bool {
_, found, err := conditions.ReadOne(t.Context(), store, key)
if err != nil {
t.Fatal(err)
}
return found
}
boom := errors.New("permissions violation")
notePut(t.Context(), "cond-node", boom)
if open() {
t.Fatal("the first failure raised the condition")
}
notePut(t.Context(), "cond-node", boom)
if !open() {
t.Fatal("the second failure in a row did not raise the condition")
}
notePut(t.Context(), "cond-node", nil)
if open() {
t.Fatal("a put that worked did not clear the condition")
}
notePut(t.Context(), "cond-node", boom)
if open() {
t.Fatal("the count did not start again after a put that worked")
}
}
// aBusThatSays is a bus that keeps the declarations published to it and, at each, asks whether the node's bucket is held.
type aBusThatSays struct {
link.Bus
published chan string
heldAtPublish func() bool
}
func (b *aBusThatSays) PublishDeclaration(_ context.Context, node string, _ []byte) error {
if b.heldAtPublish != nil && !b.heldAtPublish() {
return errors.New("published while the node's bucket was not held")
}
b.published <- node
return errors.New("stop here: nothing is recorded")
}
type aSignerOfZeros struct{}
func (aSignerOfZeros) Sign(context.Context, []byte) ([]byte, error) { return make([]byte, 64), nil }
// whileHeld says whether the node's lock is held by someone right now, without keeping it.
func whileHeld(node string) bool {
got := make(chan struct{})
go func() {
defer lockFilesOf(node)()
close(got)
}()
select {
case <-got:
return false
case <-time.After(100 * time.Millisecond):
// held: the goroutine stays waiting and takes and releases the lock once it is let go.
return true
}
}
// 1696/AC9: a send holds the node's lock from the put to the publication, so the sweep never removes an object between them. The
// test goes through overTheBus.declare: while another holds the lock the declaration is not published, and at the publication the
// lock is held.
func TestASendHoldsTheNodesLockFromThePutToThePublication(t *testing.T) { // 1696/AC9
const node = "lock-send-node"
bus := &aBusThatSays{published: make(chan string, 1)}
bus.heldAtPublish = func() bool { return whileHeld(node) }
send := overTheBus{server: link.SendingOn(bus), signer: aSignerOfZeros{}}
release := lockFilesOf(node) // a sweep is judging this node's bucket
done := make(chan error, 1)
go func() {
_, err := send.declare(t.Context(), readyNode{node: node}, []byte(`{}`))
done <- err
}()
select {
case n := <-bus.published:
t.Fatalf("%s was sent its declaration while the sweep held the node's bucket", n)
case <-time.After(150 * time.Millisecond):
}
release()
select {
case <-bus.published:
case <-time.After(5 * time.Second):
t.Fatal("the send did not go on once the sweep let go")
}
if err := <-done; err == nil || !strings.Contains(err.Error(), "nothing is recorded") {
t.Fatalf("the send did not reach the publication (a lock held at it): %v", err)
}
}
-96
View File
@@ -1,96 +0,0 @@
package main
import (
"bytes"
"context"
"encoding/json"
"fmt"
"strings"
"github.com/novox/mesh-controller/internal/conditions"
"github.com/novox/mesh-controller/internal/link"
)
// **A declaration is one bus message, and the bus's largest message is a cliff** (novox/hq issue 1689). On 2026-10-21
// the control node's declaration passed it by 2 KB and nothing reached that node until two assignments were undone.
// So the size is measured at every send: a warning condition from 75% of the limit (ADR 0486 §8, issue 1698: it was 90% while every file travelled inline), urgent over it, cleared when
// the declaration is back under 75%. The limit is the connection's own, not a constant.
// kindDeclarationNearLimit is a declaration at or near the bus's largest message.
const kindDeclarationNearLimit = "declaration-near-limit"
// declarationRoomSource is the condition's source: the sends.
const declarationRoomSource = "sends"
func declarationRoomObservation(node string, size, limit int64) conditions.Observation {
o := conditions.Observation{Scope: conditions.ScopeMachine, ID: node, Kind: kindDeclarationNearLimit,
Token: kindDeclarationNearLimit, Machine: node, Severity: conditions.Warning, Source: declarationRoomSource}
if size > limit {
o.Severity = conditions.Urgent
}
o.Summary = fmt.Sprintf("%s's declaration is %d bytes on the wire, %d%% of the bus's largest message of %d bytes",
node, size, size*100/limit, limit)
if size > limit {
o.Summary += ": over it, so it cannot be sent"
}
return o
}
func init() {
link.DeclarationMeasured = func(ctx context.Context, node string, size, limit int64) {
settleDeclarationRoom(ctx, conditionsFrom, node, size, limit)
}
}
// settleDeclarationRoom raises or clears the condition of a declaration's size. A nil keeper (a tool, not the
// serving controller) says nothing.
func settleDeclarationRoom(ctx context.Context, keeper *conditions.Keeper, node string, size, limit int64) {
if keeper == nil || limit <= 0 {
return
}
obs := declarationRoomObservation(node, size, limit)
if float64(size) < link.NearTheLimit*float64(limit) {
if _, err := keeper.Clear(ctx, obs.Key(), fmt.Sprintf("%s's declaration is %d bytes, under the limit's margin", node, size)); err != nil {
fmt.Printf("the condition %s could not be cleared: %v\n", obs.Key(), err)
}
return
}
fmt.Printf("NEAR THE LIMIT: %s\n", obs.Summary)
if _, err := keeper.Observe(ctx, obs); err != nil {
fmt.Printf("the condition %s could not be raised: %v\n", obs.Key(), err)
}
}
// compactKept writes the vault's export compact. The composition (the SDK's keptFile) writes it indented, and its 334
// sealed secrets were 164,763 bytes of it against 138,474 compact: a fifth of the declaration's room, in whitespace,
// base64-encoded again on the wire. Every reader parses JSON, so the spacing is nobody's. A resource that is not
// the export, or whose content is not JSON, is left as it is. The resources are copied, never changed in place.
func compactKept(resources []map[string]any) []map[string]any {
out := make([]map[string]any, len(resources))
for i, r := range resources {
out[i] = r
id, _ := r["id"].(string)
path, _ := r["path"].(string)
content, ok := r["content"].(string)
if !ok || (id != "kept" && !strings.HasSuffix(id, ".kept")) || !strings.HasSuffix(path, "/export.json") {
continue
}
var dst bytes.Buffer
if json.Compact(&dst, []byte(content)) != nil {
continue
}
if strings.HasSuffix(content, "\n") {
dst.WriteByte('\n')
}
if dst.String() == content {
continue
}
c := make(map[string]any, len(r))
for k, v := range r {
c[k] = v
}
c["content"] = dst.String()
out[i] = c
}
return out
}
@@ -1,71 +0,0 @@
package main
import (
"strings"
"testing"
"time"
"github.com/novox/mesh-controller/internal/conditions"
)
// The vault's export is sent compact (issue 1689): same document, no indentation, one rewrite.
func TestTheVaultsExportIsSentCompactAndNothingElseIsTouched(t *testing.T) {
indented := "{\n \"export\": 1,\n \"kept\": [\n {\n \"name\": \"a\"\n }\n ]\n}\n"
other := "{\n \"x\": 1\n}\n"
in := []map[string]any{
{"id": "mesh-vault.kept", "type": "file", "path": "/srv/vault/export.json", "mode": "0600", "content": indented},
{"id": "mesh-vault.config", "type": "file", "path": "/srv/vault/config.json", "content": other},
{"id": "other.kept", "type": "file", "path": "/srv/other/notes.json", "content": other},
}
out := compactKept(in)
want := "{\"export\":1,\"kept\":[{\"name\":\"a\"}]}\n"
if out[0]["content"] != want || out[0]["mode"] != "0600" || out[0]["path"] != "/srv/vault/export.json" {
t.Fatalf("the export is %q (%v), expected %q with its mode and path kept", out[0]["content"], out[0], want)
}
if out[1]["content"] != other || out[2]["content"] != other {
t.Fatalf("a file that is not the export was rewritten: %v %v", out[1], out[2])
}
if in[0]["content"] != indented {
t.Fatalf("the composed resource was changed in place")
}
if again := compactKept(out); again[0]["content"] != want {
t.Fatalf("compacting twice changed it: %q", again[0]["content"])
}
}
// From 75% of the bus's largest message (ADR 0486 §8, issue 1698) the declaration is a warning; over it, urgent; back under 75%, cleared.
func TestADeclarationNearTheBusLimitRaisesAConditionAndClearsWhenSmaller(t *testing.T) {
store := conditions.NewInMemory()
k := conditions.NewKeeper(t.Context(), conditions.Options{Store: store, History: store,
Teller: &conditions.Told{}, Now: func() time.Time { return time.Date(2026, 10, 21, 14, 0, 0, 0, time.UTC) }})
t.Cleanup(func() { k.Close(t.Context()) })
const limit = 1000
key := declarationRoomObservation("novox", 0, limit).Key()
read := func() (conditions.Condition, bool) {
t.Helper()
c, found, err := conditions.ReadOne(t.Context(), store, key)
if err != nil {
t.Fatal(err)
}
return c, found
}
settleDeclarationRoom(t.Context(), k, "novox", 749, limit)
if _, found := read(); found {
t.Fatal("a declaration at 74.9% raised a condition")
}
settleDeclarationRoom(t.Context(), k, "novox", 750, limit)
c, found := read()
if !found || c.Severity != conditions.Warning || !strings.Contains(c.Summary, "750 bytes") ||
!strings.Contains(c.Summary, "1000") || c.Subject.Machine != "novox" {
t.Fatalf("at 75%% the condition is %v %+v", found, c)
}
settleDeclarationRoom(t.Context(), k, "novox", 1001, limit)
if c, _ := read(); c.Severity != conditions.Urgent {
t.Fatalf("over the limit the condition is %q, expected urgent", c.Severity)
}
settleDeclarationRoom(t.Context(), k, "novox", 700, limit)
if _, found := read(); found {
t.Fatal("the condition stayed after the declaration fell under 75%")
}
}
-273
View File
@@ -1,273 +0,0 @@
package main
import (
"context"
"encoding/json"
"errors"
"fmt"
"sort"
"strings"
"time"
"github.com/nats-io/nats.go"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// **An object is removed a day after no node's latest declaration names it** (novox/hq ADR 0486 §3, issue 1696), so a node that is
// asleep and receives its latest declaration later still finds every object that declaration names. What each node's latest
// declaration names is read from the bus itself, the last message of that node's subject of the stream NODES: nothing else
// is kept of it. When an object is first found named by none, the time is written in its own metadata (declarationUnnamed);
// found named again, that is erased; found unnamed for a day, it is removed. Each node has a bucket of its own, judged against that
// node's latest declaration alone. Objects are immutable and regenerable from the
// controller's state, so a removal loses nothing.
// declarationUnnamed is the key of an object's metadata holding when it was first found named by no node's latest declaration.
const declarationUnnamed = "unnamed-since"
// sweepEvery is how often the bucket is judged: the removal is exact to within this.
const sweepEvery = time.Hour
// bucketObject is one object of the bucket as the sweep sees it.
type bucketObject struct {
Name string
// UnnamedSince is when it was first found named by none, zero when it was named at the last look.
UnnamedSince time.Time
}
// sweptBucket is what the sweep needs of the bucket.
type sweptBucket interface {
Objects(ctx context.Context) ([]bucketObject, error)
MarkUnnamed(ctx context.Context, name string, since time.Time) error
ClearUnnamed(ctx context.Context, name string) error
// Delete removes the object if it still carries the mark it was judged by, and says whether it did: a send that put it again
// since has erased the mark, and the object is left.
Delete(ctx context.Context, name string, markedAt time.Time) (bool, error)
}
// filesSweep says what a sweep did.
type filesSweep struct{ Kept, Marked, Cleared, Removed int }
// sweepDeclarationFiles judges every object against the names of the latest declarations: named ones are kept (and a
// mark of being unnamed is erased), unnamed ones are marked at the first look and removed once the mark is a day old.
func sweepDeclarationFiles(ctx context.Context, bucket sweptBucket, named map[string]bool, now time.Time) (filesSweep, error) {
var r filesSweep
objects, err := bucket.Objects(ctx)
if err != nil {
return r, err
}
sort.Slice(objects, func(i, j int) bool { return objects[i].Name < objects[j].Name })
for _, o := range objects {
switch {
case named[o.Name]:
r.Kept++
if !o.UnnamedSince.IsZero() {
if err := bucket.ClearUnnamed(ctx, o.Name); err != nil {
return r, fmt.Errorf("the object %s: %w", o.Name[:12], err)
}
r.Cleared++
}
case o.UnnamedSince.IsZero():
if err := bucket.MarkUnnamed(ctx, o.Name, now); err != nil {
return r, fmt.Errorf("the object %s: %w", o.Name[:12], err)
}
r.Marked++
case now.Sub(o.UnnamedSince) >= broker.DeclarationFilesKeptFor:
gone, err := bucket.Delete(ctx, o.Name, o.UnnamedSince)
if err != nil {
return r, fmt.Errorf("the object %s: %w", o.Name[:12], err)
}
if gone {
r.Removed++
} else {
r.Kept++
}
default:
r.Kept++
}
}
return r, nil
}
// namedByLatest is the name of every object this latest declaration of a node names. A body that cannot be read is an error, never
// "names none": an object the node still needs would be marked for removal.
func namedByLatest(node string, signed []byte) (map[string]bool, error) {
var s link.Signed
if err := json.Unmarshal(signed, &s); err != nil {
return nil, fmt.Errorf("%s's latest declaration is not readable: %w", node, err)
}
names, err := namedObjects(s.Declaration)
if err != nil {
return nil, fmt.Errorf("%s's latest declaration is not readable: %w", node, err)
}
named := make(map[string]bool, len(names))
for name := range names {
named[name] = true
}
return named, nil
}
// latestDeclarations are the last signed declarations of the nodes, by node, as the stream NODES holds them. A node with none has
// no entry.
func latestDeclarations(ctx context.Context, conn *nats.Conn, nodes []string) (map[string][]byte, error) {
js, err := conn.JetStream(nats.Context(ctx))
if err != nil {
return nil, err
}
out := map[string][]byte{}
for _, node := range nodes {
msg, err := js.GetLastMsg("NODES", link.DeclareSubject(node), nats.Context(ctx))
if errors.Is(err, nats.ErrMsgNotFound) {
continue
}
if err != nil {
return nil, fmt.Errorf("the latest declaration of %s: %w", node, err)
}
out[node] = msg.Data
}
return out, nil
}
// busSweptBucket is the bucket on the controller's connection.
type busSweptBucket struct{ store jetstream.ObjectStore }
func (b busSweptBucket) Objects(ctx context.Context) ([]bucketObject, error) {
infos, err := b.store.List(ctx)
if errors.Is(err, jetstream.ErrNoObjectsFound) {
return nil, nil
}
if err != nil {
return nil, err
}
out := make([]bucketObject, 0, len(infos))
for _, info := range infos {
o := bucketObject{Name: info.Name}
if at, err := time.Parse(time.RFC3339, info.Metadata[declarationUnnamed]); err == nil {
o.UnnamedSince = at
}
out = append(out, o)
}
return out, nil
}
func (b busSweptBucket) meta(ctx context.Context, name string, change func(map[string]string)) error {
info, err := b.store.GetInfo(ctx, name)
if err != nil {
return err
}
meta := info.ObjectMeta
kept := map[string]string{}
for k, v := range meta.Metadata {
kept[k] = v
}
change(kept)
meta.Metadata = kept
return b.store.UpdateMeta(ctx, name, meta)
}
func (b busSweptBucket) MarkUnnamed(ctx context.Context, name string, since time.Time) error {
return b.meta(ctx, name, func(m map[string]string) { m[declarationUnnamed] = since.UTC().Format(time.RFC3339) })
}
func (b busSweptBucket) ClearUnnamed(ctx context.Context, name string) error {
return b.meta(ctx, name, func(m map[string]string) { delete(m, declarationUnnamed) })
}
func (b busSweptBucket) Delete(ctx context.Context, name string, markedAt time.Time) (bool, error) {
info, err := b.store.GetInfo(ctx, name)
if errors.Is(err, jetstream.ErrObjectNotFound) {
return false, nil
}
if err != nil {
return false, err
}
if at, perr := time.Parse(time.RFC3339, info.Metadata[declarationUnnamed]); perr != nil || !at.Equal(markedAt) {
return false, nil
}
return true, b.store.Delete(ctx, name)
}
// sweepingDeclarationFiles judges the bucket every sweepEvery for as long as this controller serves and holds the lease.
func sweepingDeclarationFiles(ctx context.Context, inv *inventory.Inventory, conn *nats.Conn) {
tick := time.NewTicker(sweepEvery)
defer tick.Stop()
failing := ""
for {
select {
case <-ctx.Done():
return
case <-tick.C:
}
if epoch, err := theLease.epoch(ctx); err != nil || epoch == 0 {
continue
}
why := sweepOnce(ctx, inv, conn)
if why != failing {
if why != "" {
fmt.Printf("the bucket of declaration files cannot be swept: %s\n", why)
} else {
fmt.Printf("the bucket of declaration files is swept again\n")
}
failing = why
}
}
}
// sweepOnce judges the bucket of every node against that node's latest declaration, one bucket at a time under the node's lock, with
// the declaration read again under it: a send that put an object and published its declaration is seen, and one that has not yet
// is waiting for the lock.
func sweepOnce(ctx context.Context, inv *inventory.Inventory, conn *nats.Conn) string {
nodes, err := inv.Nodes(ctx)
if err != nil {
return fmt.Sprintf("the nodes cannot be read: %v", err)
}
var problems []string
var total filesSweep
for _, n := range nodes {
r, err := sweepNode(ctx, conn, n.Name)
total.Kept, total.Marked, total.Cleared, total.Removed = total.Kept+r.Kept, total.Marked+r.Marked, total.Cleared+r.Cleared, total.Removed+r.Removed
if err != nil {
problems = append(problems, fmt.Sprintf("%s: %v", n.Name, err))
}
}
if total.Removed+total.Marked+total.Cleared > 0 {
fmt.Printf("the buckets of declaration files: %d object(s) kept, %d found unnamed, %d named again, %d removed\n",
total.Kept, total.Marked, total.Cleared, total.Removed)
}
return strings.Join(problems, "; ")
}
func sweepNode(ctx context.Context, conn *nats.Conn, node string) (filesSweep, error) {
working, cancel := context.WithTimeout(ctx, 2*time.Minute)
defer cancel()
js, err := jetstream.New(conn)
if err != nil {
return filesSweep{}, err
}
bucket, err := broker.DeclarationFilesBucketOf(node)
if err != nil {
return filesSweep{}, fmt.Errorf("the bucket of node %q: %w", node, err)
}
store, err := js.ObjectStore(working, bucket)
if errors.Is(err, jetstream.ErrBucketNotFound) {
return filesSweep{}, nil // a node never sent a reference has no bucket yet
}
if err != nil {
return filesSweep{}, fmt.Errorf("the bucket cannot be read: %w", err)
}
defer lockFilesOf(node)()
latest, err := latestDeclarations(working, conn, []string{node})
if err != nil {
return filesSweep{}, err
}
named := map[string]bool{}
if body, ok := latest[node]; ok {
if named, err = namedByLatest(node, body); err != nil {
return filesSweep{}, err
}
}
return sweepDeclarationFiles(working, busSweptBucket{store}, named, time.Now())
}
@@ -1,89 +0,0 @@
//go:build integration
package main
import (
"encoding/json"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
)
// 1840/AC5 (ADR 0507 §5): the source of the declared-packages mirror composes each node and names, for a package two modules
// declare and one declared absent, the modules that declare it; the same records give the same bytes on a second pass, so
// the mirror writes nothing; a node with no package has a key that says so.
func TestTheDeclaredPackagesMirrorNamesWhatEachNodesModulesDeclare(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
register(t, open, catalogue.Manifest{Module: "viewer", Version: "1", Resources: []map[string]any{
{"id": "pkg-curl", "type": "package", "package": "curl"},
{"id": "pkg-nano", "type": "package", "package": "nano", "absent": true}}})
register(t, open, catalogue.Manifest{Module: "fetcher", Version: "1", Resources: []map[string]any{
{"id": "pkg-curl", "type": "package", "package": "curl"}}})
for _, module := range []string{"viewer", "fetcher"} {
if _, err := assign(ctx, open, "anchor", module); err != nil {
t.Fatalf("assign anchor %s: %v", module, err)
}
}
source := declaredPackagesMirror(open)
first, err := source(ctx)
if err != nil {
t.Fatal(err)
}
var got inventory.DeclaredPackagesMirrored
if err := json.Unmarshal(first["anchor"], &got); err != nil {
t.Fatalf("anchor has no value: %v", err)
}
// Every node also runs the mesh's own modules, which declare packages of their own: curl and nano are looked
// up by name, and no package names viewer or fetcher but those they declare.
byName := map[string]inventory.PackageDeclared{}
for _, p := range got.Packages {
byName[p.Package] = p
}
same := func(a, b []string) bool {
if len(a) != len(b) {
return false
}
for i := range a {
if a[i] != b[i] {
return false
}
}
return true
}
for _, p := range []inventory.PackageDeclared{{Package: "curl", Modules: []string{"fetcher", "viewer"}},
{Package: "nano", Modules: []string{"viewer"}, Absent: true}} {
if g, ok := byName[p.Package]; !ok || g.Absent != p.Absent || !same(g.Modules, p.Modules) {
t.Errorf("anchor declares %s as %+v, want %+v (all: %+v)", p.Package, g, p, got.Packages)
}
}
for _, p := range got.Packages {
if p.Package == "curl" || p.Package == "nano" {
continue
}
for _, m := range p.Modules {
if m == "viewer" || m == "fetcher" {
t.Errorf("%s is said to be declared by %s, which does not declare it", p.Package, m)
}
}
}
var laptop inventory.DeclaredPackagesMirrored
if err := json.Unmarshal(first["laptop"], &laptop); err != nil || laptop.Node != "laptop" || laptop.Packages == nil {
t.Errorf("a node assigned neither module has no value of its own: %s (%v)", first["laptop"], err)
}
for _, p := range laptop.Packages {
if p.Package == "curl" || p.Package == "nano" {
t.Errorf("laptop is assigned neither module and declares %+v", p)
}
}
second, err := source(ctx)
if err != nil {
t.Fatal(err)
}
for node, value := range first {
if string(second[node]) != string(value) {
t.Errorf("the value of %s changed between two passes over the same records:\n%s\n%s", node, value, second[node])
}
}
}
+153 -15
View File
@@ -815,14 +815,16 @@ func deliveryPlanDocument(m orderMember, entries []inventory.Entry, read map[str
// delivery walks [-n 50] [--plan <id>]
func deliveryCommand(ctx context.Context, args []string) error {
if len(args) == 0 {
return errors.New("delivery order|check|go|stop|walks")
return errors.New("delivery plan|order|check|go|stop|walks")
}
sub, rest := args[0], args[1:]
if sub == "plan" {
// Removed (ADR 0450 §8; issue 1318): refused by name before any flag is read, naming its replacement.
return removedByName("delivery-plan")
}
set := flag.NewFlagSet("delivery "+sub, flag.ContinueOnError)
repository := set.String("repository", "", "owner/repository")
base := set.String("base", "main", "the branch it merges into")
head := set.String("head", "", "the commit at hand")
paths := set.String("paths", "", "the files it changes, comma-separated")
moduleDirs := set.String("module-dirs", "", "the directories holding a module.json at the head, comma-separated")
removed := set.String("removed", "", "the files it deletes, comma-separated")
membersJSON := set.String("members", "", "a group's members, as JSON")
membersFile := set.String("members-file", "", "a file of a group's members, as JSON: the same as --members, for a list "+
"too long for an argument (novox/hq issue 1272)")
@@ -870,6 +872,19 @@ func deliveryCommand(ctx context.Context, args []string) error {
return enc.Encode(v)
}
switch sub {
case "plan":
if *repository == "" || *head == "" && *paths == "" {
return errors.New("delivery plan --repository owner/repo --head <commit> --paths a,b")
}
m := changeMember(*repository, *base, *head, splitList(*paths), splitList(*removed), moduleDirsOf(*moduleDirs))
entries, read, edges, err := theGraph(ctx, inv)
if err != nil {
return err
}
return answer(deliveryPlanDocument(m, entries, read, edges, func(module string) (inventory.Upgrade, bool) {
u, err := inv.UpgradeOf(ctx, module)
return u, err == nil
}))
case "order":
if len(members) == 0 {
return errors.New("delivery order --members <json>: a group has members")
@@ -899,9 +914,11 @@ func deliveryCommand(ctx context.Context, args []string) error {
if *group == "" || len(members) < 2 {
return errors.New("delivery check --group <id> --members <json> (two heads or more), or --members <one head>")
}
// **The group's composed check is the checker's** (novox/hq ADR 0503 §3; issue 1299, C2): nothing is asked of
// build-agent, and the answer says who judges it.
return answer(groupCheckAnswer(*group))
id, err := askGroupCheck(ctx, open, *group, members)
if err != nil {
return err
}
return answer(map[string]any{"asked": id, "group": *group})
case "go":
if len(positionals) != 1 {
return errors.New("delivery go <plan> [--by <who>] [--why <text>]")
@@ -1076,6 +1093,38 @@ func theGraph(ctx context.Context, inv *inventory.Inventory) ([]inventory.Entry,
return entries, read, edges, nil
}
// cycleVerdict is the failing verdict of a group whose modules are ordered in a cycle, and whether there is one: the
// group's composed check, said without a build. The manifests' change is read from the forge; an answer it cannot give
// is an error, never read as no cycle.
func cycleVerdict(ctx context.Context, group string, members []orderMember, reach map[string]orderReach,
edges []inventory.Edge, entries []inventory.Entry) (link.Checked, bool, error) {
changes, err := changesOf(ctx, changeSourcesOf(members, entries, reach))
if err != nil {
return link.Checked{}, false, fmt.Errorf("what the group's members say of their manifests cannot be read: %w", err)
}
var moved []string
for _, m := range members {
moved = appendNew(moved, reach[m.ID].Moved...)
}
said := orderEdges(moved, changes, edges).CycleSaid()
if said == "" {
return link.Checked{}, false, nil
}
first := members[0].pull()
c := link.Checked{ID: link.NewBuildID(time.Now()), Owner: first.Owner, Repo: first.Repo, Number: first.Number,
Commit: first.Commit, Verdict: "fail", Summary: said, Report: said, Group: group,
Gate: &link.CheckLayer{Verdict: "fail", Summary: said}}
for _, m := range members {
if group == "" {
break // a lone pull request's verdict names no group and no member
}
p := m.pull()
c.Members = append(c.Members, link.CheckedMember{Owner: p.Owner, Repo: p.Repo, Number: p.Number, Commit: p.Commit,
Trunk: twelve(p.BaseSHA)})
}
return c, true, nil
}
// twelve is a commit as twelve characters.
func twelve(c string) string {
if len(c) > 12 {
@@ -1084,6 +1133,93 @@ func twelve(c string) string {
return c
}
// groupPrimary is the head a group's composed check is run from: the one moving the controller, which then
// judges the group by itself; else the one moving the node-engine, whose validator judges; else the first.
func groupPrimary(members []orderMember, reach map[string]orderReach) int {
for _, want := range []string{"mesh-controller", "mesh-host"} {
for i, m := range members {
if slices.Contains(reach[m.ID].Moved, want) {
return i
}
}
}
return 0
}
// askGroupCheck asks the build seat for a group's composed check: every head laid over the mesh in turn,
// judged as one future state (novox/hq ADR 0239). The verdict comes back as `checked` with the group's id,
// for mesh-delivery; the forge's holder sets no status from it.
func askGroupCheck(ctx context.Context, open *stores, group string, members []orderMember) (string, error) {
entries, read, edges, err := theGraph(ctx, open.inventory)
if err != nil {
return "", err
}
reach := reachOfMembers(members, entries, read, edges)
// **A cycle of edges fails the group's composed check, naming every module and edge** (novox/hq ADR 0419 clause 6),
// before the build seat is asked: no composition of the heads has an order to be sent in.
if failing, cyclic, err := cycleVerdict(ctx, group, members, reach, edges, entries); err != nil {
return "", err
} else if cyclic {
sayChecked(ctx, os.Stderr, failing)
return failing.ID, nil
}
primary := groupPrimary(members, reach)
p := members[primary].pull()
scope := pullScope(p, entries, read, edges)
// The gate runs over what any member moves; a judge from the primary alone.
for _, m := range members {
r := reach[m.ID]
scope.Modules = appendNew(scope.Modules, r.Moved...)
scope.Dependents = appendNew(scope.Dependents, r.Dependents...)
}
request, err := checkRequestFor(ctx, open, p, scope, entries)
if err != nil {
return "", err
}
request.Check.Group = group
// A group's check is queued as ready when any member's pull request is marked ready.
request.Check.Ready = slices.ContainsFunc(members, func(m orderMember) bool { return m.Ready })
world, err := restOfTheMesh(ctx, open, shelfOf(entries), "")
if err != nil {
return "", err
}
for i, m := range members {
if i == primary {
continue
}
mp := m.pull()
// As the mesh clones a module built from it; a repository no module is built from, from the forge.
source := inventory.Source{Seat: gitSeat, Repository: mp.Owner + "/" + mp.Repo}
for _, e := range entries {
if !e.Provided && sameRepository(e.Source.Repository, link.SourceMoved{Owner: mp.Owner, Repo: mp.Repo}) {
source = e.Source
break
}
}
url := source.Repository
if source.Seat != "" {
url, err = clonedFromSeat(world, source.Seat, source.Repository)
}
if err != nil {
return "", fmt.Errorf("%s cannot be cloned for the group's check: %w", m.ID, err)
}
request.Check.Members = append(request.Check.Members, link.GroupMember{Owner: mp.Owner, Repo: mp.Repo,
Number: mp.Number, Repository: url, Ref: mp.Commit, Paths: mp.Paths, Base: mp.Base})
}
seat := buildSeatHeld(ctx)
ask, err := askOverOn(seat)
if err != nil {
return "", err
}
defer ask.Close()
if err := ask.Ask(ctx, request); err != nil {
return "", err
}
fmt.Fprintf(os.Stderr, "group %s: asked %s to check %d head(s) composed together, as %s\n", group, seat,
len(members), request.ID)
return request.ID, nil
}
// appendNew appends what is not there yet.
func appendNew(to []string, items ...string) []string {
for _, i := range items {
@@ -1094,6 +1230,15 @@ func appendNew(to []string, items ...string) []string {
return to
}
// shelfOf is the catalogue's manifests by name.
func shelfOf(entries []inventory.Entry) map[string]catalogue.Manifest {
shelf := map[string]catalogue.Manifest{}
for _, e := range entries {
shelf[e.Manifest.Module] = e.Manifest
}
return shelf
}
// sayPlanMoved says a walk kept in a new state as the controller's `plan-moved`, the plan whole: what the
// delivery it walks reads its steps from. Never in the way of the walk: said beside it, and a save that could
// not be said is logged — the delivery's owner, which also asks for the walks it follows, finds it by
@@ -1257,10 +1402,3 @@ func recheckOneHead(member string, answer func(any) error, ask func() (string, e
}
return answer(recheckAnswer(asked, member))
}
// groupCheckAnswer is what `delivery check` of a group answers from R4e: who judges it, and that nothing was asked
// (novox/hq ADR 0503 §3; issue 1299, C2).
func groupCheckAnswer(group string) map[string]any {
return map[string]any{"asked": "", "group": group, "judged-by": checkerModule,
"said": "the checker judges a delivery group's heads merged together (ADR 0503 §3): nothing was asked"}
}
@@ -41,7 +41,7 @@ func TestSayCheckedWritesTheSameLineToItsWriter(t *testing.T) {
var w bytes.Buffer
sayChecked(context.Background(), &w, link.Checked{ID: "c1", Owner: "novox", Repo: "x", Number: 7, Commit: "0123456789abcdef",
Verdict: "pass", Summary: "fine"})
want := "c1: novox/x#7 at 01234567 checked on a machine that did not say which: repository none\n"
want := "c1: novox/x#7 at 01234567 checked on a machine that did not say which: gate PASS — fine; repository none\n"
if w.String() != want {
t.Fatalf("sayChecked said %q; want %q", w.String(), want)
}
@@ -58,17 +58,8 @@ type stalledLine struct {
// Severity is "urgent" for a group held or half merged and a refused member (novox/hq issues 1275, 1278); empty
// or anything else is a warning, as before.
Severity string `json:"severity,omitempty"`
// Needs is "new-head" for a delivery whose cure is its author's new head (novox/hq ADR 0505 §5), Author that author
// as the operator reads it ("the author of feat/x", with the session that pushed it when its pull request names one)
// and Branch its branch. A mesh-delivery from before says none, and the words are as before.
