Compare commits

..
Author SHA1 Message Date
jschoubben 858b06a94e Take mesh-sdk at c8e3509, which counts an item's text as a placement of a config value (issue 1658)
mesh/repo-check head 858b06a94e69 merged into trunk 12752911b644 — fail: head 858b06a94e69 does not merge cleanly into trunk 12752911b644: in conflict go.m…
mesh/delivery-group group fix/1656-xterm-config rejected: a member's own check failed
mesh/merge-gate group fix/1656-xterm-config: composed again for the group's new heads
mesh/delivery superseded: a newer head of the same pull request
2026-10-21 12:55:08 +02:00
2154 changed files with 700583 additions and 26869 deletions
+1 -3
View File
@@ -2,11 +2,9 @@
# A built binary. The lab writes one here when pointed at the repository root by mistake;
# built artifacts belong in build/, which is already ignored.
/mesh-builder
/mesh-controller
/postgres-provisioner
/redis-provisioner
# And where `go build` in cmd/mesh-controller leaves it (novox/hq issue 233).
/cmd/mesh-controller/mesh-controller
# No repository commits its dependencies (issue 1305).
vendor/
+5 -3
View File
@@ -21,15 +21,17 @@ ARG GO_BASE=golang@sha256:8ac98ca534ac3f51e1f420a1dd2c15e74c75cfa0f23f3ad27eb5d7
FROM ${GO_BASE} AS build
WORKDIR /src
# **No dependency is committed** (issue 1305): go.sum names each one's version and checksum, and the build
# resolves them with -mod=mod. Dependencies first, so a change to the source does not re-copy them.
# **Nothing is fetched** (novox/hq to-be 45 Phase 1): every dependency is in vendor/, committed, so
# the image builds from this repository alone — the host's validator among them, whose module no
# public proxy is asked for. Dependencies first, so a change to the source does not re-copy them.
COPY go.mod go.sum ./
COPY vendor/ vendor/
# `COPY . .` brings `.git` along, so Go is told not to stamp the commit: the version is said by -X, and a
# rebuild of an unchanged program is the same bytes (novox/hq ADR 0267 rule 6, issue 363).
COPY . .
ARG VERSION=development
RUN CGO_ENABLED=0 GOFLAGS=-mod=mod go build -trimpath -buildvcs=false \
RUN CGO_ENABLED=0 GOFLAGS=-mod=vendor GOPROXY=off go build -trimpath -buildvcs=false \
-ldflags "-s -w -X main.version=${VERSION}" \
-o /mesh-controller ./cmd/mesh-controller
+44 -3
View File
@@ -46,6 +46,17 @@ image:
@echo
@docker image inspect $(IMAGE) --format 'built {{.RepoTags}} {{.Size}} bytes'
# The builder ships as an image too, because it is a module the mesh assigns rather than a program
# somebody starts on a machine by hand.
BUILDER_IMAGE ?= mesh-builder:$(VERSION)
BUILDER_DEV_TAG ?= mesh-builder:development
builder-image:
@test -n "$(GO_BASE)" || { echo "no GO_BASE; pass GO_BASE=<image>"; exit 1; }
docker build --build-arg GO_BASE=$(GO_BASE) -f cmd/mesh-builder/Dockerfile -t $(BUILDER_IMAGE) -t $(BUILDER_DEV_TAG) .
@echo
@docker image inspect $(BUILDER_IMAGE) --format 'built {{.RepoTags}} {{.Size}} bytes'
# The provisioner ships as an image too, because it is the thing that makes a sealed credential
# true on a machine -- and the mesh cannot, having discarded the plaintext.
PROVISIONER_IMAGE ?= mesh-provision-postgres:$(VERSION)
@@ -79,6 +90,35 @@ redis-provisioner-image:
@echo
@docker image inspect $(REDIS_PROVISIONER_IMAGE) --format 'built {{.RepoTags}} {{.Size}} bytes'
# The proxy that turns a route grant into traffic reaching a workload.
PROXY_IMAGE ?= mesh-route-proxy:$(VERSION)
PROXY_DEV_TAG ?= mesh-route-proxy:development
proxy-image:
docker build --build-arg GO_BASE=$(GO_BASE) -f modules/route-proxy/Dockerfile -t $(PROXY_IMAGE) -t $(PROXY_DEV_TAG) .
