Compare commits
24
Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
81f3c2d01b | ||
|
|
c58516f819 | ||
|
|
54b04a7604 | ||
|
|
af025562c7 | ||
|
|
51c8c7ec52 | ||
|
|
63e85b25e6 | ||
|
|
93c6ca5432 | ||
|
|
14ab2ddd9b | ||
|
|
510c91f144 | ||
|
|
33dc85d850 | ||
|
|
ea1d3ed96d | ||
|
|
eca6390d6f | ||
|
|
5438d7ff75 | ||
|
|
eb7a158086 | ||
|
|
03dac87d3a | ||
|
|
0c47f48537 | ||
|
|
2e1a9abbf7 | ||
|
|
95e7a7120a | ||
|
|
d15eee61bb | ||
|
|
1b780eae3a | ||
|
|
a5e8baf6da | ||
|
|
e168b60159 | ||
|
|
d951f22c04 | ||
|
|
b1897a3498 |
@@ -51,6 +51,9 @@ func assignWith(ctx context.Context, open *stores, node string, opts assignOptio
|
||||
if len(modules) == 0 {
|
||||
return "", fmt.Errorf("assign %s names no module", node)
|
||||
}
|
||||
if err := refusedMovingTheController(modules); err != nil {
|
||||
return "", err
|
||||
}
|
||||
// Held while it is recorded, so it cannot land between a converge's preview and its flip and
|
||||
// be taken without ever having been previewed (novox/hq ADR 0100).
|
||||
ctx, release, err := holdNodes(ctx, open, []string{node})
|
||||
@@ -209,6 +212,9 @@ func seatDependenciesOnAssign(ctx context.Context, open *stores, node string, mo
|
||||
// (novox/hq ADR 0207) — the other side of refusing that module's assignment without one. Several
|
||||
// modules in one act are judged together, so a holder and its dependents come off in one command.
|
||||
func unassign(ctx context.Context, open *stores, node string, modules ...string) (string, error) {
|
||||
if err := refusedMovingTheController(modules); err != nil {
|
||||
return "", err
|
||||
}
|
||||
if len(modules) == 0 {
|
||||
return "", fmt.Errorf("unassign %s names no module", node)
|
||||
}
|
||||
@@ -360,3 +366,17 @@ 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)
|
||||
}
|
||||
|
||||
// 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
|
||||
// change that says who may change what.
|
||||
func refusedMovingTheController(modules []string) error {
|
||||
verb, through := throughAVerb()
|
||||
if !through || !slices.Contains(modules, controllerModule) {
|
||||
return nil
|
||||
}
|
||||
return fmt.Errorf("%s is assigned and unassigned at the controller's terminal only (mesh-cli on the control-node), "+
|
||||
"never through a verb (this came through %q): the node it runs on is the control-node, whose operator is the "+
|
||||
"terminal (novox/hq ADR 0272). Nothing was assigned", controllerModule, verb)
|
||||
}
|
||||
|
||||
@@ -117,6 +117,69 @@ func judgedConfined(agent string, h inventory.NodeHealth, had bool, now time.Tim
|
||||
h.SaidAt.Local().Format("2006-01-02 15:04"))
|
||||
}
|
||||
|
||||
// searchQuietFor is how long the controller lets an agent account's verdict wait for the node-engine's setuid
|
||||
// search before that is itself the urgent condition: the engine's bound on one search (link.RootSearchBound, the
|
||||
// engine's own value), counted from when this controller first saw it waiting, never from the engine's start.
|
||||
const searchQuietFor = link.RootSearchBound
|
||||
|
||||
// The kinds of an agent account's verdict, for the quiet a search earns.
|
||||
const (
|
||||
verdictOther = iota // anything else: a stale statement, an older engine, no verdict, another unknown
|
||||
verdictPending // waiting for the search, and otherwise healthy
|
||||
verdictComplete // judged: healthy, or a way to root found
|
||||
)
|
||||
|
||||
// rootVerdictKind reads a statement for the agent account (novox/hq ADR 0266): pending when every verdict on it
|
||||
// is healthy or not judged yet because the engine's setuid search runs, at least one of them that; complete when
|
||||
// every verdict is healthy or one found a way to root. The engine's own "since" is not read: it starts again at
|
||||
// every restart of the engine.
|
||||
func rootVerdictKind(agent string, h inventory.NodeHealth, had bool, now time.Time) int {
|
||||
if !had || now.Sub(h.HeardAt) > verdictFreshFor || h.Contract < link.RootContract {
|
||||
return verdictOther
|
||||
}
|
||||
pending, any := false, false
|
||||
for _, r := range h.Resources {
|
||||
if r.Kind != link.KindAccount || r.Target != agent || r.Root != link.RootNever {
|
||||
continue
|
||||
}
|
||||
any = true
|
||||
switch {
|
||||
case r.State == link.StateHealthy:
|
||||
case r.State == link.StateUnhealthy:
|
||||
return verdictComplete
|
||||
case r.State == link.StateUnknown && strings.HasPrefix(r.Reason, link.ReasonRootPending):
|
||||
pending = true
|
||||
default:
|
||||
return verdictOther
|
||||
}
|
||||
}
|
||||
switch {
|
||||
case !any:
|
||||
return verdictOther
|
||||
case pending:
|
||||
return verdictPending
|
||||
}
|
||||
return verdictComplete
|
||||
}
|
||||
|
||||
// searchStillRunning says the one thing keeping an agent account from being judged is the node-engine's setuid
|
||||
// search, and that this controller first saw it waiting less than searchQuietFor ago — kept in the store, so a
|
||||
// node-engine restarted in a loop does not keep it quiet. A complete verdict forgets when it began.
|
||||
func searchStillRunning(ctx context.Context, inv *inventory.Inventory, node, agent string, h inventory.NodeHealth,
|
||||
had bool, now time.Time) (bool, error) {
|
||||
switch rootVerdictKind(agent, h, had, now) {
|
||||
case verdictComplete:
|
||||
return false, inv.RootSearchJudged(ctx, node)
|
||||
case verdictPending:
|
||||
since, err := inv.RootSearchPending(ctx, node, now)
|
||||
if err != nil {
|
||||
return false, err
|
||||
}
|
||||
return now.Sub(since) <= searchQuietFor, nil
|
||||
}
|
||||
return false, nil
|
||||
}
|
||||
|
||||
// probeAgentAccounts is DA: every machine that names an agent account has it judged, on its node-engine's
|
||||
// newest statement, unable to become root without a person (ADR 0266).
|
||||
func probeAgentAccounts(ctx context.Context, d *doctor) ([]conditions.Observation, error) {
|
||||
@@ -134,10 +197,22 @@ func probeAgentAccounts(ctx context.Context, d *doctor) ([]conditions.Observatio
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
confined, why := judgedConfined(n.AgentAccount, h, had, time.Now())
|
||||
now := time.Now()
|
||||
quiet, err := searchStillRunning(ctx, inv, n.Name, n.AgentAccount, h, had, now)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
confined, why := judgedConfined(n.AgentAccount, h, had, now)
|
||||
if confined {
|
||||
continue
|
||||
}
|
||||
// Not judged yet only because the first search since the node-engine started is still running: not the
|
||||
// urgent condition after every restart. The agent is still not confined — ADR 0259's router reads
|
||||
// agentConfined, not this — and `node show` still says not judged. Loud again once the search fails,
|
||||
// runs out its bound, or the statement goes stale.
|
||||
if quiet {
|
||||
continue
|
||||
}
|
||||
out = append(out, conditions.Observation{Scope: conditions.ScopeMachine, ID: n.Name, Token: "agent-root",
|
||||
Machine: n.Name, Severity: conditions.Urgent,
|
||||
Summary: fmt.Sprintf("on %s, %s (ADR 0266): an agent there may become root without a person, and "+
|
||||
|
||||
@@ -1,6 +1,7 @@
|
||||
package main
|
||||
|
||||
import (
|
||||
"github.com/novox/mesh-host/rootsearch"
|
||||
"strings"
|
||||
"testing"
|
||||
"time"
|
||||
@@ -197,3 +198,76 @@ func TestTheNodeVerbOnlyShows(t *testing.T) {
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// After every node-engine restart its search for setuid programs runs for up to its bound, and the agent account
|
||||
// is not judged until it ends. DA does not raise that as urgent while the search is within its bound, counted from
|
||||
// when this controller first saw it waiting — never from the engine's own "since", which a restart resets, so an
|
||||
// engine restarted in a loop does not keep it quiet. The account is still not confined, and `node show` still
|
||||
// says not judged; a search that failed, or a way to root found, is urgent at once.
|
||||
func TestASearchStillRunningAfterARestartIsNotUrgent(t *testing.T) {
|
||||
if searchQuietFor != rootsearch.Bound {
|
||||
t.Fatalf("the quiet is %s and the node-engine's bound %s: they are one value", searchQuietFor, rootsearch.Bound)
|
||||
}
|
||||
open := aMesh(t)
|
||||
ctx := t.Context()
|
||||
inv := open.inventory
|
||||
if _, err := inv.NodeByName(ctx, "anchor"); err != nil {
|
||||
if _, err := inv.AddNode(ctx, "anchor"); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
}
|
||||
if err := inv.SetAgentAccount(ctx, "anchor", "agent", ""); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
d := &doctor{open: open}
|
||||
say := func(state, reason string) []conditions.Observation {
|
||||
t.Helper()
|
||||
// The engine's since is always now: it was just restarted.
|
||||
v := inventory.ResourceHealth{Module: "claude-code", Resource: "claude-code.agent-account",
|
||||
Kind: link.KindAccount, Target: "agent", State: state, Reason: reason, Root: link.RootNever,
|
||||
Account: "agent", Since: time.Now()}
|
||||
if _, err := inv.RecordHealth(ctx, inventory.NodeHealth{Node: "anchor", Contract: link.RootContract,
|
||||
SaidAt: time.Now(), HeardAt: time.Now(), Resources: []inventory.ResourceHealth{v}}); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
found, err := probeAgentAccounts(ctx, d)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
return onlyMachine(found, "anchor")
|
||||
}
|
||||
running := link.ReasonRootPending + ": the search for setuid programs, started at 19:21, has not finished yet"
|
||||
healthy := func() {
|
||||
t.Helper()
|
||||
if found := say(link.StateHealthy, ""); len(found) != 0 {
|
||||
t.Fatalf("a healthy verdict raised: %+v", found)
|
||||
}
|
||||
}
|
||||
if found := say(link.StateUnknown, running); len(found) != 0 {
|
||||
t.Fatalf("a search first seen now was raised: %+v", found)
|
||||
}
|
||||
if _, confined, why, _ := agentConfined(ctx, inv, "anchor"); confined || !strings.Contains(why, "not judged") {
|
||||
t.Fatalf("an account whose search runs was read as confined: %q", why)
|
||||
}
|
||||
// The engine restarted again and again, each statement's own since fresh: the controller's clock runs on.
|
||||
if _, err := inv.RootSearchPending(ctx, "anchor", time.Now().Add(-searchQuietFor-time.Minute)); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
healthy() // a complete verdict forgets when the waiting began …
|
||||
if _, err := inv.RootSearchPending(ctx, "anchor", time.Now().Add(-searchQuietFor-time.Minute)); err != nil {
|
||||
t.Fatal(err) // … and this restart loop began past the bound
|
||||
}
|
||||
if found := say(link.StateUnknown, running); len(found) != 1 || found[0].Severity != conditions.Urgent {
|
||||
t.Fatalf("a search pending past the bound, by the controller's clock, was not urgent: %+v", found)
|
||||
}
|
||||
healthy()
|
||||
incomplete := rootsearch.ReasonIncomplete + ": the search for setuid programs did not finish (last tried at " +
|
||||
"19:23: timeout); it is tried again later"
|
||||
if found := say(link.StateUnknown, incomplete); len(found) != 1 || found[0].Severity != conditions.Urgent {
|
||||
t.Fatalf("a search that did not finish was not urgent: %+v", found)
|
||||
}
|
||||
if found := say(link.StateUnhealthy, link.ReasonRoot+": in the group docker; "+running); len(found) != 1 ||
|
||||
found[0].Severity != conditions.Urgent {
|
||||
t.Fatalf("a way to root found while the search runs was not urgent: %+v", found)
|
||||
}
|
||||
}
|
||||
|
||||
@@ -451,6 +451,10 @@ func buildOneAsked(ctx context.Context, source buildSource, path, ref string, wa
|
||||
if err != nil {
|
||||
return "", err
|
||||
}
|
||||
// Through a verb, only a repository the catalogue builds from (novox/hq ADR 0266).
|
||||
if err := verbMayAsk(ctx, source, repository); err != nil {
|
||||
return "", err
|
||||
}
|
||||
|
||||
ident, err := openIdentity(ctx)
|
||||
if err != nil {
|
||||
@@ -500,8 +504,10 @@ func buildOneAsked(ctx context.Context, source buildSource, path, ref string, wa
|
||||
// never before, so an ask that failed never reads as a build in flight. A dry run registers nothing, and
|
||||
// is not kept.
|
||||
keep := !dryRun && asker != ""
|
||||
// Asked at the terminal is what lets its outcome register a module from a repository the catalogue does
|
||||
// not build it from (novox/hq ADR 0266); never through a verb.
|
||||
asked := inventory.BuildRequest{ID: request.ID, Repository: source.Repository, Seat: source.Seat, Path: path,
|
||||
Ref: ref, For: asker}
|
||||
Ref: ref, For: asker, AtTerminal: startedAtTheTerminal()}
|
||||
|
||||
if wait == 0 {
|
||||
// Asked and not waited for (novox/hq issue 176): the outcome is the role's event, and the
|
||||
@@ -631,6 +637,21 @@ func takeIn(ctx context.Context, inv *inventory.Inventory, result link.BuildResu
|
||||
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)
|
||||
}
|
||||
if trunk == "" {
|
||||
trunk = "main"
|
||||
}
|
||||
repoID, err := mayRegisterFrom(ctx, inv, manifest.Module, recorded, result.ID, result.Path, trunk)
|
||||
if err != nil {
|
||||
return manifest, kept, fmt.Errorf("%s built %s (%s), recorded and not registered: %w", result.On,
|
||||
manifest.Module, short(result.Commit), 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.
|
||||
@@ -678,6 +699,10 @@ func takeIn(ctx context.Context, inv *inventory.Inventory, result link.BuildResu
|
||||
}
|
||||
return manifest, kept, err
|
||||
}
|
||||
// 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, 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.
|
||||
@@ -719,6 +744,9 @@ func buildAndShow(ctx context.Context, source buildSource, path, ref string, wai
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
if err := verbMayAsk(ctx, source, repository); err != nil {
|
||||
return err
|
||||
}
|
||||
ident, err := openIdentity(ctx)
|
||||
if err != nil {
|
||||
return err
|
||||
|
||||
@@ -0,0 +1,458 @@
|
||||
package main
|
||||
|
||||
import (
|
||||
"context"
|
||||
"encoding/json"
|
||||
"errors"
|
||||
"fmt"
|
||||
"os"
|
||||
"regexp"
|
||||
"strings"
|
||||
"time"
|
||||
|
||||
"github.com/novox/mesh-controller/internal/catalogue"
|
||||
"github.com/novox/mesh-controller/internal/inventory"
|
||||
"github.com/novox/mesh-controller/internal/link"
|
||||
)
|
||||
|
||||
// Where a build may register a module from (novox/hq ADR 0266).
|
||||
//
|
||||
// A build's outcome registers its module, and a registered module is what the next push sends. Before this,
|
||||
// an outcome registered whatever its manifest named, from whichever repository it was built from: an agent
|
||||
// that made a repository of its own, committed `modules/sudo/module.json` granting itself a rule without a
|
||||
// password, and asked the `build` verb for it had its repository registered as the module `sudo` — whose next
|
||||
// push made the agent root on every node. A fork of the node-engine did the same everywhere. The trunk rule
|
||||
// (ADR 0238) did not stop it: the trunk it checked was the trunk of the repository built, which was the
|
||||
// agent's own.
|
||||
//
|
||||
// So **an outcome registers a module only from the repository the catalogue already builds that module
|
||||
// from**; and a module new to the catalogue only from a repository the catalogue already builds another
|
||||
// module from — whose trunk takes a reviewed, approved merge, which is how a merge adds a module (novox/hq
|
||||
// issue 300). Anything else — a module moved to another repository, a module from a repository the
|
||||
// catalogue has never built — is the operator's, at the controller's terminal: allowed only when the build
|
||||
// request was kept as asked there. Judged at the take-in, which every outcome reaches whoever hears it and
|
||||
// whichever verb asked it (`build`, `rebuild`, `replay --register`, `assign` with build), and before the ask
|
||||
// for a call through a verb, so an agent cannot have a build node run a repository the catalogue does not
|
||||
// build from at all.
|
||||
|
||||
// errNotItsSource is an outcome refused for where it was built from.
|
||||
var errNotItsSource = errors.New("not built from the repository the catalogue builds it from")
|
||||
|
||||
// servedVar marks every process the serving controller starts — each verb's command, each child — and the serving
|
||||
// process itself, so none of them can read as the operator at the terminal (novox/hq ADR 0266). Set by serve
|
||||
// before it answers anything, inherited by every child through os.Environ.
|
||||
const servedVar = "MESH_SERVED_BY_THE_CONTROLLER"
|
||||
|
||||
// startedAtTheTerminal says this process was started at the controller's terminal: not the serving controller,
|
||||
// not anything it started, not a verb's command, not a seat call's. The serving controller marks its own
|
||||
// environment (servedVar), so a build asked in it, or by any process it starts, never reads as the terminal's;
|
||||
// runVerb also names the verb and the caller.
|
||||
//
|
||||
// **And a line mesh-cli asked as the controller's terminal** (novox/hq ADR 0272 §4): the serving controller runs it
|
||||
// without the served mark and without a verb, names its caller, and marks it with cliTerminalVar — a mark only the
|
||||
// mesh-cli path sets and every other command line the controller runs is stripped of (commandEnvironment).
|
||||
func startedAtTheTerminal() bool {
|
||||
if os.Getenv(servedVar) != "" || os.Getenv(verbVar) != "" {
|
||||
return false
|
||||
}
|
||||
return os.Getenv(link.CallerVar) == "" || os.Getenv(cliTerminalVar) != ""
|
||||
}
|
||||
|
||||
// cliTerminalVar marks a command line the serving controller runs as the controller's terminal for mesh-cli.
|
||||
const cliTerminalVar = "MESH_CLI_TERMINAL"
|
||||
|
||||
// markServed marks this process, and so everything it starts, as the serving controller's.
|
||||
func markServed() {
|
||||
if err := os.Setenv(servedVar, "1"); err != nil {
|
||||
panic("the serving controller could not mark its environment: " + err.Error())
|
||||
}
|
||||
}
|
||||
|
||||
// sourceForms compares sources however each is spelled: a path on a seat's holder (ADR 0111) or a URL — a
|
||||
// build asked of a seat's path is registered with the URL composed from it when its outcome does not echo
|
||||
// the seat. A seat's path is composed into its URL where the seat's holder is known, so both spellings of
|
||||
// one repository are one; where it is not, a seat's path matches only the same seat's same path.
|
||||
type sourceForms struct {
|
||||
bases map[string]string // the seat's clone base, `scheme://host:port`, by seat; "" where it is not known
|
||||
base func(seat string) string
|
||||
}
|
||||
|
||||
// newSourceForms reads the seats' bases from the mesh once, when first needed.
|
||||
func newSourceForms(ctx context.Context, inv *inventory.Inventory) *sourceForms {
|
||||
f := &sourceForms{bases: map[string]string{}}
|
||||
var world *catalogue.World
|
||||
f.base = func(seat string) string {
|
||||
if b, known := f.bases[seat]; known {
|
||||
return b
|
||||
}
|
||||
if world == nil {
|
||||
w := catalogue.World{}
|
||||
if shelf, err := inv.Catalogue(ctx); err == nil {
|
||||
if read, err := theRestOfTheMesh(ctx, inv, shelf, ""); err == nil {
|
||||
w = read
|
||||
}
|
||||
}
|
||||
world = &w
|
||||
}
|
||||
b, err := seatBase(*world, seat)
|
||||
if err != nil {
|
||||
b = ""
|
||||
}
|
||||
f.bases[seat] = b
|
||||
return b
|
||||
}
|
||||
return f
|
||||
}
|
||||
|
||||
// canonical is one spelling of a repository: lower case, no `.git`, no trailing slash, and a seat's path as
|
||||
// the URL its holder serves it at where that is known.