Needs string `json:"needs,omitempty"`
Author string `json:"author,omitempty"`
Branch string `json:"branch,omitempty"`
}
// needsNewHead is what a stalled line needs when its cure is its author's new head (ADR 0505 §5).
const needsNewHead = "new-head"
// severity is the condition's severity a stalled line asks: urgent when it says so, a warning otherwise.
func (l stalledLine) severity() conditions.Severity {
if l.Severity == string(conditions.Urgent) {
@@ -181,3 +181,32 @@ func TestATruncatedChangeIsReadAsChangingTheManifestsOfWhatItMoves(t *testing.T)
t.Fatalf("the member's change is read as %+v", src)
}
}
// A lone pull request's cycle fails its check, naming every module and edge, with no group and no members.
func TestALonePullRequestInACycleFailsWithNoGroup(t *testing.T) {
was, wasParent := manifestAt, parentOf
t.Cleanup(func() { manifestAt, parentOf = was, wasParent })
manifestsReadMu.Lock()
clear(manifestsRead)
manifestsReadMu.Unlock()
const head = "3333333333333333333333333333333333333333"
parentOf = func(context.Context, string, string, string) (string, error) { return "", nil }
manifestAt = func(_ context.Context, _, _, file, ref string) (catalogue.Manifest, bool, error) {
if ref == "main" {
return catalogue.Manifest{}, false, nil
}
if file == "modules/x/module.json" {
return catalogue.Manifest{Module: "mx", DefinesSeats: []catalogue.SeatDeclaration{{Name: "s", Serves: []catalogue.Verb{{Name: "a"}}}},
Invokes: []string{"seat:s.b"}}, true, nil
}
return catalogue.Manifest{Module: "my", DefinesSeats: []catalogue.SeatDeclaration{{Name: "s", Serves: []catalogue.Verb{{Name: "b"}}}},
Invokes: []string{"seat:s.a"}}, true, nil
}
m := orderMember{ID: "novox/r@" + head[:12], Repository: "novox/r", Base: "main", Head: head, Number: 5,
Paths: []string{"modules/x/module.json", "modules/y/module.json"}}
c, cyclic, err := cycleVerdict(context.Background(), "", []orderMember{m}, map[string]orderReach{m.ID: {Moved: []string{"mx", "my"}}}, nil, nil)
if err != nil || !cyclic || c.Verdict != "fail" || c.Group != "" || len(c.Members) != 0 || !strings.Contains(c.Summary, "s.a") ||
!strings.Contains(c.Summary, "my") {
t.Fatalf("the verdict is %+v (cyclic %v, %v)", c, cyclic, err)
}
}
@@ -59,18 +59,6 @@ func TestAMergeWaitsForItsDeliverysWordAndThePersonsWordWorksWithoutIt(t *testin
}
}
// heardOf has this mesh's own view of the plans take in the plan as the bucket holds it. A verb of the
// controller (deliveryCommand, plansCommand) opens an inventory of its own and writes the plan there; this
// mesh's inventory learns of the write from the bucket's watch, a moment later, and a loaded machine makes the
// moment long enough for advanceHeld to read the plan as it was, still waiting for its word, and ask nothing
// (issue 1399). Reading the plan from the bucket puts it into the view at once, with no wait to bound.
heardOf := func(id string) {
t.Helper()
if _, err := inv.PlanByID(ctx, id); err != nil {
t.Fatal(err)
}
}
// No holder on record: the merge starts its walk, as before.
p := merge("c1aaaaaaaa", "modules/app/index.ts")
if p.Waiting() || len(*asked) != 1 {
@@ -102,7 +90,6 @@ func TestAMergeWaitsForItsDeliverysWordAndThePersonsWordWorksWithoutIt(t *testin
if err := deliveryCommand(ctx, []string{"go", p.ID}); err == nil {
t.Fatal("a walk was let go twice")
}
heardOf(p.ID)
advanceHeld(ctx, open)
if len(*asked) != 2 {
t.Fatalf("the word did not start the walk: %v", *asked)
@@ -132,7 +119,6 @@ func TestAMergeWaitsForItsDeliverysWordAndThePersonsWordWorksWithoutIt(t *testin
if err := plansCommand(ctx, []string{"go", p.ID, "--why", "mesh-delivery is down"}); err != nil {
t.Fatal(err)
}
heardOf(p.ID)
advanceHeld(ctx, open)
got, _ = inv.PlanByID(ctx, p.ID)
if got.Waiting() || !strings.HasPrefix(got.Delivery.By, "a person") || len(*asked) < 4 {
+2
View File
@@ -138,6 +138,8 @@ func TestTheDeliveryVerbsComposeTheirCommands(t *testing.T) {
{"delivery-order", map[string]any{"members": "[]"}, []string{"delivery", "order", "--members", "[]"}},
{"delivery-check", map[string]any{"group": "feat/x", "members": "[]"},
[]string{"delivery", "check", "--group", "feat/x", "--members", "[]"}},
{"delivery-plan", map[string]any{"repository": "novox/a", "paths": "x", "head": "c0"},
[]string{"delivery", "plan", "--repository", "novox/a", "--paths", "x", "--head", "c0"}},
{"plans", map[string]any{"go": "plan-1", "why": "down"}, []string{"plans", "go", "plan-1", "--why", "down"}},
} {
got, err := argvFor(c.verb, c.args)
+10
View File
@@ -35,6 +35,16 @@ func TestAMergeOfABaseTakesWhatStandsOnItAlong(t *testing.T) {
t.Fatalf("%s was rebuilt and stands on nothing that moved (%v)", n, names)
}
}
// The dependents come in base order when the merge orders them: the runtime, then shop, then
// the plugin that stands on shop.
ordered := orderByBases(append([]inventory.Entry{entry("mesh-tools")}, got...), against)
pos := map[string]int{}
for i, e := range ordered {
pos[e.Manifest.Module] = i
}
if !(pos["mesh-tools"] < pos["shop"] && pos["shop"] < pos["shop-plugin"]) {
t.Fatalf("not in base order: %v", ordered)
}
// Nothing moved: nothing follows.
if more := dependentsOf(nil, entries, against); len(more) != 0 {
t.Fatalf("with nothing moved, %d module(s) were rebuilt", len(more))
@@ -0,0 +1,84 @@
//go:build integration
package main
import (
"context"
"errors"
"testing"
"time"
"git.novox.be/novox/mesh-sdk/go/asks"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/identity"
"github.com/novox/mesh-controller/internal/link"
)
// aForgeAndABuildSeat gives a test what a real ask of a build needs: the forge holding the git seat, assigned to anchor, so
// the clone URL composes; the identity store; the build seat's queue on a real bus; and the environment of the serving controller, which is
// never the controller's terminal (a verb's, in effect: startedAtTheTerminal is false).
func aForgeAndABuildSeat(t *testing.T, open *stores) *broker.JetStream {
t.Helper()
js, _ := aBuildQueue(t)
// The ask signs what it sends with the controller's identity, as the serving controller does.
identity.ForTest(t)
forge, err := catalogue.ParseManifest([]byte(`{"module":"gitea","version":"1",
"claims":[{"name":"git","scope":"mesh"}],"provides":[{"name":"git","scope":"mesh"}],
"serves":{"git":{"scheme":"http","port":3000}}}`))
if err != nil {
t.Fatal(err)
}
register(t, open, forge)
if _, err := assign(t.Context(), open, "anchor", "gitea"); err != nil {
t.Fatal(err)
}
t.Setenv(servedVar, "1")
return js
}
// 1146/AC9, through the real ask (askABuild, buildOneAsked, verbMayAsk) and a real bus: the first build of a module a
// merge adds on the operator's Approve is queued on the build seat; the same ask without that warrant, and the warrant's
// ask for a path the merge does not hold, are refused by the guard of a build through a verb.
func TestTheApprovedFirstBuildPassesTheGuardOfAVerbAndIsQueued(t *testing.T) {
const head, commit = "a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1", "0177ea7a00000000"
open := aMesh(t)
ctx := t.Context()
js := aForgeAndABuildSeat(t, open)
a := aDeliveryMergeOf(t, head, map[string]string{"novox/mesh-delivery@a1a1a1a1a1a1": head})
wasKept, wasAsk := keptDeliveryMerge, routerAskOf
keptDeliveryMerge = func(context.Context, string, string, int, string) (deliveryMerge, bool, error) {
return a.merge, true, nil
}
routerAskOf = func(context.Context, catalogueReader, string, string) (*asks.Ask, *asks.Warrant, error) {
return &a.ask, &a.warrant, nil
}
t.Cleanup(func() { keptDeliveryMerge, routerAskOf = wasKept, wasAsk })
// Without the warrant the guard refuses, whatever the repository holds.
_, err := askABuild(ctx, buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "", "main")
var policy *heldAtTheTerminal
if !errors.As(err, &policy) {
t.Fatalf("1146/AC9: a first build asked without the warrant was not refused: %v", err)
}
// And a path the merge does not hold is refused on the warrant.
warranted := withFirstBuildWarrant(ctx, "novox/mesh-delivery", []string{""}, "the Approve")
if _, err := askABuild(warranted, buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "other", "main"); !errors.As(err, &policy) {
t.Fatalf("1146/AC9: the warrant covered a path the merge does not hold: %v", err)
}
m := link.SourceMoved{Owner: "novox", Repo: "mesh-delivery", Base: "main", HeadSHA: head, Commit: commit, Number: 7,
MergedAt: time.Now().Add(-time.Hour).UTC().Format(time.RFC3339Nano), Paths: []string{"module.json", "main.go"},
ModuleDirs: []string{"."}, ModuleDirsSaid: true}
if err := (following{open: open}).SourceMoved(ctx, m); err != nil {
t.Fatal(err)
}
queue, err := link.ReadQueue(ctx, js, link.TheBuildMachine)
if err != nil {
t.Fatal(err)
}
if len(queue.Asks) != 1 {
t.Fatalf("1146/AC9: the build queue holds %+v, want the one first build", queue)
}
}
@@ -0,0 +1,29 @@
package main
import "testing"
// 1146/AC9: the warrant the controller checked covers exactly the repository on the git seat and the module paths the
// merge holds; nothing else passes verbMayAsk on it.
func TestTheFirstBuildWarrantCoversOnlyItsRepositoryAndItsModulePaths(t *testing.T) {
ctx := withFirstBuildWarrant(t.Context(), "novox/mesh-delivery", []string{"", "view"}, "the Approve")
w := firstBuildWarrantIn(ctx)
for _, c := range []struct {
source buildSource
path string
covers bool
}{
{buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "", true},
{buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "/view/", true},
{buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "other", false},
{buildSource{Repository: "agent/mesh-delivery", Seat: "git"}, "", false},
{buildSource{Repository: "novox/mesh-delivery"}, "", false},
{buildSource{Repository: "novox/mesh-delivery", Seat: "elsewhere"}, "", false},
} {
if got := w.covers(c.source, c.path); got != c.covers {
t.Errorf("1146/AC9: %v at %q covered = %v, want %v", c.source, c.path, got, c.covers)
}
}
if firstBuildWarrantIn(t.Context()).covers(buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "") {
t.Error("1146/AC9: a context with no warrant covers an ask")
}
}
+21 -77
View File
@@ -5,7 +5,6 @@ import (
"encoding/json"
"errors"
"fmt"
"regexp"
"strings"
"time"
@@ -47,7 +46,7 @@ func (f following) firstBuilds(ctx context.Context, m link.SourceMoved, entries
case len(added) == 0:
return false, nil
}
if err := askFirstBuilds(ctx, f.open, m, now, added, why); err != nil {
if err := askFirstBuilds(ctx, inv, m, now, added, why); err != nil {
return true, notNow(err)
}
return true, nil
@@ -68,35 +67,27 @@ func keepFirstMerge(ctx context.Context, inv *inventory.Inventory, m link.Source
return err
}
// askFirstBuilds keeps the merge as answered and registers its modules from the builder's done keys (novox/hq ADR 0501
// §3, §6): nothing asks their first builds. The merge kept is the registration's warrant (approvedMove reads it by the
// commit the indexer's trunk value names); a module whose key is not done yet is registered when the key's write wakes
// the take-in (new_modules.go).
func askFirstBuilds(ctx context.Context, open *stores, m link.SourceMoved, now time.Time, added []string,
// askFirstBuilds keeps the merge, asks the builds, and marks the merge as no longer waiting for them.
func askFirstBuilds(ctx context.Context, inv *inventory.Inventory, m link.SourceMoved, now time.Time, added []string,
why string) error {
inv := open.inventory
repository := m.Owner + "/" + m.Repo
if err := keepFirstMerge(ctx, inv, m, now, false); err != nil {
return err
}
fmt.Printf("%s merged into %s (%.8s) on the operator's Approve (%s): the mesh builds nothing from it yet, so %s "+
"are registered from the builder's done keys, asking no build\n", repository, m.Base, m.Commit, why,
strings.Join(added, ", "))
// Answered: it no longer waits, so the word's arrival again does nothing twice.
if err := inv.AnswerMerge(ctx, repository, m.Commit, inventory.WaitingRecord, false); err != nil {
return err
fmt.Printf("%s merged into %s (%.8s) on the operator's Approve (%s): the mesh builds nothing from it yet, so the "+
"first build of %s is asked\n", repository, m.Base, m.Commit, why, strings.Join(added, ", "))
// The ask carries the warrant just checked, for this repository and these module paths alone: the guard of a build
// through a verb (verbMayAsk) lets exactly that through.
paths := make([]string, 0, len(added))
for _, dir := range added {
if dir == "." {
dir = ""
}
paths = append(paths, dir)
}
modules, err := newModulesOf(ctx, repository)
switch {
case err != nil:
// Said, not failed: the merge is answered, and the look at every module when a controller starts, or the next
// write of the builder or the indexer for one of them, registers them.
fmt.Printf(" the indexer's modules of %s cannot be read now (%v): they are registered when the builder's next "+
"write of one wakes the take-in\n", repository, err)
case len(modules) > 0:
inBackground(func() { takeNewModules(context.WithoutCancel(ctx), open, modules...) })
}
return nil
askBuilds(withFirstBuildWarrant(ctx, repository, paths, why), inv, m, buildSource{Repository: repository, Seat: gitSeat}, added)
// Answered: it no longer waits, so the word's arrival again asks nothing twice.
return inv.AnswerMerge(ctx, repository, m.Commit, inventory.WaitingRecord, false)
}
// resumeFirstBuilds runs firstBuilds for the first merges waiting for the word mesh-delivery just announced for a pull
@@ -111,7 +102,7 @@ func (f following) resumeFirstBuilds(ctx context.Context, d deliveryMerge) error
}
for _, w := range waiting {
k := announcedOf(w)
if !k.FirstBuild || len(k.Held) > 0 || !strings.EqualFold(k.Owner+"/"+k.Repo, d.Owner+"/"+d.Repo) || k.Number != d.Number ||
if !k.FirstBuild || !strings.EqualFold(k.Owner+"/"+k.Repo, d.Owner+"/"+d.Repo) || k.Number != d.Number ||
k.HeadSHA != d.HeadSHA {
continue
}
@@ -134,7 +125,7 @@ func (f following) resumeFirstBuilds(ctx context.Context, d deliveryMerge) error
case err != nil:
return notNow(err)
case len(added) > 0:
if err := askFirstBuilds(ctx, f.open, k.SourceMoved, time.Now().UTC(), added, why); err != nil {
if err := askFirstBuilds(ctx, inv, k.SourceMoved, time.Now().UTC(), added, why); err != nil {
return notNow(err)
}
}
@@ -181,29 +172,6 @@ func firstBuildsOf(ctx context.Context, inv catalogueReader, m link.SourceMoved,
return nil, "", err
}
// heldFirstBuild says a first build was refused because its manifest uses a word the running build-agent does not read,
// and its merge is kept waiting for a build-agent that reads it (novox/hq ADR 0482 §5, issue 1173). Read from a merge kept
// before R4e, never written since: nothing asks a first build (ADR 0501 §6), and a controller that starts answers such a
// merge and registers its modules from their keys (answerHeldMerges).
type heldFirstBuild struct {
Module string `json:"module"`
Path string `json:"path"`
Word string `json:"word"`
At time.Time `json:"at"`
}
// unknownWord is how the strict reading of a manifest says a word is not one it knows: encoding/json's own words, which
// the build-agent's answer carries whole. The build's answer has no typed field for it (Failed is the builder's words).
var unknownWord = regexp.MustCompile(`json: unknown field "([^"]+)"`)
// refusedWord is the word a build's failure says the reader did not know; empty when it says none.
func refusedWord(failed string) string {
if m := unknownWord.FindStringSubmatch(failed); m != nil {
return m[1]
}
return ""
}
// probeFirstBuildsID is the probe that reads the first merges waiting for mesh-delivery's word.
const probeFirstBuildsID = "D18"
@@ -235,10 +203,9 @@ func firstBuildObservations(waiting []inventory.BatchedMerge, now time.Time) []c
Headline: fmt.Sprintf("The first build of %s is not asked: no Approve was announced", repository),
Explanation: fmt.Sprintf("The merge %.8s into %s adds modules the mesh builds nothing from yet. It registers them "+
"only when mesh-delivery announces that it merged the pull request on your Approve of that head, and no such "+
"announcement has come in %s. The merge is kept, and the modules are registered from the builder's keys the moment "+
"the announcement is kept.",
"announcement has come in %s. The merge is kept, and the builds are asked the moment the announcement is kept.",
k.Commit, repository, for_),
Resolved: "mesh-delivery announces the merge, or the operator registers the modules at the controller's terminal",
Resolved: "mesh-delivery announces the merge, or a hand `build` registers the modules",
})
}
return out
@@ -250,28 +217,5 @@ func probeFirstBuilds(ctx context.Context, d *doctor) ([]conditions.Observation,
if err != nil {
return nil, err
}
now := time.Now().UTC()
return append(firstBuildObservations(waiting, now), newModuleObservations(now)...), nil
}
// answerHeldMerges answers every merge a controller of before held for a build-agent that reads its new modules'
// manifests (issue 1173): nothing asks their first builds any more, and the look at every module when this controller
// starts registers them from their keys (ADR 0501 §6, §7).
func answerHeldMerges(ctx context.Context, inv *inventory.Inventory) {
waiting, err := inv.WaitingMerges(ctx)
if err != nil {
fmt.Printf("the merges held for a newer build-agent could not be read: %v\n", err)
return
}
for _, w := range waiting {
if k := announcedOf(w); len(k.Held) > 0 {
if err := inv.AnswerMerge(ctx, w.Repository, w.Commit, inventory.WaitingRecord, false); err != nil {
fmt.Printf("the merge %.8s of %s held for a newer build-agent could not be answered: %v\n", w.Commit,
w.Repository, err)
continue
}
fmt.Printf("the merge %.8s of %s was held for a newer build-agent; its modules are registered from the "+
"builder's keys instead\n", w.Commit, w.Repository)
}
}
return firstBuildObservations(waiting, time.Now().UTC()), nil
}
@@ -5,48 +5,71 @@ package main
import (
"context"
"encoding/json"
"slices"
"sync"
"testing"
"time"
"github.com/nats-io/nats.go/jetstream"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
"github.com/novox/mesh-controller/internal/testbus"
)
// 1146/AC7 on keys (ADR 0501 §3, §6; issue 1851, A5), whole against a real bus and the store, in both orders of the two
// events: mesh-delivery #7 merged on the operator's Approve into a repository the mesh builds nothing from. The forge's
// merge, heard as the controller hears it (SourceMoved), and mesh-delivery's word, published on the subject the
// controller consumes with the router's record of the ask in its bucket, come in either order; each way the module the
// merge commit holds is registered from the builder's done key of it, and nothing asks a build. Without the Approve in
// the record it is not registered. Store-backed: skips without one.
func TestAMergeOnTheApproveIntoANewRepositoryRegistersItsModuleFromItsKey(t *testing.T) {
// 1146/AC7, whole against a real bus and the store, in both orders of the two events (novox/hq ADR 0432, issue 1146):
// mesh-delivery #7 merged on the operator's Approve into a repository the mesh builds nothing from. The forge's merge,
// heard as the controller hears it (SourceMoved), and mesh-delivery's word, published on the subject the controller
// consumes with the router's record of the ask in its bucket, come in either order; each way the first build of the
// module the merge commit holds is asked exactly once, its outcome registers the module, and it is in the catalogue.
// Without the Approve in the record nothing is asked. Store-backed: skips without one.
func TestAMergeOnTheApproveIntoANewRepositoryAsksTheFirstBuildAndItsOutcomeRegistersTheModule(t *testing.T) {
const head, commit = "a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1", "0177ea7a00000000"
for _, c := range []struct {
what string
wordFirst bool
approved bool // the router's record holds the operator's Approve of the merged head
wantRegister bool
what string
wordFirst bool
approved bool // the router's record holds the operator's Approve of the merged head
wantRegiter bool
}{
{what: "the forge's merge before mesh-delivery's word", approved: true, wantRegister: true},
{what: "mesh-delivery's word before the forge's merge", wordFirst: true, approved: true, wantRegister: true},
{what: "the forge's merge before mesh-delivery's word", approved: true, wantRegiter: true},
{what: "mesh-delivery's word before the forge's merge", wordFirst: true, approved: true, wantRegiter: true},
{what: "a word whose record holds no Approve", wordFirst: true},
} {
t.Run(c.what, func(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
inv := open.inventory
noBuildAsked(t)
b := aBuilderOnTheBus(t)
js := b.js
// The real ask, guard and queue included: only its answers are kept, so the build's outcome can be matched.
aForgeAndABuildSeat(t, open)
// Written by the controller's merge path on its own goroutine, read by the test: under one lock.
var mu sync.Mutex
var asks [][3]string
var ids []string
asked := func() [][3]string { mu.Lock(); defer mu.Unlock(); return slices.Clone(asks) }
realAsk := askABuild
askABuild = func(c context.Context, source buildSource, path, ref string) (string, error) {
id, err := realAsk(c, source, path, ref)
if err == nil {
mu.Lock()
asks = append(asks, [3]string{source.Repository, path, ref})
ids = append(ids, id)
mu.Unlock()
}
return id, err
}
t.Cleanup(func() { askABuild = realAsk })
if err := inv.RegisterModule(ctx, catalogue.Manifest{Module: "operator-channel", Version: "1",
DefinesSeats: []catalogue.SeatDeclaration{{Name: broker.AsksSeat, Records: []string{"asks"}}}},
inventory.Source{Repository: "novox/mesh-catalog", Seat: "git", Path: "modules/operator-channel", BuiltFrom: "c0ffee"}); err != nil {
t.Fatal(err)
}
js, err := broker.Dial(testbus.URL(t))
if err != nil {
t.Fatal(err)
}
t.Cleanup(js.Close)
if err := broker.AssertMeshStreams(js); err != nil {
t.Fatal(err)
}
@@ -64,14 +87,10 @@ func TestAMergeOnTheApproveIntoANewRepositoryRegistersItsModuleFromItsKey(t *tes
a := aDeliveryMergeOf(t, head, map[string]string{"novox/mesh-delivery@a1a1a1a1a1a1": head})
if c.approved {
record, _ := json.Marshal(map[string]any{"state": "closed", "ask": a.ask, "warrant": a.warrant})
if _, err := records.Put(ctx, sdkdelivery.TellerModule+"."+a.merge.AskID, record); err != nil {
if _, err := records.Put(ctx, theDelivery+"."+a.merge.AskID, record); err != nil {
t.Fatal(err)
}
}
// The indexer's value and the builder's done key of the module the merge commit holds.
fp := fpOf('9')
manifest := b.indexed(t, "mesh-delivery-view", "1", "novox/mesh-delivery", ".", commit, fp)
b.built(t, "mesh-delivery-view", "1", fp, manifest, sdkdelivery.BuildDone)
following := following{open: open}
server := link.ConnectNats(js, nil, noListener{})
@@ -85,6 +104,8 @@ func TestAMergeOnTheApproveIntoANewRepositoryRegistersItsModuleFromItsKey(t *tes
done := make(chan struct{})
go func() { defer close(done); _ = server.Serve(serving) }()
t.Cleanup(func() { stop(); <-done })
servingBus.Store(js)
t.Cleanup(func() { servingBus.Store(nil) })
word := func() {
t.Helper()
@@ -112,27 +133,37 @@ func TestAMergeOnTheApproveIntoANewRepositoryRegistersItsModuleFromItsKey(t *tes
forge()
} else {
forge()
time.Sleep(300 * time.Millisecond)
if _, known := inventoryHolds(t, inv, "mesh-delivery-view"); known {
t.Fatal("1146/AC7: registered before mesh-delivery's word was kept")
if len(asked()) != 0 {
t.Fatalf("1146/AC7: the merge asked %v before mesh-delivery's word was kept", asked())
}
word()
}
if !c.wantRegister {
time.Sleep(time.Second)
if _, known := inventoryHolds(t, inv, "mesh-delivery-view"); known {
t.Fatal("1146/AC7: a merge whose record holds no Approve registered its module")
if !c.wantRegiter {
if len(asked()) != 0 {
t.Fatalf("1146/AC7: a merge whose record holds no Approve asked %v", asked())
}
return
}
becomes(t, 10*time.Second, "the module registered", func() bool {
_, known := inventoryHolds(t, inv, "mesh-delivery-view")
return known
})
want := [3]string{"novox/mesh-delivery", "", "main"}
becomes(t, 10*time.Second, "the first build asked", func() bool { return len(asked()) >= 1 })
if got := asked(); len(got) != 1 || got[0] != want {
t.Fatalf("1146/AC7: the merge asked %v, want %v once", got, want)
}
mu.Lock()
first := ids[0]
mu.Unlock()
build := onTrunk(first, "novox/mesh-delivery", "git", "", map[string]any{"module": "mesh-delivery-view", "version": "1"})
build.Commit = commit
if _, _, err := takeIn(ctx, inv, build); err != nil {
t.Fatalf("1146/AC7: the outcome of the first build was refused: %v", err)
}
src, err := inv.SourceOf(ctx, "mesh-delivery-view")
if err != nil || src.Repository != "novox/mesh-delivery" {
t.Fatalf("1146/AC7: the module is built from %q (%v)", src.Repository, err)
}
if _, known := inventoryHolds(t, inv, "mesh-delivery-view"); !known {
t.Fatal("1146/AC7: the module is not in the catalogue")
}
})
}
}
+2 -2
View File
@@ -64,7 +64,7 @@ func registerFrom(t *testing.T, forms *sourceForms) (int64, error) {
mesh := &aMeshRegisteredFromTheURL{anEmptyMesh{asked: true,
request: inventory.BuildRequest{ID: "build-1", Repository: aRepoOnTheForge, For: "build"}}, aRepoOnTheForge}
return mayRegisterIn(t.Context(), mesh, forms, "mesh-delivery-view", inventory.Source{Repository: aRepoOnTheForge},
byBuild("build-1"), "", "main")
"build-1", "", "main")
}
// 1277/AC2: an address unread twice, then read, is waited out and the forge is asked.
@@ -112,7 +112,7 @@ func TestAKnownAddressAndAnotherHostIsStillNotOnTheForge(t *testing.T) {
const elsewhere = "http://elsewhere.example/x/y.git"
mesh := &aMeshRegisteredFromTheURL{anEmptyMesh{asked: true, request: inventory.BuildRequest{ID: "build-1", For: "build"}}, elsewhere}
_, err := mayRegisterIn(t.Context(), mesh, sourceFormsRead(context.Background(), time.Now, read), "mesh-delivery-view",
inventory.Source{Repository: elsewhere}, byBuild("build-1"), "", "main")
inventory.Source{Repository: elsewhere}, "build-1", "", "main")
if !errors.Is(err, errNotItsSource) || !strings.Contains(err.Error(), "is not on the mesh's forge") {
t.Fatalf("1277/AC4: the refusal changed: %v", err)
}
+1 -1
View File
@@ -68,7 +68,7 @@ func TestTheForgeNotAnsweringInTimeIsNotAskedAgain(t *testing.T) {
// An ask that hangs past the bound is cut off at the bound, not at its own wait, and the failure still says it
// waited: the receive loop is held no longer than the bound (novox/hq issue 486).
func TestAnAskHangingPastTheBoundIsCutOffAtIt(t *testing.T) {
w := waitOut{bound: 100 * time.Millisecond, first: 10 * time.Millisecond, most: 50 * time.Millisecond,
w := waitOut{bound: 200 * time.Millisecond, first: 10 * time.Millisecond, most: 50 * time.Millisecond,
now: time.Now, sleep: sleepOrDone}
asked := 0
began := time.Now()
-12
View File
@@ -1,12 +0,0 @@
package main
import "testing"
// C2 (ADR 0503 §3; issue 1299): a delivery group's composed check is answered as the checker's, and nothing is asked
// of build-agent: the command answers before any ask is composed.
func TestTheGroupCheckNamesTheCheckerAndAsksNothing(t *testing.T) {
a := groupCheckAnswer("feat/x")
if a["judged-by"] != "mesh-delivery-checker" || a["asked"] != "" || a["group"] != "feat/x" {
t.Fatalf("the group check answered %v", a)
}
}
-198
View File
@@ -1,198 +0,0 @@
package main
import (
"context"
"fmt"
"sort"
"strings"
"github.com/novox/mesh-controller/internal/conditions"
"github.com/novox/mesh-controller/internal/inventory"
)
// A move held for a person is asked of the person (novox/hq issue 1732, issue 1320; ADR 0377, ADR 0259).
//
// **A wait only a person can end is said to that person at once.** ADR 0377 holds a recorded module's move that
// recreates a container or touches its data until a person pushes the machine. The walk said so in its note and
// ended "done", `status` named the machine behind, and nobody was asked: on 2026-10-21 mailu's new build waited on
// novox for 40 minutes behind a plan that read "done, 3 tier(s)", and on 2026-10-19 mssql waited on ace the same way.
// Row S24 raises one condition per machine that has such a move, for the operator, with the answer **Release**: the
// push of that machine, which the controller's own `push` verb performs on the operator's warrant (ADR 0259). It ends
// when the machine no longer waits, whoever pushed it.
//
// A hold a person chose is not asked again — a module they set to `record` with a why, and the bus, whose step has
// its own row (S17).