@echo
@docker image inspect $(PROXY_IMAGE) --format 'built {{.RepoTags}} {{.Size}} bytes'
# The whole check. Raises a database, runs the repository's own check against it -- merge-check.sh,
# the very script `mesh/repo-check` runs -- and takes the database down again, including when the
# check fails, which is why the teardown is not conditional and the script's exit status is the
# target's.
#
# **It calls the script rather than restating it** (novox/hq issue 431): this target used to list its
# own steps, its formatting step walked vendor/ where the script's does not, and it failed on
# third-party code no change could fix -- so a developer's check and the gate disagreed, and the steps
# after formatting never ran through it. The script is the one list of steps; this target only gives it
# a database (MESH_TEST_POSTGRES, exported above).
#
# **Two levels** (issue 895): the unit level, `go test ./...` untagged, timed and held to a bound; and the integration
# level, `-tags integration`, the tests that start a bus, a store or another process. Each is said with its time.
#
# **Packages in parallel, under the race detector, each test on a bus of its own** (internal/testbus),
# as the script runs them: a bus per test, of the release the mesh runs, makes the suite the same in any
# order, and the timeout bounds a hang to a failure with a stack, never a stalled gate.
check: postgres
@sh merge-check.sh ; status=$$? ; docker rm -f $(PG_CONTAINER) >/dev/null 2>&1 || true ; exit $$status
# The unit level alone (untagged: no bus, no store, no other process), which a developer runs continuously. The tests
# that need a bus or the store carry `//go:build integration` and run in `check`, not here; without a database the
# store's tests would skip rather than fail anyway.
@@ -88,13 +128,14 @@ test:
vet:
go vet ./...
# A quick step for a developer: the checker runs the build module's own formatting and vet (ADR 0501 §5, D7).
# Quick steps for a developer, not part of `check`. The directories gofmt reads must be the ones
# merge-check.sh lists, never `.`, which walks vendor/ (novox/hq issue 431).
fmt:
@unformatted=$$(gofmt -l cmd internal examples) ; \
if [ -n "$$unformatted" ] ; then echo "not gofmt'd:" ; echo "$$unformatted" ; exit 1 ; fi
postgres:
@docker rm -f -v $(PG_CONTAINER) >/dev/null 2>&1 || true
@docker rm -f $(PG_CONTAINER) >/dev/null 2>&1 || true
@docker run -d --name $(PG_CONTAINER) -e POSTGRES_PASSWORD=check \
-p 127.0.0.1:$(PG_PORT):5432 $(PG_IMAGE) >/dev/null
@printf 'waiting for postgres'
@@ -106,7 +147,7 @@ postgres:
echo ' — never came up' ; docker logs $(PG_CONTAINER) | tail -20 ; exit 1
postgres-stop:
@docker rm -f -v $(PG_CONTAINER) >/dev/null 2>&1 || true
@docker rm -f $(PG_CONTAINER) >/dev/null 2>&1 || true
clean:
rm -rf build/
+6
View File
@@ -0,0 +1,6 @@
# Release 57: config and provisioning
This branch carries the work of hq design 57 (ADR 0348, ADR 0349) in this repository. Each package
merges here through its own reviewed pull request; the branch reaches main as one pull request, in one
delivery group with the branch of the same name in mesh-controller, mesh-host, mesh-sdk, mesh-catalog
and hq, after a rehearsal in the lab. This file is deleted by the last commit before that pull request.
+29
View File
@@ -0,0 +1,29 @@
#!/bin/sh
# The unit tests of every Go module of its own below this repository's root, run from the root, for merge-check.sh
# and for the test that tries it on a tree (merge_check_nested_test.go). Its own file so that the test needs no
# particular merge-check.sh: the check that judges a change to merge-check.sh is main's.
set -eu
cgo="${cgo:-0}"
race="${race:-}"
# nested_modules vets and tests every Go module of its own below the root, `modules/<name>/go.mod` (novox/hq ADR 0420
# §5, issue 1025): `./...` stops at a nested go.mod, so a module that lives whole in this repository would have its
# unit tests run by nobody. None is a pass. Each runs in its own directory, outside the vendor directory this
# repository's own packages are built from; a nested module vendors what the forge holds (its own `vendor/`, so the
# check runs offline), as this repository's root module does (issue 1019). A failing module fails the check; the others are still run.
nested_modules() {
nested=0
for gomod in modules/*/go.mod; do
[ -e "$gomod" ] || continue
dir=$(dirname "$gomod")
echo "nested module $dir:"
if ! (cd "$dir" && GOFLAGS= CGO_ENABLED=0 go vet ./... &&
GOFLAGS= CGO_ENABLED="${cgo:-0}" go test ${race:-} -count=1 -timeout 30m ./...); then
echo "nested module $dir FAILED"
nested=1
fi
done
return $nested
}
nested_modules
+22
View File
@@ -0,0 +1,22 @@
ARG ALPINE_BASE=alpine:3
ARG GO_BASE=golang:1.25-alpine
# The builder, as a module ships one.
#
# Not FROM scratch, unlike the control plane: this one runs git and a container client, so it
# needs a filesystem with them on it. That is the honest cost of a machine whose job is to build —
# and it is why building is a MODULE on a machine that has a runtime rather than something the
# control plane does (novox/hq ADR 0005).
FROM ${GO_BASE} AS build
WORKDIR /src
COPY go.mod go.sum ./
RUN go mod download
COPY . .