|
||||
func (f *sourceForms) canonical(repository, seat string) string {
|
||||
trim := func(s string) string {
|
||||
s = strings.TrimSpace(strings.ToLower(s))
|
||||
s = strings.TrimRight(s, "/")
|
||||
return strings.TrimRight(strings.TrimSuffix(s, ".git"), "/")
|
||||
}
|
||||
if seat == "" {
|
||||
return trim(repository)
|
||||
}
|
||||
if b := f.base(seat); b != "" {
|
||||
return trim(b + "/" + strings.Trim(repository, "/"))
|
||||
}
|
||||
return "seat:" + seat + ":" + trim(strings.Trim(repository, "/"))
|
||||
}
|
||||
|
||||
// same says two sources are one repository.
|
||||
func (f *sourceForms) same(aRepository, aSeat, bRepository, bSeat string) bool {
|
||||
if aRepository == "" || bRepository == "" {
|
||||
return false
|
||||
}
|
||||
return f.canonical(aRepository, aSeat) == f.canonical(bRepository, bSeat)
|
||||
}
|
||||
|
||||
// buildsFrom says the catalogue builds some module from this repository.
|
||||
func (f *sourceForms) buildsFrom(entries []inventory.Entry, repository, seat string) bool {
|
||||
for _, e := range entries {
|
||||
if e.Provided || e.Source.Repository == "" {
|
||||
continue
|
||||
}
|
||||
if f.same(e.Source.Repository, e.Source.Seat, repository, seat) {
|
||||
return true
|
||||
}
|
||||
}
|
||||
return false
|
||||
}
|
||||
|
||||
// forgeFacts is what the mesh's forge says of a repository a module is registered from (novox/hq ADR 0266).
|
||||
type forgeFacts struct {
|
||||
// ID is the forge's own id of the repository: what tells it from one deleted and made again by its name.
|
||||
ID int64
|
||||
// Guarded is whether the branch is protected as a module's trunk must be: no direct push, at least one
|
||||
// required status, and no administrator merging past one. Why says what is missing when it is not.
|
||||
Guarded bool
|
||||
Why string
|
||||
}
|
||||
|
||||
// askTheForge asks the forge, through its module's tools on the bus, for a repository's id and its branch's
|
||||
// protection. A variable so a test needs no forge.
|
||||
var askTheForge = func(ctx context.Context, owner, repo, branch string) (forgeFacts, error) {
|
||||
js, err := aBus()
|
||||
if err != nil {
|
||||
return forgeFacts{}, err
|
||||
}
|
||||
defer js.Close()
|
||||
bus := link.OverNATS{Conn: js.Conn()}
|
||||
ask := func(tool string, args map[string]any, into any) error {
|
||||
raw, _ := json.Marshal(args)
|
||||
answer, err := link.Ask(ctx, bus, "gitea", tool, raw, 30*time.Second)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
if answer.Error != "" {
|
||||
return fmt.Errorf("gitea.%s: %s", tool, answer.Error)
|
||||
}
|
||||
return json.Unmarshal(answer.Result, into)
|
||||
}
|
||||
var found struct {
|
||||
Result struct {
|
||||
ID int64 `json:"id"`
|
||||
FullName string `json:"full_name"`
|
||||
} `json:"result"`
|
||||
}
|
||||
if err := ask("gitea_api", map[string]any{"path": "/repos/" + owner + "/" + repo}, &found); err != nil {
|
||||
return forgeFacts{}, err
|
||||
}
|
||||
if found.Result.ID == 0 {
|
||||
return forgeFacts{}, fmt.Errorf("the forge named no id for %s/%s", owner, repo)
|
||||
}
|
||||
var rules struct {
|
||||
Rules []protectionRule `json:"rules"`
|
||||
}
|
||||
if err := ask("gitea_branch_protection_get", map[string]any{"owner": owner, "repo": repo}, &rules); err != nil {
|
||||
return forgeFacts{}, err
|
||||
}
|
||||
facts := judgedRule(rules.Rules, branch)
|
||||
facts.ID = found.Result.ID
|
||||
return facts, nil
|
||||
}
|
||||
|
||||
// protectionRule is a branch protection rule as the forge's tool summarises it.
|
||||
type protectionRule struct {
|
||||
Rule string `json:"rule"`
|
||||
Push bool `json:"push"`
|
||||
RequiredStatuses []string `json:"required_statuses"`
|
||||
AdminMayOverride bool `json:"admin_may_override"`
|
||||
}
|
||||
|
||||
// judgedRule judges the one rule the forge applies to a branch, as the forge picks it: the rule named for the
|
||||
// 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 {
|
||||
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:
|
||||
return forgeFacts{Why: fmt.Sprintf("the rule %s lets a person push to %s directly", r.Rule, branch)}
|
||||
case len(r.RequiredStatuses) == 0:
|
||||
return forgeFacts{Why: fmt.Sprintf("the rule %s requires no status before a merge into %s", r.Rule, branch)}
|
||||
case r.AdminMayOverride:
|
||||
return forgeFacts{Why: fmt.Sprintf("the rule %s lets an administrator merge into %s past a status", r.Rule, branch)}
|
||||
}
|
||||
return forgeFacts{Guarded: true}
|
||||
}
|
||||
|
||||
// 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 {
|
||||
return true
|
||||
}
|
||||
if !strings.ContainsAny(rule, "*?[") {
|
||||
return false
|
||||
}
|
||||
var re strings.Builder
|
||||
re.WriteString("^")
|
||||
for i := 0; i < len(rule); i++ {
|
||||
switch c := rule[i]; {
|
||||
case c == '*' && i+1 < len(rule) && rule[i+1] == '*':
|
||||
re.WriteString(".*")
|
||||
i++
|
||||
case c == '*':
|
||||
re.WriteString("[^/]*")
|
||||
case c == '?':
|
||||
re.WriteString("[^/]")
|
||||
default:
|
||||
re.WriteString(regexp.QuoteMeta(string(c)))
|
||||
}
|
||||
}
|
||||
re.WriteString("$")
|
||||
ok, _ := regexp.MatchString(re.String(), branch)
|
||||
return ok
|
||||
}
|
||||
|
||||
// onTheForge is a source's owner and name on the mesh's own forge (the git seat's holder), and whether it is
|
||||
// there at all.
|
||||
func (f *sourceForms) onTheForge(repository, seat string) (owner, name string, ok bool) {
|
||||
var rest string
|
||||
switch {
|
||||
case seat == gitSeat:
|
||||
rest = strings.Trim(repository, "/")
|
||||
case seat == "":
|
||||
base := f.base(gitSeat)
|
||||
if base == "" {
|
||||
return "", "", false
|
||||
}
|
||||
url, prefix := f.canonical(repository, ""), f.canonical(base, "")+"/"
|
||||
if !strings.HasPrefix(url, prefix) {
|
||||
return "", "", false
|
||||
}
|
||||
// The case the forge spells it with: the URL as given, past the base.
|
||||
rest = strings.TrimSuffix(strings.Trim(repository[len(prefix):], "/"), ".git")
|
||||
default:
|
||||
return "", "", false
|
||||
}
|
||||
parts := strings.Split(rest, "/")
|
||||
if len(parts) != 2 || parts[0] == "" || parts[1] == "" {
|
||||
return "", "", false
|
||||
}
|
||||
return parts[0], parts[1], true
|
||||
}
|
||||
|
||||
// mayRegisterFrom says whether a build's outcome may register module from the source it was built from, and the
|
||||
// forge's id of that repository to record (novox/hq ADR 0266). Through any verb, only:
|
||||
//
|
||||
// - from the module's registered repository — the same repository by the forge's own id, not only its name; or,
|
||||
// for a module new to the catalogue, from a repository the catalogue builds another module from;
|
||||
// - and from a repository on the mesh's forge whose trunk is protected as a trunk must be: no direct push, at
|
||||
// least one required status, no administrator merging past one.
|
||||
//
|
||||
// Anything else only when the build request was kept as asked at the controller's terminal, for this very
|
||||
// repository and path. 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,
|
||||
buildID, path, branch string) (int64, error) {
|
||||
forms := newSourceForms(ctx, inv)
|
||||
was, err := inv.SourceOf(ctx, module)
|
||||
isNew := errors.Is(err, inventory.ErrNoSuchModule)
|
||||
if err != nil && !isNew {
|
||||
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 "+
|
||||
"controller's terminal, never through a verb: a registered module is what the next push sends, and whoever "+
|
||||
"may call a verb includes agents (novox/hq ADR 0266)", errNotItsSource, why)
|
||||
}
|
||||
|
||||
var why string
|
||||
var alongside []inventory.Entry // the modules a new one's repository already builds
|
||||
switch {
|
||||
case !isNew && forms.same(was.Repository, was.Seat, built.Repository, built.Seat):
|
||||
case !isNew:
|
||||
registered := was.Repository
|
||||
if registered == "" {
|
||||
registered = "no repository (it was handed over by hand)"
|
||||
}
|
||||
why = fmt.Sprintf("%s is built from %s, and this build is of %s", module, sourceWords(registered, was.Seat),
|
||||
sourceWords(built.Repository, built.Seat))
|
||||
default:
|
||||
entries, err := inv.Catalogued(ctx)
|
||||
if err != nil {
|
||||
return 0, err
|
||||
}
|
||||
for _, e := range entries {
|
||||
if !e.Provided && e.Source.Repository != "" &&
|
||||
forms.same(e.Source.Repository, e.Source.Seat, built.Repository, built.Seat) {
|
||||
alongside = append(alongside, e)
|
||||
}
|
||||
}
|
||||
if len(alongside) == 0 {
|
||||
why = fmt.Sprintf("%s is new to the catalogue, and %s is no repository the catalogue builds a module from",
|
||||
module, sourceWords(built.Repository, built.Seat))
|
||||
}
|
||||
}
|
||||
if why != "" && !terminal {
|
||||
return refuse(why)
|
||||
}
|
||||
|
||||
owner, name, onForge := forms.onTheForge(built.Repository, built.Seat)
|
||||
if !onForge {
|
||||
if terminal {
|
||||
fmt.Printf("%s: %s is not on the mesh's forge — registered, as asked at the controller's terminal\n",
|
||||
buildID, sourceWords(built.Repository, built.Seat))
|
||||
return 0, nil
|
||||
}
|
||||
return refuse(fmt.Sprintf("%s is not on the mesh's forge, so whether its trunk is protected cannot be read",
|
||||
sourceWords(built.Repository, built.Seat)))
|
||||
}
|
||||
facts, err := askTheForge(ctx, owner, name, branch)
|
||||
if err != nil {
|
||||
if terminal {
|
||||
fmt.Printf("%s: the forge could not be asked about %s/%s (%v) — registered, as asked at the controller's "+
|
||||
"terminal\n", buildID, owner, name, err)
|
||||
return 0, nil
|
||||
}
|
||||
return refuse(fmt.Sprintf("the forge could not say whether %s/%s's %s is protected: %v", owner, name, branch, err))
|
||||
}
|
||||
if terminal {
|
||||
if why != "" || !facts.Guarded {
|
||||
fmt.Printf("%s: %s — registered, as asked at the controller's terminal\n", buildID,
|
||||
strings.Trim(why+"; "+facts.Why, "; "))
|
||||
}
|
||||
return facts.ID, nil
|
||||
}
|
||||
if !facts.Guarded {
|
||||
return refuse(fmt.Sprintf("%s/%s's %s is not protected as a module's trunk must be: %s", owner, name, branch,
|
||||
facts.Why))
|
||||
}
|
||||
// The same repository by the forge's id, not only its name: one deleted and made again is another.
|
||||
recorded := map[string]int64{}
|
||||
if !isNew {
|
||||
id, err := inv.SourceIdentity(ctx, module)
|
||||
if err != nil {
|
||||
return 0, err
|
||||
}
|
||||
recorded[module] = id
|
||||
}
|
||||
for _, e := range alongside {
|
||||
id, err := inv.SourceIdentity(ctx, e.Manifest.Module)
|
||||
if err != nil {
|
||||
return 0, err
|
||||
}
|
||||
recorded[e.Manifest.Module] = id
|
||||
}
|
||||
for m, id := range recorded {
|
||||
if id != 0 && id != facts.ID {
|
||||
return refuse(fmt.Sprintf("%s/%s is not the repository %s was registered from: the forge knows it as "+
|
||||
"repository %d, and %s was registered from repository %d — one of that name deleted and made again",
|
||||
owner, name, m, facts.ID, m, id))
|
||||
}
|
||||
}
|
||||
return facts.ID, nil
|
||||
}
|
||||
|
||||
// askedHereFor says the build was asked at the controller's terminal, for this repository and this path: a kept
|
||||
// request marked so, whose source is the outcome's (novox/hq ADR 0266).
|
||||
func askedHereFor(ctx context.Context, inv *inventory.Inventory, forms *sourceForms, buildID string,
|
||||
built inventory.Source, path string) (bool, error) {
|
||||
r, found, err := inv.BuildRequestByID(ctx, buildID)
|
||||
if err != nil || !found || !r.AtTerminal {
|
||||
return false, err
|
||||
}
|
||||
if !forms.same(r.Repository, r.Seat, built.Repository, built.Seat) ||
|
||||
strings.Trim(r.Path, "/") != strings.Trim(path, "/") {
|
||||
fmt.Printf("%s was asked at the terminal of %s at %q, and its outcome is of %s at %q: not the terminal's\n",
|
||||
buildID, sourceWords(r.Repository, r.Seat), r.Path, sourceWords(built.Repository, built.Seat), path)
|
||||
return false, nil
|
||||
}
|
||||
return true, nil
|
||||
}
|
||||
|
||||
// sourceWords is a source as a person reads it.
|
||||
func sourceWords(repository, seat string) string {
|
||||
if seat == "" {
|
||||
return repository
|
||||
}
|
||||
return buildSource{Repository: repository, Seat: seat}.String()
|
||||
}
|
||||
|
||||
// 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.
|
||||
func verbMayAsk(ctx context.Context, source buildSource, url string) error {
|
||||
if startedAtTheTerminal() {
|
||||
return nil
|
||||
}
|
||||
open, err := openStores(ctx)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
defer open.Close()
|
||||
entries, err := open.inventory.Catalogued(ctx)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
forms := newSourceForms(ctx, open.inventory)
|
||||
if forms.buildsFrom(entries, source.Repository, source.Seat) || forms.buildsFrom(entries, url, "") {
|
||||
return nil
|
||||
}
|
||||
return terminalRefusal("%s is no repository the catalogue builds a module from, and a build of any other is "+
|
||||
"asked at the controller's terminal only, never through a verb: a build node runs what the repository "+
|
||||
"says, and its outcome would register a module the next push sends — whoever may call a verb includes "+
|
||||
"agents (novox/hq ADR 0266). Nothing was asked", source)
|
||||
}
|
||||
@@ -0,0 +1,380 @@
|
||||
package main
|
||||
|
||||
import (
|
||||
"context"
|
||||
"encoding/json"
|
||||
"errors"
|
||||
"hash/fnv"
|
||||
"os"
|
||||
"os/exec"
|
||||
"slices"
|
||||
"strings"
|
||||
"sync"
|
||||
"testing"
|
||||
|
||||
"github.com/novox/mesh-controller/internal/inventory"
|
||||
"github.com/novox/mesh-controller/internal/link"
|
||||
)
|
||||
|
||||
// The routes to root a review of ADR 0266 found (novox/hq ADR 0266 §7): an agent makes a repository of its
|
||||
// own — or forks one the mesh builds from — commits a module.json naming a module the mesh runs everywhere
|
||||
// (`sudo`, granting itself a rule without a password; `mesh-host`, the node-engine), and asks the `build` verb
|
||||
// for it. Its outcome was registered under that name from the agent's repository, and the next push sent it.
|
||||
|
||||
// onTrunk is an outcome of a commit on its repository's trunk, as the build seat says it.
|
||||
func onTrunk(id, repository, seat, path string, manifest map[string]any) link.BuildResult {
|
||||
raw, _ := json.Marshal(manifest)
|
||||
r := link.BuildResult{ID: id, Repository: "http://forge.internal:20000/" + repository + ".git", Path: path,
|
||||
Ref: "main", On: "anchor", Commit: "c0ffee0123456789", Manifest: raw,
|
||||
Trunk: "main", OnTrunk: true, Branches: []string{"main"}}
|
||||
if seat != "" {
|
||||
r.Source = &link.SourceOnSeat{Seat: seat, Repository: repository}
|
||||
}
|
||||
return r
|
||||
}
|
||||
|
||||
// keptAsked keeps a build request as the asker would: through a verb, or at the terminal.
|
||||
func keptAsked(t *testing.T, inv *inventory.Inventory, id, repository, path string, atTerminal bool) {
|
||||
t.Helper()
|
||||
if err := inv.RecordBuildRequest(t.Context(), inventory.BuildRequest{ID: id, Repository: repository, Seat: "git",
|
||||
Path: path, For: "build", AtTerminal: atTerminal}); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
}
|
||||
|
||||
// theCatalogue is a mesh whose sudo is built from the catalogue repository and whose node-engine from its own.
|
||||
func theCatalogue(t *testing.T) *stores {
|
||||
t.Helper()
|
||||
open := aMesh(t)
|
||||
ctx := t.Context()
|
||||
for _, b := range []link.BuildResult{
|
||||
onTrunk("build-sudo", "novox/mesh-catalog", "git", "modules/sudo", map[string]any{"module": "sudo", "version": "1"}),
|
||||
onTrunk("build-host", "novox/mesh-host", "git", "", map[string]any{"module": "mesh-host", "version": "1"}),
|
||||
} {
|
||||
keptAsked(t, open.inventory, b.ID, b.Source.Repository, b.Path, true)
|
||||
if _, _, err := takeIn(ctx, open.inventory, b); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
}
|
||||
return open
|
||||
}
|
||||
|
||||
func TestABuildFromAnAgentsRepositoryIsNotRegisteredAsAModuleTheMeshHolds(t *testing.T) {
|
||||
open := theCatalogue(t)
|
||||
ctx := t.Context()
|
||||
for _, c := range []struct {
|
||||
name, repository, path, module string
|
||||
}{
|
||||
{"its own repository naming sudo", "agent/sudo", "modules/sudo", "sudo"},
|
||||
{"a fork of the catalogue", "agent/mesh-catalog", "modules/sudo", "sudo"},
|
||||
{"a fork of the node-engine", "agent/mesh-host", "", "mesh-host"},
|
||||
} {
|
||||
t.Run(c.name, func(t *testing.T) {
|
||||
id := "build-" + strings.ReplaceAll(c.repository, "/", "-")
|
||||
keptAsked(t, open.inventory, id, c.repository, c.path, false) // through the build verb
|
||||
evil := onTrunk(id, c.repository, "git", c.path, map[string]any{"module": c.module, "version": "evil"})
|
||||
_, _, err := takeIn(ctx, open.inventory, evil)
|
||||
if !errors.Is(err, errNotItsSource) {
|
||||
t.Fatalf("a build of %s was taken in as %s: %v", c.repository, c.module, err)
|
||||
}
|
||||
shelf, err := open.inventory.Catalogue(ctx)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if got := shelf[c.module].Version; got != "1" {
|
||||
t.Fatalf("%s is now %q, from %s", c.module, got, c.repository)
|
||||
}
|
||||
if _, found, _ := open.inventory.BuildByID(ctx, id); !found {
|
||||
t.Errorf("the refused build %s is not recorded", id)
|
||||
}
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
func TestANewModuleFromARepositoryTheCatalogueDoesNotBuildFromIsNotRegistered(t *testing.T) {
|
||||
open := theCatalogue(t)
|
||||
ctx := t.Context()
|
||||
keptAsked(t, open.inventory, "build-new", "agent/tools", "", false)
|
||||
_, _, err := takeIn(ctx, open.inventory, onTrunk("build-new", "agent/tools", "git", "",
|
||||
map[string]any{"module": "agent-tools", "version": "1"}))
|
||||
if !errors.Is(err, errNotItsSource) {
|
||||
t.Fatalf("a new module from an agent's repository was taken in: %v", err)
|
||||
}
|
||||
// And one asked of nobody here — an outcome on the bus no request was kept for — the same.
|
||||
_, _, err = takeIn(ctx, open.inventory, onTrunk("build-unasked", "agent/tools", "git", "",
|
||||
map[string]any{"module": "agent-tools", "version": "1"}))
|
||||
if !errors.Is(err, errNotItsSource) {
|
||||
t.Fatalf("an outcome nobody asked for was taken in: %v", err)
|
||||
}
|
||||
if shelf, _ := open.inventory.Catalogue(ctx); shelf["agent-tools"].Module != "" {
|
||||
t.Fatal("the refused module is in the catalogue")
|
||||
}
|
||||
}
|
||||
|
||||
// The operator at the terminal may still move a module, or add one from a new repository.