// kindMovesHeld is S24's kind: machine.<node>.moves-held.
const kindMovesHeld = "moves-held"
// heldFacts is one machine and the moves of recorded modules that wait there for a person.
type heldFacts struct {
node string
moves []heldMove
}
// gatherHeldMoves is, per machine that holds at least one move no person chose to hold, all its held moves, by machine.
func gatherHeldMoves(ctx context.Context, inv *inventory.Inventory) ([]heldFacts, error) {
nodes, err := inv.Nodes(ctx)
if err != nil {
return nil, fmt.Errorf("the machines cannot be read: %w", err)
}
f, err := readMoveFacts(ctx, inv)
if err != nil {
return nil, err
}
var out []heldFacts
for _, n := range nodes {
moves, err := heldMovesOn(ctx, inv, f, n.Name)
if err != nil {
return nil, err
}
h := heldFacts{node: n.Name, moves: moves}
if len(h.asked()) > 0 {
out = append(out, h)
}
}
return out, nil
}
// asked are the moves no person chose to hold, and chosen the ones a person did (a module set to record with a why, the bus).
func (h heldFacts) asked() []heldMove { return h.byChoice(false) }
func (h heldFacts) chosen() []heldMove { return h.byChoice(true) }
func (h heldFacts) byChoice(chosen bool) []heldMove {
var out []heldMove
for _, m := range h.moves {
if m.Chosen == chosen {
out = append(out, m)
}
}
return out
}
// modules are the held moves' modules, in order, of all of them.
func (h heldFacts) modules() []string { return moduleNames(h.moves) }
func moduleNames(moves []heldMove) []string {
out := make([]string, 0, len(moves))
for _, m := range moves {
out = append(out, m.Module)
}
return out
}
// releaseAction is the one answer a held move offers: the push of the machine, approved by the operator. The push
// names every module whose move waits there as one that may move its data: the operator's Approve is the word
// ADR 0217 asks for, and the ask says so. Offered only where no person's own hold is on the machine too: a push
// composes every recorded module at its current build, so it would send that move as well, which the ask never named
// (ADR 0259 §6), and hold the whole machine if that module's data would move.
func (h heldFacts) releaseAction() conditions.Action {
return conditions.Action{Label: "Release", Verb: "mesh-controller.push", Level: conditions.LevelApprove,
Arguments: map[string]string{"node": h.node, "move": strings.Join(moduleNames(h.asked()), ","), "why": "",
"cause": conditions.CauseOperatorAnswer}}
}
// heldWords are the plain words of a machine's held moves: which modules wait, why in general, and what an Approve does.
func heldWords(h heldFacts) (headline, explanation, needs string, resolved string) {
names := namesWords(moduleNames(h.asked()), 3)
headline = clipWords(conditions.Capital(names)+" waits for you on "+h.node, conditions.HeadlineMax)
recreates, data := false, false
for _, m := range h.asked() {
switch {
case strings.Contains(m.Why, "recreates"):
recreates = true
default:
data = true
}
}
why := "would change what it runs"
switch {
case recreates && data:
why = "would restart a container or touch its data"
case recreates:
why = "would restart one of its containers"
case data:
why = "would touch its files or data"
}
explanation = clipWords(fmt.Sprintf("The new version of %s %s on %s, so it does not go out with its delivery; "+
"it waits for you. Release sends %s everything it is owed, these among it, and lets their data move where it "+
"has to. Nothing is sent until you answer.", names, why, h.node, h.node), conditions.ExplanationMax)
needs = "release them " + FromMeshMCPServer
if chosen := h.chosen(); len(chosen) > 0 {
// A push of the machine would send the move a person chose to hold as well: no answer is offered here.
explanation = clipWords(fmt.Sprintf("The new version of %s %s on %s, so it does not go out with its delivery; "+
"it waits for you. %s is also held there by your own choice, and a push of %s would send that too, so "+
"nothing is offered here: the push is yours at the controller's terminal.",
names, why, h.node, conditions.Capital(namesWords(moduleNames(chosen), 3)), h.node), conditions.ExplanationMax)
needs = fmt.Sprintf("push %s at the controller's terminal, once you have decided about %s.", h.node,
namesWords(moduleNames(chosen), 3))
}
resolved = "Resolved: " + names + " is no longer waiting on " + h.node
return
}
// watchMovesHeld is S24: a machine that has moves of recorded modules waiting for a person, said to the operator with the
// answer that sends them. The condition ends when the machine no longer waits: it is no longer among the facts.
func watchMovesHeld(f *signalFacts) []conditions.Observation {
var out []conditions.Observation
for _, h := range f.held {
var said []string
for _, m := range h.moves {
said = append(said, fmt.Sprintf("%s %s → %s %s", m.Module, short(m.From), short(m.To), m.Why))
}
headline, explanation, needs, resolved := heldWords(h)
if c := h.chosen(); len(c) > 0 {
said = append(said, "held by a person's own choice: "+strings.Join(moduleNames(c), ", "))
}
var actions []conditions.Action
if len(h.chosen()) == 0 {
actions = []conditions.Action{h.releaseAction()}
}
out = append(out, conditions.Observation{Scope: conditions.ScopeMachine, ID: h.node, Kind: kindMovesHeld,
Machine: h.node, Severity: conditions.Warning, Resolver: conditions.ResolverOperator,
Summary: fmt.Sprintf("%s has %d move(s) of recorded modules waiting for a person's push: %s — `push %s` sends them",
h.node, len(h.moves), strings.Join(h.modules(), ", "), h.node),
Said: strings.Join(said, "; "),
Headline: headline, Explanation: explanation, Needs: needs, Resolved: resolved,
Actions: actions})
}
return out
}
// planHeld is, for a plan that ended or is running, the moves its modules carried that a machine still holds for a
// person: "module on machine", sorted. The plan's own record says the move waited when it was judged
// (PlanModule.Recorded); the machines say whether it still does, and for the build this plan carried, so a plan older
// than the build now held is not named.
func planHeld(p inventory.Plan, held []heldFacts) []string {
var out []string
for _, h := range held {
for _, m := range h.moves {
s := p.Modules[m.Module]
if s == nil || !strings.HasPrefix(s.Recorded[h.node], "waits:") || !sameCommit(s.Commit, m.To) {
continue
}
out = append(out, m.Module+" on "+h.node)
}
}
sort.Strings(out)
return out
}
// heldSaid is what a plan's line adds when it carries a move that a machine holds for a person (novox/hq issue 1732): a
// plan whose module waits for a push is not plainly "done".
func heldSaid(p inventory.Plan, held []heldFacts) string {
waiting := planHeld(p, held)
if len(waiting) == 0 {
return ""
}
return fmt.Sprintf(" — %d held for a person: %s (the operator is asked)", len(waiting), strings.Join(waiting, ", "))
}
// heldNow reads the machines' held moves for a plan's line, nil when they cannot be read: the line then reads as before.
func heldNow(ctx context.Context, inv *inventory.Inventory) []heldFacts {
held, err := gatherHeldMoves(ctx, inv)
if err != nil {
return nil // empty-on-error: a plan's line then reads as it did before; the watchdog (S24) says it cannot read
}
return held
}
@@ -1,92 +0,0 @@
//go:build integration
package main
import (
"context"
"strings"
"testing"
"time"
"github.com/novox/mesh-controller/internal/inventory"
)
func inventoryUpgrade(why string) inventory.Upgrade { return inventory.Upgrade{Why: why, By: "jochen"} }
func mustMoveFacts(t *testing.T, rm *recordedMesh) moveFacts {
t.Helper()
f, err := readMoveFacts(t.Context(), rm.open.inventory)
if err != nil {
t.Fatal(err)
}
return f
}
// 1732/AC8 (integration): a mesh running mailu at c1 on two machines, whose registered c2 recreates a container, is read
// as two machines each holding mailu's move for a person; a bundle-only c3 on top changes nothing of it; after a machine
// is sent c2 (as a push records it) it is no longer among them, and plans that carried c2 say nothing more of it.
func TestMachinesHoldingAMoveForAPersonAreReadFromTheRecords(t *testing.T) {
rm := aRecordedMesh(t, aMailu(true))
c2 := aMailu(true)
resourceOf(&c2, "imap")["image"] = "registry.invalid/imap@sha256:2"
rm.register(t, c2, "c2", time.Now().Add(-time.Minute))
ctx := t.Context()
held, err := gatherHeldMoves(ctx, rm.open.inventory)
if err != nil {
t.Fatal(err)
}
if len(held) != 2 || held[0].node != "anchor" || held[1].node != "laptop" || len(held[0].moves) != 1 ||
held[0].moves[0].Module != "mailu" || held[0].moves[0].To != "c2" || !strings.Contains(held[0].moves[0].Why, "recreates") {
t.Fatalf("held %+v", held)
}
if err := rm.open.inventory.RecordSent(ctx, nodeID(t, rm.open, "anchor"), "d-anchor-c2", map[string]string{"mailu": "c2"}); err != nil {
t.Fatal(err)
}
held, err = gatherHeldMoves(ctx, rm.open.inventory)
if err != nil || len(held) != 1 || held[0].node != "laptop" {
t.Fatalf("after anchor was sent c2: %+v (%v)", held, err)
}
}
// 1732/AC9 (integration): a module a person set to record is no ask: they chose the hold.
func TestAHoldAPersonChoseIsNotAsked(t *testing.T) {
rm := aRecordedMesh(t, aMailu(false))
c2 := aMailu(false)
c2.Bundles[0].Digest = "sha256:bbb"
rm.register(t, c2, "c2", time.Now().Add(-time.Minute))
if err := rm.open.inventory.SetUpgradeOf(t.Context(), "mailu", inventoryUpgrade("the migration week")); err != nil {
t.Fatal(err)
}
held, err := gatherHeldMoves(t.Context(), rm.open.inventory)
if err != nil || len(held) != 0 {
t.Fatalf("a machine holding only a person's own hold is asked: %+v (%v)", held, err)
}
if moves, _ := heldMovesOn(t.Context(), rm.open.inventory, mustMoveFacts(t, rm), "anchor"); len(moves) != 1 || !moves[0].Chosen {
t.Fatalf("status still names it, marked as chosen: %+v", moves)
}
}
// 1742/AC4 (integration): a registered build naming an image of the store that the store lacks is read as a missing
// image on each machine that runs the module; one the store holds is not; and what the store holds is not asked twice.
func TestARegisteredBuildNamingAnImageTheStoreLacksIsRead(t *testing.T) {
m := aMailu(true)
resourceOf(&m, "imap")["image"] = "artifact-store://mailu/imap@sha256:" + strings.Repeat("e", 64)
rm := aRecordedMesh(t, m)
asked := 0
lacking := func(_ context.Context, reference string) (bool, error) { asked++; return false, nil }
imagesHeld.Lock()
imagesHeld.held = map[string]bool{}
imagesHeld.Unlock()
got, err := gatherMissingImages(t.Context(), rm.open.inventory, lacking)
if err != nil || len(got) != 2 || got[0].Module != "mailu" || got[0].Node != "anchor" || got[1].Node != "laptop" {
t.Fatalf("missing %+v (%v)", got, err)
}
holding := func(_ context.Context, reference string) (bool, error) { asked++; return true, nil }
if got, err := gatherMissingImages(t.Context(), rm.open.inventory, holding); err != nil || len(got) != 0 {
t.Fatalf("an image the store holds is missing: %+v (%v)", got, err)
}
before := asked
if _, err := gatherMissingImages(t.Context(), rm.open.inventory, lacking); err != nil || asked != before {
t.Fatalf("an image the store was found to hold was asked about again (%d, %v)", asked-before, err)
}
}
-237
View File
@@ -1,237 +0,0 @@
package main
import (
"context"
"errors"
"strings"
"testing"
"time"
"git.novox.be/novox/mesh-sdk/go/asks"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/conditions"
"github.com/novox/mesh-controller/internal/inventory"
)
// **A move held for a person is asked of the person** (novox/hq issue 1732, issue 1320; ADR 0377, ADR 0259).
// aHeldMailu is novox holding mailu's new build because it would recreate a container: the walk's own words.
func aHeldMailu() heldFacts {
return heldFacts{node: "novox", moves: []heldMove{{Module: "mailu", From: "b2f5ebb2aaaa", To: "3259765dbbbb",
Why: "waits: recreates 1 of mailu's 12 container(s) (with a new image: automx)"}}}
}
func heldConditionOn(f *signalFacts) conditions.Condition {
o := watchMovesHeld(f)[0]
return conditions.Condition{Key: o.Key(), Kind: o.Kind, Severity: o.Severity, Headline: o.Headline,
Explanation: conditions.Verdict(o.Needs, o.Explanation), Needs: o.Needs, Actions: o.Actions}
}
// 1732/AC1: a machine holding a move of a recorded module for a person raises one condition for the operator, with
// Release as its only answer: the machine's push, approved, naming the modules whose moves wait there.
func TestAHeldMoveIsRaisedWithReleaseAsItsOnlyAction(t *testing.T) {
f := calm(time.Date(2026, 10, 21, 17, 30, 0, 0, time.UTC))
f.held = []heldFacts{aHeldMailu()}
got := watchMovesHeld(f)
if len(got) != 1 {
t.Fatalf("%+v", got)
}
o := got[0]
if o.Kind != "moves-held" || o.Scope != conditions.ScopeMachine || o.ID != "novox" || o.Severity != conditions.Warning ||
o.Resolver != conditions.ResolverOperator || o.Key() != "machine.novox.moves-held" {
t.Fatalf("kind, scope, subject, severity or resolver: %+v (%s)", o, o.Key())
}
if len(o.Actions) != 1 {
t.Fatalf("actions: %+v", o.Actions)
}
a := o.Actions[0]
if a.Label != "Release" || a.Verb != "mesh-controller.push" || a.Level != conditions.LevelApprove ||
a.Arguments["node"] != "novox" || a.Arguments["move"] != "mailu" || a.Arguments["cause"] != conditions.CauseOperatorAnswer {
t.Fatalf("the action is not the push of novox: %+v", a)
}
if _, takes := a.Arguments["why"]; !takes {
t.Fatal("the push takes a why, which the asker fills in with the operator's answer")
}
for _, want := range []string{"mailu", "novox"} {
if !strings.Contains(o.Headline+o.Explanation, want) {
t.Errorf("the words do not name %q: %q / %q", want, o.Headline, o.Explanation)
}
}
if !strings.Contains(o.Said, "recreates 1 of mailu's 12") {
t.Errorf("the evidence does not carry the walk's own words: %q", o.Said)
}
if w := (conditions.Observation{Headline: o.Headline, Explanation: o.Explanation, Needs: o.Needs, Resolved: o.Resolved}); w.Headline == "" || len(o.Headline) > conditions.HeadlineMax {
t.Errorf("headline: %q", o.Headline)
}
}
// 1732/AC2: a machine with nothing held raises nothing, so the condition ends the look after the move is sent.
func TestAMachineHoldingNothingRaisesNoHeldMoveCondition(t *testing.T) {
f := calm(time.Date(2026, 10, 21, 17, 30, 0, 0, time.UTC))
if got := watchMovesHeld(f); len(got) != 0 {
t.Fatalf("%+v", got)
}
}
// 1732/AC3: the condition is asked on a channel at the approve level in words that say what Release does, and the
// operator's warrant calls the controller's own push with the machine and the modules, once.
func TestAHeldMovesReleaseIsAskedAndAWarrantPushesTheMachine(t *testing.T) {
f := calm(time.Date(2026, 10, 21, 17, 30, 0, 0, time.UTC))
f.held = []heldFacts{aHeldMailu()}
c := heldConditionOn(f)
r := newAskerRig(t)
r.open = []conditions.Condition{c}
if err := r.a.reconcile(context.Background()); err != nil {
t.Fatal(err)
}
sent := r.asksSent(t)
if len(sent) != 1 || len(sent[0].Options) != 1 {
t.Fatalf("asked %+v", sent)
}
o := sent[0].Options[0]
if o.Label != "Release" || o.Level != asks.Approve || o.Does != "the held builds are sent to novox" {
t.Errorf("the option is %+v", o)
}
if err := sent[0].Check(r.now); err != nil {
t.Errorf("the ask is refused as made: %v", err)
}
w := r.warrantFor(t, c.Key, "Release")
answerWith(t, r, w)
answerWith(t, r, w)
if len(r.called) != 1 || !strings.HasPrefix(r.called[0], "mesh-controller.push@ ") ||
!strings.Contains(r.called[0], `"node":"novox"`) || !strings.Contains(r.called[0], `"move":"mailu"`) ||
!strings.Contains(r.called[0], `"cause":"`+conditions.CauseOperatorAnswer+`"`) ||
!strings.Contains(r.called[0], "chose Release") {
t.Fatalf("called %v", r.called)
}
}
// 1732/AC4: the controller may call its own push and upgrade verbs on a warrant: the grant names them.
func TestTheControllerMayActOnAWarrantForPushAndUpgrade(t *testing.T) {
for _, verb := range []string{"push", "upgrade"} {
granted := false
for _, v := range broker.VerbsTheControllerActsOnAWarrant {
granted = granted || (v.Seat == broker.ControllerSeat && v.Verb == verb)
}
if !granted {
t.Errorf("the controller may not act on a warrant for %s", verb)
}
}
}
func aDoneMailuPlan() inventory.Plan {
return inventory.Plan{ID: "plan-1792585970446304628", Repository: "novox/mesh-catalog", State: inventory.PlanDone,
Tiers: [][]string{{"a"}, {"b"}, {"mailu"}},
Modules: map[string]*inventory.PlanModule{"mailu": {State: "built", Commit: "3259765dbbbbcccc",
Recorded: map[string]string{"novox": "waits: recreates 1 of mailu's 12 container(s)"}}}}
}
// 1732/AC5: a done plan that carries a move a machine still holds for a person says so on its line; once the machine
// no longer holds it, or holds another build of the module, the line says nothing more.
func TestADonePlanSaysWhatItHoldsForAPerson(t *testing.T) {
p := aDoneMailuPlan()
line := planLine(p, time.Date(2026, 10, 21, 16, 0, 0, 0, time.UTC)) + heldSaid(p, []heldFacts{aHeldMailu()})
if !strings.Contains(line, "done, 3 tier(s)") || !strings.Contains(line, "1 held for a person: mailu on novox") {
t.Errorf("the line: %q", line)
}
if said := heldSaid(p, nil); said != "" {
t.Errorf("a plan whose moves were sent says %q", said)
}
other := aHeldMailu()
other.moves[0].To = "9999999ffff"
if said := heldSaid(p, []heldFacts{other}); said != "" {
t.Errorf("a plan older than the build now held says %q", said)
}
p.Modules["mailu"].Recorded = map[string]string{"novox": recordedRollsWhy}
if said := heldSaid(p, []heldFacts{aHeldMailu()}); said != "" {
t.Errorf("a move the plan sent says %q", said)
}
}
// 1732/AC6: the builds held after a failed release are asked of the operator: the held-release condition offers
// Release, which calls the controller's own `upgrade` with release-backlog and nothing else, once, on a warrant.
func TestTheBuildsHeldAfterAFailedReleaseAreAskedAndReleasedOnAWarrant(t *testing.T) {
saved := backlogNow
t.Cleanup(func() { backlogNow = saved })
backlogNow.held = "release-1792595552499982311 failed (failed on novox); what waits is released again by a person"
backlogNow.waiting = map[string][]inventory.CarriedMove{"novox": {{Module: "indexer"}, {Module: "nextcloud"}}}
got := backlogObservation()
if len(got) != 1 || len(got[0].Actions) != 1 {
t.Fatalf("raised %+v", got)
}
a := got[0].Actions[0]
if a.Label != "Release" || a.Verb != "mesh-controller.upgrade" || a.Level != conditions.LevelApprove ||
a.Arguments["release-backlog"] != "true" || a.Arguments["cause"] != conditions.CauseOperatorAnswer {
t.Fatalf("the action: %+v", a)
}
o := got[0]
c := conditions.Condition{Key: o.Key(), Kind: o.Kind, Severity: o.Severity, Headline: o.Headline,
Explanation: conditions.Verdict(o.Needs, o.Explanation), Needs: o.Needs, Actions: o.Actions}
r := newAskerRig(t)
r.open = []conditions.Condition{c}
if err := r.a.reconcile(context.Background()); err != nil {
t.Fatal(err)
}
sent := r.asksSent(t)
if len(sent) != 1 || len(sent[0].Options) != 1 || sent[0].Options[0].Does != "the builds held after the failed delivery are released" {
t.Fatalf("asked %+v", sent)
}
w := r.warrantFor(t, c.Key, "Release")
answerWith(t, r, w)
answerWith(t, r, w)
if len(r.called) != 1 || !strings.HasPrefix(r.called[0], "mesh-controller.upgrade@ ") ||
!strings.Contains(r.called[0], `"release-backlog":"true"`) || !strings.Contains(r.called[0], "chose Release") {
t.Fatalf("called %v", r.called)
}
}
// 1732/AC7: a plan whose send waits for a gate while the builds are held after a failed release reads as waiting for
// a person to release them, not "building ... LATE"; with nothing held it still reads as the wait it is.
func TestAPlanWaitingForAGateBehindAFailedReleaseIsNotLate(t *testing.T) {
saved := backlogNow
t.Cleanup(func() { backlogNow = saved })
now := time.Date(2026, 10, 21, 18, 50, 0, 0, time.UTC)
p := inventory.Plan{ID: "plan-1", Repository: "novox/mesh-sdk", Commit: "c0ffee00", State: inventory.PlanBuilding,
Tiers: [][]string{{"indexer"}}, TierEntered: now.Add(-100 * time.Minute), Created: now.Add(-110 * time.Minute),
Modules: map[string]*inventory.PlanModule{"indexer": {State: "built"}},
Note: refusalNote(0, errors.New("a build waits there for its gate: novox"))}
backlogNow.held = ""
if line := planLine(p, now); !strings.Contains(line, "a build waits there for its gate: novox") {
t.Errorf("the wait is not said: %q", line)
}
backlogNow.held = "release-1 failed (failed on novox); what waits is released again by a person"
line := planLine(p, now)
if strings.Contains(line, "LATE") || strings.Contains(line, "building") ||
!strings.Contains(line, "until a person releases it") {
t.Errorf("a plan held behind a failed release reads %q", line)
}
}
// 1732/AC10: a machine that also holds a move a person chose to hold gets no Release, since a push would send that move
// too; the condition names both and says the push is the controller's terminal. A machine with only asked moves keeps it.
func TestAMachineAlsoHoldingAPersonsOwnHoldIsOfferedNoRelease(t *testing.T) {
f := calm(time.Date(2026, 10, 21, 17, 30, 0, 0, time.UTC))
h := aHeldMailu()
h.moves = append(h.moves, heldMove{Module: "nftables", From: "n1aaaa", To: "n2bbbb",
Why: "waits: a person set it to record, and holds every move of it until they push", Chosen: true})
f.held = []heldFacts{h}
got := watchMovesHeld(f)
if len(got) != 1 || len(got[0].Actions) != 0 {
t.Fatalf("%+v", got)
}
o := got[0]
text := o.Headline + o.Explanation + o.Needs + o.Summary + o.Said
for _, want := range []string{"mailu", "nftables", "terminal", "push novox"} {
if !strings.Contains(text, want) {
t.Errorf("does not say %q: %q", want, text)
}
}
w := conditions.Words{Headline: o.Headline, Explanation: o.Explanation, Resolved: o.Resolved, Needs: o.Needs}
if why, ok := conditions.PlainWords(w, "novox"); !ok {
t.Errorf("not plain: %s", why)
}
f.held = []heldFacts{aHeldMailu()}
if got := watchMovesHeld(f); len(got) != 1 || len(got[0].Actions) != 1 || got[0].Actions[0].Arguments["move"] != "mailu" {
t.Fatalf("a machine with only asked moves: %+v", got)
}
}
-245
View File
@@ -1,245 +0,0 @@
package main
import (
"context"
"fmt"
"slices"
"sort"
"strings"
"sync"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/artifacts"
"github.com/novox/mesh-controller/internal/conditions"
"github.com/novox/mesh-controller/internal/inventory"
)
// No declaration is sent that names an image the artifact store does not hold (novox/hq issue 1742).
//
// On 2026-10-21 at 17:13 a send to novox named `mailu/automx@sha256:c62150b8…`; the build record keeps that digest
// ("recent-build") and the store did not hold it. The node-engine removed the running container, then could not pull
// the replacement, and every later apply on novox failed for the next hour while the mail clients' autoconfiguration
// had no backend. The node-engine's half (pull first, replace after) is its own change; this is the controller's: **a
// module whose image the store lacks is left out of the declaration**, as a module whose setting nobody gave is
// (ADR 0163, rule 6) — the node-engine keeps what it has of it and touches none of its containers — and **it is said**,
// row S25: a condition naming the module, the machine and the image, until the store holds the image again (a new build
// of the module, or the copy put back).
//
// Judged only for an image the store's own address names, from the send's own resources; an image of another registry
// is no business of the store's. A store that cannot be asked judges nothing: a send is not refused for a question
// that did not answer.
// kindImageMissing is S25's kind: module.<module>.<machine>.image-missing.
const kindImageMissing = "image-missing"
// missingImage is a module held out of a machine's declaration because the store lacks an image it names.
type missingImage struct {
Node, Module, Image string
}
// storeHolds is whether the store holds the manifest an image reference at its address names.
type storeHolds func(ctx context.Context, reference string) (bool, error)
// imageInStore is the artifact-store reference of an image the store's address names, "" for any other.
func imageInStore(image, address string) string {
if address == "" || !strings.HasPrefix(image, address+"/") || !strings.Contains(image, "@sha256:") {
return ""
}
reference := catalogue.Recorded(image)
if _, ours := catalogue.InArtifactStore(reference); !ours {
return ""
}
return reference
}
// holdBackMissingImages leaves out of the declaration every module that names an image the store does not hold,
// and answers what it left out. The module's resources go, it joins LeftOut with the reason, and its build stays the
// one the machine was last sent (`was`).
func holdBackMissingImages(ctx context.Context, node string, s *sendable, address string, holds storeHolds,
was map[string]string) []missingImage {
lacks := map[string]bool{} // reference → the store does not hold it; absent: not asked, or no answer
asked := map[string]bool{}
lacking := map[string]string{} // module → the image
for _, r := range s.Resources {
if fmt.Sprint(r["type"]) != "container" {
continue
}
image, _ := r["image"].(string)
reference := imageInStore(image, address)
module := s.moduleOf[fmt.Sprint(r["id"])]
if reference == "" || module == "" || lacking[module] != "" {
continue
}
if !asked[reference] {
asked[reference] = true
held, err := holds(ctx, reference)
// empty-on-error: the store did not answer, so nothing is held back for a question unanswered
lacks[reference] = err == nil && !held
}
if lacks[reference] {
lacking[module] = image
}
}
if len(lacking) == 0 {
return nil
}
kept := s.Resources[:0:0]
for _, r := range s.Resources {
if _, out := lacking[s.moduleOf[fmt.Sprint(r["id"])]]; !out {
kept = append(kept, r)
}
}
s.Resources = kept
if s.leftOutWhy == nil {
s.leftOutWhy = map[string]string{}
}
var found []missingImage
for module, image := range lacking {
s.leftOutWhy[module] = "its image " + image + " is not in the artifact store"
if !slices.Contains(s.LeftOut, module) {
s.LeftOut = append(s.LeftOut, module)
}
if s.Builds != nil {
if prior, sent := was[module]; sent {
s.Builds[module] = prior
} else {
delete(s.Builds, module)
}
}
found = append(found, missingImage{Node: node, Module: module, Image: image})
}
sort.Strings(s.LeftOut)
sort.Slice(found, func(i, j int) bool { return found[i].Module < found[j].Module })
return found
}
// withoutMissingImages is composeSend's last step: the store asked about the images the declaration names, the modules
// whose image it lacks left out and kept in the ledger S25 reads. A store that cannot be found or asked leaves the
// declaration as composed.
func withoutMissingImages(ctx context.Context, open *stores, node string, s sendable) sendable {
inv := open.inventory
shelf, err := inv.Catalogue(ctx)
if err != nil {
return s
}
address, err := artifactStoreAddress(ctx, inv, shelf, node)
if err != nil || address == "" {
return s
}
sent, known, err := inv.SentBuilds(ctx, node)
if err != nil || !known {
sent = nil
}
store := artifacts.Store{Address: address}
found := holdBackMissingImages(ctx, node, &s, address, store.HoldsManifest, sent)
for _, m := range found {
fmt.Printf("%s: %s is left out of the declaration: its image %s is not in the artifact store\n", node, m.Module, m.Image)
}
return s
}
// imageFacts are the images S25 raises: a module whose registered build names an image the store lacks, on each machine
// that runs it.
type imageFacts struct {
missingImage
}
// imagesHeld remembers the images the store was found to hold: a stored image is named by its digest and stays what it
// is, so it is asked about once; one it lacks is asked again at every look.
var imagesHeld = struct {
sync.Mutex
held map[string]bool
}{held: map[string]bool{}}
// gatherMissingImages is every machine's module whose registered build — the one a send would carry — names an image of
// the artifact store that the store does not hold. Read from the catalogue and the store, never from what a send did:
// a push is a command of its own, and the serving controller's watchdog hears nothing of it.
func gatherMissingImages(ctx context.Context, inv *inventory.Inventory, holds storeHolds) ([]imageFacts, error) {
shelf, err := inv.Catalogue(ctx)
if err != nil {
return nil, fmt.Errorf("the catalogue cannot be read: %w", err)
}
var out []imageFacts
for _, module := range sortedKeysOf(modulesOf(shelf)) {
for _, r := range shelf[module].Resources {
if fmt.Sprint(r["type"]) != "container" {
continue
}
image, _ := r["image"].(string)
if _, ours := catalogue.InArtifactStore(image); !ours || !strings.Contains(image, "@sha256:") {
continue
}
imagesHeld.Lock()
known := imagesHeld.held[image]
imagesHeld.Unlock()
if !known {
held, err := holds(ctx, image)
if err != nil {
return nil, fmt.Errorf("the artifact store cannot be asked about %s: %w", image, err)
}
if held {
imagesHeld.Lock()
imagesHeld.held[image] = true
imagesHeld.Unlock()
continue
}
} else {
continue
}
running, err := inv.Running(ctx, module)
if err != nil {
return nil, err
}
for _, node := range running {
out = append(out, imageFacts{missingImage{Node: node, Module: module, Image: image}})
}
}
}
return out, nil
}
// modulesOf is the shelf's modules as a map of names, for sortedKeysOf.
func modulesOf(shelf map[string]catalogue.Manifest) map[string]string {
out := make(map[string]string, len(shelf))
for m := range shelf {
out[m] = m
}
return out
}
// watchMissingImages is S25: a module left out of a machine's declaration because the store lacks its image, said at
// once. The condition ends when the store holds the image, or a send names another.
func watchMissingImages(f *signalFacts) []conditions.Observation {
var out []conditions.Observation
for _, m := range f.images {
out = append(out, conditions.Observation{Scope: conditions.ScopeModule, ID: m.Module + "." + m.Node, Kind: kindImageMissing,
Machine: m.Node, Severity: conditions.Warning, Resolver: conditions.ResolverOperator,
Summary: fmt.Sprintf("%s is left out of every send to %s: its declaration names an image that the artifact store does "+
"not hold (the evidence names it); the running container is untouched", m.Module, m.Node),
Said: "image " + m.Image + " is not in the store",
Headline: clipWords(conditions.Capital(m.Module)+" is not updated on "+m.Node, conditions.HeadlineMax),
Explanation: clipWords(fmt.Sprintf("The new version of %s names a container image that the mesh's store does not "+
"hold, so it was not sent to %s. What runs there is untouched, and the rest of %s is sent as usual.",
m.Module, m.Node, m.Node), conditions.ExplanationMax),
Needs: fmt.Sprintf("build %s again; the details name the image.", m.Module),
Resolved: "Resolved: the store holds the image of " + m.Module + " again"})
}
return out
}
// gatherMissingImagesHere is gatherMissingImages asking the mesh's own artifact store; nothing to ask where the mesh has
// none on its network yet.
func gatherMissingImagesHere(ctx context.Context, inv *inventory.Inventory) ([]imageFacts, error) {
shelf, err := inv.Catalogue(ctx)
if err != nil {
return nil, fmt.Errorf("the catalogue cannot be read: %w", err)
}
address, err := artifactStoreAddress(ctx, inv, shelf, "")
if err != nil {
return nil, err
}
if address == "" {
return nil, nil
}
return gatherMissingImages(ctx, inv, artifacts.Store{Address: address}.HoldsManifest)
}
-125
View File
@@ -1,125 +0,0 @@
package main
import (
"context"
"errors"
"strings"
"testing"
"time"
"github.com/novox/mesh-controller/internal/conditions"
)
// **No declaration is sent that names an image the artifact store does not hold** (novox/hq issue 1742).
const storeAt = "novox.internal:5100"
// automx is the image the send of 2026-10-21 17:13 named and the store lacked; the others are held.
const (
automxImage = storeAt + "/mailu/automx@sha256:c62150b8e2cbcc18cb911b098c2877f928761a1e6251667f39ea18251ea35591"
imapImage = storeAt + "/mailu/imap@sha256:aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"
)
func aMailSend() sendable {
return sendable{
Resources: []map[string]any{
{"id": "mailu.automx", "type": "container", "image": automxImage},
{"id": "mailu.automx-config", "type": "file", "path": "/etc/automx.conf"},
{"id": "mailu.imap", "type": "container", "image": imapImage},
{"id": "other.web", "type": "container", "image": "docker.io/library/nginx@sha256:" + strings.Repeat("b", 64)},
{"id": "dns.dnsmasq", "type": "container", "image": storeAt + "/dnsmasq/main@sha256:" + strings.Repeat("c", 64)},
},
moduleOf: map[string]string{"mailu.automx": "automx", "mailu.automx-config": "automx", "mailu.imap": "mailu",
"other.web": "other", "dns.dnsmasq": "dnsmasq"},
Builds: map[string]string{"automx": "b2f5ebb2new", "mailu": "m2", "other": "o1", "dnsmasq": "d1"},
}
}
func storeHolding(missing ...string) (storeHolds, *[]string) {
var asked []string
return func(_ context.Context, reference string) (bool, error) {
asked = append(asked, reference)
for _, m := range missing {
if strings.Contains(reference, m) {
return false, nil
}
}
return true, nil
}, &asked
}
// 1742/AC1: given a send naming automx's image, which the store does not hold, when the declaration is judged, then
// every resource of automx is out of it, automx is among the modules left out with the reason, its build is the one the
// machine was last sent, and every other module — an image of another registry among them — is as composed.
func TestAModuleWhoseImageTheStoreLacksIsLeftOutOfTheSend(t *testing.T) {
s := aMailSend()
holds, asked := storeHolding("mailu/automx@")
found := holdBackMissingImages(t.Context(), "novox", &s, storeAt, holds, map[string]string{"automx": "b2f5ebb2old"})
if len(found) != 1 || found[0].Module != "automx" || found[0].Node != "novox" || found[0].Image != automxImage {
t.Fatalf("found %+v", found)
}
var ids []string
for _, r := range s.Resources {
ids = append(ids, r["id"].(string))
}
if got := strings.Join(ids, ","); got != "mailu.imap,other.web,dns.dnsmasq" {
t.Errorf("the declaration keeps %s", got)
}
if len(s.LeftOut) != 1 || s.LeftOut[0] != "automx" || !strings.Contains(s.leftOutWhy["automx"], "not in the artifact store") {
t.Errorf("left out %v because %v", s.LeftOut, s.leftOutWhy)
}
if s.Builds["automx"] != "b2f5ebb2old" || s.Builds["mailu"] != "m2" || s.Builds["other"] != "o1" {
t.Errorf("builds %v", s.Builds)
}
for _, a := range *asked {
if strings.Contains(a, "docker.io") {
t.Errorf("the store was asked about an image of another registry: %s", a)
}
}
}
// 1742/AC2: given a module never sent to the machine, when its image is missing, then it has no build in the record of the
// send; given the store does not answer, then nothing is held back for a question unanswered.
func TestAnUnansweredStoreHoldsNothingBackAndANewModuleHasNoBuild(t *testing.T) {
s := aMailSend()
holds, _ := storeHolding("mailu/automx@")
found := holdBackMissingImages(t.Context(), "novox", &s, storeAt, holds, nil)
if len(found) != 1 {
t.Fatalf("%+v", found)
}
if _, has := s.Builds["automx"]; has {
t.Errorf("a module never sent is recorded as sent: %v", s.Builds)
}
s = aMailSend()
silent := func(context.Context, string) (bool, error) { return false, errors.New("registry: connection refused") }
if found := holdBackMissingImages(t.Context(), "novox", &s, storeAt, silent, nil); len(found) != 0 || len(s.LeftOut) != 0 ||
len(s.Resources) != 5 {
t.Errorf("an unanswered store held back %+v, left out %v", found, s.LeftOut)
}
}
// 1742/AC3: a module left out for its image raises a condition for the operator naming the module, the machine and the
// image in the evidence, plain words, and no answer to give; a mesh whose images are all held raises nothing.
func TestAMissingImageIsRaisedForTheOperator(t *testing.T) {
f := calm(time.Date(2026, 10, 21, 17, 40, 0, 0, time.UTC))
if got := watchMissingImages(f); len(got) != 0 {
t.Fatalf("%+v", got)
}
f.images = []imageFacts{{missingImage{Node: "novox", Module: "automx", Image: automxImage}}}
got := watchMissingImages(f)
if len(got) != 1 {
t.Fatalf("%+v", got)
}
o := got[0]
if o.Kind != "image-missing" || o.Key() != "module.automx.novox.image-missing" || o.Resolver != conditions.ResolverOperator ||
o.Severity != conditions.Warning || o.Machine != "novox" {
t.Fatalf("%+v (%s)", o, o.Key())
}
if !strings.Contains(o.Said, "c62150b8") || !strings.Contains(o.Headline+o.Explanation, "automx") {
t.Errorf("words: %q / %q / %q", o.Headline, o.Explanation, o.Said)
}
w := conditions.Words{Headline: o.Headline, Explanation: o.Explanation, Resolved: o.Resolved, Needs: o.Needs}
if why, ok := conditions.PlainWords(w, "novox"); !ok {
t.Errorf("not plain: %s", why)
}
}
+1 -1
View File
@@ -35,7 +35,7 @@ func TestAMergeSaysTheModulesWhoseBuildSourceItLeftUnchanged(t *testing.T) {
{Own: true, Paths: []string{"modules/build-agent/Dockerfile", "modules/build-agent/module.json"}},
},
"route-proxy": {
{Repository: "novox/mesh-controller", Ref: "main", Paths: []string{"go.mod", "modules/route-proxy/cmd/route-proxy/"}},
{Repository: "novox/mesh-controller", Ref: "main", Paths: []string{"go.mod", "examples/route-proxy/"}},
{Own: true, Paths: []string{"modules/route-proxy/Dockerfile", "modules/route-proxy/module.json"}},
},
}
-460
View File
@@ -1,460 +0,0 @@
package main
import (
"archive/tar"
"bytes"
"compress/gzip"
"context"
"encoding/base64"
"encoding/json"
"errors"
"fmt"
"io"
"path"
"strings"
"time"
"git.novox.be/novox/mesh-sdk/go/buildstep"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/artifacts"
"github.com/novox/mesh-controller/internal/builderkeys"
"github.com/novox/mesh-controller/internal/buildreads"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// The take-in from the builder's done key (novox/hq ADR 0501, ADR 0498 §8, ADR 0453 §5; issue 1851).