RUN CGO_ENABLED=0 go build -trimpath -buildvcs=false -ldflags '-s -w' -o /mesh-builder ./cmd/mesh-builder
FROM ${ALPINE_BASE}
# git to clone what it is asked to build, and the docker client to build and push it. The daemon
# is the machine's, reached through its socket — a build machine shares the runtime it was given
# rather than running one inside itself.
RUN apk add --no-cache git docker-cli
COPY --from=build /mesh-builder /usr/local/bin/mesh-builder
ENTRYPOINT ["/usr/local/bin/mesh-builder"]
+146
View File
@@ -0,0 +1,146 @@
package main
import (
"errors"
"fmt"
"os"
"path/filepath"
"strings"
"time"
)
// **The build agent says whether it is busy, and being asked drains it** (novox/hq ADR 0335 §1, §5). Its
// container declares `busy` as `mesh-builder busy`: the node-engine runs it inside the container before it
// replaces it. It exits 0 while a build is in hand and 1 when none is; and from the first ask the agent takes
// no new build request, so its answer ends once the build in hand ends — a replaced builder finishes the check
// it runs, instead of the build queue handing it out again from the start (issue 594).
//
// Two marks in the container's own temporary directory, as the ready mark (ready.go): a new container starts
// without them, and a restarted one removes them before it takes work again.
//
// - in hand: written by the agent when it claims a request, before it reads whether it is drained; removed
// when the build ends.
// - drained: written by the ask, before it reads whether a build is in hand.
//
// Each side writes its own mark before it reads the other's, so an ask racing a fetch sees the build in hand,
// or the agent sees the drain and hands the request back; never neither.
func inHandMark() string { return filepath.Join(os.TempDir(), "mesh-builder.in-hand") }
func drainMark() string { return filepath.Join(os.TempDir(), "mesh-builder.drained") }
// The drain lapses when no ask has come for drainLapse: the node-engine asks every 20 seconds for as long as it
// really waits, so an agent whose replacement was withdrawn — the declaration put back, the walk stopped —
// takes work again rather than staying drained for ever (review of ADR 0335). And an agent drained for longer
// than the node-engine's bound and the slack says it is not ready: the replacement it was drained for never
// came, which is a fault to see.
const (
drainLapse = 2 * time.Minute
drainTooLong = 15*time.Minute + 5*time.Minute
)
// drainedAt is when the drain began and when it was last asked, from its mark; false when there is no mark.
func drainedAt() (began, asked time.Time, ok bool) {
info, err := os.Stat(drainMark())
if err != nil {
return time.Time{}, time.Time{}, false
}
asked = info.ModTime()
began = asked
if raw, err := os.ReadFile(drainMark()); err == nil {
if t, err := time.Parse(time.RFC3339, strings.TrimSpace(string(raw))); err == nil && !t.After(asked) {
began = t
}
}
return began, asked, true
}
// drained says the agent was asked whether it is busy within drainLapse: it takes no new build request.
func drained() bool { return drainedSince(time.Now()) }
func drainedSince(now time.Time) bool {
_, asked, ok := drainedAt()
return ok && now.Sub(asked) < drainLapse
}
// drainedTooLong is the health check's second half: nil unless the agent has been drained, asked all along, for
// longer than its node-engine's bound and the slack.
func drainedTooLong(now time.Time) error {
began, asked, ok := drainedAt()
if !ok || now.Sub(asked) >= drainLapse || now.Sub(began) <= drainTooLong {
return nil
}
return fmt.Errorf("drained since %s and still asked: the replacement its node-engine waits for has not come "+
"within %s (novox/hq ADR 0335)", began.UTC().Format(time.RFC3339), drainTooLong)
}
// claim marks a request in hand, unless the agent is drained: then nothing is in hand, and the request goes
// back to the queue.
func claim() bool {
// Made, never written over: one build at a time, and a mark already there is a build in hand.
f, err := os.OpenFile(inHandMark(), os.O_WRONLY|os.O_CREATE|os.O_EXCL, 0o644)
if err != nil {
// Unmarked, an ask would say idle while this build runs: the build is not taken.
fmt.Fprintf(os.Stderr, "cannot mark a build in hand (%s), so it is handed back: %v\n", inHandMark(), err)
return false
}
_, err = f.WriteString(time.Now().UTC().Format(time.RFC3339) + "\n")
if closeErr := f.Close(); err == nil {
err = closeErr
}
if err != nil {
_ = os.Remove(inHandMark())
fmt.Fprintf(os.Stderr, "cannot mark a build in hand (%s), so it is handed back: %v\n", inHandMark(), err)
return false
}
if drained() {
_ = os.Remove(inHandMark())
fmt.Fprintln(os.Stderr, "asked whether busy, so this agent takes no new build: handed back")
return false
}
return true
}
// holding says what is in hand, for the ask to print: written once the request is read.