|
||||
func TestAtTheTerminalAnotherRepositoryIsRegistered(t *testing.T) {
|
||||
open := theCatalogue(t)
|
||||
ctx := t.Context()
|
||||
keptAsked(t, open.inventory, "build-moved", "novox/sudo", "", true)
|
||||
if _, _, err := takeIn(ctx, open.inventory, onTrunk("build-moved", "novox/sudo", "git", "",
|
||||
map[string]any{"module": "sudo", "version": "2"})); err != nil {
|
||||
t.Fatalf("a move the operator asked for at the terminal was refused: %v", err)
|
||||
}
|
||||
if src, _ := open.inventory.SourceOf(ctx, "sudo"); src.Repository != "novox/sudo" {
|
||||
t.Fatalf("sudo is built from %q", src.Repository)
|
||||
}
|
||||
if err := open.inventory.RecordBuildRequest(ctx, inventory.BuildRequest{ID: "build-external",
|
||||
Repository: "http://forge.internal:20000/someone/app.git", For: "build", AtTerminal: true}); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if _, _, err := takeIn(ctx, open.inventory, onTrunk("build-external", "someone/app", "", "",
|
||||
map[string]any{"module": "app", "version": "1"})); err != nil {
|
||||
t.Fatalf("a new module the operator asked for at the terminal was refused: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
// The delivery's flow is untouched: a merge's rebuild of a module from its own repository, and a merge adding
|
||||
// a module to a repository the catalogue builds from (novox/hq issue 300), are registered with no terminal.
|
||||
func TestADeliveryFromTheRegisteredRepositoryIsRegistered(t *testing.T) {
|
||||
open := theCatalogue(t)
|
||||
ctx := t.Context()
|
||||
rebuilt := onTrunk("build-plan", "novox/mesh-catalog", "git", "modules/sudo", map[string]any{"module": "sudo", "version": "2"})
|
||||
if err := open.inventory.RecordBuildRequest(ctx, inventory.BuildRequest{ID: rebuilt.ID,
|
||||
Repository: "novox/mesh-catalog", Seat: "git", Path: "modules/sudo", For: "plan"}); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if _, _, err := takeIn(ctx, open.inventory, rebuilt); err != nil {
|
||||
t.Fatalf("a plan's build of the module's own repository was refused: %v", err)
|
||||
}
|
||||
added := onTrunk("build-merge", "novox/mesh-catalog", "git", "modules/zram", map[string]any{"module": "zram", "version": "1"})
|
||||
if _, _, err := takeIn(ctx, open.inventory, added); err != nil {
|
||||
t.Fatalf("a module a merge added to the catalogue repository was refused: %v", err)
|
||||
}
|
||||
shelf, _ := open.inventory.Catalogue(ctx)
|
||||
if shelf["sudo"].Version != "2" || shelf["zram"].Module == "" {
|
||||
t.Fatalf("not registered: sudo %q, zram %q", shelf["sudo"].Version, shelf["zram"].Module)
|
||||
}
|
||||
}
|
||||
|
||||
// Through a verb, a build of a repository the catalogue builds nothing from is not even asked: the build node
|
||||
// would run what the agent wrote.
|
||||
func TestAVerbAsksNoBuildOfARepositoryTheCatalogueDoesNotBuildFrom(t *testing.T) {
|
||||
theCatalogue(t)
|
||||
ctx := t.Context()
|
||||
t.Setenv(verbVar, "build")
|
||||
t.Setenv(link.CallerVar, "node-tools.anchor, through the mesh-controller seat")
|
||||
err := verbMayAsk(ctx, buildSource{Repository: "agent/sudo", Seat: "git"}, "http://forge.internal:20000/agent/sudo.git")
|
||||
var policy *heldAtTheTerminal
|
||||
if !errors.As(err, &policy) {
|
||||
t.Fatalf("a verb's build of an agent's repository was asked: %v", err)
|
||||
}
|
||||
if err := verbMayAsk(ctx, buildSource{Repository: "novox/mesh-catalog", Seat: "git"},
|
||||
"http://forge.internal:20000/novox/mesh-catalog.git"); err != nil {
|
||||
t.Fatalf("a verb's build of the catalogue repository was refused: %v", err)
|
||||
}
|
||||
t.Setenv(verbVar, "")
|
||||
t.Setenv(link.CallerVar, "")
|
||||
if err := verbMayAsk(ctx, buildSource{Repository: "agent/sudo", Seat: "git"}, ""); err != nil {
|
||||
t.Fatalf("the terminal was refused: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
// asTheOperator keeps a build as asked at the controller's terminal, as the operator's first build of a module
|
||||
// from a repository the catalogue does not yet build from is (novox/hq ADR 0266), and hands it back.
|
||||
func asTheOperator(t *testing.T, inv *inventory.Inventory, b link.BuildResult) link.BuildResult {
|
||||
t.Helper()
|
||||
repository, seat := b.Repository, ""
|
||||
if b.Source != nil {
|
||||
repository, seat = b.Source.Repository, b.Source.Seat
|
||||
}
|
||||
if err := inv.RecordBuildRequest(t.Context(), inventory.BuildRequest{ID: b.ID, Repository: repository, Seat: seat,
|
||||
Path: b.Path, For: "build", AtTerminal: true}); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
return b
|
||||
}
|
||||
|
||||
// A rollback puts back only a build of the module's own repository: an agent's build of the module's name,
|
||||
// recorded and refused, at the very commit the machine ran before (a fork carries it), is never registered by
|
||||
// the back door of a failed gate.
|
||||
func TestARollbackNeverPutsBackABuildFromAnotherRepository(t *testing.T) {
|
||||
open := theCatalogue(t)
|
||||
ctx := t.Context()
|
||||
inv := open.inventory
|
||||
fork := onTrunk("build-1791500000000000000", "agent/mesh-catalog", "git", "modules/sudo",
|
||||
map[string]any{"module": "sudo", "version": "evil"})
|
||||
fork.Commit = "c0ffee0123456789" // the commit sudo was registered at
|
||||
keptAsked(t, inv, fork.ID, "agent/mesh-catalog", "modules/sudo", false)
|
||||
if _, _, err := takeIn(ctx, inv, fork); !errors.Is(err, errNotItsSource) {
|
||||
t.Fatalf("the fork's build was taken in: %v", err)
|
||||
}
|
||||
failed := onTrunk("build-1791600000000000000", "novox/mesh-catalog", "git", "modules/sudo",
|
||||
map[string]any{"module": "sudo", "version": "2"})
|
||||
failed.Commit = "badbadbad0123456"
|
||||
if _, _, err := takeIn(ctx, inv, failed); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
record, _, err := inv.BuildByID(ctx, failed.ID)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
previous, found, err := inv.PreviousBuild(ctx, "sudo", "c0ffee0123456789", record)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if found && previous.ID == fork.ID {
|
||||
t.Fatalf("a rollback would put back the fork's build %s", previous.ID)
|
||||
}
|
||||
if !found || previous.ID != "build-sudo" {
|
||||
t.Fatalf("a rollback puts back %q (found %v), want the registered build-sudo", previous.ID, found)
|
||||
}
|
||||
}
|
||||
|
||||
// theForge is what the forge says in a test, by owner/name: a repository not named here is protected as a
|
||||
// trunk must be, with an id of its own.
|
||||
var theForge sync.Map
|
||||
|
||||
func init() {
|
||||
askTheForge = func(_ context.Context, owner, repo, _ string) (forgeFacts, error) {
|
||||
if said, ok := theForge.Load(owner + "/" + repo); ok {
|
||||
switch f := said.(type) {
|
||||
case error:
|
||||
return forgeFacts{}, f
|
||||
case forgeFacts:
|
||||
return f, nil
|
||||
}
|
||||
}
|
||||
h := fnv.New32a()
|
||||
_, _ = h.Write([]byte(owner + "/" + repo))
|
||||
return forgeFacts{ID: int64(h.Sum32()), Guarded: true}, nil
|
||||
}
|
||||
}
|
||||
|
||||
// A module's trunk the forge does not protect — direct pushes, no required status — or a forge that cannot say,
|
||||
// registers nothing through a verb: the trunk rule means nothing on a branch anyone pushes to.
|
||||
func TestATrunkTheForgeDoesNotProtectRegistersNothing(t *testing.T) {
|
||||
open := theCatalogue(t)
|
||||
ctx := t.Context()
|
||||
theForge.Store("novox/unguarded", forgeFacts{ID: 7, Why: "the rule main lets a person push to main directly"})
|
||||
t.Cleanup(func() { theForge.Delete("novox/unguarded") })
|
||||
first := onTrunk("build-unguarded-1", "novox/unguarded", "git", "", map[string]any{"module": "unguarded", "version": "1"})
|
||||
keptAsked(t, open.inventory, first.ID, "novox/unguarded", "", true)
|
||||
if _, _, err := takeIn(ctx, open.inventory, first); err != nil {
|
||||
t.Fatalf("at the terminal: %v", err)
|
||||
}
|
||||
again := onTrunk("build-unguarded-2", "novox/unguarded", "git", "", map[string]any{"module": "unguarded", "version": "2"})
|
||||
if _, _, err := takeIn(ctx, open.inventory, again); !errors.Is(err, errNotItsSource) ||
|
||||
!strings.Contains(err.Error(), "push to main directly") {
|
||||
t.Fatalf("a rebuild from an unprotected trunk was taken in: %v", err)
|
||||
}
|
||||
theForge.Store("novox/unguarded", errors.New("nothing serves gitea.gitea_api"))
|
||||
if _, _, err := takeIn(ctx, open.inventory, onTrunk("build-unguarded-3", "novox/unguarded", "git", "",
|
||||
map[string]any{"module": "unguarded", "version": "3"})); !errors.Is(err, errNotItsSource) {
|
||||
t.Fatalf("a forge that could not say was read as a protected trunk: %v", err)
|
||||
}
|
||||
if shelf, _ := open.inventory.Catalogue(ctx); shelf["unguarded"].Version != "1" {
|
||||
t.Fatalf("unguarded is %q", shelf["unguarded"].Version)
|
||||
}
|
||||
}
|
||||
|
||||
// A repository deleted and made again under the module's repository's name is another repository: the forge's
|
||||
// id, recorded at registration, tells them apart.
|
||||
func TestARepositoryMadeAgainUnderItsNameIsNotTheModulesSource(t *testing.T) {
|
||||
open := theCatalogue(t)
|
||||
ctx := t.Context()
|
||||
theForge.Store("novox/remade", forgeFacts{ID: 100, Guarded: true})
|
||||
t.Cleanup(func() { theForge.Delete("novox/remade") })
|
||||
first := onTrunk("build-remade-1", "novox/remade", "git", "", map[string]any{"module": "remade", "version": "1"})
|
||||
keptAsked(t, open.inventory, first.ID, "novox/remade", "", true)
|
||||
if _, _, err := takeIn(ctx, open.inventory, first); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if id, _ := open.inventory.SourceIdentity(ctx, "remade"); id != 100 {
|
||||
t.Fatalf("the forge's id was not recorded: %d", id)
|
||||
}
|
||||
theForge.Store("novox/remade", forgeFacts{ID: 101, Guarded: true}) // deleted, and made again by an agent
|
||||
if _, _, err := takeIn(ctx, open.inventory, onTrunk("build-remade-2", "novox/remade", "git", "",
|
||||
map[string]any{"module": "remade", "version": "evil"})); !errors.Is(err, errNotItsSource) ||
|
||||
!strings.Contains(err.Error(), "made again") {
|
||||
t.Fatalf("a repository made again under the name was taken in: %v", err)
|
||||
}
|
||||
// And a new module from it, beside the one registered from the first, the same.
|
||||
if _, _, err := takeIn(ctx, open.inventory, onTrunk("build-remade-3", "novox/remade", "git", "modules/other",
|
||||
map[string]any{"module": "other", "version": "1"})); !errors.Is(err, errNotItsSource) {
|
||||
t.Fatalf("a new module from a repository made again was taken in: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
// The terminal's mark is the operator's for the repository and path they asked: an outcome of another, under that
|
||||
// build's id, is not theirs.
|
||||
func TestATerminalRequestCoversOnlyWhatItAsked(t *testing.T) {
|
||||
open := theCatalogue(t)
|
||||
ctx := t.Context()
|
||||
if err := open.inventory.RecordBuildRequest(ctx, inventory.BuildRequest{ID: "build-asked", Repository: "novox/app",
|
||||
Seat: "git", Path: "modules/app", For: "build", AtTerminal: true}); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
other := onTrunk("build-asked", "agent/sudo", "git", "modules/app", map[string]any{"module": "sudo", "version": "evil"})
|
||||
if _, _, err := takeIn(ctx, open.inventory, other); !errors.Is(err, errNotItsSource) {
|
||||
t.Fatalf("an outcome of another repository under a terminal request's id was taken in: %v", err)
|
||||
}
|
||||
elsewhere := onTrunk("build-asked", "novox/app", "git", "modules/sudo", map[string]any{"module": "sudo", "version": "evil"})
|
||||
if _, _, err := takeIn(ctx, open.inventory, elsewhere); !errors.Is(err, errNotItsSource) {
|
||||
t.Fatalf("an outcome of another path under a terminal request's id was taken in: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
// The serving controller, and everything it starts, is never the terminal: a build asked in it reads as asked
|
||||
// through the mesh even when no verb and no caller is named.
|
||||
func TestTheServingControllerIsNeverTheTerminal(t *testing.T) {
|
||||
t.Setenv(verbVar, "")
|
||||
t.Setenv(link.CallerVar, "")
|
||||
t.Setenv(servedVar, "")
|
||||
if !startedAtTheTerminal() {
|
||||
t.Fatal("a process started by hand is not the terminal")
|
||||
}
|
||||
markServed()
|
||||
if startedAtTheTerminal() {
|
||||
t.Fatal("the serving controller reads as the terminal")
|
||||
}
|
||||
child := exec.Command(os.Args[0], "-test.run=^$")
|
||||
child.Env = os.Environ()
|
||||
if !slices.Contains(child.Env, servedVar+"=1") {
|
||||
t.Fatal("what the serving controller starts does not carry its mark")
|
||||
}
|
||||
}
|
||||
|
||||
// The forge applies one rule to a branch: the rule named for it, else the first glob that covers it. A stronger
|
||||
// rule later in the list is not what guards the branch, and is not read as if it were (the confirmation review).
|
||||
func TestTheRuleJudgedIsTheOneTheForgeApplies(t *testing.T) {
|
||||
guarded := protectionRule{Rule: "*", RequiredStatuses: []string{"mesh/merge-gate"}}
|
||||
open := protectionRule{Rule: "main", Push: true, RequiredStatuses: []string{"mesh/merge-gate"}}
|
||||
if f := judgedRule([]protectionRule{guarded, open}, "main"); f.Guarded {
|
||||
t.Error("an exact rule letting pushes was passed over for a glob that guards")
|
||||
}
|
||||
if f := judgedRule([]protectionRule{{Rule: "ma*", RequiredStatuses: nil}, {Rule: "*", RequiredStatuses: []string{"x"}}},
|
||||
"main"); f.Guarded || !strings.Contains(f.Why, "ma*") {
|
||||
t.Errorf("the first glob covering the branch was not the one judged: %+v", f)
|
||||
}
|
||||
if f := judgedRule([]protectionRule{{Rule: "release/*"}, {Rule: "main", RequiredStatuses: []string{"x"}}}, "main"); !f.Guarded {
|
||||
t.Errorf("the rule named for the branch was not judged: %+v", f)
|
||||
}
|
||||
if f := judgedRule(nil, "main"); f.Guarded {
|
||||
t.Error("no rule is no protection")
|
||||
}
|
||||
}
|
||||
|
||||
// An id the forge could not give keeps the id already recorded.
|
||||
func TestAnUnknownIdentityKeepsTheRecordedOne(t *testing.T) {
|
||||
open := theCatalogue(t)
|
||||
ctx := t.Context()
|
||||
if err := open.inventory.SetSourceIdentity(ctx, "sudo", 100); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if err := open.inventory.SetSourceIdentity(ctx, "sudo", 0); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if id, _ := open.inventory.SourceIdentity(ctx, "sudo"); id != 100 {
|
||||
t.Fatalf("the recorded id became %d", id)
|
||||
}
|
||||
}
|
||||
@@ -19,11 +19,11 @@ func TestABuildHeardIsRecordedAndRegistered(t *testing.T) {
|
||||
open := aMesh(t)
|
||||
ctx := t.Context()
|
||||
manifest, _ := json.Marshal(map[string]any{"module": "shop", "version": "3"})
|
||||
m, _, err := takeIn(ctx, open.inventory, link.BuildResult{
|
||||
m, _, err := takeIn(ctx, open.inventory, asTheOperator(t, open.inventory, link.BuildResult{
|
||||
ID: "b-1", Repository: "http://forge.internal:20000/novox/shop.git", Path: "modules/shop",
|
||||
Ref: "main", On: "anchor", Commit: "abcdef0123", Manifest: manifest,
|
||||
Source: &link.SourceOnSeat{Seat: "git", Repository: "novox/shop"},
|
||||
})
|
||||
}))
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
@@ -78,7 +78,7 @@ func TestABuildAtACommitKeepsTheBranchTheModuleFollows(t *testing.T) {
|
||||
Path: "modules/unifi", Ref: ref, On: "anchor", Commit: commit, Manifest: manifest,
|
||||
Source: &link.SourceOnSeat{Seat: "git", Repository: "novox/mesh-catalog"}}
|
||||
}
|
||||
if _, _, err := takeIn(ctx, open.inventory, result("b-1", "main", "1111111aaaa")); err != nil {
|
||||
if _, _, err := takeIn(ctx, open.inventory, asTheOperator(t, open.inventory, result("b-1", "main", "1111111aaaa"))); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if _, _, err := takeIn(ctx, open.inventory, result("b-2", "9c97a8a", "9c97a8a1d2c3")); err != nil {
|
||||
@@ -117,7 +117,7 @@ func TestAnOlderBuildHeardLaterDoesNotReplaceTheNewer(t *testing.T) {
|
||||
Path: "modules/postgres", Ref: "main", On: "anchor", Commit: "efff5415", Manifest: manifest,
|
||||
Source: &link.SourceOnSeat{Seat: "git", Repository: "novox/mesh-catalog"}}
|
||||
}
|
||||
if _, _, err := takeIn(ctx, open.inventory, result(newer, "4bcd5f73")); err != nil {
|
||||
if _, _, err := takeIn(ctx, open.inventory, asTheOperator(t, open.inventory, result(newer, "4bcd5f73"))); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
_, _, err := takeIn(ctx, open.inventory, result(older, "0ab07fa9"))
|
||||
|
||||
@@ -33,12 +33,14 @@ var deliveryOwnerWithin = 10 * time.Second
|
||||
|
||||
// stalledLine is one delivery past its bound, as mesh-delivery's `stalled` says it.
|
||||
type stalledLine struct {
|
||||
ID string `json:"id"`
|
||||
State string `json:"state"`
|
||||
For string `json:"for"`
|
||||
Bound string `json:"bound"`
|
||||
H2 string `json:"h2"`
|
||||
Says string `json:"says"`
|
||||
ID string `json:"id"`
|
||||
// Number is the pull request's, for a line of a head the forge never announced (novox/hq issue 347).
|
||||
Number int `json:"number,omitempty"`
|
||||
State string `json:"state"`
|
||||
For string `json:"for"`
|
||||
Bound string `json:"bound"`
|
||||
H2 string `json:"h2"`
|
||||
Says string `json:"says"`
|
||||
}
|
||||
|
||||
// operatorsOnly is whether the table leaves H2 nothing to do for the line: the state is the operator's.
|
||||
|
||||
@@ -221,7 +221,8 @@ func TestABuildThatFailsItsGateIsRolledBackOnItsFirstMachineAndGoesNoFurther(t *
|
||||
t.Fatalf("sent again after the rollback: %v", g.sent)
|
||||
}
|
||||
_, _, err = takeIn(ctx, inv, link.BuildResult{ID: "build-2", Repository: "novox/mesh-catalog", Path: "modules/app",
|
||||
Commit: "c2", Manifest: mustJSON(t, catalogue.Manifest{Module: "app", Version: "c2"})})
|
||||
Commit: "c2", Manifest: mustJSON(t, catalogue.Manifest{Module: "app", Version: "c2"}),
|
||||
Source: &link.SourceOnSeat{Seat: "git", Repository: "novox/mesh-catalog"}})
|
||||
if err == nil || !strings.Contains(err.Error(), "failed its gate") {
|
||||
t.Fatalf("the failed build was registered again: %v", err)
|
||||
}
|
||||
|
||||
@@ -216,6 +216,10 @@ func run() error {
|
||||
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
|
||||
(novox/hq ADR 0272).