//
// From R4e the controller asks no build but the builder's own module's: a module of a merge is taken from the builder's
// key for the fingerprint the indexer's trunk value of the module names (ADR 0501 §2). Only a done key whose every
// artifact names a reference is taken (ADR 0498 §8); a key not done yet, one waiting for a reference, a templates
// archive the store does not answer, or a list of predefined keys the forge does not answer is waited for and said,
// never failed and never delivered from. A failed key fails, in the builder's own words. The take-in then records the
// key as a build of today's record layout, so a controller of before reads it on a rollback, and judges it as every
// build is judged (registerJudged).
// errKeyWaits says a take-in from a key waits: for the builder, a reference, the store or the forge. Never a failure.
var errKeyWaits = errors.New("waits")
// errKeyFailed says the module's key failed: the builder's own words follow.
var errKeyFailed = errors.New("the builder's key failed")
// keyReads is what a take-in from a key reads of the builder and the indexer (builderkeys.Reader, or a test's).
type keyReads interface {
Build(ctx context.Context, module, fp string) (builderkeys.Key, error)
Manifest(ctx context.Context, digest string) (sdkdelivery.Manifest, error)
}
// keyBuildID is the id a key's build is recorded under: the key and its revision, so the same revision recorded twice
// is one record, and a rebuild of the same fingerprint (a new revision, ADR 0429 §18) another.
func keyBuildID(k builderkeys.Key) string {
part, err := sdkdelivery.FingerprintKeyPart(k.Fingerprint)
if err != nil {
part = strings.TrimPrefix(k.Fingerprint, sdkdelivery.SourcePrefix2)
}
return fmt.Sprintf("key-%s-%d", part, k.Revision)
}
// readArchive is the archive a key's reference names, read from the artifact store and checked against its digest. A
// variable so a test needs no store.
var readArchive = func(ctx context.Context, inv *inventory.Inventory, reference string) ([]byte, error) {
shelf, err := inv.Catalogue(ctx)
if err != nil {
return nil, err
}
address, err := artifactStoreAddress(ctx, inv, shelf, "")
if err != nil {
return nil, err
}
if address == "" {
return nil, errors.New("the artifact store is on no network the controller reaches")
}
return artifacts.Store{Address: address}.GetBlob(ctx, reference)
}
// predefinedHead is the list of predefined keys at the head of the catalogue repository's trunk, and that head.
type predefinedHead struct {
List catalogue.PredefinedList
Repository string
Commit string
// Generation orders two heads of the trunk: the head's commit time in seconds. A list kept from a build's
// generation (its commit count) is always older than a head read here.
Generation int
}
// predefinedAtTrunk is the list of predefined keys at the head of the catalogue repository's trunk (ADR 0501 §8), read
// from the forge. A variable so a test needs no forge.
var predefinedAtTrunk = func(ctx context.Context, inv *inventory.Inventory) (predefinedHead, error) {
entries, err := inv.Catalogued(ctx)
if err != nil {
return predefinedHead{}, err
}
repository := ""
for _, e := range entries {
if e.Source.Seat == gitSeat && path.Base(strings.Trim(e.Source.Repository, "/")) == catalogue.CatalogueRepository {
repository = strings.Trim(e.Source.Repository, "/")
break
}
}
if repository == "" {
return predefinedHead{}, fmt.Errorf("no module of the catalogue names the repository %s on the forge",
catalogue.CatalogueRepository)
}
var heads []struct {
SHA string `json:"sha"`
Commit struct {
Committer struct {
Date time.Time `json:"date"`
} `json:"committer"`
} `json:"commit"`
}
if err := askGitea(ctx, "gitea_api", map[string]any{"path": "/repos/" + repository +
"/commits?limit=1&stat=false&verification=false&files=false"}, &heads); err != nil {
return predefinedHead{}, err
}
if len(heads) == 0 || heads[0].SHA == "" {
return predefinedHead{}, fmt.Errorf("the forge named no head of %s's trunk", repository)
}
head := predefinedHead{Repository: repository, Commit: heads[0].SHA,
Generation: int(heads[0].Commit.Committer.Date.Unix())}
var file struct {
Content string `json:"content"`
Encoding string `json:"encoding"`
}
if err := askGitea(ctx, "gitea_api", map[string]any{"path": "/repos/" + repository + "/contents/" +
catalogue.PredefinedFile + "?ref=" + head.Commit}, &file); err != nil {
return predefinedHead{}, err
}
if file.Encoding != "base64" {
return predefinedHead{}, fmt.Errorf("the forge answered %s at %.8s in the encoding %q", catalogue.PredefinedFile,
head.Commit, file.Encoding)
}
raw, err := base64.StdEncoding.DecodeString(file.Content)
if err != nil {
return predefinedHead{}, err
}
if head.List, err = catalogue.ParsePredefined(raw); err != nil {
return predefinedHead{}, fmt.Errorf("%s at %.8s is refused: %w", catalogue.PredefinedFile, head.Commit, err)
}
return head, nil
}
// askGitea asks the forge's module one tool on the bus and reads its result into into.
func askGitea(ctx context.Context, tool string, args map[string]any, into any) error {
js, err := aBus()
if err != nil {
return err
}
defer js.Close()
raw, _ := json.Marshal(args)
answer, err := link.Ask(ctx, link.OverNATS{Conn: js.Conn()}, "gitea", tool, raw, 30*time.Second)
if err != nil {
return err
}
if answer.Error != "" {
return fmt.Errorf("gitea.%s: %s", tool, answer.Error)
}
var wrapped struct {
Result json.RawMessage `json:"result"`
}
if err := json.Unmarshal(answer.Result, &wrapped); err != nil {
return err
}
return json.Unmarshal(wrapped.Result, into)
}
// keyTake is one take-in from a key: the module, the indexer's trunk value naming the key's fingerprint, and the
// registration's warrant.
type keyTake struct {
Module string
Trunk sdkdelivery.Module
Ask registrationAsk
}
// takeInKey takes a module in from the builder's key for the fingerprint its trunk value names: records the key as a
// build and registers the module, or says why not. errKeyWaits when it waits, errKeyFailed when the key failed.
func takeInKey(ctx context.Context, inv *inventory.Inventory, keys keyReads, t keyTake) (
catalogue.Manifest, inventory.Build, error) {
head, err := predefinedAtTrunk(ctx, inv)
if err != nil {
return catalogue.Manifest{}, inventory.Build{}, fmt.Errorf("%w: the list of predefined keys at the catalogue's "+
"trunk head cannot be read, so %s is not judged: %v", errKeyWaits, t.Module, err)
}
keepPredefinedHead(ctx, inv, head)
previous, err := inv.Builds(ctx, t.Module, 5)
if err != nil {
return catalogue.Manifest{}, inventory.Build{}, err
}
kept, manifest, err := buildOfKey(ctx, keys, t, keyContext{list: head.List, previous: previous,
archive: func(ctx context.Context, reference string) ([]byte, error) { return readArchive(ctx, inv, reference) }})
if err != nil {
return catalogue.Manifest{}, kept, err
}
if err := inv.RecordBuild(ctx, kept); err != nil {
return catalogue.Manifest{}, kept, err
}
recorded := inventory.Source{Repository: t.Trunk.Repository, Seat: gitSeat, Path: keyPath(t.Trunk),
BuiltFrom: t.Trunk.Commit, Head: t.Trunk.Commit, Against: kept.Against}
trunk := "main"
if was, err := inv.SourceOf(ctx, manifest.Module); err == nil {
recorded.Ref = followedBranch(was.Ref)
if recorded.Ref != "" {
trunk = recorded.Ref
}
}
registered, err := registerJudged(ctx, inv, takenBuild{kept: kept, manifest: manifest, recorded: recorded,
ask: t.Ask, path: recorded.Path, trunk: trunk, by: "the builder", what: t.Trunk.Repository,
commit: t.Trunk.Commit})
return registered, kept, err
}
// keyContext is what a key's build is made with besides the key: the list of predefined keys its manifest is judged
// against, the module's recent builds (the bases its last one stood on are carried) and how an archive is read.
type keyContext struct {
list catalogue.PredefinedList
previous []inventory.Build
archive func(ctx context.Context, reference string) ([]byte, error)
}
// keyPath is a module's directory as the catalogue records it: empty for a module at its repository's root.
func keyPath(m sdkdelivery.Module) string {
if m.Path == sdkdelivery.RootPath {
return ""
}
return strings.Trim(m.Path, "/")
}
// buildOfKey reads the key of the trunk value's fingerprint and makes its build record and the manifest it resolves to
// (ADR 0498 §8, ADR 0501 §1, §8): the module's manifest by the key's digest, judged against the list of predefined keys
// at the catalogue's trunk head, resolved with the key's artifacts.
func buildOfKey(ctx context.Context, keys keyReads, t keyTake, kc keyContext) (
inventory.Build, catalogue.Manifest, error) {
waits := func(format string, a ...any) (inventory.Build, catalogue.Manifest, error) {
return inventory.Build{}, catalogue.Manifest{}, fmt.Errorf("%w: "+format, append([]any{errKeyWaits}, a...)...)
}
if t.Trunk.FP2 == "" {
why := t.Trunk.None2
if why == "" {
why = "the indexer wrote no fingerprint of the new form"
}
return waits("%s at %.8s has no fingerprint to take a build by: %s", t.Module, t.Trunk.Commit, why)
}
k, err := keys.Build(ctx, t.Module, t.Trunk.FP2)
if err != nil {
return waits("%s's key cannot be read: %v", t.Module, err)
}
switch {
case k.Failed():
return inventory.Build{}, catalogue.Manifest{}, fmt.Errorf("%w: %s", errKeyFailed, k.Says())
case !k.Done():
return waits("%s", k.Says())
case len(k.Unreferenced()) > 0:
return waits("%s", k.Says())
}
src, err := keys.Manifest(ctx, k.Build.Manifest)
if err != nil {
return waits("%s's manifest: %v", t.Module, err)
}
manifest, err := catalogue.ParseManifestWith([]byte(src.JSON), kc.list)
if err != nil {
// **A word the running controller does not read waits for the controller that reads it** (ADR 0501 §7): the
// key stays done, and the controller that reads the word registers it with no second build.
if word := refusedWord(err.Error()); word != "" {
return waits("%s's manifest uses the word %q, which this controller does not read: it is registered "+
"when a controller that reads it runs", t.Module, word)
}
return inventory.Build{}, catalogue.Manifest{}, fmt.Errorf("the builder built %s and its manifest is "+
"refused: %w", t.Module, err)
}
if manifest.Module != t.Module {
return inventory.Build{}, catalogue.Manifest{}, fmt.Errorf("the key of %s names the manifest of %s",
t.Module, manifest.Module)
}
built, err := builtOfKey(ctx, kc.archive, manifest, k)
if err != nil {
return waits("%v", err)
}
resolved, err := manifest.Resolve(built)
if err != nil {
return inventory.Build{}, catalogue.Manifest{}, fmt.Errorf("the builder built %s and it does not resolve: %w",
t.Module, err)
}
raw, err := json.Marshal(resolved)
if err != nil {
return inventory.Build{}, catalogue.Manifest{}, err
}
kept := inventory.Build{ID: keyBuildID(k), Repository: t.Trunk.Repository, Module: t.Module,
Commit: t.Trunk.Commit, On: k.Build.By, Path: keyPath(t.Trunk), Manifest: raw,
SourceFingerprint: k.Fingerprint, At: time.Now().UTC()}
// **Read from the manifest, fresh at every key** (the same rule build-agent says, internal/buildreads). Sources are
// left unsaid but for a module whose every artifact names its paths: the planner then reads the module's whole
// directory, over-planning and never under-planning, and a fingerprint that did not change moves nothing (R4e A9).
// **Its build source where its every artifact names its paths** (novox/hq ADR 0499 §1, §5; issue 1758, B1): the
// planner reads a merge's changed files against those paths and the manifest alone, so a change elsewhere in the
// module's directory touches nothing it builds from.
if own := pathsBuildSource(manifest, keyPath(t.Trunk)); own != nil {
kept.Sources = []inventory.BuildSource{{Paths: own}}
}
for _, c := range buildreads.Of(manifest) {
kept.Read = append(kept.Read, inventory.ReadRepository{Repository: c.Repository, Ref: c.Ref})
}
// **The bases its last build stood on, carried** (R4e, lead decision): what the collector keeps a mirrored base by.
// A key names none; the module's newest successful build's are carried so the keep set does not change.
for _, b := range kc.previous {
if b.Worked() && b.Module == t.Module && b.ID != kept.ID {
kept.Against = append([]string(nil), b.Against...)
break
}
}
for _, a := range k.Build.Artifacts {
kept.Made = append(kept.Made, inventory.Artifact{Name: a.Name, Kind: kindOf(manifest, a),
Reference: catalogue.Recorded(a.Reference)})
}
return kept, resolved, nil
}
// kindOf is an artifact's kind: the key's, else the manifest's.
func kindOf(m catalogue.Manifest, a sdkdelivery.BuiltArtifact) string {
if a.Kind != "" || m.Build == nil {
return a.Kind
}
for _, d := range m.Build.Artifacts {
if d.Name == a.Name {
return d.Kind
}
}
return ""
}
// builtOfKey is each of a done key's artifacts as Resolve reads it: its reference and digest, a TypeScript bundle's
// launchers by the build modules' one rule (mesh-sdk buildstep.Launchers), and, for the archive a module packs its
// templates into, each template's text read back from the archive the store holds under the key's reference.
func builtOfKey(ctx context.Context, archive func(context.Context, string) ([]byte, error), m catalogue.Manifest,
k builderkeys.Key) (
[]catalogue.Built, error) {
declared := map[string]catalogue.Artifact{}
if m.Build != nil {
for _, a := range m.Build.Artifacts {
declared[a.Name] = a
}
}
templates := ""
if m.Templated() {
templates = catalogue.TemplatesArtifact(m)
}
var out []catalogue.Built
for _, a := range k.Build.Artifacts {
d := declared[a.Name]
b := catalogue.Built{Name: a.Name, Kind: kindOf(m, a), Reference: catalogue.Recorded(a.Reference),
Digest: a.Digest}
if launchers := buildstep.Launchers(d.Language, d.Entrypoints); len(launchers) > 0 {
b.Launchers = launchers
}
if a.Name == templates && templates != "" {
body, err := archive(ctx, b.Reference)
if err != nil {
return nil, fmt.Errorf("the templates archive %q of %s cannot be read from the artifact store: %v",
a.Name, m.Module, err)
}
if b.Texts, err = unpackTexts(body); err != nil {
return nil, fmt.Errorf("the templates archive %q of %s cannot be read back: %v", a.Name, m.Module, err)
}
}
out = append(out, b)
}
return out, nil
}
// unpackTexts is every file of a packed templates archive (gzip over tar), by its path inside it.
func unpackTexts(body []byte) (map[string]string, error) {
zipped, err := gzip.NewReader(bytes.NewReader(body))
if err != nil {
return nil, err
}
defer zipped.Close()
reader := tar.NewReader(zipped)
texts := map[string]string{}
for {
header, err := reader.Next()
if errors.Is(err, io.EOF) {
return texts, nil
}
if err != nil {
return nil, err
}
if header.Typeflag != tar.TypeReg {
continue
}
text, err := io.ReadAll(io.LimitReader(reader, 16<<20))
if err != nil {
return nil, err
}
texts[strings.TrimPrefix(header.Name, "./")] = string(text)
}
}
// keepPredefinedHead keeps the list read at the catalogue's trunk head when it is newer than the one kept, as a build of
// the catalogue's did (ADR 0348 §7), and judges every manifest read again against it. Said, never fatal: the list read
// is the one this take-in judges against either way.
func keepPredefinedHead(ctx context.Context, inv *inventory.Inventory, head predefinedHead) {
kept, err := inv.KeepPredefined(ctx, inventory.PredefinedRecord{Keys: head.List, Repository: head.Repository,
Commit: head.Commit, Generation: head.Generation, Build: "the trunk head " + short(head.Commit)})
if err != nil {
fmt.Printf("the list of predefined keys at %s's trunk head %s is not kept: %v\n", head.Repository,
short(head.Commit), err)
return
}
if kept {
fmt.Printf("the list of predefined keys at %s's trunk head %s is published: %s\n", head.Repository,
short(head.Commit), strings.Join(head.List.Keys(), ", "))
}
if err := loadPredefined(ctx, inv); err != nil {
fmt.Printf("the list of predefined keys kept cannot be read: %v\n", err)
}
}
// openBuilderKeys is the builder's and the indexer's buckets, read on a connection for the moment.
func openBuilderKeys() (builderkeys.Reader, func(), error) {
js, err := aBus()
if err != nil {
return builderkeys.Reader{}, nil, err
}
api, err := jetstream.New(js.Conn())
if err != nil {
js.Close()
return builderkeys.Reader{}, nil, err
}
return builderkeys.Reader{JS: api}, js.Close, nil
}
// pathsBuildSource is the build source of a module whose every artifact names the paths it is built from (ADR 0499
// §1): its manifest and each path, relative to its repository's root, a path read as a file or everything below a
// directory of that name (`<path>/**` holds both). Nil when an artifact names none: the module reads its directory.
func pathsBuildSource(m catalogue.Manifest, dir string) []string {
if m.Build == nil || len(m.Build.Artifacts) == 0 {
return nil
}
in := func(p string) string {
if dir == "" {
return p
}
return dir + "/" + p
}
seen := map[string]bool{}
out := []string{in("module.json")}
for _, a := range m.Build.Artifacts {
if len(a.Paths) == 0 {
return nil
}
for _, p := range a.Paths {
entry := in(strings.Trim(path.Clean(p), "/")) + "/**"
if !seen[entry] {
seen[entry] = true
out = append(out, entry)
}
}
}
return out
}
@@ -1,251 +0,0 @@
//go:build integration
package main
import (
"context"
"encoding/json"
"errors"
"reflect"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/novox/mesh-controller/internal/builderkeys"
"github.com/novox/mesh-controller/internal/inventory"
)
// withTrunkList stands in for the catalogue's trunk head: an empty list of predefined keys, readable, or not.
func withTrunkList(t *testing.T, readable bool) {
t.Helper()
was := predefinedAtTrunk
predefinedAtTrunk = func(context.Context, *inventory.Inventory) (predefinedHead, error) {
if !readable {
return predefinedHead{}, errors.New("the forge did not answer")
}
list, err := catalogue.ParsePredefined([]byte(`{}`))
return predefinedHead{List: list, Repository: "novox/mesh-catalog", Commit: "abcdef0123456789",
Generation: 1 << 40}, err
}
t.Cleanup(func() { predefinedAtTrunk = was })
}
// aKeyFor is a done key of module at fp, of one archive with a reference, and the indexer's trunk value naming it at
// commit in repository/path.
func aKeyFor(module, version, fp, repository, path, commit string) (keyBuckets, sdkdelivery.Module) {
raw := `{"module":"` + module + `","version":"` + version + `","build":{"artifacts":[{"name":"a","kind":"archive","from":"files"}]},` +
`"resources":[{"id":"a","type":"archive","path":"/opt/` + module + `","artifact":"a"}]}`
digest := sdkdelivery.ManifestDigest([]byte(raw))
blob := shaOf(module + version)
k := builderkeys.Key{Module: module, Fingerprint: fp, Found: true, Revision: 1,
Build: sdkdelivery.BuilderBuild{State: sdkdelivery.BuildDone, By: "ace", Manifest: digest,
Artifacts: []sdkdelivery.BuiltArtifact{{Name: "a", Kind: "archive", Digest: blob,
Reference: "artifact-store://" + module + "/a/blobs/" + blob}}}}
return keyBuckets{keys: map[string]builderkeys.Key{module: k}, manifests: map[string]string{digest: raw}},
sdkdelivery.Module{FP2: fp, Manifest: digest, Repository: repository, Commit: commit, Path: path}
}
func fpOf(c byte) string { return "src2:" + strings.Repeat(string(c), 64) }
// A3, A4 (ADR 0498 §8, ADR 0501 §1-§3, §8): a done key of a trunk fingerprint records one build of today's layout,
// whose artifacts are the key's, and registers the module under the same rules of who may register as before.
func TestADoneKeyOfTheTrunkRegistersUnderTheSameRules(t *testing.T) {
open := theCatalogue(t) // sudo from novox/mesh-catalog modules/sudo, mesh-host from novox/mesh-host
ctx := t.Context()
inv := open.inventory
withTrunkList(t, true)
// A known module from the repository and path the catalogue builds it from: registered from its key.
keys, trunk := aKeyFor("sudo", "2", fpOf('a'), "novox/mesh-catalog", "modules/sudo", "c0ffee0123456789")
_, kept, err := takeInKey(ctx, inv, keys, keyTake{Module: "sudo", Trunk: trunk, Ask: registrationAsk{Merge: trunk.Commit}})
if err != nil {
t.Fatalf("0501 §1: a done key of sudo's trunk fingerprint was not taken in: %v", err)
}
if shelf, _ := inv.Catalogue(ctx); shelf["sudo"].Version != "2" {
t.Fatalf("0501 §1: sudo is %q", shelf["sudo"].Version)
}
got, found, err := inv.BuildByID(ctx, kept.ID)
if err != nil || !found || len(got.Made) != 1 || got.Made[0].Reference != kept.Made[0].Reference {
t.Fatalf("0498 §8: the recorded build is %+v (%v)", got, err)
}
if fps, err := inv.BuildSourceFingerprints(ctx, "sudo"); err != nil || fps["c0ffee0123456789"] != fpOf('a') {
t.Fatalf("0498 §8: the recorded build's fingerprint is %v (%v), not the key's", fps, err)
}
// Recording the same key again is one record.
if _, _, err := takeInKey(ctx, inv, keys, keyTake{Module: "sudo", Trunk: trunk, Ask: registrationAsk{Merge: trunk.Commit}}); err != nil {
t.Fatal(err)
}
builds, _ := inv.Builds(ctx, "sudo", 20)
n := 0
for _, b := range builds {
if b.ID == kept.ID {
n++
}
}
if n != 1 {
t.Fatalf("0498 §8: one key taken in twice is %d records", n)
}
// A known module from another repository: refused, as a build of it was (ADR 0266).
other, otherTrunk := aKeyFor("sudo", "evil", fpOf('b'), "agent/sudo", "modules/sudo", "d00d0123456789ab")
if _, _, err := takeInKey(ctx, inv, other, keyTake{Module: "sudo", Trunk: otherTrunk,
Ask: registrationAsk{Merge: otherTrunk.Commit}}); !errors.Is(err, errNotItsSource) {
t.Fatalf("0501 §3: a known module from another repository was taken in: %v", err)
}
// A new module from a repository the catalogue builds from: registered.
zram, zramTrunk := aKeyFor("zram", "1", fpOf('c'), "novox/mesh-catalog", "modules/zram", "c0ffee0123456789")
if _, _, err := takeInKey(ctx, inv, zram, keyTake{Module: "zram", Trunk: zramTrunk, Ask: registrationAsk{Merge: zramTrunk.Commit}}); err != nil {
t.Fatalf("0501 §3: a new module from a built repository was refused: %v", err)
}
// A new module from a repository nothing is built from, with no Approve: refused.
tools, toolsTrunk := aKeyFor("agent-tools", "1", fpOf('d'), "agent/tools", "", "e0e0e0e0e0e0e0e0")
if _, _, err := takeInKey(ctx, inv, tools, keyTake{Module: "agent-tools", Trunk: toolsTrunk,
Ask: registrationAsk{Merge: toolsTrunk.Commit}}); !errors.Is(err, errNotItsSource) {
t.Fatalf("0501 §3: a new repository's module was taken in with no warrant: %v", err)
}
// At the terminal the operator registers it (ADR 0501 §4).
if _, _, err := takeInKey(ctx, inv, tools, keyTake{Module: "agent-tools", Trunk: toolsTrunk,
Ask: registrationAsk{Terminal: true}}); err != nil {
t.Fatalf("0501 §4: the terminal's registration was refused: %v", err)
}
}
// ADR 0501 §2: a key only a candidate names registers nothing: the trunk value names another fingerprint, whose key is
// not there, and the take-in waits.
func TestAKeyOnlyACandidateNamesRegistersNothing(t *testing.T) {
open := theCatalogue(t)
ctx := t.Context()
withTrunkList(t, true)
keys, trunk := aKeyFor("sudo", "3", fpOf('e'), "novox/mesh-catalog", "modules/sudo", "c0ffee0123456789")
trunk.FP2 = fpOf('f') // the trunk names another fingerprint than the candidate's done key
_, _, err := takeInKey(ctx, open.inventory, keys, keyTake{Module: "sudo", Trunk: trunk, Ask: registrationAsk{Merge: trunk.Commit}})
if !errors.Is(err, errKeyWaits) {
t.Fatalf("0501 §2: %v", err)
}
if shelf, _ := open.inventory.Catalogue(ctx); shelf["sudo"].Version != "1" {
t.Fatalf("0501 §2: a candidate's key registered sudo %q", shelf["sudo"].Version)
}
}
// ADR 0501 §8: a take-in that cannot read the list of predefined keys at the catalogue's trunk head waits, says so and
// registers nothing.
func TestATakeInWithoutTheTrunksListWaits(t *testing.T) {
open := theCatalogue(t)
ctx := t.Context()
withTrunkList(t, false)
keys, trunk := aKeyFor("sudo", "4", fpOf('1'), "novox/mesh-catalog", "modules/sudo", "c0ffee0123456789")
_, _, err := takeInKey(ctx, open.inventory, keys, keyTake{Module: "sudo", Trunk: trunk, Ask: registrationAsk{Merge: trunk.Commit}})
if !errors.Is(err, errKeyWaits) || !strings.Contains(err.Error(), "predefined keys") {
t.Fatalf("0501 §8: %v", err)
}
if shelf, _ := open.inventory.Catalogue(ctx); shelf["sudo"].Version != "1" {
t.Fatalf("0501 §8: registered %q", shelf["sudo"].Version)
}
}
// A3 (the lead's decision on Against and Read): a key's record keeps the collector's keep set as it was and its read
// repositories are its manifest's, fresh at every key.
func TestAKeysRecordKeepsTheKeepSetAndReadsItsManifest(t *testing.T) {
open := theCatalogue(t)
ctx := t.Context()
inv := open.inventory
withTrunkList(t, true)
before, err := inv.KeepSet(ctx)
if err != nil {
t.Fatal(err)
}
raw := `{"module":"sudo","version":"5","build":{"artifacts":[{"name":"img","kind":"image","from":"Dockerfile",` +
`"context":{"repository":"novox/elsewhere","seat":"git"}}]},"resources":[{"id":"run","type":"container","artifact":"img"}]}`
digest := sdkdelivery.ManifestDigest([]byte(raw))
k := builderkeys.Key{Module: "sudo", Fingerprint: fpOf('2'), Found: true, Revision: 1,
Build: sdkdelivery.BuilderBuild{State: sdkdelivery.BuildDone, By: "g14", Manifest: digest,
Artifacts: []sdkdelivery.BuiltArtifact{{Name: "img", Kind: "image", Digest: shaOf("img5"),
Reference: "artifact-store://sudo/img@" + shaOf("img5")}}}}
keys := keyBuckets{keys: map[string]builderkeys.Key{"sudo": k}, manifests: map[string]string{digest: raw}}
trunk := sdkdelivery.Module{FP2: fpOf('2'), Manifest: digest, Repository: "novox/mesh-catalog",
Commit: "c0ffee0123456789", Path: "modules/sudo"}
if _, _, err := takeInKey(ctx, inv, keys, keyTake{Module: "sudo", Trunk: trunk, Ask: registrationAsk{Merge: trunk.Commit}}); err != nil {
t.Fatal(err)
}
after, err := inv.KeepSet(ctx)
if err != nil {
t.Fatal(err)
}
for reference, why := range before {
if !reflect.DeepEqual(after[reference], why) && !strings.Contains(strings.Join(after[reference], ","), strings.Join(why, ",")) {
t.Errorf("the keep set changed for %s: %v, was %v", reference, after[reference], why)
}
}
read, err := inv.ReadRepositories(ctx)
if err != nil {
t.Fatal(err)
}
found := false
for _, r := range read["sudo"] {
if !r.Own && r.Repository == "novox/elsewhere" {
found = true
}
}
if !found {
text, _ := json.Marshal(read["sudo"])
t.Fatalf("the key's record does not read its manifest's context: %s", text)
}
}
// A9 (ADR 0453 §6, ADR 0499 consequences): at the switch, a key's build whose every artifact has the digest of the
// build-agent build the mesh holds (`src1:`) is no move: it is registered with that build's artifacts, and a walk sends
// nothing for it.