func holding(what string) {
_ = os.WriteFile(inHandMark(), []byte(time.Now().UTC().Format(time.RFC3339)+" "+what+"\n"), 0o644)
}
// released ends the build in hand.
func released() {
if err := os.Remove(inHandMark()); err != nil && !errors.Is(err, os.ErrNotExist) {
fmt.Fprintf(os.Stderr, "cannot remove the mark of the build in hand (%s): %v\n", inHandMark(), err)
}
}
// undrained clears both marks at the start of a holder: a container started again is asked afresh.
func undrained() error {
for _, mark := range []string{drainMark(), inHandMark()} {
if err := os.Remove(mark); err != nil && !errors.Is(err, os.ErrNotExist) {
return fmt.Errorf("cannot remove %s: %w", mark, err)
}
}
return nil
}
// busy is the ask: it drains the agent, then says whether a build is in hand. Nil — exit 0 — while one is.
func busy() (string, error) {
// The mark keeps when the drain began while it is asked within drainLapse, and is written again on every
// ask: its time is the last ask.
began := time.Now().UTC()
if was, _, ok := drainedAt(); ok && drained() {
began = was.UTC()
}
if err := os.WriteFile(drainMark(), []byte(began.Format(time.RFC3339)+"\n"), 0o644); err != nil {
// Not drained, it could take the next build: said, and answered busy only for what is in hand now.
fmt.Fprintf(os.Stderr, "cannot mark this agent drained (%s): %v\n", drainMark(), err)
}
held, err := os.ReadFile(inHandMark())
if err != nil {
return "", errors.New("idle: no build in hand, and this agent takes no new one")
}
since, what, _ := strings.Cut(strings.TrimSpace(string(held)), " ")
if what == "" {
what = "a build"
}
return fmt.Sprintf("busy: %s, in hand since %s; taking no new build", what, since), nil
}
+162
View File
@@ -0,0 +1,162 @@
package main
import (
"os"
"strings"
"sync"
"testing"
"time"
)
// `mesh-builder busy` answers 0 while a build is in hand and non-zero when idle, and after its first ask the
// agent takes no new build request (novox/hq ADR 0335 §5).
func TestBusyAnswersWhileABuildIsInHandAndDrainsTheAgent(t *testing.T) {
t.Setenv("TMPDIR", t.TempDir())
if err := undrained(); err != nil {
t.Fatal(err)
}
if drained() {
t.Fatal("an agent never asked is drained")
}
// A build is claimed and runs.
if !claim() {
t.Fatal("an agent never asked refused a build")
}
holding("novox/mesh-controller (check-280)")
said, err := busy()
if err != nil || !strings.Contains(said, "novox/mesh-controller (check-280)") {
t.Fatalf("an agent with a build in hand did not answer busy with what it holds: %q %v", said, err)
}
// From that ask on it takes nothing new: the fetch is not made, and one fetched is handed back.
if !drained() {
t.Fatal("the ask did not drain the agent")
}
if claim() {
t.Fatal("a drained agent took a new build")
}
// The build in hand is still in hand after the refused claim, and busy goes on saying so.
if _, err := busy(); err != nil {
t.Fatalf("a refused claim ended the build in hand: %v", err)
}
// It ends: idle, and still drained.
released()
if _, err := busy(); err == nil {
t.Fatal("an agent with nothing in hand answered busy")
}
if claim() {
t.Fatal("a drained agent took a build after its last one ended")
}
// Started again, it is asked afresh.
if err := undrained(); err != nil {
t.Fatal(err)
}
if drained() || !claim() {
t.Fatal("a holder started again is still drained")
}
released()
}
// An ask racing a fetch sees the build in hand, or the agent sees the drain and hands the request back: never
// neither, which would replace the container under a build just taken.
func TestAnAskRacingAFetchNeverMissesTheBuild(t *testing.T) {
t.Setenv("TMPDIR", t.TempDir())
for i := 0; i < 200; i++ {
if err := undrained(); err != nil {
t.Fatal(err)
}
var claimed bool
var askErr error
var wg sync.WaitGroup
wg.Add(2)
go func() { defer wg.Done(); claimed = claim() }()
go func() { defer wg.Done(); _, askErr = busy() }()
wg.Wait()
if claimed && askErr != nil {
t.Fatalf("round %d: the agent took a build and the ask said idle", i)
}
released()
}
}
// A drain no longer asked lapses: an agent whose replacement was withdrawn after the first ask (a put-back, a
// revert, a stopped walk) takes work again after drainLapse, rather than staying drained with its health green.