|
||||
|
||||
migrate bring each context's schema up to date
|
||||
prepare the same, asked the way the mesh asks any module (ADR 0135)
|
||||
node add <name> [--adopted] create a node record; --adopted: the machine is in use
|
||||
|
||||
@@ -0,0 +1,261 @@
|
||||
package main
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"context"
|
||||
"errors"
|
||||
"fmt"
|
||||
"os/exec"
|
||||
"slices"
|
||||
"strings"
|
||||
|
||||
"github.com/novox/mesh-controller/internal/inventory"
|
||||
"github.com/novox/mesh-controller/internal/link"
|
||||
)
|
||||
|
||||
// The operator's command, `mesh-cli` (alias `nox`), answered here (novox/hq ADR 0272).
|
||||
//
|
||||
// mesh-cli asks the node-engine on its own machine; the engine reads the asking account from the kernel and asks
|
||||
// this controller on its own node's subject. Here it is judged — the controller's terminal, an ordinary call, or
|
||||
// nothing — and run as every verb's line is: a fresh process of this binary, with this process's environment.
|
||||
//
|
||||
// **The terminal** is a line from a node's operator account, on the control-node (§4). Phase 2 adds a node whose
|
||||
// agents run under an account of their own, judged unable to become root (ADR 0266, not built yet); until then no
|
||||
// other node is the terminal, because agents there run as its operator. **An ordinary call** is a line from that
|
||||
// account elsewhere: it meets every refusal of the generic `command` verb and runs with `MESH_VERB=mesh-cli`, so a
|
||||
// terminal-only change is refused with its reason. **Any other account is refused**, root included: those two
|
||||
// accounts already reach the mesh's verbs through the mesh MCP server, and no other account gains anything here.
|
||||
|
||||
// cliVerb is what a line from mesh-cli that is not the terminal runs as: the verb its process names, so every
|
||||
// terminal-only refusal applies and says it came through mesh-cli.
|
||||
const cliVerb = "mesh-cli"
|
||||
|
||||
// cliServers are commands that serve until stopped. Run for mesh-cli they would hold a second server under this
|
||||
// one until the call's bound killed it.
|
||||
var cliServers = map[string]bool{"serve": true, "api": true, "board": true}
|
||||
|
||||
// cliVerdict is how a line is run, or why it is not.
|
||||
type cliVerdict struct {
|
||||
terminal bool
|
||||
// why says why the line is or is not the terminal, in words the operator reads under the answer.
|
||||
why string
|
||||
// refused says why nothing runs.
|
||||
refused string
|
||||
}
|
||||
|
||||
// judgeCLI decides how a line from mesh-cli runs (ADR 0272 §4). nodes is every node the mesh knows; control is
|
||||
// every node the controller's module is assigned to, which is exactly one in a mesh that is well.
|
||||
func judgeCLI(node string, asked link.CLIAsked, nodes []inventory.Node, control []string) cliVerdict {
|
||||
var record *inventory.Node
|
||||
for i := range nodes {
|
||||
if nodes[i].Name == node {
|
||||
record = &nodes[i]
|
||||
break
|
||||
}
|
||||
}
|
||||
switch {
|
||||
case record == nil:
|
||||
return cliVerdict{refused: fmt.Sprintf("%s is not a node this mesh knows, so nothing ran", node)}
|
||||
case asked.UID == 0 || asked.Account == "root":
|
||||
return cliVerdict{refused: "mesh-cli answers the operator's own account, never root: run it as yourself, " +
|
||||
"not under sudo. Nothing ran"}
|
||||
case record.Account == "":
|
||||
return cliVerdict{refused: fmt.Sprintf("the mesh does not know %s's operator account (`node account <node> <login>`), "+
|
||||
"so no account there is answered. Nothing ran", node)}
|
||||
case asked.Account != record.Account:
|
||||
return cliVerdict{refused: fmt.Sprintf("mesh-cli answers %s's operator account (%s) only, and this line came "+
|
||||
"from %s. Nothing ran", node, record.Account, asked.Account)}
|
||||
}
|
||||
if len(control) != 1 {
|
||||
return cliVerdict{why: fmt.Sprintf("not the controller's terminal: the mesh names %d control-nodes, and the "+
|
||||
"terminal is the one control-node's operator account", len(control))}
|
||||
}
|
||||
if control[0] == node {
|
||||
// **A person at a terminal, not a service of the operator's** (review of ADR 0272): the tool runner and the
|
||||
// account's user units run as the operator too, and only a login session is the terminal.
|
||||
if asked.Session == "" {
|
||||
return cliVerdict{why: fmt.Sprintf("not the controller's terminal: %s on %s asked from no login session — a "+
|
||||
"service or a user unit running as the operator, not a person at a terminal", asked.Account, node)}
|
||||
}
|
||||
return cliVerdict{terminal: true, why: fmt.Sprintf("the controller's terminal: %s on the control-node %s, "+
|
||||
"login session %s", asked.Account, node, asked.Session)}
|
||||
}
|
||||
return cliVerdict{why: fmt.Sprintf("not the controller's terminal: agents on %s may run as %s, so a "+
|
||||
"terminal-only change is made from the control-node %s (novox/hq ADR 0272)", node, asked.Account, control[0])}
|
||||
}
|
||||
|
||||
// controlNodes is every node the controller's module is assigned to.
|
||||
func controlNodes(ctx context.Context, inv *inventory.Inventory, nodes []inventory.Node) ([]string, error) {
|
||||
var out []string
|
||||
for _, n := range nodes {
|
||||
modules, err := inv.Assigned(ctx, n.Name)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
if slices.Contains(modules, controllerModule) {
|
||||
out = append(out, n.Name)
|
||||
}
|
||||
}
|
||||
return out, nil
|
||||
}
|
||||
|
||||
// controllerModule is the module that runs the controller, and so marks the control-node.
|
||||
const controllerModule = "mesh-controller"
|
||||
|
||||
// answerMeshCLI is the serving controller's answer to mesh-cli.
|
||||
func answerMeshCLI(inv *inventory.Inventory) link.CLIHandler {
|
||||
return func(ctx context.Context, node string, asked link.CLIAsked) link.CLIAnswer {
|
||||
if handingOver.Load() {
|
||||
return link.CLIRefusal("this controller is stopping and runs no new command; ask again in a moment. Nothing ran")
|
||||
}
|
||||
nodes, err := inv.Nodes(ctx)
|
||||
if err != nil {
|
||||
return link.CLIRefusal("the mesh's nodes cannot be read, so nothing ran: " + err.Error())
|
||||
}
|
||||
control, err := controlNodes(ctx, inv, nodes)
|
||||
if err != nil {
|
||||
return link.CLIRefusal("which node is the control-node cannot be read, so nothing ran: " + err.Error())
|
||||
}
|
||||
return runForMeshCLI(ctx, node, asked, judgeCLI(node, asked, nodes, control))
|
||||
}
|
||||
}
|
||||
|
||||
// cliJournal says one line in the controller's journal (its standard output); a variable so a test can read it.
|
||||
var cliJournal = func(line string) { fmt.Println(line) }
|
||||
|
||||
// runForMeshCLI runs a judged line and answers what it said. Every line is said in the journal first: its call,
|
||||
// who asked on which node, its command word only, and how it runs (review of ADR 0272).
|
||||
func runForMeshCLI(ctx context.Context, node string, asked link.CLIAsked, v cliVerdict) link.CLIAnswer {
|
||||
how := "as an ordinary call"
|
||||
switch {
|
||||
case v.refused != "":
|
||||
how = "refused: " + v.refused
|
||||
case v.terminal:
|
||||
how = "as the controller's terminal"
|
||||
}
|
||||
call := link.CallIDIn(ctx)
|
||||
if call == "" {
|
||||
call = "(no call)"
|
||||
}
|
||||
cliJournal(fmt.Sprintf("mesh-cli %s: %s on %s asked %q, %s", call, asked.Account, node, asked.Line[0], how))
|
||||
if v.refused != "" {
|
||||
return link.CLIRefusal(v.refused)
|
||||
}
|
||||
if cliServers[asked.Line[0]] {
|
||||
return link.CLIAnswer{Exit: 1, Why: v.why, Refused: fmt.Sprintf("%s serves until stopped, and is not a "+
|
||||
"command line mesh-cli runs. Nothing ran", asked.Line[0])}
|
||||
}
|
||||
verb, line := "", asked.Line
|
||||
if !v.terminal {
|
||||
composed, err := ordinaryLine(asked.Line)
|
||||
if err != nil {
|
||||
return link.CLIAnswer{Exit: 1, Why: v.why, Refused: err.Error()}
|
||||
}
|
||||
verb, line = cliVerb, composed
|
||||
}
|
||||
cmd := selfCommand(ctx, line)
|
||||
cmd.Env = commandEnvironment(fmt.Sprintf("%s through mesh-cli on %s", asked.Account, node), verb, v.terminal)
|
||||
// No standard input: a command that reads one gets nothing, and fails saying so (ADR 0272 §5).
|
||||
cmd.Stdin = nil
|
||||
var stdout, stderr bytes.Buffer
|
||||
cmd.Stdout, cmd.Stderr = &stdout, &stderr
|
||||
err := cmd.Run()
|
||||
answer := link.CLIAnswer{Stdout: stdout.Bytes(), Stderr: stderr.Bytes(), Terminal: v.terminal, Why: v.why}
|
||||
var exit *exec.ExitError
|
||||
switch {
|
||||
case err == nil:
|
||||
case errors.As(err, &exit):
|
||||
answer.Exit = exit.ExitCode()
|
||||
if answer.Exit < 0 {
|
||||
// Killed: by the call's bound, or by this controller stopping.
|
||||
answer.Exit = 1
|
||||
answer.Stderr = append(answer.Stderr, []byte(fmt.Sprintf("\nmesh-cli: the command was stopped (%v)\n",
|
||||
exit))...)
|
||||
}
|
||||
default:
|
||||
answer.Exit = 1
|
||||
answer.Refused = fmt.Sprintf("could not run %s from this controller's own build: %v", strings.Join(asked.Line, " "), err)
|
||||
}
|
||||
return answer
|
||||
}
|
||||
|
||||
// settingsForms is what an ordinary `settings` line may say: the settings verb's own forms.
|
||||
const settingsForms = "settings set <module> <values> [--replace] [--node <node>], settings clear <module> " +
|
||||
"[--node <node>], settings show <module> [--history] [--node <node>], or settings preferences [<module>] " +
|
||||
"[--node <node>]"
|
||||
|
||||
// ordinaryLine is the command line an ordinary call runs (ADR 0272 §4): a `settings` line composed exactly as the
|
||||
// settings verb composes its own, so its refusals — the terminal-only keys among them — are that verb's (review of
|
||||
// ADR 0272); any other line as the generic `command` verb takes it, with its refusals.
|
||||
func ordinaryLine(argv []string) ([]string, error) {
|
||||
if len(argv) == 0 || argv[0] != "settings" {
|
||||
// What the generic verb runs, and nothing it would refuse: its reading forms only, and never a command that
|
||||
// is the terminal's alone (novox/hq ADR 0266) — the two checks argvFor makes of the `command` verb's line,
|
||||
// made of the words as given rather than re-split from one string.
|
||||
if err := refusedAsTheGenericCommand(argv); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
if err := terminalOnly(argv); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
return argv, nil
|
||||
}
|
||||
args := map[string]any{}
|
||||
var words []string
|
||||
rest := argv[1:]
|
||||
for i := 0; i < len(rest); i++ {
|
||||
w := rest[i]
|
||||
switch {
|
||||
case w == "--replace" || w == "-replace":
|
||||
args["replace"] = "true"
|
||||
case w == "--history" || w == "-history":
|
||||
args["history"] = "true"
|
||||
case w == "--node" || w == "-node":
|
||||
if i+1 >= len(rest) {
|
||||
return nil, fmt.Errorf("--node names no node; settings takes %s. Nothing ran", settingsForms)
|
||||
}
|
||||
i++
|
||||
args["node"] = rest[i]
|
||||
case strings.HasPrefix(w, "--node=") || strings.HasPrefix(w, "-node="):
|
||||
_, args["node"], _ = strings.Cut(w, "=")
|
||||
case strings.HasPrefix(w, "-"):
|
||||
return nil, fmt.Errorf("settings takes no %s through mesh-cli outside the terminal; it takes %s. "+
|
||||
"Nothing ran", w, settingsForms)
|
||||
default:
|
||||
words = append(words, w)
|
||||
}
|
||||
}
|
||||
wrong := fmt.Errorf("through mesh-cli outside the terminal, settings takes the settings verb's forms: %s. "+
|
||||
"Nothing ran", settingsForms)
|
||||
if len(words) == 0 {
|
||||
return nil, wrong
|
||||
}
|
||||
switch words[0] {
|
||||
case "set":
|
||||
if len(words) != 3 {
|
||||
return nil, wrong
|
||||
}
|
||||
args["module"], args["values"] = words[1], words[2]
|
||||
case "clear":
|
||||
if len(words) != 2 {
|
||||
return nil, wrong
|
||||
}
|
||||
args["module"], args["clear"] = words[1], "true"
|
||||
case "show":
|
||||
if len(words) != 2 {
|
||||
return nil, wrong
|
||||
}
|
||||
args["module"] = words[1]
|
||||
case "preferences":
|
||||
if len(words) > 2 {
|
||||
return nil, wrong
|
||||
}
|
||||
args["list"] = "preferences"
|
||||
if len(words) == 2 {
|
||||
args["module"] = words[1]
|
||||
}
|
||||
default:
|
||||
return nil, wrong
|
||||
}
|
||||
return argvFor("settings", args)
|
||||
}
|
||||
@@ -0,0 +1,305 @@
|
||||
package main
|
||||
|
||||
import (
|
||||
"context"
|
||||
"encoding/base64"
|
||||
"encoding/json"
|
||||
"strings"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/nats-io/nats.go"
|
||||
|
||||
"github.com/novox/mesh-controller/internal/inventory"
|
||||
"github.com/novox/mesh-controller/internal/link"
|
||||
"github.com/novox/mesh-controller/internal/testbus"
|
||||
)
|
||||
|
||||
// echoEnvironment makes the test binary, run as a command line, say the verb and caller it was given (TestMain).
|
||||
const echoEnvironment = "MESH_TEST_ECHO_ENVIRONMENT"
|
||||
|
||||
var cliNodes = []inventory.Node{
|
||||
{Name: "control", Account: "operator"},
|
||||
{Name: "laptop", Account: "operator"},
|
||||
{Name: "unnamed"},
|
||||
}
|
||||
|
||||
func asked(account string, uid uint32, line ...string) link.CLIAsked {
|
||||
return link.CLIAsked{Line: line, Account: account, UID: uid, Session: "session-1.scope"}
|
||||
}
|
||||
|
||||
// Who is the controller's terminal, and who is an ordinary call or refused (novox/hq ADR 0272 §4).
|
||||
func TestMeshCLIIsTheTerminalOnlyForTheControlNodesOperator(t *testing.T) {
|
||||
control := []string{"control"}
|
||||
cases := []struct {
|
||||
name, node string
|
||||
asked link.CLIAsked
|
||||
control []string
|
||||
terminal bool
|
||||
refused string
|
||||
why string
|
||||
}{
|
||||
{"the control-node's operator", "control", asked("operator", 1000, "status"), control, true, "", "the controller's terminal"},
|
||||
{"another node's operator", "laptop", asked("operator", 1000, "status"), control, false, "", "agents on laptop may run as operator"},
|
||||
{"another account", "control", asked("agent", 1001, "status"), control, false, "operator account (operator) only", ""},
|
||||
{"root", "control", asked("root", 0, "status"), control, false, "never root", ""},
|
||||
{"a node with no operator account", "unnamed", asked("operator", 1000, "status"), control, false, "does not know unnamed's operator account", ""},
|
||||
{"a node the mesh does not know", "elsewhere", asked("operator", 1000, "status"), control, false, "not a node this mesh knows", ""},
|
||||
{"two control-nodes", "control", asked("operator", 1000, "status"), []string{"control", "laptop"}, false, "", "2 control-nodes"},
|
||||
}
|
||||
for _, c := range cases {
|
||||
v := judgeCLI(c.node, c.asked, cliNodes, c.control)
|
||||
if v.terminal != c.terminal {
|
||||
t.Errorf("%s: terminal %v, want %v (%+v)", c.name, v.terminal, c.terminal, v)
|
||||
}
|
||||
if c.refused == "" && v.refused != "" || c.refused != "" && !strings.Contains(v.refused, c.refused) {
|
||||
t.Errorf("%s: refused %q, want %q", c.name, v.refused, c.refused)
|
||||
}
|
||||
if c.why != "" && !strings.Contains(v.why, c.why) {
|
||||
t.Errorf("%s: why %q, want %q", c.name, v.why, c.why)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// The terminal runs its line without MESH_VERB, even where this process carries one; an ordinary call names
|
||||
// mesh-cli; both record who asked through mesh-cli.
|
||||
func TestTheTerminalRunsWithoutAVerbAndAnOrdinaryCallNamesMeshCLI(t *testing.T) {
|
||||
t.Setenv(echoEnvironment, "1")
|
||||
t.Setenv("MESH_VERB", "leaked-from-the-serving-process")
|
||||
// The serving controller marks itself (ADR 0266); its terminal line for mesh-cli is still the terminal's.
|
||||
t.Setenv(servedVar, "1")
|
||||
ctx := context.Background()
|
||||
|
||||
a := runForMeshCLI(ctx, "control", asked("operator", 1000, "status"), cliVerdict{terminal: true, why: "the terminal"})
|
||||
if a.Exit != 0 || a.Refused != "" || !a.Terminal {
|
||||
t.Fatalf("the terminal's line did not run: %+v", a)
|
||||
}
|
||||
if got := string(a.Stdout); !strings.Contains(got, `verb=""`) || !strings.Contains(got, "operator through mesh-cli on control") ||
|
||||
!strings.Contains(got, "terminal=true") {
|
||||
t.Fatalf("the terminal's line ran with %s", got)
|
||||
}
|
||||
|
||||
a = runForMeshCLI(ctx, "laptop", asked("operator", 1000, "status"), cliVerdict{why: "not the terminal"})
|
||||
if a.Exit != 0 || a.Terminal || a.Why != "not the terminal" {
|
||||
t.Fatalf("an ordinary line did not run as one: %+v", a)
|
||||
}
|
||||
if got := string(a.Stdout); !strings.Contains(got, `verb="mesh-cli"`) || !strings.Contains(got, "terminal=false") {
|
||||
t.Fatalf("an ordinary line ran with %s", got)
|
||||
}
|
||||
}
|
||||
|
||||
// An ordinary call meets every refusal of the generic command verb, and nothing runs; a server is never run.
|
||||
func TestAnOrdinaryCallMeetsTheCommandVerbsRefusals(t *testing.T) {
|
||||
t.Setenv(echoEnvironment, "1")
|
||||
ctx := context.Background()
|
||||
ordinary := cliVerdict{why: "not the terminal"}
|
||||
a := runForMeshCLI(ctx, "laptop", asked("operator", 1000, "cleanup", "delete", "x"), ordinary)
|
||||
if a.Refused == "" || len(a.Stdout) != 0 || a.Exit != 1 || a.Why != "not the terminal" {
|
||||
t.Fatalf("a repair without --why ran as an ordinary call: %+v", a)
|
||||
}
|
||||
a = runForMeshCLI(ctx, "laptop", asked("operator", 1000, "settings", "set", "claude-code", "{}"), ordinary)
|
||||
if a.Refused != "" || !strings.Contains(string(a.Stdout), `verb="mesh-cli"`) {
|
||||
t.Fatalf("an ordinary settings set did not run through the settings verb's path with MESH_VERB set: %+v", a)
|
||||
}
|
||||
for _, server := range []string{"serve", "api", "board"} {
|
||||
a := runForMeshCLI(ctx, "control", asked("operator", 1000, server), cliVerdict{terminal: true})
|
||||
if a.Refused == "" || len(a.Stdout) != 0 {
|
||||
t.Fatalf("%s was run for mesh-cli: %+v", server, a)
|
||||
}
|
||||
}
|
||||
a = runForMeshCLI(ctx, "control", asked("agent", 1001, "status"), cliVerdict{refused: "agent is not answered"})
|
||||
if a.Refused != "agent is not answered" || len(a.Stdout) != 0 {
|
||||
t.Fatalf("a refused line ran: %+v", a)
|
||||
}
|
||||
}
|
||||
|
||||
// **Which node is the terminal does not follow a verb** (review of ADR 0272): the controller's module is assigned
|
||||
// and unassigned at the terminal alone, so no caller of `assign` can move the terminal to a node of its choosing.