func TestAKeyWithTheHeldBuildsDigestsIsNoMove(t *testing.T) {
open := theCatalogue(t)
ctx := t.Context()
inv := open.inventory
withTrunkList(t, true)
keys, trunk := aKeyFor("sudo", "6", fpOf('6'), "novox/mesh-catalog", "modules/sudo", "c0ffee0123456789")
k := keys.keys["sudo"]
held := inventory.Build{ID: "build-sudo-old", Module: "sudo", Repository: "novox/mesh-catalog", Path: "modules/sudo",
Commit: "0ld0ld0ld0ld0ld0", SourceFingerprint: "src1:" + strings.Repeat("9", 64),
Made: []inventory.Artifact{{Name: "a", Kind: "archive", Reference: k.Build.Artifacts[0].Reference}}}
held.Manifest, _ = json.Marshal(catalogue.Manifest{Module: "sudo", Version: "held"})
if err := inv.RecordBuild(ctx, held); err != nil {
t.Fatal(err)
}
_, kept, err := takeInKey(ctx, inv, keys, keyTake{Module: "sudo", Trunk: trunk, Ask: registrationAsk{Merge: trunk.Commit}})
if err != nil {
t.Fatal(err)
}
same, err := inv.SameSourceCommits(ctx, "sudo", kept.ID)
if err != nil || !same["0ld0ld0ld0ld0ld0"] {
t.Fatalf("A9: the key's build does not stand on the held build of the same digests: %v %v", same, err)
}
if shelf, _ := inv.Catalogue(ctx); shelf["sudo"].Version != "held" {
t.Fatalf("A9: sudo is registered as %q, not with the held build's artifacts", shelf["sudo"].Version)
}
}
// A13 against the store: the switch's list reads the build the mesh holds of each module and the builder's key of its
// trunk fingerprint, and names each module unmoved, moved or not known.
func TestTheSwitchsListReadsTheHeldBuildsAndTheKeys(t *testing.T) {
open := theCatalogue(t)
ctx := t.Context()
inv := open.inventory
keys, trunk := aKeyFor("sudo", "1", fpOf('b'), "novox/mesh-catalog", "modules/sudo", "c0ffee0123456789")
k := keys.keys["sudo"]
if err := inv.RecordBuild(ctx, inventory.Build{ID: "build-sudo-held", Module: "sudo", Repository: "novox/mesh-catalog",
Commit: "c0ffee0123456789", Made: []inventory.Artifact{{Name: "a", Kind: "archive",
Reference: k.Build.Artifacts[0].Reference}}}); err != nil {
t.Fatal(err)
}
reads := planKeysOf{values: map[string]sdkdelivery.Module{"sudo@": trunk}, keys: map[string]builderkeys.Key{"sudo": k}}
lines, err := switchList(ctx, inv, reads)
if err != nil {
t.Fatal(err)
}
got := map[string]string{}
for _, l := range lines {
got[l.Module] = l.Verdict
}
if got["sudo"] != switchUnmoved || got["mesh-host"] != switchUnknown {
t.Fatalf("A13: the switch's list is %+v", lines)
}
}
-231
View File
@@ -1,231 +0,0 @@
package main
import (
"archive/tar"
"bytes"
"compress/gzip"
"context"
"crypto/sha256"
"encoding/hex"
"encoding/json"
"errors"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/novox/mesh-controller/internal/builderkeys"
"github.com/novox/mesh-controller/internal/inventory"
)
// keyBuckets is the builder's and the indexer's buckets as a take-in reads them, for a test: keys by module, manifests by
// digest.
type keyBuckets struct {
keys map[string]builderkeys.Key
manifests map[string]string
}
func (k keyBuckets) Build(_ context.Context, module, fp string) (builderkeys.Key, error) {
if key, ok := k.keys[module]; ok && key.Fingerprint == fp {
return key, nil
}
return builderkeys.Key{Module: module, Fingerprint: fp}, nil
}
func (k keyBuckets) Manifest(_ context.Context, digest string) (sdkdelivery.Manifest, error) {
raw, ok := k.manifests[digest]
if !ok {
return sdkdelivery.Manifest{}, errors.New("the indexer holds no manifest " + digest)
}
return sdkdelivery.Manifest{Module: "x", JSON: raw}, nil
}
const keyFP = "src2:" + "1111111111111111111111111111111111111111111111111111111111111111"
// aKeyOf is a done key of module x at keyFP for the manifest raw, with these artifacts.
func aKeyOf(raw string, revision uint64, made ...sdkdelivery.BuiltArtifact) (keyBuckets, sdkdelivery.Module) {
digest := sdkdelivery.ManifestDigest([]byte(raw))
k := builderkeys.Key{Module: "x", Fingerprint: keyFP, Found: true, Revision: revision,
Build: sdkdelivery.BuilderBuild{State: sdkdelivery.BuildDone, By: "ace", Manifest: digest, Artifacts: made}}
trunk := sdkdelivery.Module{FP2: keyFP, Manifest: digest, Repository: "novox/mesh-catalog", Commit: "c0ffee0123456789",
Path: "modules/x"}
return keyBuckets{keys: map[string]builderkeys.Key{"x": k}, manifests: map[string]string{digest: raw}}, trunk
}
const tsBundle = `{"module":"x","version":"1","tools":["x_read"],
"build":{"artifacts":[{"name":"tools","kind":"bundle","language":"typescript","entrypoints":["dist/index.js"]},
{"name":"img","kind":"image","from":"Dockerfile","context":{"repository":"novox/other","seat":"git"}}]},
"resources":[{"id":"run","type":"container","artifact":"img"}]}`
func shaOf(b string) string {
sum := sha256.Sum256([]byte(b))
return "sha256:" + hex.EncodeToString(sum[:])
}
// A3 (ADR 0498 §8, ADR 0501 §1): a done key whose every artifact names a reference is one recorded build, its artifacts
// the key's, its launchers by the build modules' rule, its reads from its manifest, the bases its last build stood on
// carried; recorded twice it is one record.
func TestADoneKeyIsOneRecordedBuildOfItsArtifacts(t *testing.T) {
keys, trunk := aKeyOf(tsBundle, 3,
sdkdelivery.BuiltArtifact{Name: "tools", Kind: "bundle", Digest: shaOf("t"),
Reference: "artifact-store://x/tools/blobs/" + shaOf("t")},
sdkdelivery.BuiltArtifact{Name: "img", Kind: "image", Digest: shaOf("i"),
Reference: "artifact-store://x/img@" + shaOf("i")})
previous := []inventory.Build{{ID: "build-1", Module: "x", Against: []string{"artifact-store://base@sha256:aa"}}}
take := keyTake{Module: "x", Trunk: trunk}
kept, resolved, err := buildOfKey(t.Context(), keys, take, keyContext{previous: previous})
if err != nil {
t.Fatal(err)
}
if kept.ID != "key-1111111111111111111111111111111111111111111111111111111111111111-3" {
t.Errorf("the record's id is %s", kept.ID)
}
again, _, _ := buildOfKey(t.Context(), keys, take, keyContext{previous: previous})
if again.ID != kept.ID {
t.Errorf("one key recorded twice is two records: %s and %s", kept.ID, again.ID)
}
if kept.SourceFingerprint != keyFP || kept.Commit != trunk.Commit || kept.Path != "modules/x" || kept.On != "ace" ||
kept.Module != "x" || kept.Repository != "novox/mesh-catalog" {
t.Errorf("the record: %+v", kept)
}
if len(kept.Made) != 2 || kept.Made[0].Reference != "artifact-store://x/tools/blobs/"+shaOf("t") ||
kept.Made[1].Reference != "artifact-store://x/img@"+shaOf("i") || kept.Made[1].Kind != "image" {
t.Errorf("Made is not the key's artifacts: %+v", kept.Made)
}
if len(kept.Read) != 1 || kept.Read[0].Repository != "novox/other" {
t.Errorf("Read is not the manifest's contexts: %+v", kept.Read)
}
if kept.Sources != nil || kept.Mirrored != nil {
t.Errorf("a key's record says build sources or mirrored bases it cannot know: %+v %+v", kept.Sources, kept.Mirrored)
}
if len(kept.Against) != 1 || kept.Against[0] != "artifact-store://base@sha256:aa" {
t.Errorf("the bases the last build stood on are not carried: %v", kept.Against)
}
if len(resolved.Bundles) != 1 || resolved.Bundles[0].Launchers["dist/index.js"] != "dist/index.serve.mjs" {
t.Errorf("the bundle's launchers: %+v", resolved.Bundles)
}
var m map[string]any
if err := json.Unmarshal(kept.Manifest, &m); err != nil || !strings.Contains(string(kept.Manifest), shaOf("i")) {
t.Errorf("the recorded manifest names no image digest: %s", kept.Manifest)
}
}
// A3 (ADR 0498 §8): a key with an artifact naming no reference, a key not done and a missing key are waited for, said;
// a failed key fails in the builder's words; a module with no fingerprint of the new form waits, naming why.
func TestAKeyNotReadyWaitsAndAFailedKeyFails(t *testing.T) {
noRef, trunk := aKeyOf(tsBundle, 1,
sdkdelivery.BuiltArtifact{Name: "tools", Kind: "bundle", Digest: shaOf("t"),
Reference: "artifact-store://x/tools/blobs/" + shaOf("t")},
sdkdelivery.BuiltArtifact{Name: "img", Kind: "image", Digest: shaOf("i")})
take := keyTake{Module: "x", Trunk: trunk}
_, _, err := buildOfKey(t.Context(), noRef, take, keyContext{})
if !errors.Is(err, errKeyWaits) || !strings.Contains(err.Error(), "waits for the reference of img") {
t.Errorf("a key without a reference: %v", err)
}
building := noRef
k := building.keys["x"]
k.Build = sdkdelivery.BuilderBuild{State: sdkdelivery.BuildBuilding, By: "g14"}
building.keys = map[string]builderkeys.Key{"x": k}
if _, _, err := buildOfKey(t.Context(), building, take, keyContext{}); !errors.Is(err, errKeyWaits) ||
!strings.Contains(err.Error(), "building on g14") {
t.Errorf("a building key: %v", err)
}
if _, _, err := buildOfKey(t.Context(), keyBuckets{}, take, keyContext{}); !errors.Is(err, errKeyWaits) ||
!strings.Contains(err.Error(), "no build of x") {
t.Errorf("no key: %v", err)
}
k.Build = sdkdelivery.BuilderBuild{State: sdkdelivery.BuildFailed, Error: "digests differ: ace sha256:1, g14 sha256:2"}
failed := keyBuckets{keys: map[string]builderkeys.Key{"x": k}}
if _, _, err := buildOfKey(t.Context(), failed, take, keyContext{}); !errors.Is(err, errKeyFailed) ||
!strings.Contains(err.Error(), "digests differ: ace sha256:1, g14 sha256:2") {
t.Errorf("a failed key: %v", err)
}
none := take
none.Trunk.FP2, none.Trunk.None2 = "", "no lock"
if _, _, err := buildOfKey(t.Context(), noRef, none, keyContext{}); !errors.Is(err, errKeyWaits) ||
!strings.Contains(err.Error(), "no lock") {
t.Errorf("no fingerprint: %v", err)
}
}
// ADR 0501 §7: a manifest word this controller does not read waits for the controller that reads it; never fails.
func TestAManifestWordTheControllerDoesNotReadWaits(t *testing.T) {
raw := `{"module":"x","version":"1","not-a-word-yet":true,"build":{"artifacts":[{"name":"a","kind":"archive","from":"files"}]},
"resources":[{"id":"a","type":"archive","path":"/opt/x","artifact":"a"}]}`
keys, trunk := aKeyOf(raw, 1, sdkdelivery.BuiltArtifact{Name: "a", Kind: "archive", Digest: shaOf("a"),
Reference: "artifact-store://x/a/blobs/" + shaOf("a")})
_, _, err := buildOfKey(t.Context(), keys, keyTake{Module: "x", Trunk: trunk}, keyContext{})
if !errors.Is(err, errKeyWaits) || !strings.Contains(err.Error(), `"not-a-word-yet"`) {
t.Fatalf("an unread word: %v", err)
}
}
func packed(t *testing.T, files map[string]string) []byte {
t.Helper()
var buf bytes.Buffer
z := gzip.NewWriter(&buf)
w := tar.NewWriter(z)
for name, text := range files {
if err := w.WriteHeader(&tar.Header{Name: name, Mode: 0o644, Size: int64(len(text)), Typeflag: tar.TypeReg}); err != nil {
t.Fatal(err)
}
if _, err := w.Write([]byte(text)); err != nil {
t.Fatal(err)
}
}
if err := w.Close(); err != nil {
t.Fatal(err)
}
if err := z.Close(); err != nil {
t.Fatal(err)
}
return buf.Bytes()
}
// A templated module's texts are read back from the archive the store holds under the key's reference; an archive the
// store does not answer is waited for (the lead's Q1b).
func TestATemplatedModulesTextsAreReadFromItsArchive(t *testing.T) {
raw := `{"module":"x","version":"1","templates":"templates",
"build":{"artifacts":[{"name":"templates","kind":"archive","from":"templates"}]},
"resources":[{"id":"main","type":"file","path":"/etc/x.conf","mode":"0644","template":"templates/x.conf"}]}`
body := packed(t, map[string]string{"x.conf": "listen 1\n"})
sum := sha256.Sum256(body)
digest := "sha256:" + hex.EncodeToString(sum[:])
keys, trunk := aKeyOf(raw, 1, sdkdelivery.BuiltArtifact{Name: "templates", Kind: "archive", Digest: digest,
Reference: "artifact-store://x/templates/blobs/" + digest})
asked := ""
archive := func(_ context.Context, reference string) ([]byte, error) { asked = reference; return body, nil }
_, resolved, err := buildOfKey(t.Context(), keys, keyTake{Module: "x", Trunk: trunk}, keyContext{archive: archive})
if err != nil {
t.Fatal(err)
}
if asked != "artifact-store://x/templates/blobs/"+digest {
t.Errorf("the archive asked: %q", asked)
}
if text, _ := json.Marshal(resolved.Resources); !strings.Contains(string(text), "listen 1") {
t.Errorf("the template's text is not in the file: %s", text)
}
gone := func(context.Context, string) ([]byte, error) { return nil, errors.New("the store does not hold it") }
if _, _, err := buildOfKey(t.Context(), keys, keyTake{Module: "x", Trunk: trunk}, keyContext{archive: gone}); !errors.Is(err, errKeyWaits) ||
!strings.Contains(err.Error(), "cannot be read from the artifact store") {
t.Errorf("an archive the store does not answer: %v", err)
}
}
// The launchers a key's bundle is resolved with are mesh-sdk's one rule (buildstep.Launchers), the names
// mesh-build-node writes: a TypeScript bundle's entrypoint x.js runs as x.serve.mjs, a Go bundle has none.
func TestALaunchersRuleIsTheBuildModules(t *testing.T) {
m, err := catalogue.ParseManifest([]byte(tsBundle))
if err != nil {
t.Fatal(err)
}
k := builderkeys.Key{Build: sdkdelivery.BuilderBuild{Artifacts: []sdkdelivery.BuiltArtifact{{Name: "tools", Digest: "sha256:1",
Reference: "artifact-store://x/tools/blobs/sha256:1"}}}}
built, err := builtOfKey(t.Context(), nil, m, k)
if err != nil {
t.Fatal(err)
}
if built[0].Launchers["dist/index.js"] != "dist/index.serve.mjs" || built[0].Kind != "bundle" {
t.Fatalf("built: %+v", built[0])
}
}
-117
View File
@@ -1,117 +0,0 @@
package main
import (
"context"
"time"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/inventory"
)
// Backing off before standing (novox/hq issue 906).
//
// A controller loses its lease when its bus stops answering for longer than the key's age: on the control-node,
// a bus whose disks are saturated has taken 30 to 42 seconds to answer (2026-10-23 05:52). The process then
// exits, as ADR 0229 rule 4 says, and its service manager starts it again five seconds later. **A start is the
// heaviest thing a controller asks of the bus** — every bucket, stream and consumer asserted again — and a start
// that cannot assert the lease's bucket used to exit at once and be started again five seconds later, on the
// same overloaded bus, nine times in twelve minutes (2026-10-16). So a candidate backs off twice:
//
// - **After a loss**, before it touches the bus: from the newest epoch lost in the last ten minutes (the
// controller's store says it, rule 7), it waits one key's age, doubled for every further loss in that
// window, at most two minutes. A single loss costs a key's age; a controller losing it again and again
// stands less and less often.
// - **While the bus does not answer it**, in its own process: asserting the lease's bucket, opening it or
// reading who holds it is tried again after 5 s, doubling to at most 60 s, instead of exiting into a restart
// every five seconds. Nothing acts meanwhile: the process holds no lease, and the gate refuses every act.
const (
// lossWindow is how far back lost epochs count towards the backoff.
lossWindow = 10 * time.Minute
// lossMost is the longest a candidate waits after losses.
lossMost = 2 * time.Minute
// standFirst and standMost are the waits between tries of a candidate the bus does not answer.
standFirst = 5 * time.Second
standMost = time.Minute
)
// lossBackoff is how long a candidate waits now before it stands, after the epochs it read: one key's age from
// the newest loss in the window, doubled for every further loss in it, at most lossMost.
func lossBackoff(epochs []inventory.Epoch, now time.Time) time.Duration {
return lossBackoffFrom(epochs, now, broker.LeaseTTL)
}
func lossBackoffFrom(epochs []inventory.Epoch, now time.Time, base time.Duration) time.Duration {
lost, newest := lostSince(epochs, now)
if lost == 0 {
return 0
}
delay := base
for i := 1; i < lost && delay < lossMost; i++ {
delay *= 2
}
delay = min(delay, lossMost)
return max(newest.Add(delay).Sub(now), 0)
}
// standUp runs try until it answers nil, waiting standFirst between tries and doubling to standMost; it answers
// ctx's error once ctx ends.
func standUp(ctx context.Context, sleep func(context.Context, time.Duration) error, say func(string, ...any),
try func(context.Context) error) error {
wait := standFirst
for {
err := try(ctx)
if err == nil {
return nil
}
if ctx.Err() != nil {
return ctx.Err()
}
say("this controller cannot stand for the lease yet: %v — it tries again in %s, in this process, rather "+
"than start again on a bus that does not answer (novox/hq issue 906)", err, wait)
if err := sleep(ctx, wait); err != nil {
return err
}
wait = min(wait*2, standMost)
}
}
// backoffAfterLosses is how long this candidate waits before it stands, read from the epochs its store kept,
// and said. A store that cannot be read makes it wait nothing: the backoff spares the bus, and never keeps a
// controller from standing.
func (a *actor) backoffAfterLosses(ctx context.Context, inv *inventory.Inventory, say func(string, ...any)) time.Duration {
now := time.Now()
reading, cancel := context.WithTimeout(ctx, 10*time.Second)
epochs, err := inv.EpochsSince(reading, now.Add(-lossWindow))
cancel()
if err != nil {
say("whether the lease was lost lately cannot be read, so this controller stands without backing off: %v", err)
return 0
}
base := a.lossBase
if base <= 0 {
base = broker.LeaseTTL
}
wait := lossBackoffFrom(epochs, now, base)
if wait > 0 {
lost, _ := lostSince(epochs, now)
say("the controller lease was lost lately (%d time(s) in the last %s): this controller backs off %s "+
"before it stands, so its start does not add to the load that lost it (novox/hq issue 906)",
lost, lossWindow, wait.Round(time.Second))
}
return wait
}
// lostSince is how many epochs were lost in the window before now, and when the newest of them was.
func lostSince(epochs []inventory.Epoch, now time.Time) (lost int, newest time.Time) {
for _, e := range epochs {
if e.Ended == nil || e.How != inventory.EpochLost || now.Sub(*e.Ended) > lossWindow {
continue
}
lost++
if e.Ended.After(newest) {
newest = *e.Ended
}
}
return lost, newest
}
-97
View File
@@ -1,97 +0,0 @@
package main
import (
"context"
"errors"
"slices"
"testing"
"time"
"github.com/novox/mesh-controller/internal/inventory"
)
// A controller that lost its lease to a slow bus backs off before it stands again (novox/hq issue 906): it
// does not add its start — every bucket and consumer asserted again — to the load that lost it the lease.
func lostAt(epoch uint64, ended time.Time, how string) inventory.Epoch {
return inventory.Epoch{Epoch: epoch, Instance: "controller@anchor", Taken: ended.Add(-time.Hour), Ended: &ended,
How: how}
}
// 906/AC1: given epochs lost in the last ten minutes, when a candidate starts, then it waits from the newest
// loss one key's age, doubled for each further loss, at most two minutes; nothing else makes it wait.
func TestACandidateAfterALostLeaseBacksOff(t *testing.T) {
now := time.Date(2026, 10, 23, 5, 52, 40, 0, time.UTC)
for _, c := range []struct {
name string
epochs []inventory.Epoch
want time.Duration
}{
{"nothing lost", nil, 0},
{"one held, one given back", []inventory.Epoch{
{Epoch: 3, Taken: now.Add(-time.Minute)},
lostAt(2, now.Add(-2*time.Second), inventory.EpochReleased),
}, 0},
{"one lost five seconds ago", []inventory.Epoch{lostAt(7, now.Add(-5*time.Second), inventory.EpochLost)},
10 * time.Second},
{"one lost long ago enough", []inventory.Epoch{lostAt(7, now.Add(-20*time.Second), inventory.EpochLost)}, 0},
{"three lost, the newest a second ago", []inventory.Epoch{
lostAt(9, now.Add(-time.Second), inventory.EpochLost),
lostAt(8, now.Add(-3*time.Minute), inventory.EpochLost),
lostAt(7, now.Add(-6*time.Minute), inventory.EpochLost),
}, 59 * time.Second},
{"six lost: at most two minutes", []inventory.Epoch{
lostAt(12, now, inventory.EpochLost), lostAt(11, now.Add(-time.Minute), inventory.EpochLost),
lostAt(10, now.Add(-2*time.Minute), inventory.EpochLost), lostAt(9, now.Add(-3*time.Minute), inventory.EpochLost),
lostAt(8, now.Add(-4*time.Minute), inventory.EpochLost), lostAt(7, now.Add(-5*time.Minute), inventory.EpochLost),
}, 2 * time.Minute},
{"a loss older than ten minutes does not count", []inventory.Epoch{
lostAt(9, now.Add(-time.Second), inventory.EpochLost),
lostAt(8, now.Add(-11*time.Minute), inventory.EpochLost),
}, 14 * time.Second},
{"expired is not lost", []inventory.Epoch{lostAt(9, now.Add(-time.Second), inventory.EpochExpired)}, 0},
} {
t.Run(c.name, func(t *testing.T) {
if got := lossBackoff(c.epochs, now); got != c.want {
t.Fatalf("a candidate after %d epoch(s) waits %s, want %s", len(c.epochs), got, c.want)
}
})
}
}
// 906/AC2: given a bus that does not answer a candidate, when it asserts the lease's bucket or reads who holds
// it, then the candidate tries again in its own process after 5 s, doubling to at most 60 s, instead of exiting
// to be restarted by its service manager every five seconds.
func TestACandidateTheBusDoesNotAnswerTriesAgainWithBackoff(t *testing.T) {
var waited []time.Duration
sleep := func(ctx context.Context, d time.Duration) error { waited = append(waited, d); return nil }
tries := 0
slow := errors.New("asserting bucket mesh-controller_lease: context deadline exceeded")
err := standUp(t.Context(), sleep, func(string, ...any) {}, func(context.Context) error {
tries++
if tries < 7 {
return slow
}
return nil
})
if err != nil {
t.Fatalf("a candidate whose bus answered at the seventh try gave up: %v", err)
}
want := []time.Duration{5 * time.Second, 10 * time.Second, 20 * time.Second, 40 * time.Second, time.Minute,
time.Minute}
if tries != 7 || !slices.Equal(waited, want) {
t.Fatalf("a candidate tried %d time(s), waiting %v between; want 7, waiting %v", tries, waited, want)
}
}
// 906/AC2: and a candidate stopping while it waits stops, with why.
func TestACandidateStoppingWhileItBacksOffStops(t *testing.T) {
ctx, cancel := context.WithCancel(t.Context())
cancel()
err := standUp(ctx, sleepOrDone, func(string, ...any) {}, func(context.Context) error {
return errors.New("who holds the lease cannot be read: context deadline exceeded")
})
if !errors.Is(err, context.Canceled) {
t.Fatalf("a candidate stopped while backing off answered %v", err)
}
}
@@ -152,33 +152,3 @@ func TestTwoControllersOneActs(t *testing.T) {
t.Fatalf("B's epoch does not say it was lost: %+v", ended)
}
}
// 906/AC1, on a real bus and store: a controller starting right after an epoch was lost backs off before it
// stands — it takes the lease no sooner than the backoff after the loss, and then holds it as usual.
func TestAControllerAfterALostLeaseBacksOffBeforeItStands(t *testing.T) {
url, _ := aBusForTheLease(t)
inv := inventory.ForTest(t)
ctx := t.Context()
// The controller before it lost epoch 5, and said so in the store, as `took` does on a loss.
if _, err := inv.TookEpoch(ctx, inventory.Epoch{Epoch: 5, Instance: "controller@anchor pid 1", Host: "anchor",
Taken: time.Now()}); err != nil {
t.Fatal(err)
}
if err := inv.EndEpoch(ctx, 5, inventory.EpochLost); err != nil {
t.Fatal(err)
}
// The key's age stands in as a backoff of two seconds, so the case runs in seconds (issue 591).
const backoff = 2 * time.Second
a := &actor{renewEvery: lease.RenewEvery / 5, poll: lease.Poll / 10, lossBase: backoff}
start := time.Now()
if _, err := a.serveUnderTheLease(ctx, inv, url); err != nil {
t.Fatal(err)
}
defer a.release()
if waited := time.Since(start); waited < backoff-300*time.Millisecond {
t.Fatalf("a controller took the lease %s after an epoch was lost, without backing off %s", waited, backoff)
}
if epoch, err := a.epoch(ctx); err != nil || epoch <= 5 {
t.Fatalf("after backing off the controller holds epoch %d (%v), want one above the lost 5", epoch, err)
}
}
+14 -44
View File
@@ -54,8 +54,6 @@ func run() error {
return fmt.Errorf("no command given")
}
// The bus's users are composed from what each module announced it keeps (novox/hq ADR 0495, issue 1790).
inventory.Announcements = readAnnouncements
// Every connection this process dials says what it is (novox/hq issue 327).
broker.ConnectionName = connectionName(args[0], os.Getenv(verbVar))
@@ -75,13 +73,8 @@ func run() error {
}
switch args[0] {
case "switch":
return switchCommand(ctx, args[1:])
case "register":
return registerCommand(ctx, args[1:])
case "build", "rebuild":
// **The hand build goes with the asks** (novox/hq ADR 0501 §5): refused by name, naming the replacement.
return removedByName(args[0])
case "build":
return buildCommand(ctx, args[1:])
case "builder":
return builderCommand(ctx, args[1:])
case "board":
@@ -105,12 +98,14 @@ func run() error {
return cancelCommand(ctx, args[1:])
case "clear":
return clearCommand(ctx, args[1:])
case "rebuild":
return rebuildCommand(ctx, args[1:])
case "replay":
return replayCommand(ctx, args[1:])
case "kill":
return killCommand(ctx, args[1:])
case "forget-checks":
return removedByName("forget-checks")
return forgetChecksCommand(ctx, args[1:])
case "try":
return tryCommand(ctx, args[1:])
case "pause", "resume":
@@ -237,8 +232,8 @@ func run() error {
}
}
// usageText is the terminal's help: every command it has, and none it refuses as removed (removedByName).
var usageText = `mesh-controller — the control plane
func usage() {
fmt.Fprint(os.Stderr, `mesh-controller — the control plane
On a node the operator types these as 'mesh-cli <command>' (in zsh, 'nox <command>'): asked
through the node's engine, and run as the controller's terminal only on the control-node
@@ -312,14 +307,15 @@ var usageText = `mesh-controller — the control plane
operator key make [--out <file>] make the operator's sealing key, off the mesh; private half to the file only
operator key set <public> [--replace] tell the mesh which operator key to seal to
operator key show the operator key, and what it can recover
register <module> <owner>/<repository> <path> register a module from a repository the catalogue does not
build, from the builder's done key of its fingerprint (ADR 0501 §4)
switch [--json] the switch's list: each module, the build it runs, and what the builder holds
build <repository> [--ref R] have a build machine build it, and record what came out
build --behind build every module the mesh holds older than its source
build --on <module> rebuild every module that stands on this module's artifacts, bases first
builds [<module>] what has been built lately, and what came of it
queue [--json] every ask in the build queue: waiting, in flight (where, how long), dead
cancel <id> drop a waiting or dead ask; recorded failed, cancelled by hand
clear [--dead] cancel every waiting ask (and the dead ones); never one in flight
replay <build-id> that build's commit again, a dry run: what a build of it makes
rebuild <module|build-id> ask the module's source again, or that build's, under a new id
replay <build-id> [--register [--older]] that build's commit again; a dry run unless --register
kill <id> end a build where it runs; recorded failed, killed by hand
pause [<node>] / resume [<node>] the build seat's holder there, or every holder, takes nothing new / again
plans retry <id> ask a failed plan's failed builds again, and carry the plan on
@@ -354,13 +350,10 @@ var usageText = `mesh-controller — the control plane
version what this binary is
Each context reaches its own store through its own credential (novox/hq ADR 0008), named
` + store.Variable("<context>") + ` — or ` + store.FileVariable("<context>") + `, naming a file that holds
`+store.Variable("<context>")+` — or `+store.FileVariable("<context>")+`, naming a file that holds
the same thing and keeps the password out of the environment. This process holds:
`
func usage() {
fmt.Fprint(os.Stderr, usageText)
`)
for _, c := range held {
fmt.Fprintf(os.Stderr, " %-12s database %-12s from %s\n",
c.name, store.Database(c.name), store.Variable(c.name))
@@ -482,26 +475,3 @@ func connectionName(command, verb string) string {
}
return "mesh-controller command " + command
}
// removedByName refuses a command the controller no longer has, naming what replaced it (novox/hq ADR 0501 §4, §5;
// ADR 0429 §18; ADR 0503 §3; issue 1851, A11, and issue 1318).
func removedByName(command string) error {
replacement := map[string]string{
"build": "the controller asks no build (ADR 0501 §5): the builder builds every fingerprint the indexer writes, a " +
"module is registered from its done key, and a branch is tried with `try`; a module from a repository the " +
"catalogue does not build is registered at the controller's terminal with `register`",
"rebuild": "the controller asks no build (ADR 0501 §5): a failed key is built again by the builder as a new " +
"revision (ADR 0429 §18), and a walk's Retry waits for it",
"replay --register": "a build is registered only from the builder's done key of a trunk fingerprint (ADR 0501 " +
"§1, §2); `replay` without --register still shows what a build of a recorded commit makes",
"forget-checks": "build-agent makes no check build from R4e (ADR 0503 §3), so there is nothing to let go of",
"delivery-plan": "a delivery's plan is read with `delivery-order` and the checker's verdict (ADR 0503 §3)",
}[command]
return &removedCommand{msg: fmt.Sprintf("`%s` is removed: %s", command, replacement)}
}
// removedCommand is a command or verb the controller no longer has: a refusal that is the verb's whole answer, not a
// verb of a newer build this one cannot run.
type removedCommand struct{ msg string }
func (r *removedCommand) Error() string { return r.msg }
+2 -20
View File
@@ -65,9 +65,6 @@ type mergeCheckInput struct {
// group are the other heads of a delivery group composed with this one (novox/hq ADR 0239), each laid
// over the mesh after the one before: the group judged as one future state.
group []groupChange
// held are the approved config values held for the keys the pull requests introduce (novox/hq ADR 0502 clause 4),
// laid over the composition with the change only.
held []heldValue
// admin is a PostgreSQL the gate may create and drop databases in.
admin string
// beside is the directory holding the core clones beside the check (MESH_CHECK_BESIDE), for finding moves.