func TestADrainNoLongerAskedLapses(t *testing.T) {
t.Setenv("TMPDIR", t.TempDir())
if err := undrained(); err != nil {
t.Fatal(err)
}
if _, err := busy(); err == nil {
t.Fatal("an idle agent answered busy")
}
if !drained() || claim() {
t.Fatal("an agent just asked takes work")
}
// The last ask was three minutes ago.
old := time.Now().Add(-3 * time.Minute)
if err := os.Chtimes(drainMark(), old, old); err != nil {
t.Fatal(err)
}
if drained() {
t.Fatal("a drain no longer asked for three minutes still holds")
}
if !claim() {
t.Fatal("an agent whose drain lapsed takes no work")
}
released()
}
// An agent drained, and asked all along, for longer than its node-engine's bound and the slack is not ready: the
// replacement it was drained for never came. One asked again after a lapse starts a new drain.
func TestAnAgentDrainedTooLongIsNotReady(t *testing.T) {
t.Setenv("TMPDIR", t.TempDir())
if err := undrained(); err != nil {
t.Fatal(err)
}
taking("mesh-build-machine")
if _, err := busy(); err == nil {
t.Fatal("an idle agent answered busy")
}
now := time.Now()
if err := drainedTooLong(now); err != nil {
t.Fatalf("an agent drained a moment ago is not ready: %v", err)
}
// The drain began 25 minutes ago, and the node-engine asked a moment ago.
began := now.Add(-25 * time.Minute).UTC().Format(time.RFC3339)
if err := os.WriteFile(drainMark(), []byte(began+"\n"), 0o644); err != nil {
t.Fatal(err)
}
err := drainedTooLong(time.Now())
if err == nil || !strings.Contains(err.Error(), "has not come") {
t.Fatalf("an agent drained for 25 minutes reads ready: %v", err)
}
// Each ask keeps when the drain began.
if _, err := busy(); err == nil {
t.Fatal("an idle agent answered busy")
}
if err := drainedTooLong(time.Now()); err == nil {
t.Fatal("an ask started the drain again")
}
// The asks stopped: the drain lapsed, and so did the fault.
old := time.Now().Add(-3 * time.Minute)
if err := os.Chtimes(drainMark(), old, old); err != nil {
t.Fatal(err)
}
if err := drainedTooLong(time.Now()); err != nil {
t.Fatalf("a lapsed drain still reads not ready: %v", err)
}
}
// An ask this build does not know exits non-zero: an older image asked `busy` reads idle, not busy for the
// node-engine's whole bound.
func TestAnUnknownCommandFails(t *testing.T) {
was := os.Args
t.Cleanup(func() { os.Args = was })
os.Args = []string{"mesh-builder", "some-later-word"}
if err := run(); err == nil {
t.Fatal("an unknown command exited 0")
}
os.Args = []string{"mesh-builder", "help"}
if err := run(); err != nil {
t.Fatalf("help failed: %v", err)
}
}
+38
View File
@@ -0,0 +1,38 @@
package main
import (
"testing"
"github.com/novox/mesh-controller/internal/builder"
"github.com/novox/mesh-controller/internal/link"
)
// The build seat hands the check the judge cache the build agent declares (MESH_JUDGE_CACHE), never a path of its own.
func TestTheBuildSeatHandsTheCheckTheJudgeCacheItDeclares(t *testing.T) {
t.Setenv("MESH_JUDGE_CACHE", "/var/lib/mesh/build-agent/judge-cache")
if got := judgeCache(); got != "/var/lib/mesh/build-agent/judge-cache" {
t.Fatalf("the judge cache the manifest declares is read as %q", got)
}
t.Setenv("MESH_JUDGE_CACHE", "")
if got := judgeCache(); got != "" {
t.Fatalf("with no judge cache declared, %q is used", got)
}
}
// The siblings a check is asked to read reach the build seat's spec and, as read, the outcome (novox/hq ADR 0382).