|
||||
// Asked through the acts themselves, as the verbs ask them; refused before any store is touched (none is given).
|
||||
func TestTheControllersModuleIsMovedAtTheTerminalAlone(t *testing.T) {
|
||||
t.Setenv("MESH_VERB", "assign")
|
||||
ctx := context.Background()
|
||||
if _, err := assignWith(ctx, nil, "laptop", assignOptions{}, "zsh", "mesh-controller"); err == nil ||
|
||||
!strings.Contains(err.Error(), "terminal") {
|
||||
t.Fatalf("assigning the controller through a verb was not refused: %v", err)
|
||||
}
|
||||
if _, err := assign(ctx, nil, "laptop", "mesh-controller"); err == nil || !strings.Contains(err.Error(), "terminal") {
|
||||
t.Fatalf("assigning the controller through a verb was not refused: %v", err)
|
||||
}
|
||||
t.Setenv("MESH_VERB", "unassign")
|
||||
if _, err := unassign(ctx, nil, "control", "mesh-controller"); err == nil || !strings.Contains(err.Error(), "terminal") {
|
||||
t.Fatalf("unassigning the controller through a verb was not refused: %v", err)
|
||||
}
|
||||
// Another module through a verb, and the controller's at the terminal, are not refused for it.
|
||||
if err := refusedMovingTheController([]string{"zsh"}); err != nil {
|
||||
t.Fatalf("another module was refused: %v", err)
|
||||
}
|
||||
t.Setenv("MESH_VERB", "")
|
||||
if err := refusedMovingTheController([]string{"mesh-controller"}); err != nil {
|
||||
t.Fatalf("the terminal was refused: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
// An ordinary `settings set|clear` goes down the settings verb's own path: composed as that verb composes it, and
|
||||
// run with MESH_VERB set, so its refusals — the terminal-only keys among them — are the settings command's own,
|
||||
// not a blanket refusal of the generic command (review of ADR 0272).
|
||||
func TestAnOrdinarySettingsLineTakesTheSettingsVerbsPath(t *testing.T) {
|
||||
cases := []struct {
|
||||
line []string
|
||||
want string
|
||||
err string
|
||||
}{
|
||||
{[]string{"settings", "set", "zsh", `{"execute":"withhold"}`, "--node", "laptop"}, `settings set zsh {"execute":"withhold"} --node laptop`, ""},
|
||||
{[]string{"settings", "set", "zsh", "{}", "--replace"}, "settings set zsh {} --replace", ""},
|
||||
{[]string{"settings", "clear", "zsh", "--node", "laptop"}, "settings clear zsh --node laptop", ""},
|
||||
{[]string{"settings", "show", "zsh", "--history"}, "settings show zsh --history", ""},
|
||||
{[]string{"settings", "preferences"}, "settings preferences", ""},
|
||||
{[]string{"settings", "set", "zsh"}, "", "settings"},
|
||||
{[]string{"settings", "set", "zsh", "{}", "--sideways"}, "", "--sideways"},
|
||||
}
|
||||
for _, c := range cases {
|
||||
argv, err := ordinaryLine(c.line)
|
||||
if c.err != "" {
|
||||
if err == nil || !strings.Contains(err.Error(), c.err) {
|
||||
t.Errorf("%q: refused with %v, want %q", c.line, err, c.err)
|
||||
}
|
||||
continue
|
||||
}
|
||||
if err != nil || strings.Join(argv, " ") != c.want {
|
||||
t.Errorf("%q: composed %q (%v), want %q", c.line, argv, err, c.want)
|
||||
}
|
||||
}
|
||||
// Anything else still meets the generic command verb's refusals.
|
||||
if _, err := ordinaryLine([]string{"retire", "delete", "x"}); err == nil {
|
||||
t.Error("a repair composed as an ordinary line")
|
||||
}
|
||||
}
|
||||
|
||||
// Every line is said in the controller's journal, with its call, who asked where and how it ran — its command word
|
||||
// only, never the rest of the line (review of ADR 0272).
|
||||
func TestEveryMeshCLILineIsSaidInTheJournal(t *testing.T) {
|
||||
t.Setenv(echoEnvironment, "1")
|
||||
var said []string
|
||||
was := cliJournal
|
||||
cliJournal = func(line string) { said = append(said, line) }
|
||||
t.Cleanup(func() { cliJournal = was })
|
||||
ctx := link.WithCallID(context.Background(), "call-1")
|
||||
runForMeshCLI(ctx, "control", asked("operator", 1000, "settings", "set", "x", `{"password":"s3cret"}`),
|
||||
cliVerdict{terminal: true, why: "the terminal"})
|
||||
runForMeshCLI(ctx, "control", asked("agent", 1001, "status"), cliVerdict{refused: "agent is not answered"})
|
||||
all := strings.Join(said, "\n")
|
||||
if len(said) != 2 || !strings.Contains(all, "call-1") || !strings.Contains(all, "operator on control") ||
|
||||
!strings.Contains(all, "as the controller's terminal") || !strings.Contains(all, "refused") {
|
||||
t.Fatalf("the journal said %q", said)
|
||||
}
|
||||
if strings.Contains(all, "s3cret") {
|
||||
t.Fatalf("the journal carries the line's values: %q", said)
|
||||
}
|
||||
}
|
||||
|
||||
// **An ordinary line runs only what the generic command verb would** (review of ADR 0272, ADR 0266): its reading
|
||||
// forms, and never a command that is the terminal's alone. Each of these, from another node's operator, is refused
|
||||
// and runs nothing.
|
||||
func TestAnOrdinaryLineRunsNothingTheCommandVerbWouldRefuse(t *testing.T) {
|
||||
t.Setenv(echoEnvironment, "1")
|
||||
for _, line := range [][]string{
|
||||
{"node", "account", "control", "x"},
|
||||
{"node", "agent-account", "control", "x", "--clear"},
|
||||
{"token", "issue", "laptop"},
|
||||
{"secret", "export", "x"},
|
||||
{"operator", "key", "set", "x"},
|
||||
{"assign", "laptop", "zsh"},
|
||||
} {
|
||||
if _, err := ordinaryLine(line); err == nil {
|
||||
t.Errorf("%q composed as an ordinary line", line)
|
||||
}
|
||||
a := runForMeshCLI(context.Background(), "laptop", asked("operator", 1000, line...), cliVerdict{why: "not the terminal"})
|
||||
if a.Refused == "" || len(a.Stdout) != 0 {
|
||||
t.Errorf("%q ran as an ordinary line: %+v", line, a)
|
||||
}
|
||||
}
|
||||
if argv, err := ordinaryLine([]string{"status"}); err != nil || argv[0] != "status" {
|
||||
t.Fatalf("a read was refused: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
// **`calls` never shows a mesh-cli line's answer** (review of ADR 0272): the verb's own answer is read for a line
|
||||
// served over a bus, and neither the answer's text nor the base64 JSON writes its bytes in is there.
|
||||
func TestTheCallsVerbNeverShowsAMeshCLILinesAnswer(t *testing.T) {
|
||||
conn, err := nats.Connect(testbus.URL(t))
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
defer conn.Close()
|
||||
stop, err := link.OverNATS{Conn: conn}.ServeCLI(func(context.Context, string, link.CLIAsked) link.CLIAnswer {
|
||||
return link.CLIAnswer{Stdout: []byte("s3cret-join")}
|
||||
}, nil)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
defer stop()
|
||||
body, _ := json.Marshal(link.CLIAsked{Line: []string{"token", "issue", "x"}, Account: "operator"})
|
||||
msg, err := conn.Request(link.CLISubject("control"), body, 5*time.Second)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if !strings.Contains(string(msg.Data), base64.StdEncoding.EncodeToString([]byte("s3cret-join"))) {
|
||||
t.Fatalf("the asker was not given its answer: %s", msg.Data)
|
||||
}
|
||||
recent, _ := link.Calls.Recent()
|
||||
var id string
|
||||
for _, c := range recent {
|
||||
if c.Seat == link.CLISeat {
|
||||
id = c.ID
|
||||
break
|
||||
}
|
||||
}
|
||||
shown, err := callsAnswer(link.Calls, id)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
said, _ := json.Marshal(shown)
|
||||
if strings.Contains(string(said), "s3cret-join") ||
|
||||
strings.Contains(string(said), base64.StdEncoding.EncodeToString([]byte("s3cret-join"))) {
|
||||
t.Fatalf("calls showed a mesh-cli line's answer: %s", said)
|
||||
}
|
||||
}
|
||||
|
||||
// **Only the terminal's line carries the terminal's mark** (review of ADR 0272): a mark the serving controller's
|
||||
// environment holds — leaked, or set by anything — is stripped from every other command line it runs, a verb's and
|
||||
// an ordinary mesh-cli line alike, so neither reads as the terminal.
|
||||
func TestTheTerminalsMarkIsStrippedFromEveryOtherLine(t *testing.T) {
|
||||
t.Setenv(echoEnvironment, "1")
|
||||
t.Setenv(cliTerminalVar, "1")
|
||||
t.Setenv(servedVar, "1")
|
||||
for _, env := range [][]string{commandEnvironment("someone", "status", false)} {
|
||||
for _, kv := range env {
|
||||
if strings.HasPrefix(kv, cliTerminalVar+"=") {
|
||||
t.Fatalf("a verb's line carries the terminal's mark: %s", kv)
|
||||
}
|
||||
}
|
||||
}
|
||||
a := runForMeshCLI(context.Background(), "laptop", asked("operator", 1000, "status"), cliVerdict{why: "not the terminal"})
|
||||
if got := string(a.Stdout); !strings.Contains(got, "terminal=false") || !strings.Contains(got, `verb="mesh-cli"`) {
|
||||
t.Fatalf("an ordinary line with the mark in the serving environment ran as %s", got)
|
||||
}
|
||||
a = runForMeshCLI(context.Background(), "control", asked("operator", 1000, "status"), cliVerdict{terminal: true})
|
||||
if got := string(a.Stdout); !strings.Contains(got, "terminal=true") {
|
||||
t.Fatalf("the terminal's line ran as %s", got)
|
||||
}
|
||||
}
|
||||
|
||||
// **Only a login session is the terminal** (review of ADR 0272): the operator's account on the control-node, asking
|
||||
// from a service — the tool runner, a user unit — and not a login session, is an ordinary call.
|
||||
func TestOnlyALoginSessionIsTheTerminal(t *testing.T) {
|
||||
control := []string{"control"}
|
||||
v := judgeCLI("control", link.CLIAsked{Line: []string{"status"}, Account: "operator", UID: 1000}, cliNodes, control)
|
||||
if v.terminal || v.refused != "" || !strings.Contains(v.why, "login session") {
|
||||
t.Fatalf("a line from no login session reads %+v", v)
|
||||
}
|
||||
v = judgeCLI("control", link.CLIAsked{Line: []string{"status"}, Account: "operator", UID: 1000, Session: "session-3.scope"},
|
||||
cliNodes, control)
|
||||
if !v.terminal {
|
||||
t.Fatalf("a line from a login session on the control-node reads %+v", v)
|
||||
}
|
||||
}
|
||||
@@ -1107,7 +1107,7 @@ func refuseTerminalSettingsThroughAVerb(ctx context.Context, inv *inventory.Inve
|
||||
}
|
||||
// A trusted mergeable file takes any key, so its module's whole layer is the terminal's (novox/hq issue 340).
|
||||
if files := catalogue.TrustedMergeable(shelf[module]); len(files) > 0 && !sameLayer(before, after) {
|
||||
return fmt.Errorf("the settings of %s on %s are set at the controller's terminal only, never through a verb (this "+
|
||||
return fmt.Errorf("the settings of %s on %s are set at the controller's terminal only (`mesh-cli` on the control-node), never through a verb (this "+
|
||||
"line came through %q): %s merges whatever key a layer sets into a file root or a consumer trusts, so "+
|
||||
"any key could point the module at a listener of the caller's, and whoever may call a verb includes "+
|
||||
"agents (novox/hq issue 340; a file nothing trusts says \"trusted\": false). Nothing was changed",
|
||||
@@ -1119,7 +1119,7 @@ func refuseTerminalSettingsThroughAVerb(ctx context.Context, inv *inventory.Inve
|
||||
if string(was) == string(now) {
|
||||
continue
|
||||
}
|
||||
return fmt.Errorf("%s of %s on %s is set at the controller's terminal only, never through a verb (this "+
|
||||
return fmt.Errorf("%s of %s on %s is set at the controller's terminal only (`mesh-cli` on the control-node), never through a verb (this "+
|
||||
"line came through %q): it says where root creates and owns a module's directories, which of "+
|
||||
"the machine's paths are mounted into its container, what the mesh's consumers trust, or what a file "+
|
||||
"root or a person's session obeys takes, and whoever may call a verb includes agents (novox/hq issue 339; "+
|
||||
|
||||
@@ -748,6 +748,25 @@ func stalledWords(l stalledLine, o conditions.Observation) (headline, explanatio
|
||||
actions []conditions.Action) {
|
||||
repository, _, _ := strings.Cut(l.ID, "@")
|
||||
name := repoName(repository)
|
||||
if l.State == "unannounced" {
|
||||
// A pull request the forge never announced (novox/hq issue 347): no delivery exists, so there is nothing to
|
||||
// stop, release or close; a new commit on its branch is announced and checked.
|
||||
long := "for too long"
|
||||
if d, err := time.ParseDuration(l.For); err == nil {
|
||||
long = "for " + humanDuration(d)
|
||||
}
|
||||
if o.Resolver == conditions.ResolverOperator {
|
||||
needs = "push a new commit to its branch; the forge announces it and the mesh checks it."
|
||||
}
|
||||
pull := "A pull request of " + name
|
||||
if l.Number > 0 {
|
||||
pull = fmt.Sprintf("Pull request %s #%d", name, l.Number)
|
||||
}
|
||||
return fmt.Sprintf("%s has had no merge check %s", pull, long),
|
||||
fmt.Sprintf("%s has been open %s on a branch that requires the merge check, and the mesh was never asked "+
|
||||
"to check it: the forge never announced it. It cannot merge until it is checked.", pull, long),
|
||||
fmt.Sprintf("%s has a merge check now, or is closed", pull), needs, nil
|
||||
}
|
||||
held := l.State
|
||||
if held == "" {
|
||||
held = "held"
|
||||
|
||||
@@ -309,3 +309,24 @@ func TestANeedNoNotificationAnswersNamesTheMeshMCPServer(t *testing.T) {
|
||||
}
|
||||
check("updates held", releaseHeldWords([]string{"openrazer"}, []string{"g14"}))
|
||||
}
|
||||
|
||||
// **A pull request the forge never announced says what to do about it** (novox/hq issue 347): mesh-delivery says one
|
||||
// as a stalled line in state `unannounced` — no delivery exists, so there is nothing to stop, release or close —
|
||||
// and the operator's one act is a new commit on its branch, which the forge announces.
|
||||
func TestAnUnannouncedPullRequestSaysToPushANewCommit(t *testing.T) {
|
||||
l := stalledLine{ID: "novox/hq@bfe82315f42c", Number: 243, State: "unannounced", For: "11m0s", Bound: "10m0s",
|
||||
H2: "none: the forge never announced it, so there is no delivery to close — the operator's",
|
||||
Says: "novox/hq#243 is open on main, which requires the merge check, and its head has had no merge check"}
|
||||
obs := stalledObservations([]stalledLine{l})
|
||||
if len(obs) != 1 || obs[0].Resolver != conditions.ResolverOperator {
|
||||
t.Fatalf("an unannounced pull request is not the operator's: %+v", obs)
|
||||
}
|
||||
o := obs[0]
|
||||
if strings.Contains(o.Needs, "stop") || strings.Contains(o.Needs, "release") || !strings.Contains(o.Needs, "commit") {
|
||||
t.Fatalf("it says to %q", o.Needs)
|
||||
}
|
||||
if !strings.Contains(o.Headline, "no merge check") || !strings.Contains(o.Headline, "#243") ||
|
||||
!strings.Contains(o.Explanation, "never") {
|
||||
t.Fatalf("it reads %q / %q", o.Headline, o.Explanation)
|
||||
}
|
||||
}
|
||||
|
||||
@@ -512,6 +512,15 @@ func sortedKeysOf(m map[string]string) []string {
|
||||
return out
|
||||
}
|
||||
|
||||
// sayPlanDiff is `plan --diff`: what the declaration leaves out, then the diff. A module left out is not in
|
||||
// the body, so the diff alone would say "nothing would change" for a module just assigned whose settings
|
||||
// cannot compose — success-shaped silence. Said first, with why, as push and the plain plan say it
|
||||
// (novox/hq ADR 0163, rule 6).
|
||||
func sayPlanDiff(node string, declared sendable, diff func() error) error {
|
||||
reportLeftOut(node, declared)
|
||||
return diff()
|
||||
}
|
||||
|
||||
// reportLeftOut says which of a machine's modules its declaration leaves out and why (novox/hq ADR
|
||||
// 0163, rule 6), one line each: the machine is told everything else, and is told it was left out.
|
||||
func reportLeftOut(node string, declared sendable) {
|
||||
@@ -1239,7 +1248,9 @@ func planCommand(ctx context.Context, args []string) error {
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
return writePlanDiff(ctx, open.inventory, args[0], body)
|
||||
return sayPlanDiff(args[0], declared, func() error {
|
||||
return writePlanDiff(ctx, open.inventory, args[0], body)
|
||||
})
|
||||
}
|
||||
if *asJSON {
|
||||
declared, err := declarationFor(ctx, open, args[0], plan, settings)
|
||||
|
||||
@@ -0,0 +1,25 @@
|
||||
package main
|
||||
|
||||
import (
|
||||
"os"
|
||||
"strings"
|
||||
"testing"
|
||||
)
|
||||
|
||||
// 2026-10-09: nfs-server was assigned to the home server, its file asked for a setting nothing set, and
|
||||
// `plan --diff` said "nothing would change". The diff now says what is left out, and why, before the diff.
|
||||
func TestPlanDiffSaysAModuleLeftOutBeforeTheDiff(t *testing.T) {
|
||||
declared := sendable{LeftOut: []string{"nfs-server"},
|
||||
leftOutWhy: map[string]string{"nfs-server": `nothing sets "shares" for it`}}
|
||||
said := printed(t, func() error {
|
||||
return sayPlanDiff("home", declared, func() error {
|
||||
writeDiff(os.Stdout, "home", sentDiff{}, nil)
|
||||
return nil
|
||||
})
|
||||
})
|
||||
left := strings.Index(said, "home: nfs-server left out")
|
||||
nothing := strings.Index(said, "home: nothing would change")
|
||||
if left < 0 || !strings.Contains(said, `nothing sets "shares" for it`) || nothing < left {
|
||||
t.Errorf("the left-out module and why, then the diff:\n%s", said)
|
||||
}
|
||||
}
|
||||
@@ -65,6 +65,8 @@ func connectLink(ctx context.Context, inv *inventory.Inventory, enroller link.En
|
||||
}
|
||||
|
||||
func serve(ctx context.Context) (err error) {
|
||||
// Nothing this process does, or starts, is the operator at the terminal (novox/hq ADR 0266).
|
||||
markServed()
|
||||
// The one process whose log is read over time, so the one that says each change to a node's
|
||||
// unmet seat dependencies once (novox/hq ADR 0207).
|
||||
logUnheldChanges = true
|
||||
@@ -264,6 +266,12 @@ func serve(ctx context.Context) (err error) {
|
||||
return err
|
||||
}
|
||||
defer stopServing()
|
||||
// And the operator's command on every node, asked through each node's engine (novox/hq ADR 0272).
|
||||
stopCLI, err := bus.ServeCLI(answerMeshCLI(open.inventory), log.New(os.Stdout, "", log.LstdFlags))
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
defer stopCLI()
|
||||
// And says so on the bus (novox/hq ADR 0197): what it serves, as the NATS services protocol asks.