@@ -192,8 +189,6 @@ func mergeGateCommand(ctx context.Context, args []string) error {
"--changed, for a list too long for an argument (novox/hq issue 1247)")
group := set.String("group", "", "a delivery group's other heads, as JSON [{repository, tree, changed}], composed "+
"with this one as one future state (novox/hq ADR 0239)")
heldFile := set.String("held", "", "a file of the config values approved for keys the change's pull requests introduce, as JSON "+
"[{pull, node, module, key, value, approved_by}], laid over the composition with the change (novox/hq ADR 0502)")
asJSON := set.Bool("json", false, "print the verdict as JSON")
if _, err := parseAround(set, args); err != nil {
return err
@@ -213,10 +208,6 @@ func mergeGateCommand(ctx context.Context, args []string) error {
if err != nil {
return err
}
held, err := readHeld(*heldFile)
if err != nil {
return err
}
out := io.Writer(os.Stdout)
stdout := os.Stdout
if *asJSON {
@@ -232,7 +223,7 @@ func mergeGateCommand(ctx context.Context, args []string) error {
return err
}
v, err := judgeChange(ctx, mergeCheckInput{facts: f, repository: *repository, tree: *tree, changed: changedList,
group: others, held: held, admin: *admin, say: out, beside: os.Getenv("MESH_CHECK_BESIDE")})
group: others, admin: *admin, say: out, beside: os.Getenv("MESH_CHECK_BESIDE")})
os.Stdout = stdout
if err != nil {
return err
@@ -396,16 +387,7 @@ func judgeChange(ctx context.Context, in mergeCheckInput) (mergeVerdict, error)
}
}
sort.Strings(added)
// **The held values are laid in this composition only** (novox/hq ADR 0502 clause 4): the mesh as it is does not
// declare their keys, so the base composition would refuse them.
withHeld := in
var heldNotes []string
withHeld.facts, heldNotes = layHeld(in.facts, in.held, change)
for _, n := range heldNotes {
say("%s", n)
}
v.Notes = append(v.Notes, heldNotes...)
after, tried, notes, err := composeEveryMachine(ctx, withHeld, change, changeSources, added)
after, tried, notes, err := composeEveryMachine(ctx, in, change, changeSources, added)
if err != nil {
return v, fmt.Errorf("the mesh with the change could not be raised: %w", err)
}
-135
View File
@@ -1,135 +0,0 @@
package main
import (
"encoding/json"
"fmt"
"os"
"sort"
"strings"
"git.novox.be/novox/mesh-sdk/go/catalogue"
snapshot "git.novox.be/novox/mesh-sdk/go/facts"
)
// Config values held for the keys an open pull request introduces (novox/hq ADR 0502 clause 4, issue 1826): the gate
// lays each approved one over its module's values on its node in the composition with the change, and in that one
// only. In the mesh as it is the key is not declared, so the value would be refused there.
// heldValue is one approved held value as the build agent writes it (--held): the value of Key for Module on Node,
// a machine's real name, which the snapshot holds only as a pseudonym.
type heldValue struct {
Pull string `json:"pull"`
Head string `json:"head,omitempty"`
Node string `json:"node"`
Module string `json:"module"`
Key string `json:"key"`
Value json.RawMessage `json:"value"`
Approved string `json:"approved,omitempty"`
By string `json:"by,omitempty"`
At string `json:"at,omitempty"`
}
// readHeld reads the file --held names; none named, none held.
func readHeld(path string) ([]heldValue, error) {
if path == "" {
return nil, nil
}
raw, err := os.ReadFile(path)
if err != nil {
return nil, err
}
var held []heldValue
if err := json.Unmarshal(raw, &held); err != nil {
return nil, fmt.Errorf("the held values are not readable: %w", err)
}
for _, h := range held {
if h.Node == "" || h.Module == "" || h.Key == "" || len(h.Value) == 0 {
return nil, fmt.Errorf("a held value of %s names no node, module, key or value", h.Pull)
}
}
return held, nil
}
// layHeld is the snapshot with each held value laid as the value of its key in its node's layer of its module, and what
// it said of them. The snapshot given is not changed. **A held value for a machine the snapshot does not hold, or a
// module the change does not define, is said and not laid**, never dropped silently; one for a module not assigned to
// its machine is laid and said. The notes name machines as the snapshot does (roles and pseudonym), not by their names.
func layHeld(f snapshot.Facts, held []heldValue, shelf map[string]catalogue.Manifest) (snapshot.Facts, []string) {
if len(held) == 0 {
return f, nil
}
out := f
out.Machines = make([]snapshot.Machine, len(f.Machines))
copy(out.Machines, f.Machines)
var notes, laid []string
pulls := map[string]bool{}
for _, h := range held {
at := -1
pseudonym := snapshot.Pseudonym("machine", h.Node)
for i, m := range out.Machines {
if m.Name == pseudonym {
at = i
}
}
if at < 0 {
notes = append(notes, fmt.Sprintf("the value held for %s of %s (%s=%s) is not composed: the snapshot holds no machine %s",
h.Module, h.Pull, h.Key, h.Value, pseudonym))
continue
}
if _, defined := shelf[h.Module]; !defined {
notes = append(notes, fmt.Sprintf("the value held for %s of %s (%s=%s) is not composed: the change defines no module %s",
h.Module, h.Pull, h.Key, h.Value, h.Module))
continue
}
var value any
if err := json.Unmarshal(h.Value, &value); err != nil {
notes = append(notes, fmt.Sprintf("the value held for %s of %s (%s) is not composed: it is not a value: %v",
h.Module, h.Pull, h.Key, err))
continue
}
m := &out.Machines[at]
layers := make([]snapshot.Settings, len(m.Settings))
copy(layers, m.Settings)
found := false
for i := range layers {
if layers[i].Module == h.Module {
values := make(map[string]any, len(layers[i].Values)+1)
for k, v := range layers[i].Values {
values[k] = v
}
values[h.Key] = value
layers[i].Values = values
found = true
}
}
if !found {
layers = append(layers, snapshot.Settings{Module: h.Module, Values: map[string]any{h.Key: value}})
sort.Slice(layers, func(a, b int) bool { return layers[a].Module < layers[b].Module })
}
m.Settings = layers
assigned := false
for _, a := range m.Assigned {
assigned = assigned || a == h.Module
}
by := ""
if h.Approved != "" {
by = ", approved by " + h.Approved
}
line := fmt.Sprintf("%s on %s %s=%s%s", h.Module, m.Described(), h.Key, h.Value, by)
if !assigned {
line += " (not assigned there in the snapshot)"
}
laid = append(laid, line)
pulls[h.Pull] = true
}
if len(laid) > 0 {
var names []string
for p := range pulls {
names = append(names, p)
}
sort.Strings(names)
notes = append([]string{fmt.Sprintf("composed with %d value(s) held for %s: %s", len(laid), strings.Join(names, ", "),
strings.Join(laid, "; "))}, notes...)
}
return out, notes
}
@@ -1,85 +0,0 @@
package main
import (
"os"
"path/filepath"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
snapshot "git.novox.be/novox/mesh-sdk/go/facts"
)
func heldGateFacts() snapshot.Facts {
return snapshot.Facts{Machines: []snapshot.Machine{
{Name: snapshot.Pseudonym("machine", "ace"), Roles: []string{"the build node"}, Assigned: []string{"n8n"},
Settings: []snapshot.Settings{{Module: "n8n", Values: map[string]any{"port": 5678.0}}}},
{Name: snapshot.Pseudonym("machine", "g14")},
}}
}
var heldShelf = map[string]catalogue.Manifest{"n8n": {Module: "n8n"}}
// 1826/AC3: a held value is laid in its node's layer of its module, beside what the layer holds, and said.
func TestAHeldValueIsLaidOverItsNodesLayerAndSaid(t *testing.T) {
f := heldGateFacts()
got, notes := layHeld(f, []heldValue{{Pull: "novox/mesh-catalog#434", Node: "ace", Module: "n8n", Key: "media-gid",
Value: []byte("2000"), Approved: "the operator"}}, heldShelf)
layer := got.Machines[0].Settings[0].Values
if layer["media-gid"] != 2000.0 || layer["port"] != 5678.0 {
t.Fatalf("the layer holds %v", layer)
}
if _, has := f.Machines[0].Settings[0].Values["media-gid"]; has {
t.Fatal("the snapshot given was changed")
}
if len(notes) != 1 || !strings.Contains(notes[0], "composed with 1 value(s) held for novox/mesh-catalog#434") ||
!strings.Contains(notes[0], "media-gid=2000") || !strings.Contains(notes[0], "approved by the operator") ||
strings.Contains(notes[0], "ace") {
t.Fatalf("said %q", notes)
}
}
// 1826/AC3: a machine with no layer of the module gets one.
func TestAHeldValueMakesTheLayerAMachineLacks(t *testing.T) {
got, _ := layHeld(heldGateFacts(), []heldValue{{Pull: "p", Node: "g14", Module: "n8n", Key: "k", Value: []byte(`"x"`)}}, heldShelf)
s := got.Machines[1].Settings
if len(s) != 1 || s[0].Module != "n8n" || s[0].Values["k"] != "x" {
t.Fatalf("layers %+v", s)
}
}
// 1826/AC3: a held value for a machine or a module the snapshot or the change does not hold is said, never laid.
func TestAHeldValueForWhatIsNotHeldIsSaidNotLaid(t *testing.T) {
f := heldGateFacts()
got, notes := layHeld(f, []heldValue{
{Pull: "p", Node: "nowhere", Module: "n8n", Key: "k", Value: []byte("1")},
{Pull: "p", Node: "ace", Module: "ghost", Key: "k", Value: []byte("1")},
}, heldShelf)
if len(notes) != 2 || !strings.Contains(notes[0], "holds no machine") || !strings.Contains(notes[1], "defines no module ghost") {
t.Fatalf("said %q", notes)
}
if _, has := got.Machines[0].Settings[0].Values["k"]; has {
t.Fatal("laid")
}
}
// 1826/AC3: none held lays nothing and says nothing; the file is read and refuses a value without a node.
func TestNoHeldValuesLayNothingAndTheFileIsRead(t *testing.T) {
if got, notes := layHeld(heldGateFacts(), nil, heldShelf); len(notes) != 0 || len(got.Machines) != 2 {
t.Fatalf("%v %v", got, notes)
}
dir := t.TempDir()
good := filepath.Join(dir, "good.json")
os.WriteFile(good, []byte(`[{"pull":"p","node":"ace","module":"n8n","key":"k","value":1}]`), 0o644)
if h, err := readHeld(good); err != nil || len(h) != 1 || h[0].Key != "k" {
t.Fatalf("%v %v", h, err)
}
bad := filepath.Join(dir, "bad.json")
os.WriteFile(bad, []byte(`[{"pull":"p","module":"n8n","key":"k","value":1}]`), 0o644)
if _, err := readHeld(bad); err == nil {
t.Fatal("a value without a node was read")
}
if h, err := readHeld(""); h != nil || err != nil {
t.Fatalf("%v %v", h, err)
}
}
@@ -526,57 +526,3 @@ func TestAMemberJudgedAloneSeesAMoveThroughTheClonesBeside(t *testing.T) {
t.Errorf("nothing beside, yet moves %v", got)
}
}
// 1826/AC3 (integration level, a throwaway store): a change whose container says ${config:media-gid} fails the gate for
// want of the value, passes it with the value the operator approved held for its pull request, and the mesh as it is
// composes the same either way.
func TestAChangeNeedingAHeldValueFailsWithoutItAndPassesWithIt(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
trunk := `{"module":"n8n","version":"1","resources":[{"id":"server","type":"container","name":"n8n",
"image":"n8nio/n8n@sha256:4846eb2f4b874ab04cde7fc1e249d2ddaec66e9aea64439beb2972cfea88e3c0"}]}`
m, err := catalogue.ParseManifest([]byte(trunk))
if err != nil {
t.Fatal(err)
}
if err := open.inventory.RegisterModule(ctx, m, inventory.Source{Repository: "novox/mesh-catalog",
Path: "modules/n8n", BuiltFrom: "c0ffee"}); err != nil {
t.Fatal(err)
}
if _, err := assign(ctx, open, "laptop", "n8n"); err != nil {
t.Fatal(err)
}
f, err := gatherFacts(ctx, open, "2.11.17")
if err != nil {
t.Fatal(err)
}
for i := range f.Machines {
f.Machines[i].Capabilities = append(f.Machines[i].Capabilities, snapshot.Capability{Name: "container-groups", Present: true})
}
tree := aTree(t, map[string]string{"n8n": n8nWithAGroup})
judge := func(held []heldValue) mergeVerdict {
v, err := judgeChange(t.Context(), mergeCheckInput{facts: f, admin: os.Getenv("MESH_TEST_POSTGRES"),
repository: "novox/mesh-catalog", tree: tree, changed: []string{"modules/n8n/module.json"}, held: held})
if err != nil {
t.Fatal(err)
}
return v
}
without := judge(nil)
if without.Verdict != "fail" || !strings.Contains(without.Report(), "media-gid") {
t.Fatalf("a required value nothing sets passed, or failed for another reason:\n%s", without.Report())
}
with := judge([]heldValue{{Pull: "novox/mesh-catalog#434", Node: "laptop", Module: "n8n", Key: "media-gid",
Value: []byte("2000"), Approved: "the operator"}})
if with.Verdict == "fail" || with.Verdict == "error" {
t.Fatalf("the held value did not make it compose:\n%s", with.Report())
}
if !strings.Contains(with.Report(), "composed with 1 value(s) held for novox/mesh-catalog#434") {
t.Fatalf("the value laid is not said:\n%s", with.Report())
}
for i := range with.Machines {
if with.Machines[i].Base.Composes != without.Machines[i].Base.Composes {
t.Fatalf("the mesh as it is composes differently with a held value: %+v / %+v", with.Machines[i].Base, without.Machines[i].Base)
}
}
}
-10
View File
@@ -13,7 +13,6 @@ import (
"github.com/novox/mesh-controller/internal/conditions"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"git.novox.be/novox/mesh-sdk/go/overlay"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/licences"
@@ -56,18 +55,9 @@ func aMeshOnTheStore(t *testing.T) *stores {
t.Fatal(err)
}
t.Cleanup(open.Close)
// The work this mesh started in the background ends before its stores close (cleanups run last first).
t.Cleanup(background.Wait)
// A condition store of its own, held in memory (novox/hq to-be 45 §2): status leads with what is
// open, and a mesh with no bus would otherwise read as one whose conditions cannot be read.
withConditionsInMemory(t)
// The indexer holds no module unless a test puts one on its bus (aBuilderOnTheBus): an assignment's look at it
// reads "not held" on a mesh with no bus of its own.
wasIndexed := indexedModule
indexedModule = func(context.Context, string) (sdkdelivery.Module, bool, error) {
return sdkdelivery.Module{}, false, nil
}
t.Cleanup(func() { indexedModule = wasIndexed })
for _, m := range provided {
if err := open.inventory.Provide(t.Context(), m); err != nil {
t.Fatal(err)
+20 -26
View File
@@ -52,22 +52,23 @@ type deliveryMerge struct {
Asker string `json:"asker,omitempty"`
}
// askerOf is the asker whose warrant the merge is read under: the one the announcement names. One that names none is
// from mesh-delivery of before the gate, whose warrant is no longer taken (novox/hq ADR 0450 §8; issue 1318).
func (d deliveryMerge) askerOf() string { return d.Asker }
// askerOf is the asker whose warrant the merge is read under: the one the announcement names, mesh-delivery when it
// names none.
func (d deliveryMerge) askerOf() string {
if d.Asker == "" {
return theDelivery
}
return d.Asker
}
// askerAccepted is whether the controller takes a warrant from this asker for a merge: the teller alone, since the gate
// took the delivery seat over (novox/hq ADR 0437 §6, ADR 0450 §8; issue 1318). Any other registers no move.
// askerAccepted is whether the controller takes a warrant from this asker for a merge: the teller, and mesh-delivery
// until the gate takes the seat over (novox/hq ADR 0437 §6). Any other registers no move.
func askerAccepted(asker string) error {
if asker == sdkdelivery.TellerModule {
if asker == theDelivery || asker == sdkdelivery.TellerModule {
return nil
}
if asker == "" {
return fmt.Errorf("the merge names no asker, as mesh-delivery's of before the gate did, and the controller takes "+
"a warrant only from %s (ADR 0450 §8)", sdkdelivery.TellerModule)
}
return fmt.Errorf("the merge names the asker %q, and the controller takes a warrant only from %s (ADR 0450 §8)",
asker, sdkdelivery.TellerModule)
return fmt.Errorf("the merge names the asker %q, and the controller takes a warrant only from %s and %s",
asker, sdkdelivery.TellerModule, theDelivery)
}
// parseDeliveryMerge reads an announcement, refusing one that names no pull request, head or ask.
@@ -264,32 +265,25 @@ var errNotAMerge = errors.New("the build is not of a merge")
// (ADR 0420 §7): the build is of a merge, that merge is a pull request mesh-delivery announced it made, and the
// ask it made it on holds an Approve by the operator of that head. Answers why it was accepted; an error says why
// it was not, and the refusal of ADR 0266 stands.
func approvedMove(ctx context.Context, inv registrations, forms *sourceForms, ask registrationAsk,
func approvedMove(ctx context.Context, inv registrations, forms *sourceForms, buildID string,
built inventory.Source) (string, error) {
commit := ask.Merge
if ask.Build != "" {
r, found, err := inv.BuildRequestByID(ctx, ask.Build)
if err != nil {
return "", err
}
if !found || r.For != "merge" || r.Commit == "" {
return "", errNotAMerge
}
commit = r.Commit
r, found, err := inv.BuildRequestByID(ctx, buildID)
if err != nil {
return "", err
}
if commit == "" {
if !found || r.For != "merge" || r.Commit == "" {
return "", errNotAMerge
}
owner, name, onForge := forms.onTheForge(built.Repository, built.Seat)
if !onForge {
return "", errors.New("the repository is not on the mesh's forge")
}
merged, known, err := inv.MergeOf(ctx, owner+"/"+name, commit)
merged, known, err := inv.MergeOf(ctx, owner+"/"+name, r.Commit)
if err != nil {
return "", err
}
if !known {
return "", fmt.Errorf("the controller kept no merge of %s/%s at %.8s", owner, name, commit)
return "", fmt.Errorf("the controller kept no merge of %s/%s at %.8s", owner, name, r.Commit)
}
return approvedAnnounced(ctx, inv, owner, name, announcedOf(merged))
}
+7 -8
View File
@@ -48,19 +48,18 @@ func TestAMergeNamingTheTellerIsMovedOnTheTellersWarrant(t *testing.T) {
}
}
// 1318 (ADR 0450 §8): a merge naming no asker, or the asker mesh-delivery of before the gate, registers no move; the
// teller's is taken, and its warrant read under its name.
func TestAMergeOfMeshDeliveryOrNoAskerMovesNothing(t *testing.T) {
// 1213/AC2: an announcement naming mesh-delivery, or none (today's module): read under mesh-delivery, registered.
func TestAMergeNamingNoAskerIsMovedOnMeshDeliverysWarrant(t *testing.T) {
for _, named := range []string{"", theDelivery} {
a := askedBy(t, named, theDelivery)
if _, err := movedOnApprove(a.merge, &a.ask, &a.warrant); err == nil || !strings.Contains(err.Error(), "ADR 0450 §8") {
t.Errorf("asker %q: %v", named, err)
if _, err := movedOnApprove(a.merge, &a.ask, &a.warrant); err != nil {
t.Errorf("asker %q: refused: %v", named, err)
}
}
was := routerAskOf
wasKept := keptDeliveryMerge
t.Cleanup(func() { routerAskOf, keptDeliveryMerge = was, wasKept })
a := askedBy(t, sdkdelivery.TellerModule, sdkdelivery.TellerModule)
a := askedBy(t, "", theDelivery)
keptDeliveryMerge = func(context.Context, string, string, int, string) (deliveryMerge, bool, error) {
return a.merge, true, nil
}
@@ -72,8 +71,8 @@ func TestAMergeOfMeshDeliveryOrNoAskerMovesNothing(t *testing.T) {
if _, err := approvedAnnounced(t.Context(), nil, "novox", "mesh-controller", announcedPull(7, a.merge.HeadSHA)); err != nil {
t.Fatal(err)
}
if readUnder != sdkdelivery.TellerModule {
t.Errorf("the warrant was read under %q, not %q", readUnder, sdkdelivery.TellerModule)
if readUnder != theDelivery {
t.Errorf("with no asker named the warrant was read under %q, not %q", readUnder, theDelivery)
}
}
+1 -2
View File
@@ -7,7 +7,6 @@ import (
"encoding/json"
"errors"
"fmt"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"testing"
"time"
@@ -69,7 +68,7 @@ func TestAModuleBuiltFromANewRepositoryIsRegisteredOnlyOnTheOperatorsApproveOnTh
a.merge.Owner, a.merge.Repo, a.merge.Number = "novox", "mesh-host", 7
if c.answer {
record, _ := json.Marshal(map[string]any{"state": "closed", "ask": a.ask, "warrant": a.warrant})
if _, err := records.Put(ctx, sdkdelivery.TellerModule+"."+a.merge.AskID, record); err != nil {
if _, err := records.Put(ctx, theDelivery+"."+a.merge.AskID, record); err != nil {
t.Fatal(err)
}
}
+2 -3
View File
@@ -1,7 +1,6 @@
package main
import (
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"strings"
"testing"
"time"
@@ -22,7 +21,7 @@ func approvedMerge(t *testing.T) anApprovedMerge {
heads := map[string]string{"novox/mesh-controller@aaaaaaaaaaaa": "a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1",
"novox/mesh-catalog@bbbbbbbbbbbb": "b2b2b2b2b2b2b2b2b2b2b2b2b2b2b2b2b2b2b2b2"}
m := deliveryMerge{Owner: "novox", Repo: "mesh-controller", Number: 7, HeadSHA: "a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1",
Group: "g1", AskID: "merge-g1", Option: optionApprove, Heads: heads, Asker: sdkdelivery.TellerModule}
Group: "g1", AskID: "merge-g1", Option: optionApprove, Heads: heads}
bind := func(option string, h map[string]string) string {
d, err := asks.ActDigest(mergeAct(m.AskID, option, h))
if err != nil {
@@ -36,7 +35,7 @@ func approvedMerge(t *testing.T) anApprovedMerge {
Options: []asks.Option{
{ID: optionApprove, Label: "Approve", Does: "merges the group", Level: asks.Approve, Binds: bind(optionApprove, heads)},
{ID: "decline", Label: "Decline", Does: "leaves it", Level: asks.Approve, Binds: bind("decline", heads)}}}
w := asks.Warrant{Ask: m.AskID, Asker: sdkdelivery.TellerModule, Outcome: asks.OutcomeChosen, Option: optionApprove, Label: "Approve",
w := asks.Warrant{Ask: m.AskID, Asker: theDelivery, Outcome: asks.OutcomeChosen, Option: optionApprove, Label: "Approve",
Level: asks.Approve, By: &asks.Person{Who: asks.Operator, Kind: "telegram", Identity: "1", Verified: "user id"},
At: at, AskDigest: ask.Digest()}
return anApprovedMerge{merge: m, ask: ask, warrant: w}
@@ -1,40 +0,0 @@
package main
import (
"strings"
"testing"
"github.com/novox/mesh-controller/internal/conditions"
)
// novox/hq ADR 0505 §5: a delivery whose cure is its author's new head reaches the operator in words that name its
// author and say a new head is needed, offer no Stop and never call it the operator's; the line as mesh-delivery-gate
// says it (ready.go readyLines: needs "new-head", author, branch).
func TestADeliveryThatNeedsANewHeadNamesItsAuthorAndOffersNoStop(t *testing.T) {
l := stalledLine{ID: "novox/mesh-controller@499ff5c21861", Number: 432, State: "rejected", For: "2h1m0s", Bound: "2h0m0s",
H2: "none: a new head from its author, or a recheck that passes", Needs: needsNewHead,
Author: "the author of feat/quiet-deliveries (quiet-deliveries on shanks)", Branch: "feat/quiet-deliveries",
Says: "novox/mesh-controller#432 passed its checks at 499ff5c21861 and is not merged: a conflict"}
o := stalledObservations([]stalledLine{l})[0]
for _, words := range []string{o.Headline, o.Explanation, o.Resolved, o.Needs} {
if why, ok := conditions.Plain(words); !ok {
t.Errorf("%q is not plain: %s", words, why)
}
for _, never := range []string{"stop it", "operator's"} {
if strings.Contains(words, never) {
t.Errorf("%q says %q", words, never)
}
}
}
if o.Headline != "Pull request 432 of mesh-controller needs a new head from its author" ||
!strings.Contains(o.Explanation, "until the author of feat/quiet-deliveries (quiet-deliveries on shanks) pushes a new head") ||
!strings.Contains(o.Explanation, "for 2 hours") || len(o.Actions) != 0 ||
o.Needs != "ask the author of feat/quiet-deliveries (quiet-deliveries on shanks) for a new head." {
t.Fatalf("words: %q / %q / %q / %v", o.Headline, o.Explanation, o.Needs, o.Actions)
}
// A branch in code is not said; the author is then "its author".
l.Author = "the author of fix/a1b2c3d4e5f6"
if o := stalledObservations([]stalledLine{l})[0]; !strings.Contains(o.Needs, "ask its author") {
t.Fatalf("a dotted branch was said: %q", o.Needs)
}
}
@@ -116,7 +116,7 @@ func TestANewModuleIsRegisteredOnTheOperatorsApproveOfItsMergeAndRefusedOtherwis
if c.built != nil {
src = *c.built
}
id, err := mayRegisterIn(t.Context(), mesh, forms, "mesh-delivery-view", src, byBuild("build-1"), "", "main")
id, err := mayRegisterIn(t.Context(), mesh, forms, "mesh-delivery-view", src, "build-1", "", "main")
switch {
case c.accepted && err != nil:
t.Fatalf("%s: the Approved merge's new module was refused: %v", c.ac, err)
@@ -157,7 +157,7 @@ func TestASecondNewModuleFromARepositoryIsJudgedAlongsideAndAsksNoWarrant(t *tes
request: inventory.BuildRequest{ID: "build-2", Repository: "novox/mesh-delivery", Seat: "git", For: "build"}}}
forms := &sourceForms{bases: map[string]string{}, base: func(string) string { return "http://forge.internal:20000" }}
built := inventory.Source{Repository: "novox/mesh-delivery", Seat: "git"}
if id, err := mayRegisterIn(t.Context(), mesh, forms, "mesh-delivery-watchman", built, byBuild("build-2"), "", "main"); err != nil || id == 0 {
if id, err := mayRegisterIn(t.Context(), mesh, forms, "mesh-delivery-watchman", built, "build-2", "", "main"); err != nil || id == 0 {
t.Fatalf("1146/AC5: a second module from a repository the catalogue builds from was refused (id %d): %v", id, err)
}
}
+1 -2
View File
@@ -7,7 +7,6 @@ import (
"encoding/json"
"errors"
"fmt"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"testing"
"time"
@@ -70,7 +69,7 @@ func TestANewModuleFromANewRepositoryIsRegisteredOnlyOnTheOperatorsApproveOnTheB
map[string]string{"novox/mesh-delivery@a1a1a1a1a1a1": "a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1"})
if c.answer {
record, _ := json.Marshal(map[string]any{"state": "closed", "ask": a.ask, "warrant": a.warrant})
if _, err := records.Put(ctx, sdkdelivery.TellerModule+"."+a.merge.AskID, record); err != nil {
if _, err := records.Put(ctx, theDelivery+"."+a.merge.AskID, record); err != nil {
t.Fatal(err)
}
}
+58 -70
View File
@@ -4,110 +4,98 @@ package main
import (
"context"
"strings"
"sync"
"testing"
"time"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// ADR 0501 §6 (issue 1851, A5): the live case of 2026-10-07 (novox/hq issue 300), on keys. A merge of the catalogue adds
// one module's directory and touches nothing else the mesh holds: nothing asks its first build, it opens no plan, and
// the module is registered from the builder's first done key of it, assigned nowhere, with no build request written.
func TestAMergeAddingAModuleEndsRegisteredFromItsKeyAndAsksNoBuild(t *testing.T) {
// asksWithPaths records every build a merge asks for, as repository, path and ref. The merge window's goroutine and
// the test's own may ask at once, so the record is written under a lock (issue 1543).
func asksWithPaths(t *testing.T) *[][3]string {
t.Helper()
var asked [][3]string
var mu sync.Mutex
was := askABuild
askABuild = func(_ context.Context, source buildSource, path, ref string) (string, error) {
mu.Lock()
defer mu.Unlock()
asked = append(asked, [3]string{source.Repository, path, ref})
return "b-" + path, nil
}
t.Cleanup(func() { askABuild = was })
return &asked
}
// The live case of 2026-10-07 (novox/hq issue 300), replayed: a merge of the catalogue adds one module's
// directory and touches nothing else the mesh holds. Its delivery plan said "builds new: …, sent nowhere";
// the merge said it changed nothing any module the mesh holds is built from, and built nothing, so the
// module was never registered and `assign` refused it. The merge asks for its build, at the branch merged
// into, from the repository as the mesh spells it — and opens no plan, since nothing it holds moved.
func TestAMergeBuildsTheNewModuleItAdds(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
inv := open.inventory
noBuildAsked(t)
b := aBuilderOnTheBus(t)
asked := asksWithPaths(t)
if err := inv.RegisterModule(ctx, catalogue.Manifest{Module: "networkmanager", Version: "1"},
inventory.Source{Repository: "novox/mesh-catalog", Seat: "git", Path: "modules/networkmanager", Ref: "main",
BuiltFrom: "c0", Head: "c0"}); err != nil {
t.Fatal(err)
}
watching, stop := context.WithCancel(context.Background())
t.Cleanup(stop)
watchNewModules(watching, open)
const commit = "3da80a4b00000000"
m := link.SourceMoved{Owner: "novox", Repo: "mesh-catalog", Base: "main", Commit: commit,
m := link.SourceMoved{Owner: "novox", Repo: "mesh-catalog", Base: "main", Commit: "3da80a4b00",
Paths: []string{"modules/systemd-resolved/module.json", "modules/systemd-resolved/cmd/main.go"},
ModuleDirs: []string{"modules/systemd-resolved"}, ModuleDirsSaid: true}
if err := (following{open: open}).SourceMoved(ctx, m); err != nil {
t.Fatal(err)
}
want := [3]string{"novox/mesh-catalog", "modules/systemd-resolved", "main"}
if len(*asked) != 1 || (*asked)[0] != want {
t.Fatalf("the merge adding modules/systemd-resolved asked for %v, not %v", *asked, want)
}
if plans, err := inv.OpenPlans(ctx); err != nil || len(plans) != 0 {
t.Fatalf("a merge moving nothing the mesh holds opened a plan: %+v %v", plans, err)
}
// The indexer writes the module's trunk value; the builder builds it, then its key is done.
fp := fpOf('7')
manifest := b.indexed(t, "systemd-resolved", "1", "novox/mesh-catalog", "modules/systemd-resolved", commit, fp)
b.built(t, "systemd-resolved", "1", fp, manifest, sdkdelivery.BuildBuilding)
time.Sleep(300 * time.Millisecond)
if _, known := inventoryHolds(t, inv, "systemd-resolved"); known {
t.Fatal("0501 §1: registered from a key still building")
}
if obs := newModuleObservations(time.Now().Add(time.Hour)); len(obs) != 1 || obs[0].Kind != kindNewModuleWaits {
t.Fatalf("0501 §6: the new module waiting on its key past its bound is named by %+v", obs)
}
b.built(t, "systemd-resolved", "1", fp, manifest, sdkdelivery.BuildDone)
becomes(t, 10*time.Second, "the new module registered", func() bool {
_, known := inventoryHolds(t, inv, "systemd-resolved")
return known
})
// The take-in forgets the wait just after the registration it made is seen.
becomes(t, 5*time.Second, "the registered module no longer named waiting", func() bool {
return len(newModuleObservations(time.Now().Add(time.Hour))) == 0
})
requests, err := inv.RequestsNamed(ctx, inventory.BuildRequest{Repository: "novox/mesh-catalog",
Path: "modules/systemd-resolved"}.Name())
if err != nil || len(requests) != 0 {
t.Fatalf("0501 §6: a build request was written: %+v (%v)", requests, err)
}
}
// ADR 0501 §7: a new module whose manifest uses a word this controller does not read is named waiting at once, never
// failed, and registered from the same key, with no second build, when a controller that reads it looks.
func TestANewModuleWithAnUnreadWordWaitsForTheControllerThatReadsIt(t *testing.T) {
// A merge that changes a held module and adds a new one does both: the held one walks its plan, the new
// one is asked for beside it. A merge adding nothing builds nothing new.
func TestAMergeBuildsItsNewModuleBesideItsPlan(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
inv := open.inventory
noBuildAsked(t)
b := aBuilderOnTheBus(t)
if err := inv.RegisterModule(ctx, catalogue.Manifest{Module: "networkmanager", Version: "1"},
inventory.Source{Repository: "novox/mesh-catalog", Seat: "git", Path: "modules/networkmanager", Ref: "main",
asked := asksWithPaths(t)
if err := inv.RegisterModule(ctx, catalogue.Manifest{Module: "app", Version: "1"},
inventory.Source{Repository: "novox/mesh-catalog", Seat: "git", Path: "modules/app", Ref: "main",
BuiltFrom: "c0", Head: "c0"}); err != nil {
t.Fatal(err)
}
fp := fpOf('8')
raw := `{"module":"watchman","version":"1","not-a-word-yet":"5m","build":{"artifacts":[{"name":"a","kind":"archive","from":"files"}]},` +
`"resources":[{"id":"a","type":"archive","path":"/opt/watchman","artifact":"a"}]}`
digest := sdkdelivery.ManifestDigest([]byte(raw))
value, _ := sdkdelivery.EncodeManifest(sdkdelivery.Manifest{Module: "watchman", JSON: raw})
if _, err := b.manifests.Put(ctx, digest, value); err != nil {
m := link.SourceMoved{Owner: "novox", Repo: "mesh-catalog", Base: "main", Commit: "c1aaaaaaaa",
Paths: []string{"modules/app/index.ts", "modules/fresh/module.json"},
ModuleDirs: []string{"modules/app", "modules/fresh"}, ModuleDirsSaid: true}
if err := (following{open: open}).SourceMoved(ctx, m); err != nil {
t.Fatal(err)
}
trunk, _ := sdkdelivery.EncodeModule(sdkdelivery.Module{FP: "src1:" + fp[5:], FP2: fp, Manifest: digest,
Repository: "novox/mesh-catalog", Commit: "4444444400000000", Path: "modules/watchman"})
if _, err := b.modules.Put(ctx, "watchman", trunk); err != nil {
paths := map[string]bool{}
for _, a := range *asked {
paths[a[1]] = true
}
if !paths["modules/fresh"] || !paths["modules/app"] || len(*asked) != 2 {
t.Fatalf("asked %v; want the new module and the plan's first tier", *asked)
}
plans, err := inv.OpenPlans(ctx)
if err != nil || len(plans) != 1 || plans[0].Modules["fresh"] != nil || plans[0].Modules["app"] == nil {
t.Fatalf("the plan should walk app alone: %+v %v", plans, err)
}
*asked = nil
m2 := link.SourceMoved{Owner: "novox", Repo: "mesh-catalog", Base: "main", Commit: "c2aaaaaaaa",
Paths: []string{"README.md"}, ModuleDirsSaid: true}
if err := (following{open: open}).SourceMoved(ctx, m2); err != nil {
t.Fatal(err)
}
b.built(t, "watchman", "1", fp, digest, sdkdelivery.BuildDone)
takeNewModules(ctx, open, "watchman")
if _, known := inventoryHolds(t, inv, "watchman"); known {
t.Fatal("0501 §7: a manifest with an unread word was registered")
}
obs := newModuleObservations(time.Now())
if len(obs) != 1 || obs[0].Kind != kindNewModuleWaits || !strings.Contains(obs[0].Summary, `"not-a-word-yet"`) {
t.Fatalf("0501 §7: the module waiting for a reader of its word is named by %+v", obs)
}
builds, _ := inv.Builds(ctx, "watchman", 5)
for _, b := range builds {
if !b.Worked() {
t.Fatalf("0501 §7: a failed build was recorded: %+v", b)
}
if len(*asked) != 0 {
t.Fatalf("a merge adding no module asked for %v", *asked)
}
}
-309
View File
@@ -1,309 +0,0 @@
package main
import (
"context"
"errors"
"fmt"
"slices"
"sort"
"strings"
"sync"
"time"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/builderkeys"
"github.com/novox/mesh-controller/internal/conditions"
"github.com/novox/mesh-controller/internal/inventory"
)
// A new module is registered from its first done key (novox/hq ADR 0501 §1, §6; issue 1851, A5).