func TestTheSiblingsAnAskNamesReachTheSpecAndTheOutcome(t *testing.T) {
spec := checkSpecOf(link.BuildRequest{Check: &link.CheckRequest{Owner: "novox", Repo: "mesh-host", Base: "main",
Siblings: []link.CheckSibling{{Repository: "novox/mesh-controller", Number: 346, Head: "a", Base: "main", Dir: "mesh-controller"}},
SiblingsUnknown: "why"}})
if len(spec.Siblings) != 1 || spec.Siblings[0] != (builder.Sibling{Repository: "novox/mesh-controller", Number: 346,
Head: "a", Base: "main", Dir: "mesh-controller"}) || spec.SiblingsUnknown != "why" {
t.Fatalf("the spec reads %+v", spec)
}
out := siblingsOut([]builder.JudgedSibling{{Repository: "novox/mesh-controller", Number: 346, Head: "a", Base: "main", BaseCommit: "b"}})
if len(out) != 1 || out[0].BaseCommit != "b" || out[0].Number != 346 {
t.Fatalf("the outcome reads %+v", out)
}
if plain := checkSpecOf(link.BuildRequest{Check: &link.CheckRequest{Owner: "o", Repo: "r"}}); plain.Siblings != nil {
t.Fatalf("a request that names none has siblings %+v", plain.Siblings)
}
}
+433
View File
@@ -0,0 +1,433 @@
package main
import (
"context"
"encoding/json"
"errors"
"fmt"
"os"
"os/exec"
"path/filepath"
"strings"
"sync"
"time"
"github.com/nats-io/nats.go/micro"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/builder"
"github.com/novox/mesh-controller/internal/link"
)
// What a holder of the build seat answers for, on its own machine (novox/hq ADR 0219).
//
// **The queue is the controller's; the build running here is this machine's.** The controller can
// see and change what waits in the seat's queue, but an ask a holder already took is a process tree
// on this machine and containers in this machine's runtime, and only this machine can end them. So
// the holder serves four verbs on the seat's subjects for this machine: what it is building, kill
// it, pause, resume.
//
// **One build at a time** (ADR 0190), which is also what builder.Said assumes — a package global
// set per build — so "the build running here" is one or none, and kill names it by id so a call
// that arrives as one build ends and the next begins cannot end the wrong one.
// holder is this machine's state as a holder of the build seat.
type holder struct {
on, seat string
// workspace is where the paused flag is kept, so a holder restarted while paused stays paused
// rather than silently taking work again.
workspace string
// say publishes this machine's state: whether it takes work (link.HolderState).
say func(link.HolderState) error
// remove runs what a kill needs outside the build: the containers left behind.
remove builder.Runner
// scratch is where check builds are kept (novox/hq ADR 0350 §5): what the week's sweep lets go of. Nil where
// this machine publishes to no artifact store.
scratch builder.Scratch
mu sync.Mutex
paused bool
running *running
}
// running is the build this machine is doing.
type running struct {
request link.BuildRequest
step string
started time.Time
cancel context.CancelFunc
killed bool
// returned is the build's own work having ended, before a kill or not; outcome is what was then
// announced, and announced whether it went out.
returned bool
outcome string
announced bool
done chan struct{}
}
// pausedFile is where the flag lives in the workspace.
func pausedFile(workspace string) string { return filepath.Join(workspace, ".mesh-builder-paused") }
// newHolder reads the paused flag the workspace keeps.
func newHolder(on, seat, workspace string, say func(link.HolderState) error) *holder {
h := &holder{on: on, seat: seat, workspace: workspace, say: say, remove: plainRun}
if _, err := os.Stat(pausedFile(workspace)); err == nil {
h.paused = true
}
return h
}
// Paused is asked by the taking loop before every fetch.
func (h *holder) Paused() bool {
h.mu.Lock()
defer h.mu.Unlock()
return h.paused
}
// setPaused records the flag in the workspace first and then in memory, so what this holder says
// it is and what it would be after a restart never differ.
func (h *holder) setPaused(paused bool) error {
path := pausedFile(h.workspace)
if paused {
if err := os.MkdirAll(h.workspace, 0o755); err != nil {
return err
}
if err := os.WriteFile(path, []byte(time.Now().UTC().Format(time.RFC3339)+"\n"), 0o644); err != nil {
return fmt.Errorf("cannot keep the paused flag in %s: %w", path, err)
}
} else if err := os.Remove(path); err != nil && !errors.Is(err, os.ErrNotExist) {
return fmt.Errorf("cannot remove the paused flag %s: %w", path, err)
}
h.mu.Lock()
h.paused = paused
h.mu.Unlock()
h.announce()
return nil
}
// announce says whether this machine takes work. Never fatal: the flag is kept either way, and the
// controller reading an older state is a plan read as late rather than a build lost.
func (h *holder) announce() {
if h.say == nil {
return
}
if err := h.say(link.HolderState{On: h.on, Paused: h.Paused(), At: time.Now().UTC().Format(time.RFC3339Nano)}); err != nil {
fmt.Fprintf(os.Stderr, "cannot say whether this machine takes builds: %v\n", err)
}
}
// stateWhilePaused says this machine's state at start, and again every while it stays paused, so
// a pause outlives the events stream's retention.
func (h *holder) stateWhilePaused(ctx context.Context, every time.Duration) {
h.announce()
tick := time.NewTicker(every)
defer tick.Stop()
for {
select {
case <-ctx.Done():
return
case <-tick.C:
if h.Paused() {
h.announce()
}
}
}
}
// expireChecks lets go of every check build older than a week (novox/hq ADR 0350 §5), at start and every while
// after. Every holder sweeps the one store; what one let go of, another finds gone.