|
||||
stopAnnouncing, err := bus.Announce(seatAnnouncement(handlers), log.New(os.Stdout, "", log.LstdFlags))
|
||||
if err != nil {
|
||||
|
||||
@@ -26,7 +26,7 @@ func TestAPushSaysWhatItRecreates(t *testing.T) {
|
||||
aContainerBuild(t, "postgres", "c1111111", "", start.Add(time.Second),
|
||||
map[string][2]string{"server": {image("e"), ""}}),
|
||||
} {
|
||||
if _, _, err := takeIn(ctx, inv, b); err != nil {
|
||||
if _, _, err := takeIn(ctx, inv, asTheOperator(t, inv, b)); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
}
|
||||
|
||||
@@ -71,7 +71,7 @@ func TestARecordedBuildIsCarriedOnlyByAPersonsPush(t *testing.T) {
|
||||
aContainerBuild(t, "mailu", "c1111111", "", start.Add(time.Second),
|
||||
map[string][2]string{"smtp": {mailImage, ""}, "imap": {mailImage, ""}}),
|
||||
} {
|
||||
if _, _, err := takeIn(ctx, inv, b); err != nil {
|
||||
if _, _, err := takeIn(ctx, inv, asTheOperator(t, inv, b)); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
}
|
||||
|
||||
@@ -539,8 +539,8 @@ func TestReplay301APersonsPushOfAHeldRecordedBuildIsNoRepair(t *testing.T) {
|
||||
inv := open.inventory
|
||||
start := time.Now().Add(-time.Hour)
|
||||
image := "registry.invalid:5000/resolver/server@sha256:" + strings.Repeat("e", 64)
|
||||
if _, _, err := takeIn(ctx, inv, aContainerBuild(t, "resolver", "c1111111", catalogue.PolicyRecord, start,
|
||||
map[string][2]string{"server": {image, ""}})); err != nil {
|
||||
if _, _, err := takeIn(ctx, inv, asTheOperator(t, inv, aContainerBuild(t, "resolver", "c1111111", catalogue.PolicyRecord, start,
|
||||
map[string][2]string{"server": {image, ""}}))); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
for _, node := range []string{"anchor", "laptop"} {
|
||||
|
||||
@@ -54,7 +54,7 @@ func TestARebuildOfAnUnchangedSourceKeepsItsArtifacts(t *testing.T) {
|
||||
// Another module's merge rebuilt it: a new commit, a new image digest, the same source.
|
||||
anImageBuild(t, "app", "", "c2bbbbbb", strings.Repeat("b", 64), "src1:same", start.Add(time.Minute)),
|
||||
} {
|
||||
if _, _, err := takeIn(ctx, inv, b); err != nil {
|
||||
if _, _, err := takeIn(ctx, inv, asTheOperator(t, inv, b)); err != nil {
|
||||
t.Fatalf("build %d: %v", i, err)
|
||||
}
|
||||
}
|
||||
@@ -114,7 +114,7 @@ func TestABusRebuiltFromAnUnchangedSourceDemandsNoBusStep(t *testing.T) {
|
||||
b.Manifest = manifest
|
||||
return b
|
||||
}
|
||||
if _, _, err := takeIn(ctx, inv, bus("", "n1111111", strings.Repeat("a", 64), "src1:bus", start)); err != nil {
|
||||
if _, _, err := takeIn(ctx, inv, asTheOperator(t, inv, bus("", "n1111111", strings.Repeat("a", 64), "src1:bus", start))); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if _, err := inv.Assign(ctx, "anchor", "nats"); err != nil {
|
||||
|
||||
@@ -26,6 +26,12 @@ import (
|
||||
// shape fails the suite, naming its source. The keeper would say such a summary in words at run time;
|
||||
// this is where the producer is made to say it rightly in the first place.
|
||||
func TestMain(m *testing.M) {
|
||||
// The process a mesh-cli test runs as a command line: it says the verb and the caller it was given, and
|
||||
// ends (meshcli_test.go).
|
||||
if os.Getenv(echoEnvironment) != "" {
|
||||
fmt.Printf("verb=%q caller=%q terminal=%v\n", os.Getenv("MESH_VERB"), os.Getenv("MESH_CALLER"), startedAtTheTerminal())
|
||||
os.Exit(0)
|
||||
}
|
||||
conditions.Unsayable = func(o conditions.Observation, field string, r outward.Refusal) {
|
||||
unsaid.note(o, field, r)
|
||||
}
|
||||
|
||||
@@ -229,6 +229,35 @@ func quoteAll(xs []string) string {
|
||||
return strings.Join(q, ", ")
|
||||
}
|
||||
|
||||
// refusedAsTheGenericCommand is what the generic `command` verb refuses of a command line, and so what every
|
||||
// line asked through a verb's route refuses: the `command` verb's, and an ordinary line from mesh-cli (novox/hq ADR
|
||||
// 0272 §4). One function, so the two routes cannot drift apart.
|
||||
func refusedAsTheGenericCommand(argv []string) error {
|
||||
if len(argv) == 0 {
|
||||
return errors.New("command names no command")
|
||||
}
|
||||
// A layer is written through the settings verb, never the generic one (novox/hq issue 339): the
|
||||
// settings verb is where what a verb may not set is refused, and one route is one set of words.
|
||||
// The command refuses places and accesses through any verb as well; this says so before it runs.
|
||||
if argv[0] == "settings" && slices.ContainsFunc(argv[1:], func(w string) bool { return w == "set" || w == "clear" }) {
|
||||
return &heldAtTheTerminal{msg: "settings are set and cleared through the settings verb, not the " +
|
||||
"generic command; and places and accesses only at the controller's terminal (novox/hq issue 339). " +
|
||||
"Nothing was done"}
|
||||
}
|
||||
// The generic verb only reads (novox/hq ADR 0266): what writes has a named verb that composes its own
|
||||
// line, or is the operator's at the controller's terminal.
|
||||
if err := commandReads(argv); err != nil {
|
||||
return err
|
||||
}
|
||||
// The generic verb is no way round the hand-act log (novox/hq to-be 45 §7): a repair through
|
||||
// it says why, as it would through its own verb.
|
||||
if repair := repairingCommand(argv); repair != "" && !slices.ContainsFunc(argv, isWhyFlag) {
|
||||
return fmt.Errorf("%s is a repair done by hand, and says why: add --why <text> to the command "+
|
||||
"line (recorded in the hand-act log). Nothing was done", repair)
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
// commandLine composes the command. Every argument it reads is one it uses: a branch that reads an
|
||||
// argument and then drops it would pass it over, which is what the check after it exists to refuse.
|
||||
func (a *verbArguments) commandLine() ([]string, error) {
|
||||
@@ -245,28 +274,9 @@ func (a *verbArguments) commandLine() ([]string, error) {
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
if len(argv) == 0 {
|
||||
return nil, errors.New("command names no command")
|
||||
}
|
||||
// A layer is written through the settings verb, never the generic one (novox/hq issue 339): the
|
||||
// settings verb is where what a verb may not set is refused, and one route is one set of words.
|
||||
// The command refuses places and accesses through any verb as well; this says so before it runs.
|
||||
if argv[0] == "settings" && slices.ContainsFunc(argv[1:], func(w string) bool { return w == "set" || w == "clear" }) {
|
||||
return nil, &heldAtTheTerminal{msg: "settings are set and cleared through the settings verb, not the " +
|
||||
"generic command; and places and accesses only at the controller's terminal (novox/hq issue 339). " +
|
||||
"Nothing was done"}
|
||||
}
|
||||
// The generic verb only reads (novox/hq ADR 0266): what writes has a named verb that composes its own
|
||||
// line, or is the operator's at the controller's terminal.
|
||||
if err := commandReads(argv); err != nil {
|
||||
if err := refusedAsTheGenericCommand(argv); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
// The generic verb is no way round the hand-act log (novox/hq to-be 45 §7): a repair through
|
||||
// it says why, as it would through its own verb.
|
||||
if repair := repairingCommand(argv); repair != "" && !slices.ContainsFunc(argv, isWhyFlag) {
|
||||
return nil, fmt.Errorf("%s is a repair done by hand, and says why: add --why <text> to the command "+
|
||||
"line (recorded in the hand-act log). Nothing was done", repair)
|
||||
}
|
||||
return argv, nil
|
||||
case "tools":
|
||||
return nil, errors.New("tools is answered from the records, not by a command")
|
||||
@@ -919,6 +929,30 @@ func isWhyFlag(word string) bool {
|
||||
return word == "--why" || word == "-why" || strings.HasPrefix(word, "--why=") || strings.HasPrefix(word, "-why=")
|
||||
}
|
||||
|
||||
// commandEnvironment is the environment of a command line this controller runs for someone: its own — the
|
||||
// stores' credentials, the bus, the broker, everything a command run from a shell beside it would have, because it
|
||||
// is that — with who asked, and how it came.
|
||||
//
|
||||
// **terminal** is said, never inferred (novox/hq ADR 0272): only a line mesh-cli asked as the controller's terminal
|
||||
// passes true. Its line has no `MESH_VERB`, not the served mark ADR 0266 puts on everything the serving controller
|
||||
// starts, and the terminal's mark (cliTerminalVar), so startedAtTheTerminal reads yes. Every other line names its
|
||||
// verb, and is stripped of the terminal's mark whatever this process's environment holds.
|
||||
func commandEnvironment(caller, verb string, terminal bool) []string {
|
||||
env := make([]string, 0, len(os.Environ())+3)
|
||||
for _, kv := range os.Environ() {
|
||||
if strings.HasPrefix(kv, verbVar+"=") || strings.HasPrefix(kv, link.CallerVar+"=") ||
|
||||
strings.HasPrefix(kv, cliTerminalVar+"=") || (terminal && strings.HasPrefix(kv, servedVar+"=")) {
|
||||
continue
|
||||
}
|
||||
env = append(env, kv)
|
||||
}
|
||||
env = append(env, link.CallerVar+"="+caller)
|
||||
if terminal {
|
||||
return append(env, cliTerminalVar+"=1")
|
||||
}
|
||||
return append(env, verbVar+"="+verb)
|
||||
}
|
||||
|
||||
// runVerb runs this binary with the given command line and gathers what it said.
|
||||
//
|
||||
// **This binary is the image this process runs, never the file at the path it started from**
|
||||
@@ -932,21 +966,17 @@ func runVerb(ctx context.Context, argv []string) (verbAnswer, error) {
|
||||
return verbAnswer{}, fmt.Errorf("%w: this controller is stopping and runs no new command", link.ErrHandingOver)
|
||||
}
|
||||
cmd := selfCommand(ctx, argv)
|
||||
// The same environment: the stores' credentials, the bus, the broker — everything a command run
|
||||
// from a shell in this container would have, because it is that.
|
||||
cmd.Env = os.Environ()
|
||||
// And who asked, so an act it does by hand is recorded as theirs (novox/hq to-be 45 §7).
|
||||
caller := link.CallerIn(ctx)
|
||||
if caller == "" {
|
||||
caller = "a seat call whose caller the bus did not name"
|
||||
}
|
||||
cmd.Env = append(cmd.Env, link.CallerVar+"="+caller+", through the "+catalogue.ControllerSeatName+" seat")
|
||||
// And which verb, so the connection it dials says so in the bus's list (novox/hq issue 327).
|
||||
verb, _ := ctx.Value(verbKey{}).(string)
|
||||
if verb == "" {
|
||||
verb = argv[0]
|
||||
}
|
||||
cmd.Env = append(cmd.Env, verbVar+"="+verb)
|
||||
cmd.Env = commandEnvironment(caller+", through the "+catalogue.ControllerSeatName+" seat", verb, false)
|
||||
// Two buffers, one answer. What the command *says* is both streams, in the order a person at
|
||||
// a shell would read them; what it *answers as data* is standard output alone — `status --json`
|
||||
// prints its warnings beside the document, and a JSON parsed from the two together parsed
|
||||
@@ -1168,7 +1198,7 @@ func answersFirst(argv []string) bool {
|
||||
// the reason said beside it.
|
||||
func callsAnswer(log *link.CallLog, id string) (any, error) {
|
||||
if id != "" {
|
||||
c, ok, err := log.Get(id)
|
||||
c, ok, err := log.Shown(id)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("call %s is not in this controller's memory, and the calls kept on the "+
|
||||
"bus could not be read: %w", id, err)
|
||||
|
||||
@@ -35,4 +35,4 @@ require (
|
||||
// committed. Every build (the build agent's `go build`, the Dockerfile) compiles from vendor/ and
|
||||
// fetches nothing; go refuses to build when vendor/ and this file disagree, so a pin moved without
|
||||
// `go mod vendor` fails loudly, at once, everywhere.
|
||||
replace github.com/novox/mesh-host => git.novox.be/novox/mesh-host v0.0.0-20261009081005-b28d7bbcbff4
|
||||
replace github.com/novox/mesh-host => git.novox.be/novox/mesh-host v0.0.0-20261009103656-1c61b72f354d
|
||||
|
||||
@@ -6,6 +6,10 @@ git.novox.be/novox/mesh-host v0.0.0-20261007162834-56e2ebec4bac h1:yLtFS0pDCCqIE
|
||||
git.novox.be/novox/mesh-host v0.0.0-20261007162834-56e2ebec4bac/go.mod h1:VlilMCRZ5yyNXg7SNigNBLr0Gt32jrGw5KSNq5JAVYs=
|
||||
git.novox.be/novox/mesh-host v0.0.0-20261009081005-b28d7bbcbff4 h1:f4rBnKSemuN0Z9dTtRJMigIGfEs6ltFPOILJGHGab74=
|
||||
git.novox.be/novox/mesh-host v0.0.0-20261009081005-b28d7bbcbff4/go.mod h1:72ZATZjxMLaJfWdvlSDJrygIoBzCmKIjCDMhEXxVzTo=
|
||||
git.novox.be/novox/mesh-host v0.0.0-20261009101157-2673e7a2c95e h1:H7eVqDILL6e9cMbWSLHTbCqu9ZxDOmyeQhUmWl9QBV0=
|
||||
git.novox.be/novox/mesh-host v0.0.0-20261009101157-2673e7a2c95e/go.mod h1:72ZATZjxMLaJfWdvlSDJrygIoBzCmKIjCDMhEXxVzTo=
|
||||
git.novox.be/novox/mesh-host v0.0.0-20261009103656-1c61b72f354d h1:IrmJ+lz21n+eSqKrmXREtR/7raUCBJ+fZvs+BNhuXVI=
|
||||
git.novox.be/novox/mesh-host v0.0.0-20261009103656-1c61b72f354d/go.mod h1:72ZATZjxMLaJfWdvlSDJrygIoBzCmKIjCDMhEXxVzTo=
|
||||
github.com/antithesishq/antithesis-sdk-go v0.7.0-default-no-op h1:Z/MZK75wC/NSrkgqeNIa7jexam9uWzhLmFTSCPI/kn0=
|
||||
github.com/antithesishq/antithesis-sdk-go v0.7.0-default-no-op/go.mod h1:FQyySiasQQM8735Ddel3MRojmy4dA1IqCeyJ5jmPMbI=
|
||||
github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
|
||||
|
||||
@@ -87,8 +87,10 @@ var WritersTable = []WriterRow{
|
||||
{State: "a machine's applied state and its report", Writer: "the node-engine's apply queue",
|
||||
KeptIn: "the machine; the report on the bus", Others: "the reconcile and a delivery enqueue, never apply",
|
||||
// And its health statement between reports (novox/hq ADR 0240): the same writer stating the same
|
||||
// machine, inside the grant it already had (`mesh.control.<its own>.>`).
|
||||
Subjects: []string{"mesh.control.*.report", "mesh.control.*.health"}, Writes: ownMachine},
|
||||
// machine, inside the grant it already had (`mesh.control.<its own>.>`). And a line mesh-cli was given
|
||||
// on the machine (novox/hq ADR 0272): the node is what the controller judges the terminal by, so that
|
||||
// subject is this machine's engine's alone.
|
||||
Subjects: []string{"mesh.control.*.report", "mesh.control.*.health", "mesh.control.*.cli"}, Writes: ownMachine},
|
||||
{State: "the controller lease", Writer: "the controller instance holding it", KeptIn: "key-value " + LeaseBucket,
|
||||
Others: "a candidate waits", Subjects: kvOf(LeaseBucket), Writes: isController},
|
||||
{State: "plans and their tiers", Writer: "controller (lease holder), compare-and-set on the plan's revision",
|
||||
|
||||
@@ -86,6 +86,10 @@ func TestASecondWriterIsRefusedAtComposition(t *testing.T) {
|
||||
[]string{"mesh.control.>"}, "a machine's applied state and its report"},
|
||||
{"a machine publishing another's report", Principal{Kind: KindNode, Node: "one"},
|
||||
[]string{"mesh.control.two.report"}, "a machine's applied state and its report"},
|
||||
{"a machine asking the controller as another (mesh-cli, ADR 0272)", Principal{Kind: KindNode, Node: "one"},
|
||||
[]string{"mesh.control.two.cli"}, "a machine's applied state and its report"},
|
||||
{"a module asking the controller as a machine (mesh-cli, ADR 0272)", Principal{Kind: KindModule, Module: "shop"},
|
||||
[]string{"mesh.control.one.cli"}, "a machine's applied state and its report"},
|
||||
{"a machine publishing every machine's", Principal{Kind: KindNode, Node: "one"},
|
||||
[]string{"mesh.control.*.>"}, "a machine's applied state and its report"},
|
||||
{"a module writing the lease", Principal{Kind: KindModule, Module: "shop"},
|
||||
@@ -115,6 +119,7 @@ func TestASecondWriterIsRefusedAtComposition(t *testing.T) {
|
||||
publish []string
|
||||
}{
|
||||
{Principal{Kind: KindNode, Node: "one"}, []string{"mesh.control.one.>"}},
|
||||
{Principal{Kind: KindNode, Node: "one"}, []string{"mesh.control.one.cli"}},
|
||||
{Principal{Kind: KindController}, []string{"mesh.node.>", "$JS.API.>", "$KV.mesh-controller_lease.>"}},
|
||||
{Principal{Kind: KindModule, Module: "gitea"}, []string{"mesh.mod.gitea.event.pull.merged"}},
|
||||
{Principal{Kind: KindModule, Module: "shop"}, []string{"$JS.API.CONSUMER.CREATE.KV_shop_carts.>"}},
|
||||
|
||||
@@ -169,6 +169,27 @@ func (i *Inventory) RegisterModule(ctx context.Context, m catalogue.Manifest, fr
|
||||
return nil
|
||||
}
|
||||
|
||||
// SourceIdentity is the forge's id of the repository a module is registered from, 0 when none is recorded.
|
||||
func (i *Inventory) SourceIdentity(ctx context.Context, module string) (int64, error) {
|
||||
var id *int64
|
||||
err := i.store.Pool().QueryRow(ctx, `select source_repo_id from module where name = $1`, module).Scan(&id)
|
||||
if errors.Is(err, pgx.ErrNoRows) {
|
||||
return 0, fmt.Errorf("%w: %s", ErrNoSuchModule, module)
|
||||
}
|
||||
if err != nil || id == nil {
|
||||
return 0, err
|
||||
}
|
||||
return *id, nil
|
||||
}
|
||||
|
||||
// SetSourceIdentity records the forge's id of the repository a module is registered from. 0 — the forge could not
|
||||
// say, at the terminal — keeps the id already recorded: an unknown is never a reason to forget what was known.
|
||||
func (i *Inventory) SetSourceIdentity(ctx context.Context, module string, id int64) error {
|
||||
_, err := i.store.Pool().Exec(ctx,
|
||||
`update module set source_repo_id = coalesce(nullif($2::bigint, 0), source_repo_id) where name = $1`, module, id)
|
||||
return err
|
||||
}
|
||||
|
||||
// hasModule is whether the catalogue holds a module of that name.
|
||||
func (i *Inventory) hasModule(ctx context.Context, name string) (bool, error) {
|
||||
var one int
|
||||
|
||||
@@ -7,6 +7,7 @@ import (
|
||||
"encoding/json"
|
||||
"errors"
|
||||
"fmt"
|
||||
"strings"
|
||||
"time"
|
||||
|
||||
"github.com/jackc/pgx/v5"
|
||||
@@ -169,6 +170,12 @@ func (i *Inventory) PreviousBuild(ctx context.Context, module, commit string, fa
|
||||
if b.ID == failed.ID || (commit != "" && b.Commit != commit) {
|
||||
continue
|
||||
}
|
||||
// Only a build of the repository the failed build was made from — the module's, since its take-in
|
||||
// registered it (novox/hq ADR 0266): a build of the module's name from another repository is recorded
|
||||
// and was never registered, and putting it back would register it now.
|
||||
if failed.Repository != "" && !sameRepositoryAs(b.Repository, failed.Repository) {
|
||||
continue
|
||||
}
|
||||
if !failed.AskedOrAt().IsZero() && !b.AskedOrAt().Before(failed.AskedOrAt()) {
|
||||
continue
|
||||
}
|
||||
@@ -190,6 +197,14 @@ func (i *Inventory) PreviousBuild(ctx context.Context, module, commit string, fa
|
||||
return Build{}, false, nil
|
||||
}
|
||||
|
||||
// sameRepositoryAs says two recorded repositories are one, however their case or `.git` is spelled.