//
// Nothing asks a new module's first build: the builder builds every fingerprint the indexer writes, a new module's
// included (ADR 0429 §5). The controller registers a module the catalogue does not hold from the builder's done key of the
// fingerprint the indexer's trunk value names, under the same rules of who may register as before (ADR 0501 §3): from a
// repository the catalogue builds another module from, or on the operator's Approve of the merge the trunk value names
// (a first merge, kept with mesh-delivery's word), or at the terminal (ADR 0501 §4). It is assigned nowhere.
//
// **No timer** (ADR 0426): a write of the builder or the indexer wakes it for that module (builderkeys.Watch), a first
// merge's word wakes it for that repository, and a controller that starts looks at every module once (ADR 0501 §7: a
// manifest the controller before could not read is registered when the controller that reads it runs).
// newModulesMu keeps one take-in of new modules at a time in this process.
var newModulesMu sync.Mutex
// newModuleWait is a new module the controller could not register yet, and why: what the first-builds condition names.
type newModuleWait struct {
Module, Repository, Commit, Why string
// Word is the manifest word this controller does not read, when that is why (ADR 0501 §7).
Word string
Since time.Time
}
// newModulesWaiting are the new modules waiting, by module, in this process: a restart looks at every module again.
var newModulesWaiting = struct {
sync.Mutex
by map[string]newModuleWait
}{by: map[string]newModuleWait{}}
// takeNewModules looks at the modules the indexer holds that the catalogue does not, all of them or those named, and
// registers each whose done key may register it. Each outcome is said once: a module still waiting for the same reason
// is not said again.
func takeNewModules(ctx context.Context, open *stores, only ...string) {
newModulesMu.Lock()
defer newModulesMu.Unlock()
keys, closeKeys, err := openBuilderKeys()
if err != nil {
fmt.Printf("new modules: the builder's keys cannot be read: %v\n", err)
return
}
defer closeKeys()
for _, module := range takeNewModulesWith(ctx, open.inventory, keys, only...) {
// An assignment waiting for it is made now (ADR 0501 §6; issue 1851, A10).
for _, line := range settlePendingOnKey(ctx, open, module) {
fmt.Println(line)
}
}
}
// takeNewModulesWith is takeNewModules on the buckets handed in; it answers the modules it registered.
func takeNewModulesWith(ctx context.Context, inv *inventory.Inventory, keys newModuleReads, only ...string) []string {
entries, err := inv.Catalogued(ctx)
if err != nil {
fmt.Printf("new modules: the catalogue cannot be read: %v\n", err)
// empty-on-error: said above; nothing is registered this time, and the next write of a key looks again
return nil
}
held := map[string]bool{}
for _, e := range entries {
held[e.Manifest.Module] = true
}
values, unread, err := keys.Modules(ctx)
if err != nil {
fmt.Printf("new modules: the indexer's modules cannot be read: %v\n", err)
// empty-on-error: said above; nothing is registered this time, and the next write of a key looks again
return nil
}
for module, err := range unread {
if !held[module] && (len(only) == 0 || slices.Contains(only, module)) {
sayNewModuleWait(newModuleWait{Module: module, Why: "the indexer's value of it cannot be read: " + err.Error()})
}
}
names := make([]string, 0, len(values))
for module := range values {
if !held[module] && (len(only) == 0 || slices.Contains(only, module)) {
names = append(names, module)
}
}
sort.Strings(names)
var registered []string
for _, module := range names {
v := values[module]
_, _, err := takeInKey(ctx, inv, keys, keyTake{Module: module, Trunk: v, Ask: registrationAsk{Merge: v.Commit}})
switch {
case err == nil:
forgetNewModuleWait(module)
registered = append(registered, module)
fmt.Printf("%s is new in %s at %.8s: registered from the builder's done key, assigned nowhere\n", module,
v.Repository, v.Commit)
case errors.Is(err, errKeyWaits):
w := newModuleWait{Module: module, Repository: v.Repository, Commit: v.Commit, Why: err.Error()}
if word := wordIn(err.Error()); word != "" {
w.Word = word
}
sayNewModuleWait(w)
default:
sayNewModuleWait(newModuleWait{Module: module, Repository: v.Repository, Commit: v.Commit,
Why: "not registered: " + err.Error()})
}
}
return registered
}
// newModuleReads is what the take-in of new modules reads of the builder and the indexer.
type newModuleReads interface {
keyReads
Modules(ctx context.Context) (map[string]sdkdelivery.Module, map[string]error, error)
}
var _ newModuleReads = builderkeys.Reader{}
// wordIn is the manifest word a take-in's words say this controller does not read.
func wordIn(said string) string {
_, rest, ok := strings.Cut(said, "uses the word \"")
if !ok {
return ""
}
word, _, _ := strings.Cut(rest, "\"")
return word
}
// sayNewModuleWait keeps a new module waiting and says it, once for each reason.
func sayNewModuleWait(w newModuleWait) {
newModulesWaiting.Lock()
defer newModulesWaiting.Unlock()
was, known := newModulesWaiting.by[w.Module]
if known && was.Why == w.Why && was.Commit == w.Commit {
return
}
w.Since = time.Now().UTC()
if known && was.Commit == w.Commit {
w.Since = was.Since
}
newModulesWaiting.by[w.Module] = w
fmt.Printf("%s is new at %.8s and not registered yet: %s\n", w.Module, w.Commit, firstLine(w.Why))
}
func forgetNewModuleWait(module string) {
newModulesWaiting.Lock()
defer newModulesWaiting.Unlock()
delete(newModulesWaiting.by, module)
}
// kindNewModuleWaits is what a new module waiting for this controller's reader of its manifest, or waiting past its
// bound, raises (ADR 0501 §6, §7: the first-builds condition reads the key, not a build request).
const kindNewModuleWaits = "new-module-waits"
// newModuleWaitBound is how long a new module waits for its key before it is said: the builder builds a fingerprint in
// minutes. Not a delay: its key's write registers it the moment it is done.
const newModuleWaitBound = 30 * time.Minute
// newModuleObservations names the new modules waiting: one whose manifest uses a word this controller does not read at
// once (nothing is lost, but the operator sees what waits for which controller), any other past the bound.
func newModuleObservations(now time.Time) []conditions.Observation {
newModulesWaiting.Lock()
defer newModulesWaiting.Unlock()
var out []conditions.Observation
for _, w := range newModulesWaiting.by {
if w.Word == "" && now.Sub(w.Since) <= newModuleWaitBound {
continue
}
head := fmt.Sprintf("%s is not registered: %s", w.Module, firstLine(w.Why))
resolved := "its key is done and its take-in registers it"
if w.Word != "" {
head = fmt.Sprintf("%s waits for a controller that reads the manifest word %q", w.Module, w.Word)
resolved = "a controller that reads the word runs and registers it from the same key, with no second build"
}
out = append(out, conditions.Observation{Scope: conditions.ScopeMerge,
ID: fmt.Sprintf("%s_%s", w.Module, short(w.Commit)), Kind: kindNewModuleWaits, Severity: conditions.Warning,
Summary: head,
Said: fmt.Sprintf("%s at %s waits since %s: %s", w.Module, w.Commit, w.Since.Format(time.RFC3339), w.Why),
Headline: fmt.Sprintf("The new module %s is not registered yet", w.Module),
Explanation: fmt.Sprintf("The merge %.8s of %s adds %s. The mesh registers it from the builder's done key of its "+
"fingerprint, and has not yet: %s", w.Commit, w.Repository, w.Module, w.Why),
Resolved: resolved,
})
}
sort.Slice(out, func(i, j int) bool { return out[i].ID < out[j].ID })
return out
}
// watchNewModules wakes the take-in of new modules for each module the builder or the indexer writes, until ctx ends, and
// looks at every module once now. A bucket not there yet is said: the look at start and a first merge's word still work.
func watchNewModules(ctx context.Context, open *stores) {
inBackground(func() {
answerHeldMerges(ctx, open.inventory)
takeNewModules(ctx, open)
})
keys, closeKeys, err := openBuilderKeys()
if err != nil {
fmt.Printf("new modules: the builder's keys are not watched: %v\n", err)
return
}
says := watchingSays{js: keys.JS}
says.watch(ctx)
if err := builderkeys.Watch(ctx, keys.JS, func(module string) {
says.watch(ctx)
wakeNewModule(ctx, open, module)
}); err != nil {
closeKeys()
fmt.Printf("new modules: the builder's keys are not watched: %v\n", err)
return
}
go func() {
<-ctx.Done()
closeKeys()
}()
}
// wakeNewModule takes module in when the catalogue does not hold it: never on the watch's own goroutine, which must keep
// draining.
func wakeNewModule(ctx context.Context, open *stores, module string) {
inBackground(func() {
entries, err := open.inventory.Catalogued(ctx)
if err != nil {
return
}
for _, e := range entries {
if e.Manifest.Module == module {
// Known: a plan waiting on its key is woken by the plans' own look.
askToLook()
return
}
}
takeNewModules(ctx, open, module)
})
}
// background is the work the take-ins of keys and of announcements start beside the event that woke them, so a test
// ends only once the work its mesh started has.
var background sync.WaitGroup
// inBackground runs work beside the caller, counted in background.
func inBackground(work func()) {
background.Add(1)
go func() {
defer background.Done()
work()
}()
}
// newModulesOf are the modules the indexer holds in a repository: what a first merge's word wakes.
func newModulesOf(ctx context.Context, repository string) ([]string, error) {
keys, closeKeys, err := openBuilderKeys()
if err != nil {
return nil, err
}
defer closeKeys()
values, _, err := keys.Modules(ctx)
if err != nil {
return nil, err
}
var out []string
for module, v := range values {
if strings.EqualFold(v.Repository, repository) {
out = append(out, module)
}
}
sort.Strings(out)
return out, nil
}
// watchingSays watches the builder's bucket of instances once it is there (issue 1886): a change of what the builder says
// asks the look, which judges every open walk again. A builder from before R4e makes no such bucket; a write of the
// builder's keys asks again whether it is there now, so nothing waits on a timer.
type watchingSays struct {
js jetstream.JetStream
mu *sync.Mutex
on *bool
}
func (w *watchingSays) watch(ctx context.Context) {
if w.mu == nil {
w.mu, w.on = &sync.Mutex{}, new(bool)
}
w.mu.Lock()
defer w.mu.Unlock()
if *w.on {
return
}
on, err := builderkeys.WatchSays(ctx, w.js, askToLook)
if err != nil {
fmt.Printf("plans: what the builder says it writes is not watched now: %v\n", err)
return
}
if on {
*w.on = true
askToLook() // what it says now may differ from what the last look read
}
}
-33
View File
@@ -1,33 +0,0 @@
package main
import (
"strings"
"testing"
"time"
)
// ADR 0501 §6, §7 (issue 1851, A5b): the condition of a new module reads its key's take-in: one waiting for a controller
// that reads its manifest's word is named at once, one waiting for its key only past the bound, each reason said once.
func TestANewModuleWaitingIsNamedByWhatItWaitsFor(t *testing.T) {
t.Cleanup(func() {
newModulesWaiting.Lock()
newModulesWaiting.by = map[string]newModuleWait{}
newModulesWaiting.Unlock()
})
word := `waits: watchman's manifest uses the word "key-ttl", which this controller does not read: it is registered when a controller that reads it runs`
sayNewModuleWait(newModuleWait{Module: "watchman", Commit: "c1", Why: word, Word: wordIn(word)})
sayNewModuleWait(newModuleWait{Module: "view", Commit: "c1", Why: "waits: the build of view at aaaa is building on ace"})
now := time.Now()
obs := newModuleObservations(now)
if len(obs) != 1 || !strings.Contains(obs[0].Summary, `"key-ttl"`) || obs[0].Kind != kindNewModuleWaits {
t.Fatalf("at once, only the unread word is named: %+v", obs)
}
if obs := newModuleObservations(now.Add(newModuleWaitBound + time.Minute)); len(obs) != 2 {
t.Fatalf("past the bound, both are named: %+v", obs)
}
forgetNewModuleWait("watchman")
forgetNewModuleWait("view")
if obs := newModuleObservations(now.Add(time.Hour)); len(obs) != 0 {
t.Fatalf("registered modules are still named: %+v", obs)
}
}
+24
View File
@@ -0,0 +1,24 @@
package main
import (
"testing"
"github.com/novox/mesh-controller/internal/chain"
)
// 1146: a merge's ask that failed is kept under an id the bus takes as a key, for a module at the repository's root
// (".") as for one in a folder.
func TestTheIdOfABuildNotAskedIsAKeyTheBusTakes(t *testing.T) {
for dir, want := range map[string]string{
".": "not-asked-0177ea7a",
"modules/mesh-delivery-view": "not-asked-0177ea7a-modules-mesh-delivery-view",
} {
id := notAskedID("0177ea7a00000000", dir)
if id != want {
t.Errorf("notAskedID(%q) = %q, want %q", dir, id, want)
}
if !chain.ValidKey("req." + id) {
t.Errorf("notAskedID(%q) = %q: the bus takes no key req.%s", dir, id, id)
}
}
}
+11 -6
View File
@@ -109,8 +109,8 @@ func TestAFeatureWithANewSeatVerbIsOneBatchAndOnePlanInOrderWithNoAfterLine(t *t
}
// Issue 1049: a move delivered half before its second merge. The controller adds a module the catalogue removes, in one
// group: the merges are one batch and one plan in either order, no build of the module is asked (it is registered from
// the builder's key, ADR 0501 §6), and the catalogue's removal does not forget it.
// group: the merges are one batch and one plan in either order, the module is built once and from the controller's
// repository, and the catalogue's removal does not forget it.
func TestAMoveOfAModuleAcrossTwoRepositoriesIsOnePlanAndBuildsItOnceFromTheNewOne(t *testing.T) {
for name, controllerFirst := range map[string]bool{"controller_first": true, "catalogue_first": false} {
t.Run(name, func(t *testing.T) {
@@ -149,12 +149,17 @@ func TestAMoveOfAModuleAcrossTwoRepositoriesIsOnePlanAndBuildsItOnceFromTheNewOn
if len(ws) != 1 || len(bs) != 0 {
t.Fatalf("a move: %d walk(s), %d batch(es); want exactly one plan", len(ws), len(bs))
}
// **No build of it is asked from either repository** (ADR 0501 §6, issue 1851): the move is registered from the
// builder's done key of route-proxy in the controller's repository, on the operator's Approve of the group.
var fromController int
for _, a := range *asked {
if a[1] == "modules/route-proxy" {
t.Errorf("a build of route-proxy was asked: %v", a)
if strings.Contains(a[0], "mesh-controller") && a[1] == "modules/route-proxy" {
fromController++
}
if strings.Contains(a[0], "mesh-catalog") && a[1] == "modules/route-proxy" {
t.Errorf("route-proxy was built from the repository it left: %v", a)
}
}
if fromController != 1 {
t.Fatalf("route-proxy was built %d time(s) from the controller's repository: %v", fromController, *asked)
}
entries, err := open.inventory.Catalogued(ctx)
if err != nil {
+42
View File
@@ -28,6 +28,48 @@ func stoodOn(edges map[string][]string) map[string][]string {
return out
}
// A module built before the module it stands on is built against the old one and reports success
// (novox/hq 04-ISSUES/131). So bases come first, however the set arrived.
func TestBasesAreBuiltBeforeWhatStandsOnThem(t *testing.T) {
in := []inventory.Entry{entry("app"), entry("runtime"), entry("other"), entry("base")}
edges := stoodOn(map[string][]string{"app": {"runtime"}, "runtime": {"base"}})
got := orderByBases(in, edges)
pos := map[string]int{}
for i, e := range got {
pos[e.Manifest.Module] = i
}
if !(pos["base"] < pos["runtime"] && pos["runtime"] < pos["app"]) {
t.Fatalf("bases not first: %v", pos)
}
if len(got) != 4 {
t.Fatalf("an entry was lost or doubled: %d", len(got))
}
// A base outside the set is not waited for: it is not being rebuilt.
got = orderByBases([]inventory.Entry{entry("app")}, stoodOn(map[string][]string{"app": {"elsewhere"}}))
if len(got) != 1 {
t.Fatalf("a dependency outside the set changed the set: %v", got)
}
}
// A module registered from its manifest and never built still names its bases there; once built,
// the recorded edge is what says so. Both are read, and a module never stands on itself.
func TestWhatStandsOnAModuleIsReadFromItsBuildOrItsManifest(t *testing.T) {
built := entry("gitea")
edges := stoodOn(map[string][]string{"gitea": {"mesh-tools"}})
if !standsOnModule(built, "mesh-tools", edges) {
t.Fatal("a recorded edge was not read")
}
if standsOnModule(built, "gitea", edges) || standsOnModule(built, "postgres", edges) {
t.Fatal("an edge was invented")
}
fresh := inventory.Entry{Manifest: catalogue.Manifest{Module: "plex", Build: &catalogue.Build{
On: []catalogue.BuildsOn{{Arg: "RUNTIME_BASE", Module: "mesh-tools", Artifact: "runtime"}},
}}}
if !standsOnModule(fresh, "mesh-tools", nil) {
t.Fatal("a manifest's own base was not read")
}
}
// A merge names a repository the way the forge does; a source is recorded the way a build was
// asked for. The two meet on owner/repo and branch, whichever form the record took.
func TestAMergeMatchesTheSourcesBuiltFromIt(t *testing.T) {
@@ -1,198 +0,0 @@
package main
import (
"context"
"strings"
"testing"
"time"
"github.com/novox/mesh-controller/internal/inventory"
)
// heldElsewhere is the plans held by another act for longer than any test runs: waiting for them ends only with the
// context, and a look that does not wait is told at once that they are busy.
func heldElsewhere(ctx context.Context, wait bool) (func(), error) {
if !wait {
return nil, inventory.ErrPlansBusy
}
<-ctx.Done()
return nil, inventory.ErrPlansBusy
}
// 1708/AC1: given the plans held by another act, when the event loop takes a build's outcome, then it returns at
// once, without waiting the plans' two minutes (on 2026-10-21 at 14:20:45 nodered's outcome held the loop until
// 14:22:45, and the operator was told the controller stopped listening).
func TestAnOutcomeDoesNotWaitForThePlansAnotherActHolds(t *testing.T) {
ctx, cancel := context.WithTimeout(t.Context(), 3*time.Second)
defer cancel()
start := time.Now()
release, err := holdPlansForOutcome(ctx, heldElsewhere)
if took := time.Since(start); took > time.Second {
t.Fatalf("taking an outcome waited %s for plans another act holds; the event loop takes nothing meanwhile", took)
}
if err == nil || release != nil {
t.Fatalf("plans another act holds were said to be held: release %v, err %v", release != nil, err)
}
}
// 1708/AC4: given the plans free, when the event loop takes a build's outcome, then it holds them and marks the plan
// from the outcome directly, as before.
func TestAnOutcomeHoldsThePlansWhenTheyAreFree(t *testing.T) {
let := false
release, err := holdPlansForOutcome(t.Context(), func(context.Context, bool) (func(), error) {
return func() { let = true }, nil
})
if err != nil || release == nil {
t.Fatalf("free plans were not held: %v", err)
}
release()
if !let {
t.Fatal("the hold given was not the one taken")
}
}
// withLooker installs a looker that is never run, and answers it, so a test can read whether a look was asked.
func withLooker(t *testing.T) *looker {
t.Helper()
l := newLooker(func(context.Context) bool { return false })
lookerMu.Lock()
before := theLooker
theLooker = l
lookerMu.Unlock()
t.Cleanup(func() {
lookerMu.Lock()
theLooker = before
lookerMu.Unlock()
})
return l
}
// 1708/AC2: given an outcome left to the build records because the plans were held, when it is left, then the look
// at the plans is asked for, so the records settle it once the holder lets go rather than at the next event.
func TestAnOutcomeLeftToTheRecordsAsksForTheLook(t *testing.T) {
l := withLooker(t)
leaveToRecords("nodered", "build-1708-ac2", "", "", inventory.ErrPlansBusy)
select {
case <-l.signal:
default:
t.Fatal("an outcome left to the build records asked for no look: the plan waits for whatever event comes next")
}
}
// 1708/AC3: given an outcome left to the records that said what the build did (ADR 0350 §4), when the look settles
// the plan from the build records by that build's id, then the module keeps those words, once.
func TestTheRecordsSettleAnOutcomeWithWhatTheBuildDid(t *testing.T) {
withLooker(t)
asked := time.Date(2026, 10, 21, 12, 15, 0, 0, time.UTC)
plan := func() inventory.Plan {
return inventory.Plan{ID: "plan-1708", Tiers: [][]string{{"nodered"}},
Modules: map[string]*inventory.PlanModule{"nodered": {State: "asked", AskedAt: &asked, Build: "build-1708-ac3"}}}
}
byID := map[string]inventory.Build{"build-1708-ac3": {ID: "build-1708-ac3", Module: "nodered", Commit: "4efac366",
At: asked.Add(5 * time.Minute)}}
leaveToRecords("nodered", "build-1708-ac3", "reused from build-1708-check", "", inventory.ErrPlansBusy)
p := plan()
if !settleFromRecords(&p, p.Tiers[0], nil, byID) || p.Modules["nodered"].State != "built" {
t.Fatalf("the records did not settle the outcome: %+v", p.Modules["nodered"])
}
if got := p.Modules["nodered"].Made; got != "reused from build-1708-check" {
t.Fatalf("settled from the records, the module says %q of what its build did, not the outcome's words", got)
}
again := plan()
settleFromRecords(&again, again.Tiers[0], nil, byID)
if got := again.Modules["nodered"].Made; got != "" {
t.Fatalf("an outcome's words were kept for a second settle: %q", got)
}
}
// 1708/AC5: given a serving controller (its look at the plans running), when an outcome was taken into the plans,
// then moving them (a release's sends, which may retry for minutes) is left to the look, off the event loop
// (2026-10-21 14:28:40 to 14:32:10: the loop sent okioki's release to g14 after mesh-delivery-builder's outcome).
func TestAnOutcomeLeavesMovingThePlansToTheLook(t *testing.T) {
l := withLooker(t)
moved := false
advanceAfterOutcome(func() { moved = true })
if moved {
t.Fatal("the plans were moved on the event loop after an outcome, where a send that retries holds every message")
}
select {
case <-l.signal:
default:
t.Fatal("the look was not asked to move the plans after an outcome")
}
}
// 1708/AC5: given no look running (a command, which has no event loop), when an outcome was taken, then the plans are
// moved at once, as before.
func TestAnOutcomeMovesThePlansWhereNoLookRuns(t *testing.T) {
lookerMu.Lock()
before := theLooker
theLooker = nil
lookerMu.Unlock()
t.Cleanup(func() {
lookerMu.Lock()
theLooker = before
lookerMu.Unlock()
})
moved := false
advanceAfterOutcome(func() { moved = true })
if !moved {
t.Fatal("with no look running, an outcome left the plans unmoved")
}
}
// 1708/AC6: given a registration the mesh refused (its build recorded as worked, the outcome failed), left to the
// records because the plans were held, when the look settles the plan from the records, then the module and the plan
// are failed, never built: settled as built, the tier would be sent the version from before (reviewer-1708@shanks).
func TestARefusedRegistrationLeftToTheRecordsFailsThePlan(t *testing.T) {
withLooker(t)
asked := time.Date(2026, 10, 21, 12, 15, 0, 0, time.UTC)
p := inventory.Plan{ID: "plan-1708-ac6", State: inventory.PlanBuilding, Tiers: [][]string{{"nodered"}},
Modules: map[string]*inventory.PlanModule{"nodered": {State: "asked", AskedAt: &asked, Build: "build-1708-ac6"}}}
byID := map[string]inventory.Build{"build-1708-ac6": {ID: "build-1708-ac6", Module: "nodered", Commit: "4efac366",
At: asked.Add(5 * time.Minute)}}
leaveToRecords("nodered", "build-1708-ac6", "", "the mesh does not register it: a port is taken", inventory.ErrPlansBusy)
overlayRefusals(nil, byID, refusalsKeptFor([]string{"build-1708-ac6"}))
settleFromRecords(&p, p.Tiers[0], nil, byID)
if s := p.Modules["nodered"]; s.State != "failed" || p.State != inventory.PlanFailed {
t.Fatalf("a refused registration left to the records settled as %q, the plan %q", s.State, p.State)
}
if !strings.Contains(p.Modules["nodered"].Why, "a port is taken") {
t.Fatalf("the refusal's words were lost: %q", p.Modules["nodered"].Why)
}
}
// 1708/AC6: a refusal kept in the store (a controller replaced before the settle) fails the record as one kept in
// the process does.
func TestARefusalFromTheStoreFailsTheRecord(t *testing.T) {
recorded := map[string][]inventory.Build{"nodered": {{ID: "b-store", Module: "nodered"}, {ID: "b-other", Module: "nodered"}}}
byID := map[string]inventory.Build{"b-store": {ID: "b-store", Module: "nodered"}}
overlayRefusals(recorded, byID, map[string]string{"b-store": "not registered"})
if recorded["nodered"][0].Worked() || byID["b-store"].Worked() {
t.Fatal("a build whose registration was refused still reads as worked")
}
if !recorded["nodered"][1].Worked() {
t.Fatal("a build nobody refused was failed")
}
}
// 1708/AC7: given the records settle a module as deleted at its source, or fail the plan, when the look settles, then
// it does what taking the outcome directly does: the deleted module is named to be forgotten, and the plan's failure
// is said (reviewer-1708@shanks).
func TestSettlingFromTheRecordsDoesWhatTakingTheOutcomeDoes(t *testing.T) {
p := inventory.Plan{State: inventory.PlanFailed, Modules: map[string]*inventory.PlanModule{
"gone": {State: planDeleted}, "x": {State: "failed"}, "y": {State: "built"}}}
before := map[string]string{"gone": "asked", "x": "asked", "y": "built"}
deleted, failed := settledEffects(before, &p, inventory.PlanBuilding)
if len(deleted) != 1 || deleted[0] != "gone" {
t.Fatalf("the module the records found deleted is not forgotten: %v", deleted)
}
if !failed {
t.Fatal("the plan the records failed is not said as failed")
}
if d, f := settledEffects(map[string]string{"gone": planDeleted}, &p, inventory.PlanFailed); len(d) != 0 || f {
t.Fatalf("what was deleted or failed before the settle is acted on again: %v %v", d, f)
}
}
-69
View File
@@ -1,69 +0,0 @@
package main
import (
"context"
"fmt"
)
// A new controller's own grants reach the bus before it writes (novox/hq issue 1800).
//
// A controller's grants are composed by the controller binary itself (broker.PermissionsFor), so the controller that
// walks a delivery cannot put its successor's grants in the bus's user list: the list that carries them was sent only
// once the new controller passed its gate on its first machine (carryUserList). Until then the new controller wrote
// buckets its account was not granted, and the bus refused them: plans-building on 2026-10-21, declared-packages on
// 2026-10-22, and R4e's states-announced would be the third.
//
// So a controller that starts, holding the lease and before it starts anything that writes its buckets, sends the
// machine holding the bus its user list when the list it composes is not the one that machine was last sent: the grants
// step's send (release.go grantsStep), every build there kept at the one it runs, refused unsent when the declaration
// would change more than the user list (errNotGrantsOnly). No build of a running plan reaches that machine before the
// plan's own tier sends it (ADR 0503 §8). A send that cannot be made is said once, and the controller serves on.
//
// It does not ask heldMachines, as carryUserList does: every reason that gives is a build that would move with the
// list (a policy, a plan not yet sent there, no gate passed), and here none can. The running plan that brought this
// controller holds the bus's machine until its tier sends it; asking would leave the list behind exactly when 1800
// meets it. The refusal unsent (errNotGrantsOnly) is the guard instead.