func (h *holder) expireChecks(ctx context.Context, every time.Duration, now func() time.Time) {
sweep := func() {
if h.scratch == nil {
return
}
n, err := h.scratch.Expire(ctx, builder.CheckBuildsKept, now())
switch {
case err != nil:
fmt.Fprintf(os.Stderr, "check builds older than %s: %d let go of, and not all: %v\n", builder.CheckBuildsKept, n, err)
case n > 0:
fmt.Fprintf(os.Stderr, "check builds older than %s: %d check(s)' builds let go of\n", builder.CheckBuildsKept, n)
}
}
sweep()
tick := time.NewTicker(every)
defer tick.Stop()
for {
select {
case <-ctx.Done():
return
case <-tick.C:
sweep()
}
}
}
// begin records the build this machine took, with the cancel that ends it.
func (h *holder) begin(request link.BuildRequest, cancel context.CancelFunc) *running {
r := &running{request: request, started: time.Now(), cancel: cancel, done: make(chan struct{})}
h.mu.Lock()
h.running = r
h.mu.Unlock()
return r
}
// end clears it, and lets a kill waiting on it know it is over.
func (h *holder) end(r *running) {
h.mu.Lock()
if h.running == r {
h.running = nil
}
h.mu.Unlock()
close(r.done)
}
// stepped records the step a build is at, for `current`.
func (h *holder) stepped(r *running, step string) {
h.mu.Lock()
r.step = step
h.mu.Unlock()
}
// currentBuild is what `current` answers.
type currentBuild struct {
On string `json:"on"`
Paused bool `json:"paused"`
Running *struct {
ID string `json:"id"`
Repository string `json:"repository"`
Path string `json:"path,omitempty"`
Ref string `json:"ref,omitempty"`
Step string `json:"step,omitempty"`
Started string `json:"started"`
Elapsed string `json:"elapsed"`
} `json:"running,omitempty"`
Said string `json:"said"`
}
func (h *holder) current() currentBuild {
h.mu.Lock()
defer h.mu.Unlock()
out := currentBuild{On: h.on, Paused: h.paused}
taking := "taking builds"
if h.paused {
taking = "paused, taking no new build"
}
if h.running == nil {
out.Said = fmt.Sprintf("%s is building nothing; %s", h.on, taking)
return out
}
r := h.running
elapsed := time.Since(r.started).Round(time.Second)
out.Running = &struct {
ID string `json:"id"`
Repository string `json:"repository"`
Path string `json:"path,omitempty"`
Ref string `json:"ref,omitempty"`
Step string `json:"step,omitempty"`
Started string `json:"started"`
Elapsed string `json:"elapsed"`
}{r.request.ID, r.request.Repository, r.request.Path, r.request.Ref, r.step,
r.started.UTC().Format(time.RFC3339), elapsed.String()}
out.Said = fmt.Sprintf("%s is building %s (%s) at %s for %s; %s",
h.on, r.request.Repository, r.request.ID, orNothing(r.step), elapsed, taking)
return out
}
// The bounds a kill keeps: each pass removing containers, and the wait for the build to end between
// them. The worst case — 15s, 20s, 15s — is inside what the controller waits for the answer
// (killAnswer in the controller's queue.go, 75s).
var (
killRemoves = 15 * time.Second
killWaits = 20 * time.Second
)
// kill ends the build with this id, if it is the one running here and still working: its context
// cancelled — which kills each command's process group — and the containers it started removed by
// their label, once at once and again after the build has ended, so one created while it was being
// killed is not left. The build's own goroutine announces it failed, killed by hand, and settles the
// ask so it is not redelivered; the answer says whether that happened.
func (h *holder) kill(id string) (string, error) {
h.mu.Lock()
r := h.running
if r == nil || r.request.ID != id {
doing := "nothing"
if r != nil {
doing = r.request.ID
}
h.mu.Unlock()
return "", fmt.Errorf("%s is not building %s; it is building %s. `queue` says where an ask is", h.on, id, doing)
}
if r.returned {
h.mu.Unlock()
return "", fmt.Errorf("%s on %s has already ended on its own and is saying how; `builds` shows it", id, h.on)
}
r.killed = true
cancel, done := r.cancel, r.done
h.mu.Unlock()
cancel()
removed, removeErr := h.removeContainers(id)
ended := false
select {
case <-done:
ended = true
case <-time.After(killWaits):
}
again, againErr := h.removeContainers(id)
removed += again
if removeErr == nil {
removeErr = againErr
}
containers := fmt.Sprintf("%d container(s) it started removed", removed)
if removeErr != nil {
containers = "its containers could not all be listed or removed: " + removeErr.Error()
}
if !ended {
return fmt.Sprintf("killed %s on %s: %s; the build has not finished ending yet — `builds` says when "+
"its outcome is in", id, h.on, containers), nil
}
h.mu.Lock()
outcome, announced := r.outcome, r.announced
h.mu.Unlock()
if !announced {
return fmt.Sprintf("killed %s (%s) on %s: its commands ended and %s, and its outcome could not be "+
"announced — the ask is not settled and will be handed out again", id, r.request.Repository, h.on,
containers), nil
}
return fmt.Sprintf("killed %s (%s) on %s: its commands ended, %s, and its outcome announced as failed, %s — "+
"settled, so it is not handed to another machine", id, r.request.Repository, h.on, containers, outcome), nil
}
// removeContainers is one pass of removing what the build left, bounded.