|
||||
func sameRepositoryAs(a, b string) bool {
|
||||
trim := func(s string) string {
|
||||
return strings.TrimSuffix(strings.TrimRight(strings.ToLower(strings.TrimSpace(s)), "/"), ".git")
|
||||
}
|
||||
return trim(a) == trim(b)
|
||||
}
|
||||
|
||||
// RestoreModule puts a module's registered build back to an earlier one: its manifest, the commit it
|
||||
// was built from, and when it was asked — as now, so the build that failed its gate, asked before, can
|
||||
// never register over it again (issue 219's order). The source's head is left where the merge moved it:
|
||||
|
||||
@@ -0,0 +1,18 @@
|
||||
-- A build asked at the controller's terminal says so (novox/hq ADR 0266).
|
||||
--
|
||||
-- A build's outcome registers its module, and a module is what the next push sends: a module named `sudo`
|
||||
-- built from a repository an agent made would grant whoever wrote it root on every node it is assigned.
|
||||
-- So an outcome may register a module only from the repository the catalogue already builds it from — or,
|
||||
-- for a module new to the catalogue, from a repository the catalogue already builds another module from.
|
||||
-- Anything else — a module moved to another repository, a new module from a new repository — is the
|
||||
-- operator's, at the controller's terminal. This column is how the take-in tells: true only for a build
|
||||
-- request kept by a `build` or `replay --register` run at the terminal, never through a verb (whoever may
|
||||
-- call a verb includes agents). False for every request kept before this column existed.
|
||||
alter table build_request add column at_terminal boolean not null default false;
|
||||
|
||||
-- And which repository a module is registered from, by the forge's own id for it (novox/hq ADR 0266): a name
|
||||
-- is not an identity. A repository deleted and made again under the same name is another repository, with
|
||||
-- none of the protection the first had until someone sets it; its outcome must not register as the module's.
|
||||
-- Recorded when a build of it is registered from the mesh's own forge; null until then, and for a source the
|
||||
-- forge does not hold.
|
||||
alter table module add column source_repo_id bigint;
|
||||
+9
@@ -0,0 +1,9 @@
|
||||
-- The controller keeps when it first saw an agent account's root verdict waiting for the node-engine's search
|
||||
-- for setuid programs (novox/hq ADR 0266), cleared by the next complete verdict.
|
||||
--
|
||||
-- The self-check does not raise `agent-can-become-root` while that search is within its bound, so a restart is
|
||||
-- not an urgent condition each time. The node-engine's own "since" starts again at every restart: an agent
|
||||
-- that restarted the engine in a loop kept the condition quiet for ever. This time is the controller's, kept
|
||||
-- across the engine's restarts and its own, so the quiet ends once the bound has passed since the search first
|
||||
-- read as pending, however often the engine started again.
|
||||
alter table node add column agent_root_pending_since timestamptz;
|
||||
@@ -1292,3 +1292,24 @@ func (i *Inventory) Sequence(ctx context.Context, id string) (int64, error) {
|
||||
}
|
||||
return *n, nil
|
||||
}
|
||||
|
||||
// RootSearchPending keeps when the controller first saw this node's agent account waiting for the node-engine's
|
||||
// setuid search (novox/hq ADR 0266) and answers it: the first time it is seen since the last complete verdict,
|
||||
// at, kept; seen again, what was kept. Kept across the engine's restarts and the controller's own.
|
||||
func (i *Inventory) RootSearchPending(ctx context.Context, node string, at time.Time) (time.Time, error) {
|
||||
var since time.Time
|
||||
err := i.store.Pool().QueryRow(ctx,
|
||||
`update node set agent_root_pending_since = coalesce(agent_root_pending_since, $2)
|
||||
where name = $1 returning agent_root_pending_since`, node, at).Scan(&since)
|
||||
if errors.Is(err, pgx.ErrNoRows) {
|
||||
return time.Time{}, fmt.Errorf("%w: %s", ErrNoSuchNode, node)
|
||||
}
|
||||
return since, err
|
||||
}
|
||||
|
||||
// RootSearchJudged forgets when a search began pending: a complete verdict came.
|
||||
func (i *Inventory) RootSearchJudged(ctx context.Context, node string) error {
|
||||
_, err := i.store.Pool().Exec(ctx,
|
||||
`update node set agent_root_pending_since = null where name = $1 and agent_root_pending_since is not null`, node)
|
||||
return err
|
||||
}
|
||||
|
||||
@@ -38,6 +38,9 @@ type BuildRequest struct {
|
||||
// unknown. Read as in flight until its outcome or its bound.
|
||||
OutcomeUnknown string
|
||||
At time.Time
|
||||
// AtTerminal says the request was asked at the controller's terminal, never through a verb (novox/hq ADR
|
||||
// 0266): what lets its outcome register a module from a repository the catalogue does not build it from.
|
||||
AtTerminal bool
|
||||
}
|
||||
|
||||
// Name is the module this request is expected to register, read from its directory: the last element of
|
||||
@@ -73,16 +76,31 @@ func (i *Inventory) RecordBuildRequest(ctx context.Context, a BuildRequest) erro
|
||||
notAsked = &a.NotAsked
|
||||
}
|
||||
if _, err := i.store.Pool().Exec(ctx,
|
||||
`insert into build_request (id, repository, seat, source_path, ref, commit_hash, asked_for, not_asked, asked_at)
|
||||
values ($1, $2, $3, $4, $5, $6, $7, $8, $9)
|
||||
`insert into build_request (id, repository, seat, source_path, ref, commit_hash, asked_for, not_asked, asked_at,
|
||||
at_terminal)
|
||||
values ($1, $2, $3, $4, $5, $6, $7, $8, $9, $10)
|
||||
on conflict (id) do nothing`,
|
||||
a.ID, a.Repository, a.Seat, strings.Trim(a.Path, "/"), a.Ref, a.Commit, a.For, notAsked, at); err != nil {
|
||||
a.ID, a.Repository, a.Seat, strings.Trim(a.Path, "/"), a.Ref, a.Commit, a.For, notAsked, at,
|
||||
a.AtTerminal); err != nil {
|
||||
return err
|
||||
}
|
||||
_, err := i.store.Pool().Exec(ctx, `delete from build_request where asked_at < $1`, time.Now().Add(-KeptFor))
|
||||
return err
|
||||
}
|
||||
|
||||
// BuildRequestByID is the build request kept under this id, and whether one was kept.
|
||||
func (i *Inventory) BuildRequestByID(ctx context.Context, id string) (BuildRequest, bool, error) {
|
||||
var a BuildRequest
|
||||
err := i.store.Pool().QueryRow(ctx,
|
||||
`select id, repository, seat, source_path, ref, commit_hash, asked_for, at_terminal, asked_at
|
||||
from build_request where id = $1`, id).Scan(&a.ID, &a.Repository, &a.Seat, &a.Path, &a.Ref, &a.Commit,
|
||||
&a.For, &a.AtTerminal, &a.At)
|
||||
if errors.Is(err, pgx.ErrNoRows) {
|
||||
return BuildRequest{}, false, nil
|
||||
}
|
||||
return a, err == nil, err
|
||||
}
|
||||
|
||||
// MarkNotAsked says a kept build request was never handed over: the words are kept, unless its outcome was
|
||||
// heard first.
|
||||
func (i *Inventory) MarkNotAsked(ctx context.Context, id, why string) error {
|
||||
|
||||
@@ -324,6 +324,15 @@ func (l *CallLog) finish(c *Call, answer []byte, failed, answeredAlready bool) {
|
||||
return
|
||||
}
|
||||
onTheBus := *c
|
||||
if c.Seat == CLISeat {
|
||||
// **A mesh-cli line's answer is its asker's alone** (ADR 0272): what a command at the controller's terminal
|
||||
// printed — a token, a secret's reference — and `calls` answers anyone who may call the seat. Kept in this
|
||||
// process's memory for the asker to follow, and never on the bus.
|
||||
onTheBus.Answer, _ = json.Marshal(map[string]any{"not kept": "a mesh-cli line's answer, its asker's alone: " +
|
||||
"kept in the memory of the controller that ran it, for the asker to follow"})
|
||||
l.keep(onTheBus)
|
||||
return
|
||||
}
|
||||
if len(onTheBus.Answer) > keptOnTheBusAtMost {
|
||||
// A record on the bus is one message too, and one larger than the bus carries is refused whole —
|
||||
// the call's state with it (novox/hq issue 314). The answer stays in this process's memory, and
|
||||
@@ -391,6 +400,29 @@ func (l *CallLog) Recent() ([]Call, error) {
|
||||
return out, nil
|
||||
}
|
||||
|
||||
// inMemory is one call this process served and still holds, whole.
|
||||
func (l *CallLog) inMemory(id string) (Call, bool) {
|
||||
l.mu.Lock()
|
||||
defer l.mu.Unlock()
|
||||
for _, c := range l.calls {
|
||||
if c.ID == id {
|
||||
return *c, true
|
||||
}
|
||||
}
|
||||
return Call{}, false
|
||||
}
|
||||
|
||||
// Shown is one kept call as `calls` shows it: whole, but for a mesh-cli line's answer, which is its asker's alone
|
||||
// and followed by it (ADR 0272) — `calls` answers anyone who may call the seat.
|
||||
func (l *CallLog) Shown(id string) (Call, bool, error) {
|
||||
c, found, err := l.Get(id)
|
||||
if found && c.Seat == CLISeat {
|
||||
c.Answer, _ = json.Marshal(map[string]any{"not shown": "a mesh-cli line's answer is its asker's alone: " +
|
||||
"mesh-cli follows it"})
|
||||
}
|
||||
return c, found, err
|
||||
}
|
||||
|
||||
// Get is one kept call: from memory, or from the bus when this process did not serve it.
|
||||
func (l *CallLog) Get(id string) (Call, bool, error) {
|
||||
l.mu.Lock()
|
||||
@@ -431,6 +463,11 @@ func kept(args json.RawMessage) json.RawMessage {
|
||||
switch {
|
||||
case k == "values" || k == "secret":
|
||||
out[k] = "(given, not kept)"
|
||||
case k == "line":
|
||||
// A mesh-cli line (ADR 0272): its command word, never the rest, which may carry settings.
|
||||
if words, ok := v.([]any); ok && len(words) > 0 {
|
||||
out[k] = []any{words[0], "(the rest given, not kept)"}
|
||||
}
|
||||
case isString && len(s) <= 120:
|
||||
out[k] = s
|
||||
case isString:
|
||||
@@ -513,6 +550,19 @@ var watched struct {
|
||||
conns map[*nats.Conn]bool
|
||||
}
|
||||
|
||||
type callIDKey struct{}
|
||||
|
||||
// WithCallID is ctx carrying the call it serves; CallIDIn reads it back, empty outside one.
|
||||
func WithCallID(ctx context.Context, id string) context.Context {
|
||||
return context.WithValue(ctx, callIDKey{}, id)
|
||||
}
|
||||
|
||||
// CallIDIn is the call ctx serves, or empty.
|
||||
func CallIDIn(ctx context.Context) string {
|
||||
id, _ := ctx.Value(callIDKey{}).(string)
|
||||
return id
|
||||
}
|
||||
|
||||
// answerNow is how a handler asks for its caller to be answered before it goes on (Acknowledge).
|
||||
type answerNowKey struct{}
|
||||
|
||||
@@ -565,6 +615,8 @@ func (l *CallLog) serveCallWithin(seat, verb string, args json.RawMessage, reply
|
||||
c := l.begin(seat, verb, kept(args), reply)
|
||||
// Who asked travels with the call, so an act it does by hand says so (novox/hq to-be 45 §7).
|
||||
ctx = context.WithValue(ctx, callerKey{}, c.Caller)
|
||||
// And which call it is, for what the handler says of it in the journal.
|
||||
ctx = WithCallID(ctx, c.ID)
|
||||
acknowledged := make(chan struct{})
|
||||
var once sync.Once
|
||||
ctx = context.WithValue(ctx, answerNowKey{}, func() { once.Do(func() { close(acknowledged) }) })
|
||||
|
||||
@@ -0,0 +1,220 @@
|
||||
package link
|
||||
|
||||
import (
|
||||
"context"
|
||||
"encoding/json"
|
||||
"fmt"
|
||||
"log"
|
||||
"strings"
|
||||
|
||||
"github.com/nats-io/nats.go"
|
||||
)
|
||||
|
||||
// mesh-cli asks the controller through the node-engine of the machine it runs on (novox/hq ADR 0272 §3). The engine
|
||||
// reads the asking account from the kernel and asks on its own node's subject, which only that node's engine may
|
||||
// publish (its grant is `mesh.control.<node>.>`): so the node is a fact the bus server enforces, and the account a
|
||||
// fact the kernel gave. The engine's side holds the same field names (mesh-host internal/meshcli, a test on each
|
||||
// side), as for its health statement.
|
||||
|
||||
// CLISubjects is every node's mesh-cli subject, which the serving controller answers.
|
||||
const CLISubjects = "mesh.control.*.cli"
|
||||
|
||||
// CLISubject is one node's.
|
||||
func CLISubject(node string) string { return "mesh.control." + node + ".cli" }
|
||||
|
||||
// CLISeat is what a mesh-cli line is recorded under in the calls record, beside the seats' verbs: its verb there is
|
||||
// the node it was asked on.
|
||||
const CLISeat = "mesh-cli"
|
||||
|
||||
// CLIAtOnce bounds the mesh-cli lines running at once, from every node together. A person types one at a time; one
|
||||
// over the bound is answered busy at once, and nothing runs.
|
||||
var CLIAtOnce = 8
|
||||
|
||||
// CLIAsked is a line mesh-cli was given on a node, and the account the node-engine says is asking — or, with
|
||||
// Follow, the call a line runs as, asked again by the same account on the same node until it ends.
|
||||
type CLIAsked struct {
|
||||
Line []string `json:"line,omitempty"`
|
||||
Account string `json:"account"`
|
||||
UID uint32 `json:"uid"`
|
||||
Follow string `json:"follow,omitempty"`
|
||||
// Session is the login session the asking process runs in, as the node-engine read it from the kernel's cgroup
|
||||
// (`session-<id>.scope` under the account's own slice), or empty: a service, a user unit. Only a login session is
|
||||
// the terminal (novox/hq ADR 0272).
|
||||
Session string `json:"session,omitempty"`
|
||||
}
|
||||
|
||||
// CLIAnswer is what the controller answers, as the `result` of a call's answer: what the command printed, how it
|
||||
// exited, whether it ran as the controller's terminal and why, or why nothing ran. A line still running is answered
|
||||
// as every call is (`running`, `call`), and followed.
|
||||
type CLIAnswer struct {
|
||||
Stdout []byte `json:"stdout,omitempty"`
|
||||
Stderr []byte `json:"stderr,omitempty"`
|
||||
Exit int `json:"exit"`
|
||||
Terminal bool `json:"terminal"`
|
||||
Why string `json:"why,omitempty"`
|
||||
Refused string `json:"refused,omitempty"`
|
||||
Cut bool `json:"cut,omitempty"`
|
||||
}
|
||||
|
||||
// CLIRefusal is an answer saying nothing ran, and why.
|
||||
func CLIRefusal(why string) CLIAnswer { return CLIAnswer{Exit: 1, Refused: why} }
|
||||
|
||||
// CLINode is the node a mesh-cli subject names, or false for any other subject.
|
||||
func CLINode(subject string) (string, bool) {
|
||||
rest, ok := strings.CutPrefix(subject, "mesh.control.")
|
||||
if !ok {
|
||||
return "", false
|
||||
}
|
||||
node, tail, ok := strings.Cut(rest, ".")
|
||||
if !ok || tail != "cli" || node == "" {
|
||||
return "", false
|
||||
}
|
||||
return node, true
|
||||
}
|
||||
|
||||
// CLIHandler answers one line from one node.
|
||||
type CLIHandler func(ctx context.Context, node string, asked CLIAsked) CLIAnswer
|
||||
|
||||
// ServeCLI answers mesh-cli for every node until stopped: one queue group, so of two controllers during a handover
|
||||
// one answers.
|
||||
//
|
||||
// **Every line is a call** (review of ADR 0272): run, recorded and answered as a seat's verb is (calls.go) — its
|
||||
// caller answered within AnswerWithin that it is still running, with its call, and the line's answer kept for
|
||||
// the asker to follow. The bus permits an answer for a minute only (broker.ResponseTTL); a line answered when it
|
||||
// ends lost every answer after that, and mesh-cli said nothing ran of a line that had.
|
||||
func (b OverNATS) ServeCLI(handle CLIHandler, logger *log.Logger) (func(), error) {
|
||||
return b.serveCLI(Calls, CLIAtOnce, handle, logger)
|
||||
}
|
||||
|
||||
func (b OverNATS) serveCLI(calls *CallLog, atOnce int, handle CLIHandler, logger *log.Logger) (func(), error) {
|
||||
done := make(chan struct{})
|
||||
slots := make(chan struct{}, atOnce)
|
||||
bind := func() (*nats.Subscription, error) {
|
||||
return b.Conn.QueueSubscribe(CLISubjects, "mesh-cli", func(msg *nats.Msg) {
|
||||
go b.answerCLI(msg, calls, slots, handle, logger)
|
||||
})
|
||||
}
|
||||
sub, err := bind()
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("serving %s: %w", CLISubjects, err)
|
||||
}
|
||||
go keepBound(sub, bind, CLISubjects, done, logger)
|
||||
return func() {
|
||||
close(done)
|
||||
_ = sub.Unsubscribe()
|
||||
}, nil
|
||||
}
|
||||
|
||||
// cliEnvelope is an answer as every call's is: its result.
|
||||
func cliEnvelope(a CLIAnswer) []byte {
|
||||
body, _ := json.Marshal(map[string]any{"result": a})
|
||||
return body
|
||||
}
|
||||
|
||||
func (b OverNATS) answerCLI(msg *nats.Msg, calls *CallLog, slots chan struct{}, handle CLIHandler, logger *log.Logger) {
|
||||
if msg.Reply == "" {
|
||||
return
|
||||
}
|
||||
respond := func(body []byte) {
|
||||
if err := msg.Respond(body); err != nil && logger != nil {
|
||||
logger.Printf("mesh-cli on %s: the answer could not be sent: %v", msg.Subject, err)
|
||||
}
|
||||
}
|
||||
node, ok := CLINode(msg.Subject)
|
||||
var asked CLIAsked
|
||||
switch {
|
||||
case !ok:
|
||||
respond(cliEnvelope(CLIRefusal("not a mesh-cli subject: " + msg.Subject)))
|
||||
return
|
||||
case json.Unmarshal(msg.Data, &asked) != nil:
|
||||
respond(cliEnvelope(CLIRefusal("the node-engine's request could not be read, so nothing ran")))
|
||||
return
|
||||
case asked.Follow != "":
|
||||
respond(calls.followCLI(asked.Follow, node, asked.Account))
|
||||
return
|
||||
case len(asked.Line) == 0:
|
||||
respond(cliEnvelope(CLIRefusal("the request names no command, so nothing ran")))
|
||||
return
|
||||
}
|
||||
select {
|
||||
case slots <- struct{}{}:
|
||||
defer func() { <-slots }()
|
||||
default:
|
||||
respond(cliEnvelope(CLIRefusal(fmt.Sprintf("busy: %d lines from mesh-cli are running already; ask again "+
|
||||
"when one has ended. Nothing ran", cap(slots)))))
|
||||
return
|
||||
}
|
||||
limit := b.Conn.MaxPayload()
|
||||
calls.serveCallWithin(CLISeat, node, msg.Data, msg.Reply, func(ctx context.Context, _ json.RawMessage) (any, error) {
|
||||
return fitCLI(handle(ctx, node, asked), limit-4096), nil
|
||||
}, msg.Respond, limit, logger)
|
||||
}
|
||||
|
||||
// followCLI answers the asker of a line what came of it: running still, its answer, or why that is not known. Only
|
||||
// the account that asked it, on the node it was asked on: the answer is what the command printed, and `calls`
|
||||
// keeps it from everyone else.