// ownGrantsCarrier is what the start's send reads and does: whether the bus's list is behind and on which machine, and
// the grants-only send to it. Variables so a test needs no store and no bus.
type ownGrantsCarrier struct {
behind func(ctx context.Context) (holder string, behind bool, err error)
send func(ctx context.Context, holder string) error
}
// carry sends the machine holding the bus the controller's own user list when it is behind, and says what it did.
func (c ownGrantsCarrier) carry(ctx context.Context) string {
holder, behind, err := c.behind(ctx)
switch {
case err != nil:
return fmt.Sprintf("whether the bus's user list carries this controller's grants cannot be read, so it is not sent "+
"at start; a bucket it newly writes may be refused until the gate carries it: %v", err)
case holder == "" || !behind:
return ""
}
if err := c.send(ctx, holder); err != nil {
return fmt.Sprintf("this controller's grants could not be sent to %s, which holds the bus, at its start; a bucket "+
"it newly writes may be refused until the gate carries them: %v", holder, err)
}
return fmt.Sprintf("sent %s, which holds the bus, this controller's user list alone, every build there kept, "+
"before it writes", holder)
}
// startWithOwnGrants carries the controller's own grants to the bus and only then starts what writes (then).
func startWithOwnGrants(ctx context.Context, c ownGrantsCarrier, then func()) {
if said := c.carry(ctx); said != "" {
fmt.Println(said)
}
then()
}
// ownGrantsOf is the carrier of the serving controller: the grants step's reading and its grants-only send.
func ownGrantsOf(open *stores) ownGrantsCarrier {
return ownGrantsCarrier{
behind: func(ctx context.Context) (string, bool, error) { return brokerBehindOf(ctx, open, nil) },
send: func(ctx context.Context, holder string) error {
_, err := sendRollout(keepingEveryBuild(withScope(ctx, sendScope{}), holder), open, []string{holder})
return err
},
}
}
-73
View File
@@ -1,73 +0,0 @@
package main
import (
"context"
"errors"
"strings"
"testing"
)
// 1800 (1): a controller starting with the bus's list behind its own sends the bus's machine once, before anything
// that writes its buckets starts.
func TestAStartingControllerSendsItsGrantsBeforeItWrites(t *testing.T) {
var order []string
c := ownGrantsCarrier{
behind: func(context.Context) (string, bool, error) { return "novox", true, nil },
send: func(_ context.Context, holder string) error { order = append(order, "send "+holder); return nil },
}
startWithOwnGrants(t.Context(), c, func() { order = append(order, "writers") })
if strings.Join(order, ", ") != "send novox, writers" {
t.Fatalf("the start did %v", order)
}
}
// 1800 (3), (4): no send when the list is current or there is no bus with a list; one that cannot be read or sent is
// said once, and the controller goes on.
func TestAStartingControllerSendsNothingWhenTheListIsCurrentAndGoesOnWhenRefused(t *testing.T) {
sends := 0
send := func(context.Context, string) error { sends++; return nil }
for _, c := range []ownGrantsCarrier{
{behind: func(context.Context) (string, bool, error) { return "novox", false, nil }, send: send},
{behind: func(context.Context) (string, bool, error) { return "", true, nil }, send: send},
} {
if said := c.carry(t.Context()); said != "" || sends != 0 {
t.Fatalf("a current list sent %d and said %q", sends, said)
}
}
refused := ownGrantsCarrier{behind: func(context.Context) (string, bool, error) { return "novox", true, nil },
send: func(context.Context, string) error { return errNotGrantsOnly }}
went := false
startWithOwnGrants(t.Context(), refused, func() { went = true })
if !went {
t.Fatal("a refused send stopped the controller")
}
if said := refused.carry(t.Context()); !strings.Contains(said, "could not be sent to novox") ||
!strings.Contains(said, errNotGrantsOnly.Error()) {
t.Fatalf("a refused send is said as %q", said)
}
unread := ownGrantsCarrier{behind: func(context.Context) (string, bool, error) { return "", false, errors.New("no store") },
send: send}
if said := unread.carry(t.Context()); !strings.Contains(said, "cannot be read") || sends != 0 {
t.Fatalf("an unread list: sent %d, said %q", sends, said)
}
}
// 1800 (2): the start's send is the grants step's: composed with every build of the bus's machine kept at the one it
// runs, so the declaration differs from its last only by the user list (and is refused unsent otherwise,
// errNotGrantsOnly, release.go), whatever a running plan registered since.
func TestAStartingControllersSendKeepsEveryBuildOfTheBusMachine(t *testing.T) {
was := sendRollout
t.Cleanup(func() { sendRollout = was })
var kept bool
var named []string
sendRollout = func(ctx context.Context, _ *stores, names []string) ([]string, error) {
kept, named = everyBuildKept(ctx, "novox"), names
return names, nil
}
if err := ownGrantsOf(nil).send(t.Context(), "novox"); err != nil {
t.Fatal(err)
}
if !kept || len(named) != 1 || named[0] != "novox" {
t.Fatalf("the start's send keeps every build: %v, names %v", kept, named)
}
}
-40
View File
@@ -1,40 +0,0 @@
package main
import (
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
)
// B1 (ADR 0499 §1, §5; issue 1758): a module whose every artifact names the paths it is built from is read from those
// paths and its manifest alone. mesh-sdk's TypeScript package at its repository's root: a change to its go/ touches no
// module, one to its src/ or its package.json touches it. A module with an artifact naming no paths reads its directory.
func TestAModuleNamingItsPathsIsReadFromThemAlone(t *testing.T) {
m, err := catalogue.ParseManifest([]byte(`{"module":"mesh-sdk","version":"1","build":{"artifacts":[{"name":"lib",
"kind":"package","language":"typescript","paths":["src","package.json","package-lock.json","tsconfig.json"]}]},
"resources":[]}`))
if err != nil {
t.Fatal(err)
}
own := pathsBuildSource(m, "")
read := inventory.WithBuildSources(nil, []inventory.BuildSource{{Paths: own}})
e := inventory.Entry{Manifest: m, Source: inventory.Source{Repository: "novox/mesh-sdk", Path: ""}}
for file, reads := range map[string]bool{
"go/delivery/builder.go": false, "go/go.mod": false, "README.md": false,
"src/index.ts": true, "src/deep/x.ts": true, "package.json": true, "module.json": true,
} {
if got := readsFile(e, read, file); got != reads {
t.Errorf("%s: read %v, want %v", file, got, reads)
}
}
// In a module's directory, the paths are under it.
if got := pathsBuildSource(m, "modules/sdk"); got[0] != "modules/sdk/module.json" || got[1] != "modules/sdk/src/**" {
t.Errorf("paths under a directory: %v", got)
}
// An artifact naming no paths: the whole directory, as before.
m.Build.Artifacts = append(m.Build.Artifacts, catalogue.Artifact{Name: "files", Kind: "archive", From: "files"})
if got := pathsBuildSource(m, ""); got != nil {
t.Errorf("a module with an artifact naming no paths is narrowed to %v", got)
}
}
+22 -116
View File
@@ -9,9 +9,6 @@ import (
"strings"
"time"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/novox/mesh-controller/internal/builderkeys"
"github.com/novox/mesh-controller/internal/conditions"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
@@ -88,30 +85,8 @@ const (
unknownModule = iota
buildInFlight
knownNotBuilt
// keyComing is a module the indexer holds and the catalogue does not yet: registered from the builder's done key
// of it (ADR 0501 §6), which nobody asks.
keyComing
)
// keyPendingPrefix opens the build a pending assignment waits for when it waits for a module's key: no build is asked
// (ADR 0501 §5, §6; issue 1851, A10), so it names the module's key, not a build id.
const keyPendingPrefix = "key:"
// indexedModule is the indexer's trunk value of a module the catalogue does not hold, found when it holds one. A
// variable so a test needs no bus.
var indexedModule = func(ctx context.Context, module string) (sdkdelivery.Module, bool, error) {
keys, closeKeys, err := openBuilderKeys()
if err != nil {
return sdkdelivery.Module{}, false, err
}
defer closeKeys()
v, err := keys.ModuleAt(ctx, module, "")
if errors.Is(err, builderkeys.ErrNoValue) {
return sdkdelivery.Module{}, false, nil
}
return v, err == nil, err
}
// notHeld is where a module the catalogue does not hold stands: its case, the build request it was read
// from, and the sentence that says it.
type notHeld struct {
@@ -121,24 +96,7 @@ type notHeld struct {
}
// whereIs looks for a module the catalogue does not hold among the build requests the controller keeps.
// notHeldReads is what whereIs reads of the records (*inventory.Inventory, or a test's).
type notHeldReads interface {
RequestsNamed(ctx context.Context, name string) ([]inventory.RequestOutcome, error)
Catalogue(ctx context.Context) (map[string]catalogue.Manifest, error)
RequestedNames(ctx context.Context) ([]string, error)
}
func whereIs(ctx context.Context, inv notHeldReads, module string, now time.Time) (notHeld, error) {
// **Its key is coming** (ADR 0501 §6): the indexer holds it, and the controller registers it from the builder's done
// key of it. Read before the build requests of before R4e, which nothing writes any more.
if v, held, err := indexedModule(ctx, module); err != nil {
return notHeld{}, fmt.Errorf("whether the indexer holds %s cannot be read: %w", module, err)
} else if held {
return notHeld{kind: keyComing, request: inventory.RequestOutcome{BuildRequest: inventory.BuildRequest{
ID: keyPendingPrefix + module, Repository: v.Repository, Seat: gitSeat, Path: keyPath(v)}},
said: fmt.Sprintf("%s is not registered yet: the indexer holds it in %s at %.8s, and it is registered from "+
"the builder's done key of it, which no one asks", module, v.Repository, v.Commit)}, nil
}
func whereIs(ctx context.Context, inv *inventory.Inventory, module string, now time.Time) (notHeld, error) {
requests, err := inv.RequestsNamed(ctx, module)
if err != nil {
return notHeld{}, err
@@ -211,12 +169,6 @@ func notRegistered(ctx context.Context, open *stores, node, module string, opts
return "", err
}
switch where.kind {
case keyComing:
lead := where.said
if opts.Build {
lead += "\n build \"true\" asks nothing: the builder builds every fingerprint the indexer writes (ADR 0501 §6)"
}
return keepPending(ctx, open, node, module, where.request, waiting, lead)
case buildInFlight:
lead := where.said
if opts.Build {
@@ -231,20 +183,30 @@ func notRegistered(ctx context.Context, open *stores, node, module string, opts
also = fmt.Sprintf("\n %s already has a pending assignment of %s, waiting for build %s; build "+
"\"true\" asks for a new build and makes it wait for that one", node, module, waiting.Build)
}
return "", fmt.Errorf("%w: %s%s\n it is registered from the builder's done key once a merge adds it to %s "+
"(path %s) and the indexer holds it, or at the controller's terminal; assign it then",
inventory.ErrNoSuchModule, where.said, also, last.Repository, orRoot(last.Path))
return "", fmt.Errorf("%w: %s%s\n assign it again with build \"true\" to ask for its build and keep this "+
"assignment until the build registers it; or `build` it (repository %s, path %s) and assign it "+
"once it is registered", inventory.ErrNoSuchModule, where.said, also, last.Repository, orRoot(last.Path))
}
// **No build is asked** (ADR 0501 §5, §6): a module the indexer does not hold has no key coming.
return "", fmt.Errorf("%w: %s\n build \"true\" asks no build: the controller asks none (ADR 0501 §5); a module "+
"is registered from the builder's done key once a merge adds it to a repository the catalogue builds from, or "+
"at the controller's terminal, and the indexer holds no %s", inventory.ErrNoSuchModule, where.said, module)
if waiting != nil && waiting.State == inventory.PendingApplying {
return "", fmt.Errorf("%s is being assigned %s now; nothing was asked", node, module)
}
source := buildSource{Repository: last.Repository, Seat: last.Seat}
id, err := askABuild(ctx, source, last.Path, last.Ref)
if err != nil {
return "", fmt.Errorf("%s\n its build could not be asked for: %w; nothing was assigned", where.said, err)
}
asked := inventory.RequestOutcome{BuildRequest: inventory.BuildRequest{ID: id, Repository: last.Repository,
Seat: last.Seat, Path: last.Path, Ref: last.Ref, For: "assign", At: now.UTC()}}
recordBuildRequest(ctx, inv, asked.BuildRequest)
lead := fmt.Sprintf("%s\n build %s of %s is asked for now; %s can be assigned once it registers it",
where.said, id, requestSource(asked), module)
return keepPending(ctx, open, node, module, asked, waiting, lead)
}
answer := strings.TrimPrefix(where.said, inventory.ErrNoSuchModule.Error()) +
"\n a module in a repository is registered from the builder's done key of it, then assigned"
"\n a module in a repository is built with `build` (its repository and path), then assigned"
if opts.Build {
answer += "\n build \"true\" asks for nothing: the controller asks no build (ADR 0501 §5), and the indexer " +
"holds no module of that name"
answer += "\n build \"true\" asks for nothing here: the controller has no build request saying where a " +
"module of that name is"
}
return "", fmt.Errorf("%w%s", inventory.ErrNoSuchModule, answer)
}
@@ -354,7 +316,7 @@ func clock(t time.Time) string { return t.Local().Format("2006-01-02 15:04:05 MS
// closestNames is up to three names near the one asked: registered modules and directories the controller
// asked to build.
func closestNames(ctx context.Context, inv notHeldReads, name string) []string {
func closestNames(ctx context.Context, inv *inventory.Inventory, name string) []string {
var candidates []string
if shelf, err := inv.Catalogue(ctx); err == nil {
for m := range shelf {
@@ -573,19 +535,6 @@ func settlePending(ctx context.Context, open *stores, now time.Time) []string {
}
continue
}
if strings.HasPrefix(p.Build, keyPendingPrefix) {
why, err := keyPendingEnded(ctx, p.Module)
if err != nil {
say("the key %s waits for cannot be read: %v", p.Module, err)
continue
}
if why != "" {
if line := expirePending(ctx, inv, p, why); line != "" {
said = append(said, line)
}
}
continue
}
requests, err := inv.RequestsNamed(ctx, p.Module)
if err != nil {
say("the build requests for %s could not be read: %v", p.Module, err)
@@ -795,46 +744,3 @@ func withdrawPending(ctx context.Context, inv *inventory.Inventory, node, module
}
return "", false, nil
}
// keyPendingEnded is why a pending assignment waiting for a module's key ends: the indexer no longer holds the module,
// or the builder's key of its fingerprint failed; empty while it is coming. A variable so a test needs no bus.
var keyPendingEnded = func(ctx context.Context, module string) (string, error) {
keys, closeKeys, err := openBuilderKeys()
if err != nil {
return "", err
}
defer closeKeys()
v, err := keys.ModuleAt(ctx, module, "")
if errors.Is(err, builderkeys.ErrNoValue) {
return fmt.Sprintf("the indexer no longer holds %s", module), nil
}
if err != nil || v.FP2 == "" {
return "", err
}
k, err := keys.Build(ctx, module, v.FP2)
if err != nil {
return "", err
}
if k.Failed() {
return k.Says(), nil
}
return "", nil
}
// settlePendingOnKey assigns, now, every pending assignment of a module its done key just registered (ADR 0501 §6):
// what an outcome of a build did for one waiting on that build.
func settlePendingOnKey(ctx context.Context, open *stores, module string) []string {
waiting, err := open.inventory.Pending(ctx, time.Time{})
if err != nil {
return []string{fmt.Sprintf("the pending assignments could not be read after %s was registered: %v", module, err)}
}
var said []string
for _, p := range waiting {
if p.Module == module {
if line := applyPending(ctx, open, p, keyPendingPrefix+module); line != "" {
said = append(said, line)
}
}
}
return said
}
+150 -37
View File
@@ -3,6 +3,7 @@
package main
import (
"context"
"encoding/json"
"errors"
"fmt"
@@ -12,7 +13,6 @@ import (
"time"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
@@ -36,13 +36,11 @@ func aCatalogueMesh(t *testing.T) *stores {
return open
}
// mergeAdding keeps the build request a merge adding a module asked before R4e (issue 300): nothing asks one since (ADR
// 0501 §6), and the pending assignments still read the ones kept. These tests are of what an assignment does with one.
func mergeAdding(t *testing.T, open *stores, dir, commit string) {
t.Helper()
if err := open.inventory.RecordBuildRequest(t.Context(), inventory.BuildRequest{ID: "b-" + dir,
Repository: "novox/mesh-catalog", Seat: "git", Path: dir, Ref: "main", Commit: commit, For: "merge",
At: time.Now().UTC()}); err != nil {
m := link.SourceMoved{Owner: "novox", Repo: "mesh-catalog", Base: "main", Commit: commit,
Paths: []string{dir + "/module.json"}, ModuleDirs: []string{dir}, ModuleDirsSaid: true}
if err := (following{open: open}).SourceMoved(t.Context(), m); err != nil {
t.Fatal(err)
}
}
@@ -177,7 +175,7 @@ func TestAPendingAssignmentExpiresWhenItsBuildFails(t *testing.T) {
t.Fatalf("a module whose build failed: %v", err)
}
for _, want := range []string{"sensors is known and not registered", "its last build, b-modules/sensors",
"failed: go build: undefined: sensorsRead", "registered from the builder's done key", "novox/mesh-catalog (path modules/sensors)"} {
"failed: go build: undefined: sensorsRead", "build \"true\"", "repository novox/mesh-catalog, path modules/sensors"} {
if !strings.Contains(err.Error(), want) {
t.Fatalf("the refusal does not say %q:\n%v", want, err)
}
@@ -227,10 +225,100 @@ func TestTheConditionClearsWhenTheModuleIsAssignedLater(t *testing.T) {
}
}
func TestAssignWithBuildAsksForAKnownModule(t *testing.T) {
open := aCatalogueMesh(t)
ctx := t.Context()
inv := open.inventory
if err := inv.RecordBuildRequest(ctx, inventory.BuildRequest{ID: "build-old", Repository: "novox/mesh-catalog",
Seat: "git", Path: "modules/sensors", Ref: "main", For: "build", At: time.Now().Add(-time.Hour)}); err != nil {
t.Fatal(err)
}
if err := inv.RecordBuild(ctx, inventory.Build{ID: "build-old", Repository: "novox/mesh-catalog", Ref: "main",
Path: "modules/sensors", On: "anchor", Failed: "no space left on device"}); err != nil {
t.Fatal(err)
}
asked := asksWithPaths(t)
said, err := assignWith(ctx, open, "laptop", assignOptions{Build: true}, "sensors")
if err != nil {
t.Fatal(err)
}
if len(*asked) != 1 || (*asked)[0] != [3]string{"novox/mesh-catalog", "modules/sensors", "main"} {
t.Fatalf("asked %v", *asked)
}
for _, want := range []string{"known and not registered", "no space left on device",
"build b-modules/sensors of novox/mesh-catalog@main (modules/sensors) is asked for now", "kept as pending"} {
if !strings.Contains(said, want) {
t.Fatalf("the answer does not say %q:\n%s", want, said)
}
}
pending := pendingOf(t, open)
if len(pending) != 1 || pending[0].Build != "b-modules/sensors" {
t.Fatalf("pending: %+v", pending)
}
requests, _ := inv.RequestsNamed(ctx, "sensors")
if len(requests) != 2 || requests[0].ID != "b-modules/sensors" || requests[0].For != "assign" {
t.Fatalf("the request is not kept as assign's: %+v", requests)
}
}
func TestAssignWithBuildRepointsTheWaitingAssignment(t *testing.T) {
open := aCatalogueMesh(t)
ctx := t.Context()
inv := open.inventory
if err := inv.RecordBuildRequest(ctx, inventory.BuildRequest{ID: "build-old", Repository: "novox/mesh-catalog",
Seat: "git", Path: "modules/sensors", Ref: "main", For: "merge", At: time.Now().Add(-time.Hour)}); err != nil {
t.Fatal(err)
}
if err := inv.RecordBuild(ctx, inventory.Build{ID: "build-old", Repository: "novox/mesh-catalog", Ref: "main",
Path: "modules/sensors", On: "anchor", Failed: "boom"}); err != nil {
t.Fatal(err)
}
// Waiting on the failed build, its expiry not yet settled.
if _, err := inv.RecordPending(ctx, inventory.PendingAssignment{Node: "laptop", Module: "sensors",
Build: "build-old", Repository: "novox/mesh-catalog", Path: "modules/sensors"}); err != nil {
t.Fatal(err)
}
_, err := assign(ctx, open, "laptop", "sensors")
if err == nil || !strings.Contains(err.Error(), "already has a pending assignment of sensors, waiting for build build-old") {
t.Fatalf("known and not built with a pending assignment waiting: %v", err)
}
asked := asksWithPaths(t)
said, err := assignWith(ctx, open, "laptop", assignOptions{Build: true}, "sensors")
if err != nil || !strings.Contains(said, "which waited for build build-old, now waits for build b-modules/sensors") {
t.Fatalf("assign with build: %q %v", said, err)
}
if len(*asked) != 1 {
t.Fatalf("asked %v", *asked)
}
pending := pendingOf(t, open)
if len(pending) != 1 || pending[0].Build != "b-modules/sensors" || pending[0].State != inventory.PendingWaiting {
t.Fatalf("pending: %+v", pending)
}
// The new build in flight now: build "true" again asks nothing.
if said, err := assignWith(ctx, open, "laptop", assignOptions{Build: true}, "sensors"); err != nil ||
!strings.Contains(said, "no second build was asked") || len(*asked) != 1 {
t.Fatalf("a second build true while the build runs: %q %v, asked %v", said, err, *asked)
}
}
func TestAFailedAskIsNotABuildInFlight(t *testing.T) {
open := aCatalogueMesh(t)
ctx := t.Context()
inv := open.inventory
was := askABuild
askABuild = func(context.Context, buildSource, string, string) (string, error) {
return "", errors.New("cannot submit a build: nats: timeout")
}
t.Cleanup(func() { askABuild = was })
mergeAdding(t, open, "modules/sensors", "3da80a4b00aa")
_, err := assign(ctx, open, "laptop", "sensors")
if err == nil || !strings.Contains(err.Error(), "the merge that added it (3da80a4b) could not ask for its build: cannot submit") {
t.Fatalf("a merge that could not ask: %v", err)
}
if len(pendingOf(t, open)) != 0 {
t.Fatal("a pending assignment waits on a build never asked")
}
if err := inv.RecordBuildRequest(ctx, inventory.BuildRequest{ID: "build-waited", Repository: "novox/mesh-catalog",
Seat: "git", Path: "modules/gauges", Ref: "main", For: "build"}); err != nil {
t.Fatal(err)
@@ -238,7 +326,7 @@ func TestAFailedAskIsNotABuildInFlight(t *testing.T) {
if err := inv.MarkNotAsked(ctx, "build-waited", "cannot submit a build: no responders"); err != nil {
t.Fatal(err)
}
_, err := assign(ctx, open, "laptop", "gauges")
_, err = assign(ctx, open, "laptop", "gauges")
if err == nil || !strings.Contains(err.Error(), "build build-waited, asked by build, was not handed over") {
t.Fatalf("a build not handed over: %v", err)
}
@@ -307,7 +395,7 @@ func TestAnUnknownModuleIsRefusedWithTheClosestNames(t *testing.T) {
}
for _, want := range []string{"no module of that name: sensors", "the catalogue holds no module of that name",
"no build request the controller kept (the last 30 days)", "the closest names it holds: sensord",
"asks for nothing: the controller asks no build"} {
"asks for nothing here"} {
if !strings.Contains(err.Error(), want) {
t.Fatalf("the refusal does not say %q:\n%v", want, err)
}
@@ -477,6 +565,59 @@ func TestAPendingAssignmentNeverWaitsSilently(t *testing.T) {
}
}
func TestASecondAssignmentNotMadeKeepsItsCondition(t *testing.T) {
open := aCatalogueMesh(t)
ctx := t.Context()
asked := 0
was := askABuild
askABuild = func(context.Context, buildSource, string, string) (string, error) {
asked++
return fmt.Sprintf("b-%d", asked), nil
}
t.Cleanup(func() { askABuild = was })
mergeAdding(t, open, "modules/sensors", "3da80a4b00aa")
if _, err := assign(ctx, open, "laptop", "sensors"); err != nil {
t.Fatal(err)
}
failedBuild(t, open, "b-1")
settlePending(ctx, open, time.Now())
first := pendingOf(t, open)[0]
if !conditionOpen(t, first) {
t.Fatal("the first assignment not made raised nothing")
}
if _, err := assignWith(ctx, open, "laptop", assignOptions{Build: true}, "sensors"); err != nil {
t.Fatal(err)
}
failedBuild(t, open, "b-2")
settlePending(ctx, open, time.Now())
var second inventory.PendingAssignment
for _, p := range pendingOf(t, open) {
if p.Build == "b-2" {
second = p
}
}
if second.State != inventory.PendingExpired {
t.Fatalf("the second: %+v", second)
}
if !conditionOpen(t, second) {
t.Fatal("the second assignment's condition was cleared in the tick that raised it")
}
if !conditionOpen(t, rowOf(t, open, first.ID)) {
t.Fatal("the first was judged answered by a second that itself was not made")
}
// unassign takes both back, and the next tick clears both.
if said, err := unassign(ctx, open, "laptop", "sensors"); err != nil || !strings.Contains(said, "b-1") ||
!strings.Contains(said, "b-2") {
t.Fatalf("unassign: %q %v", said, err)
}
settlePending(ctx, open, time.Now())
if conditionOpen(t, first) || conditionOpen(t, second) {
t.Fatal("a condition stayed open after both were taken back")
}
}
func TestARaisedRowIsKeptUntilItsConditionClears(t *testing.T) {
open := aCatalogueMesh(t)
ctx := t.Context()
@@ -594,31 +735,3 @@ func TestABuildNoLongerWaitedForIsStillInFlight(t *testing.T) {
t.Fatalf("past the bound: %+v", rows)
}
}
// A10 (ADR 0501 §6; issue 1851): an assignment of a module the indexer holds and the catalogue does not yet waits for
// the builder's done key of it, asking nothing, and is made when that key registers the module.
func TestAnAssignmentWaitsForTheKeyAndAsksNothing(t *testing.T) {
open := aCatalogueMesh(t)
ctx := t.Context()
noBuildAsked(t)
b := aBuilderOnTheBus(t)
fp := fpOf('a')
manifest := b.indexed(t, "sensors", "1", "novox/mesh-catalog", "modules/sensors", "3da80a4b00aa0000", fp)
b.built(t, "sensors", "1", fp, manifest, sdkdelivery.BuildBuilding)
said, err := assignWith(ctx, open, "laptop", assignOptions{Build: true}, "sensors")
if err != nil || !strings.Contains(said, "registered from the builder's done key") || !strings.Contains(said, "kept as pending") {
t.Fatalf("A10: the assignment of a module whose key is coming: %q %v", said, err)
}
if p := pendingOf(t, open); len(p) != 1 || p[0].Build != keyPendingPrefix+"sensors" {
t.Fatalf("A10: pending %+v", p)
}
settlePending(ctx, open, time.Now().Add(time.Hour))
if p := pendingOf(t, open); p[0].State != inventory.PendingWaiting {
t.Fatalf("A10: a pending assignment waiting for a building key expired: %+v", p)
}
b.built(t, "sensors", "1", fp, manifest, sdkdelivery.BuildDone)
takeNewModules(ctx, open, "sensors")
if !slices.Contains(assignedTo(t, open, "laptop"), "sensors") {
t.Fatalf("A10: the key registered sensors and the pending assignment was not made: %+v", pendingOf(t, open))
}
}
-39
View File
@@ -1,39 +0,0 @@
package main
import (
"context"
"strings"
"testing"
"time"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/novox/mesh-controller/internal/inventory"
)
// A10 (ADR 0501 §6; issue 1851): a module the catalogue does not hold and the indexer does is one whose key is coming:
// an assignment of it waits for that key, named by the module, with no build asked and no build request read.
func TestAModuleTheIndexerHoldsIsWaitedForByItsKey(t *testing.T) {
was := indexedModule
indexedModule = func(_ context.Context, module string) (sdkdelivery.Module, bool, error) {
return sdkdelivery.Module{Repository: "novox/mesh-catalog", Commit: "3da80a4b00aa", Path: "modules/" + module}, true, nil
}
t.Cleanup(func() { indexedModule = was })
where, err := whereIs(t.Context(), noRequests{}, "sensors", time.Now())
if err != nil {
t.Fatal(err)
}
if where.kind != keyComing || where.request.ID != keyPendingPrefix+"sensors" || where.request.Path != "modules/sensors" ||
!strings.Contains(where.said, "registered from the builder's done key") {
t.Fatalf("a module whose key is coming is read as %+v", where)
}
}
// noRequests is records with no build request and no module: what a mesh after R4e reads of a new module.
type noRequests struct{}
func (noRequests) RequestsNamed(context.Context, string) ([]inventory.RequestOutcome, error) {
return nil, nil
}
func (noRequests) Catalogue(context.Context) (map[string]catalogue.Manifest, error) { return nil, nil }
func (noRequests) RequestedNames(context.Context) ([]string, error) { return nil, nil }
-61
View File
@@ -195,29 +195,6 @@ var plainWordings = map[string]func(conditions.Observation) words{
"there yet is sealed by the controller until the vault holds it.", m, m),
Resolved: "Resolved: the vault gave " + m + " the core's credentials"}
}),
kindFilesNotPut: worded(func(o conditions.Observation) words {
m := machineOr(o, "a machine")
return words{Headline: "The big files for " + m + " cannot be put on the bus",
Needs: "look at the bus's permissions for the controller; the details give the reason.",
Explanation: fmt.Sprintf("The mesh could not put the large files of %s's instructions in their place on the bus, "+
"send after send, so %s is sent them inside the instructions instead, which may make them too big.", m, m),
Resolved: "Resolved: the big files for " + m + " reach the bus again"}
}),
kindDeclarationNearLimit: worded(func(o conditions.Observation) words {
m := machineOr(o, "a machine")
if o.Severity == conditions.Urgent {
return words{Headline: m + " cannot be sent its instructions",
Needs: "make the instructions for " + m + " smaller, for example by taking a module off it.",
Explanation: fmt.Sprintf("The instructions for %s have grown past the largest message the bus takes, so "+
"nothing new reaches %s until they are smaller.", m, m),
Resolved: "Resolved: the instructions for " + m + " fit the bus again"}
}
return words{Headline: "The instructions for " + m + " are nearly too big",
Needs: "do not add more to " + m + " until this is dealt with; the details give the size.",
Explanation: fmt.Sprintf("The instructions for %s are close to the largest message the bus takes. "+
"One more module or secret may stop anything new from reaching %s.", m, m),
Resolved: "Resolved: the instructions for " + m + " have room again"}
}),
kindCoreDowngrade: worded(func(o conditions.Observation) words {
m := machineOr(o, "a machine")
component := idPart(o, 0)
@@ -457,18 +434,6 @@ var plainWordings = map[string]func(conditions.Observation) words{
Explanation: "A delivery did not finish. Nothing is tried again until you say so.",
Resolved: "Resolved: the delivery is no longer failed"}
}),
kindMovesHeld: worded(func(o conditions.Observation) words {
return words{Headline: "An update waits for you",
Explanation: "A new version of a module would restart a container or touch its data, so it waits for you.",
Resolved: "Resolved: the update is no longer waiting"}
}),
kindImageMissing: worded(func(o conditions.Observation) words {
return words{Headline: "A module is not updated",
Explanation: "The new version of a module names a container image that the mesh's store does not hold, so it " +
"was not sent. What runs is untouched.",
Needs: "build the module again; the details name the image.",
Resolved: "Resolved: the store holds the image again"}
}),
"ask-lost": worded(func(o conditions.Observation) words {
return words{Headline: "A build has no result",
Explanation: "A build was asked for and never answered. It is asked again with the next delivery.",
@@ -966,9 +931,6 @@ func stalledWords(l stalledLine, o conditions.Observation) (headline, explanatio
if l.State == "unanswered" {
return unansweredWords(l, o)
}
if l.Needs == needsNewHead {
return newHeadWords(l, name)
}
held := l.State
if held == "" {
held = "held"
@@ -1002,29 +964,6 @@ func stalledWords(l stalledLine, o conditions.Observation) (headline, explanatio
fmt.Sprintf("Delivery of %s is no longer %s", name, held), needs, actions
}
// newHeadWords are the words of a delivery whose cure is its author's new head (novox/hq ADR 0505 §5): it names the
// author and says a new head is needed; it offers no Stop and never calls it the operator's, because the operator
// hears of it only once the author's own watch has failed, after two hours with no new head.
func newHeadWords(l stalledLine, name string) (headline, explanation, resolved, needs string, actions []conditions.Action) {
pull := "A pull request of " + name
if l.Number > 0 {
pull = fmt.Sprintf("Pull request %d of %s", l.Number, name)
}
author := l.Author
if _, plain := conditions.Plain(author); !plain {
author = "its author" // a branch named in code (a dotted name, a number) is not said
}
long := "for a long time"
if d, err := time.ParseDuration(l.For); err == nil {
long = "for " + humanDuration(d)
}
return pull + " needs a new head from its author",
fmt.Sprintf("%s cannot merge until %s pushes a new head: its check failed after its Approve, and no new head "+
"came %s.", pull, author, long),
pull + " has a new head, or is closed",
"ask " + author + " for a new head.", nil
}
// stoppedNeeds is what ends a delivery that is already stopped, from the stalled line's healer words ("none: a
// merge that delivers its modules ends it — the operator's"): the part that says what ends it.
func stoppedNeeds(l stalledLine) string {
+1 -1
View File
@@ -260,7 +260,7 @@ func TestUpdatesHeldNameWhatWaitsAndWhereItIsReleased(t *testing.T) {
plainExample(t, got[0], "Updates of openrazer and sensors wait for your release",
"Needs you: release them from the mesh MCP server; this notification cannot do it. Updates of openrazer and "+
"sensors on g14 and shanks wait for a person to release them, because the last walk failed. "+
"They are not delivered until then, and stay held if you leave them.", "Release")
"They are not delivered until then, and stay held if you leave them.")
}
// **What held them is said in the glossary's words** (2026-10-08 review): the backlog is held after any

Some files were not shown because too many files have changed in this diff Show More