|
||||
func (l *CallLog) followCLI(id, node, account string) []byte {
|
||||
noSuch := func() []byte {
|
||||
body, _ := json.Marshal(map[string]any{"error": fmt.Sprintf("no mesh-cli call %s was asked by %s on %s", id,
|
||||
account, node)})
|
||||
return body
|
||||
}
|
||||
c, inMemory := l.inMemory(id)
|
||||
if !inMemory {
|
||||
kept, found, err := l.Get(id)
|
||||
if err != nil || !found {
|
||||
return noSuch()
|
||||
}
|
||||
c = kept
|
||||
}
|
||||
var args CLIAsked
|
||||
_ = json.Unmarshal(c.Args, &args)
|
||||
if c.Seat != CLISeat || c.Verb != node || args.Account != account {
|
||||
return noSuch()
|
||||
}
|
||||
switch {
|
||||
case c.State == CallRunning:
|
||||
return running(&c, AnswerWithin, false, l.Follow)
|
||||
case c.State == CallAbandoned:
|
||||
body, _ := json.Marshal(map[string]any{"error": fmt.Sprintf("call %s was running under a controller that "+
|
||||
"stopped before it finished: it may have done part of what it was asked, and nothing will finish it", id)})
|
||||
return body
|
||||
case !inMemory:
|
||||
body, _ := json.Marshal(map[string]any{"error": fmt.Sprintf("call %s finished (%s), and its answer was kept "+
|
||||
"only in the memory of the controller that ran it, which has stopped; whether it took effect, the "+
|
||||
"mesh says (status, the hand-act log)", id, c.State)})
|
||||
return body
|
||||
}
|
||||
return c.Answer
|
||||
}
|
||||
|
||||
// fitCLI is an answer whose streams fit in limit bytes once written: what the command printed is cut, standard
|
||||
// output first, and the cut is said (ADR 0272 §5) — never silently short.
|
||||
func fitCLI(a CLIAnswer, limit int64) CLIAnswer {
|
||||
body, _ := json.Marshal(a)
|
||||
if limit <= 0 || int64(len(body)) <= limit {
|
||||
return a
|
||||
}
|
||||
a.Cut = true
|
||||
// Room for the rest of the answer and JSON's base64 of the streams (4 bytes for every 3).
|
||||
room := (limit - 4096) * 3 / 4
|
||||
if room < 0 {
|
||||
room = 0
|
||||
}
|
||||
if int64(len(a.Stderr)) > room/2 {
|
||||
a.Stderr = a.Stderr[:room/2]
|
||||
}
|
||||
if left := room - int64(len(a.Stderr)); int64(len(a.Stdout)) > left {
|
||||
if left < 0 {
|
||||
left = 0
|
||||
}
|
||||
a.Stdout = a.Stdout[:left]
|
||||
}
|
||||
return a
|
||||
}
|
||||
|
||||
// FitCLIAnswer is the answer as one bus message of at most limit bytes, written.
|
||||
func FitCLIAnswer(a CLIAnswer, limit int64) []byte {
|
||||
body, _ := json.Marshal(fitCLI(a, limit))
|
||||
return body
|
||||
}
|
||||
@@ -0,0 +1,269 @@
|
||||
package link
|
||||
|
||||
import (
|
||||
"context"
|
||||
"encoding/base64"
|
||||
"encoding/json"
|
||||
"sort"
|
||||
"strings"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/nats-io/nats.go"
|
||||
|
||||
"github.com/novox/mesh-controller/internal/testbus"
|
||||
)
|
||||
|
||||
// The node-engine's request and the answer it hands mesh-cli hold these field names; the engine's side holds the
|
||||
// same list (mesh-host internal/meshcli, TestTheRequestToTheControllerKeepsItsFieldNames).
|
||||
func TestTheMeshCLIRequestAndAnswerKeepTheirFieldNames(t *testing.T) {
|
||||
body, _ := json.Marshal(CLIAsked{Line: []string{"status"}, Account: "a", UID: 1, Session: "session-3.scope"})
|
||||
if got := keysIn(t, body); got != "account line session uid" {
|
||||
t.Fatalf("the request's fields are %q", got)
|
||||
}
|
||||
body, _ = json.Marshal(CLIAsked{Line: []string{"status"}, Account: "a", UID: 1})
|
||||
if got := keysIn(t, body); got != "account line uid" {
|
||||
t.Fatalf("the request's fields are %q", got)
|
||||
}
|
||||
body, _ = json.Marshal(CLIAnswer{Stdout: []byte("o"), Stderr: []byte("e"), Exit: 1, Terminal: true, Why: "w",
|
||||
Refused: "r", Cut: true})
|
||||
if got := keysIn(t, body); got != "cut exit refused stderr stdout terminal why" {
|
||||
t.Fatalf("the answer's fields are %q", got)
|
||||
}
|
||||
body, _ = json.Marshal(CLIAsked{Follow: "call-1", Account: "a", UID: 1})
|
||||
if got := keysIn(t, body); got != "account follow uid" {
|
||||
t.Fatalf("a follow's fields are %q", got)
|
||||
}
|
||||
// What a line still running answers, in the envelope every call's answer is in: `result`, then these.
|
||||
var env struct {
|
||||
Result json.RawMessage `json:"result"`
|
||||
}
|
||||
_ = json.Unmarshal(running(&Call{ID: "call-1", Seat: CLISeat, Verb: "a"}, AnswerWithin, false, ""), &env)
|
||||
if got := keysIn(t, env.Result); got != "call output running started" {
|
||||
t.Fatalf("a running answer's fields are %q", got)
|
||||
}
|
||||
}
|
||||
|
||||
func keysIn(t *testing.T, body []byte) string {
|
||||
t.Helper()
|
||||
var m map[string]any
|
||||
if err := json.Unmarshal(body, &m); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
var keys []string
|
||||
for k := range m {
|
||||
keys = append(keys, k)
|
||||
}
|
||||
sort.Strings(keys)
|
||||
return strings.Join(keys, " ")
|
||||
}
|
||||
|
||||
// A node's mesh-cli subject names the node, and no other subject does.
|
||||
func TestTheMeshCLISubjectNamesItsNode(t *testing.T) {
|
||||
if node, ok := CLINode(CLISubject("laptop")); !ok || node != "laptop" {
|
||||
t.Fatalf("CLINode(%q) = %q %v", CLISubject("laptop"), node, ok)
|
||||
}
|
||||
for _, s := range []string{"mesh.control.laptop.report", "mesh.control..cli", "mesh.control.a.b.cli", "mesh.node.a.cli"} {
|
||||
if _, ok := CLINode(s); ok {
|
||||
t.Fatalf("%s was read as a mesh-cli subject", s)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// An answer larger than one bus message is cut to fit, and the cut is said.
|
||||
func TestALargeMeshCLIAnswerIsCutAndSaysSo(t *testing.T) {
|
||||
a := CLIAnswer{Stdout: []byte(strings.Repeat("o", 200000)), Stderr: []byte(strings.Repeat("e", 1000)), Why: "the terminal"}
|
||||
body := FitCLIAnswer(a, 64<<10)
|
||||
if len(body) > 64<<10 {
|
||||
t.Fatalf("the answer is %d bytes, over the bound", len(body))
|
||||
}
|
||||
var got CLIAnswer
|
||||
if err := json.Unmarshal(body, &got); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if !got.Cut || got.Why != "the terminal" || len(got.Stdout) == 0 || string(got.Stderr) != strings.Repeat("e", 1000) {
|
||||
t.Fatalf("the cut answer is %+v", got)
|
||||
}
|
||||
if small := FitCLIAnswer(CLIAnswer{Stdout: []byte("ok")}, 64<<10); strings.Contains(string(small), `"cut"`) {
|
||||
t.Fatal("an answer that fits was said to be cut")
|
||||
}
|
||||
}
|
||||
|
||||
// cliBus serves mesh-cli on a bus of the test's own with a call log of its own, and returns a connection to ask on.
|
||||
func cliBus(t *testing.T, l *CallLog, atOnce int, handle CLIHandler) *nats.Conn {
|
||||
t.Helper()
|
||||
conn, err := nats.Connect(testbus.URL(t))
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
t.Cleanup(conn.Close)
|
||||
stop, err := OverNATS{Conn: conn}.serveCLI(l, atOnce, handle, nil)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
t.Cleanup(stop)
|
||||
return conn
|
||||
}
|
||||
|
||||
// askCLI asks one request on a node's subject and reads the envelope.
|
||||
func askCLI(t *testing.T, conn *nats.Conn, node string, asked CLIAsked) (CLIAnswer, map[string]any, string) {
|
||||
t.Helper()
|
||||
body, _ := json.Marshal(asked)
|
||||
msg, err := conn.Request(CLISubject(node), body, 5*time.Second)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
var env struct {
|
||||
Result json.RawMessage `json:"result"`
|
||||
Error string `json:"error"`
|
||||
}
|
||||
if err := json.Unmarshal(msg.Data, &env); err != nil {
|
||||
t.Fatalf("not an envelope: %s", msg.Data)
|
||||
}
|
||||
var a CLIAnswer
|
||||
var raw map[string]any
|
||||
_ = json.Unmarshal(env.Result, &a)
|
||||
_ = json.Unmarshal(env.Result, &raw)
|
||||
return a, raw, env.Error
|
||||
}
|
||||
|
||||
// **A line that outlasts the bus's window for an answer is followed to its answer, never lost** (review of ADR
|
||||
// 0272): the first answer says it is running and names its call, and following the call by the same account on
|
||||
// the same node gives what the line printed once it ends. Another account, or another node, is told no such call.
|
||||
func TestALongMeshCLILineIsFollowedToItsAnswer(t *testing.T) {
|
||||
was := AnswerWithin
|
||||
AnswerWithin = 50 * time.Millisecond
|
||||
t.Cleanup(func() { AnswerWithin = was })
|
||||
release := make(chan struct{})
|
||||
conn := cliBus(t, NewCallLog(), 4, func(ctx context.Context, node string, asked CLIAsked) CLIAnswer {
|
||||
<-release
|
||||
return CLIAnswer{Stdout: []byte("done on " + node), Terminal: true}
|
||||
})
|
||||
_, first, failed := askCLI(t, conn, "laptop", CLIAsked{Line: []string{"push", "laptop"}, Account: "op", UID: 1000})
|
||||
call, _ := first["call"].(string)
|
||||
if failed != "" || first["running"] != true || call == "" {
|
||||
t.Fatalf("a long line was not answered as running with its call: %v %q", first, failed)
|
||||
}
|
||||
_, still, _ := askCLI(t, conn, "laptop", CLIAsked{Follow: call, Account: "op", UID: 1000})
|
||||
if still["running"] != true {
|
||||
t.Fatalf("following a running line answered %v", still)
|
||||
}
|
||||
close(release)
|
||||
deadline := time.Now().Add(5 * time.Second)
|
||||
for {
|
||||
a, raw, failed := askCLI(t, conn, "laptop", CLIAsked{Follow: call, Account: "op", UID: 1000})
|
||||
if failed != "" {
|
||||
t.Fatalf("following answered %q", failed)
|
||||
}
|
||||
if raw["running"] != true {
|
||||
if string(a.Stdout) != "done on laptop" || !a.Terminal {
|
||||
t.Fatalf("followed to %+v", a)
|
||||
}
|
||||
break
|
||||
}
|
||||
if time.Now().After(deadline) {
|
||||
t.Fatal("the line never finished for its follower")
|
||||
}
|
||||
time.Sleep(20 * time.Millisecond)
|
||||
}
|
||||
if _, _, failed := askCLI(t, conn, "laptop", CLIAsked{Follow: call, Account: "agent", UID: 1001}); failed == "" {
|
||||
t.Fatal("another account followed the operator's line")
|
||||
}
|
||||
if _, _, failed := askCLI(t, conn, "desktop", CLIAsked{Follow: call, Account: "op", UID: 1000}); failed == "" {
|
||||
t.Fatal("another node followed the line")
|
||||
}
|
||||
}
|
||||
|
||||
// Lines running at once are bounded: one over the bound is answered busy at once, and nothing ran.
|
||||
func TestMeshCLILinesAtOnceAreBounded(t *testing.T) {
|
||||
was := AnswerWithin
|
||||
AnswerWithin = 50 * time.Millisecond
|
||||
t.Cleanup(func() { AnswerWithin = was })
|
||||
release := make(chan struct{})
|
||||
t.Cleanup(func() { close(release) })
|
||||
ran := make(chan struct{}, 4)
|
||||
conn := cliBus(t, NewCallLog(), 1, func(context.Context, string, CLIAsked) CLIAnswer {
|
||||
ran <- struct{}{}
|
||||
<-release
|
||||
return CLIAnswer{}
|
||||
})
|
||||
if _, first, _ := askCLI(t, conn, "laptop", CLIAsked{Line: []string{"status"}, Account: "op"}); first["running"] != true {
|
||||
t.Fatalf("the first line answered %v", first)
|
||||
}
|
||||
a, _, _ := askCLI(t, conn, "laptop", CLIAsked{Line: []string{"status"}, Account: "op"})
|
||||
if !strings.Contains(a.Refused, "busy") || a.Exit == 0 {
|
||||
t.Fatalf("a line over the bound answered %+v", a)
|
||||
}
|
||||
if len(ran) != 1 {
|
||||
t.Fatalf("%d lines ran, one was bound", len(ran))
|
||||
}
|
||||
}
|
||||
|
||||
// A mesh-cli line's record keeps its first word, never the rest of its line, and its answer is never kept on the
|
||||
// bus, where `calls` answers anyone who may call the seat.
|
||||
func TestAMeshCLIRecordKeepsNeitherItsLineNorItsAnswerOnTheBus(t *testing.T) {
|
||||
l, a := NewCallLog(), newAnswers(t)
|
||||
writes := make(chan Call, 4)
|
||||
l.writes = writes
|
||||
asked, _ := json.Marshal(CLIAsked{Line: []string{"settings", "set", "x", `{"password":"s3cret"}`}, Account: "op"})
|
||||
l.serveCall(CLISeat, "laptop", asked, "_INBOX.node.laptop.abcdefghijklmnopqrstuv",
|
||||
func(context.Context, json.RawMessage) (any, error) {
|
||||
return CLIAnswer{Stdout: []byte("s3cret-join")}, nil
|
||||
},
|
||||
a.respond, nil)
|
||||
_ = a.only()
|
||||
close(writes)
|
||||
for c := range writes {
|
||||
if carries(c.Args, "s3cret") || carries(c.Answer, "s3cret-join") {
|
||||
t.Fatalf("the bus was sent %s / %s", c.Args, c.Answer)
|
||||
}
|
||||
if !strings.Contains(string(c.Args), "settings") || !strings.Contains(string(c.Args), `"op"`) {
|
||||
t.Fatalf("the record does not say what was asked and by whom: %s", c.Args)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// `calls` answers anyone who may call the seat, agents among them: a mesh-cli line's answer is never shown there,
|
||||
// even from the memory of the controller that ran it; every other call's is (review of ADR 0272).
|
||||
func TestCallsNeverShowsAMeshCLILinesAnswer(t *testing.T) {
|
||||
l, a := NewCallLog(), newAnswers(t)
|
||||
asked, _ := json.Marshal(CLIAsked{Line: []string{"token", "issue", "x"}, Account: "operator"})
|
||||
l.serveCall(CLISeat, "control", asked, "_INBOX.node.control.abcdefghijklmnopqrstuv",
|
||||
func(context.Context, json.RawMessage) (any, error) {
|
||||
return CLIAnswer{Stdout: []byte("s3cret-join")}, nil
|
||||
},
|
||||
a.respond, nil)
|
||||
_ = a.only()
|
||||
recent, _ := l.Recent()
|
||||
shown, found, err := l.Shown(recent[0].ID)
|
||||
if err != nil || !found {
|
||||
t.Fatalf("the line is not shown at all: %v %v", found, err)
|
||||
}
|
||||
if carries(shown.Answer, "s3cret-join") {
|
||||
t.Fatalf("calls shows a mesh-cli line's answer: %s", shown.Answer)
|
||||
}
|
||||
b := newAnswers(t)
|
||||
l.serveCall("mesh-controller", "status", nil, "_INBOX.x.abcdefghijklmnopqrstuv",
|
||||
func(context.Context, json.RawMessage) (any, error) { return "all well", nil }, b.respond, nil)
|
||||
_ = b.only()
|
||||
recent, _ = l.Recent()
|
||||
if shown, _, _ := l.Shown(recent[0].ID); !strings.Contains(string(shown.Answer), "all well") {
|
||||
t.Fatalf("another call's answer is withheld: %s", shown.Answer)
|
||||
}
|
||||
}
|
||||
|
||||
// carries says whether a record holds a secret as text or as the base64 JSON writes bytes in: a line's output is
|
||||
// bytes, so a search for its text alone finds nothing whatever the record keeps (review of ADR 0272).
|
||||
func carries(body []byte, secret string) bool {
|
||||
return strings.Contains(string(body), secret) ||
|
||||
strings.Contains(string(body), base64.StdEncoding.EncodeToString([]byte(secret)))
|
||||
}
|
||||
|
||||
// The two tests above hold what they claim: each fails when the protection it names is taken away.
|
||||
func TestTheWithholdingTestsHoldSomething(t *testing.T) {
|
||||
// The answer as a mesh-cli line's record would carry it, were it kept: the search finds it.
|
||||
body, _ := json.Marshal(map[string]any{"result": CLIAnswer{Stdout: []byte("s3cret-join")}})
|
||||
if !carries(body, "s3cret-join") {
|
||||
t.Fatalf("a record carrying the answer is not found carrying it: %s", body)
|
||||
}
|
||||
}
|
||||
@@ -8,6 +8,7 @@ package link
|
||||
|
||||
import (
|
||||
"encoding/base64"
|
||||
"github.com/novox/mesh-host/rootsearch"
|
||||
"strconv"
|
||||
"strings"
|
||||
"time"
|
||||
@@ -430,6 +431,15 @@ const RootContract = 3
|
||||
// own words (mesh-host internal/accounts ReasonRoot).
|
||||
const ReasonRoot = "can become root without a person"
|
||||
|
||||
// ReasonRootPending starts the reason of an account verdict the node-engine cannot give yet because its search
|
||||
// for setuid programs, started when the engine started, has not finished: not judged yet, said as such, never a
|
||||
// pass. The node-engine's own words (mesh-host rootsearch.ReasonPending), read from it rather than copied.
|
||||
const ReasonRootPending = rootsearch.ReasonPending
|
||||
|
||||
// RootSearchBound is the longest the node-engine lets one search for setuid programs run (mesh-host
|
||||
// rootsearch.Bound): the one value both read.
|
||||
const RootSearchBound = rootsearch.Bound
|
||||
|
||||
// RootNever is the value of a user's `root`, and of a verdict's Root, that the account must never become
|
||||
// root without a person (ADR 0266).
|
||||
const RootNever = "never"
|
||||
|
||||
+19
@@ -0,0 +1,19 @@
|
||||
// Package rootsearch holds what the node-engine's search for setuid programs and the controller that reads its
|
||||
// verdicts must agree on (novox/hq ADR 0266): how long one search may run, and the words a verdict starts with
|
||||
// while it has no answer. Public, so the controller imports these rather than keeps copies that could drift.
|
||||
package rootsearch
|
||||
|
||||
import "time"
|
||||
|
||||
// Bound is how long one search for setuid programs may run. It governs the whole search, the walk and the
|
||||
// package manager's answers together; the controller does not raise an agent account as not judged while the
|
||||
// first search after a start is within it.
|
||||
const Bound = 15 * time.Minute
|
||||
|
||||
// The reasons of a root verdict the search could not answer yet.
|
||||
const (
|
||||
// ReasonPending starts the reason while the first search since the engine started runs.
|
||||
ReasonPending = "not judged yet (search running)"
|
||||
// ReasonIncomplete starts the reason when no search has finished: one failed or ran out its bound.
|
||||
ReasonIncomplete = "not judged (search incomplete)"
|
||||
)
|
||||
Vendored
+3
-2
@@ -75,9 +75,10 @@ github.com/nats-io/nkeys
|
||||
# github.com/nats-io/nuid v1.0.1
|
||||
## explicit
|
||||
github.com/nats-io/nuid
|
||||
# github.com/novox/mesh-host v0.0.0 => git.novox.be/novox/mesh-host v0.0.0-20261009081005-b28d7bbcbff4
|
||||
# github.com/novox/mesh-host v0.0.0 => git.novox.be/novox/mesh-host v0.0.0-20261009103656-1c61b72f354d
|
||||
## explicit; go 1.26.0
|
||||
github.com/novox/mesh-host/internal/declaration
|
||||
github.com/novox/mesh-host/rootsearch
|
||||
github.com/novox/mesh-host/validate
|
||||
# go.uber.org/automaxprocs v1.6.0
|
||||
## explicit; go 1.20
|
||||
@@ -133,4 +134,4 @@ golang.org/x/text/width
|
||||
# golang.org/x/time v0.15.0
|
||||
## explicit; go 1.25.0
|
||||
golang.org/x/time/rate
|
||||
# github.com/novox/mesh-host => git.novox.be/novox/mesh-host v0.0.0-20261009081005-b28d7bbcbff4
|
||||
# github.com/novox/mesh-host => git.novox.be/novox/mesh-host v0.0.0-20261009103656-1c61b72f354d
|
||||
|
||||
Reference in New Issue
Block a user