func (h *holder) removeContainers(id string) (int, error) {
cleanup, stop := context.WithTimeout(context.Background(), killRemoves)
defer stop()
return builder.RemoveContainersOf(cleanup, h.remove, id)
}
// removeLeftBehind removes, once, what earlier holders on this machine left — called before anything is
// taken, so none of it is this holder's — and says what it did. A runtime that cannot be asked is said
// and the holder takes work all the same: what is left is waste, not a reason to stop building.
func (h *holder) removeLeftBehind() int {
cleanup, stop := context.WithTimeout(context.Background(), killRemoves)
defer stop()
n, err := builder.RemoveLeftBehind(cleanup, h.remove)
switch {
case err != nil:
fmt.Fprintf(os.Stderr, "could not look for containers earlier builds left on %s: %v\n", h.on, err)
case n > 0:
fmt.Fprintf(os.Stderr, "removed %d container(s) earlier builds left on %s\n", n, h.on)
}
return n
}
// returned records that the build's work ended, and says whether a kill came first — only then is
// the build killed; an error it ended with on its own is its own outcome.
func (h *holder) returned(r *running) bool {
h.mu.Lock()
defer h.mu.Unlock()
r.returned = true
return r.killed
}
// said records the outcome announced, and whether it went out, for a kill to answer with.
func (h *holder) said(r *running, outcome string, announced bool) {
h.mu.Lock()
r.outcome, r.announced = outcome, announced
h.mu.Unlock()
}
// handlers are the seat's verbs, as this machine answers them.
func (h *holder) handlers() map[string]link.ToolHandler {
return map[string]link.ToolHandler{
"current": func(context.Context, json.RawMessage) (any, error) { return h.current(), nil },
"kill": func(_ context.Context, raw json.RawMessage) (any, error) {
var args struct {
ID string `json:"id"`
}
if err := json.Unmarshal(raw, &args); err != nil || strings.TrimSpace(args.ID) == "" {
return nil, errors.New("kill needs the build's id")
}
said, err := h.kill(strings.TrimSpace(args.ID))
if err != nil {
return nil, err
}
return map[string]any{"said": said}, nil
},
"pause": func(context.Context, json.RawMessage) (any, error) {
if err := h.setPaused(true); err != nil {
return nil, err
}
said := h.on + " is paused: it takes no new build until resumed"
if c := h.current(); c.Running != nil {
said += "; " + c.Running.ID + " runs on and finishes"
}
return map[string]any{"said": said, "paused": true}, nil
},
"resume": func(context.Context, json.RawMessage) (any, error) {
if err := h.setPaused(false); err != nil {
return nil, err
}
return map[string]any{"said": h.on + " takes builds again", "paused": false}, nil
},
}
}
func orNothing(s string) string {
if s == "" {
return "its start"
}
return s
}
// plainRun runs a command outside any build: no line reaches a build's log, because the build whose
// log it would be is the one being ended.
func plainRun(ctx context.Context, dir, name string, args ...string) (string, error) {
cmd := exec.CommandContext(ctx, name, args...)
cmd.Dir = dir
out, err := cmd.CombinedOutput()
if err != nil {
return string(out), fmt.Errorf("%s %s: %w: %s", name, strings.Join(args, " "), err, strings.TrimSpace(string(out)))
}
return string(out), nil
}
// announcement is what this holder answers discovery with (novox/hq ADR 0195, ADR 0197): this
// machine's verbs of the seat, in the shape every tool runtime announces a seat's verb — kind seat,
// the module answering, the seat, scope node, the machine, its description and argument schema — so
// the console finds `<node>/node-build-agent.kill` by searching, as it finds any seat's verb.
//
// One service per machine, named for the seat and identified by the machine, so the answer is this
// machine's four verbs and nothing more. The verbs are the compiled seat row's: a build machine has no
// store, and what it serves is what this binary was built to serve.
func announcement(seat, module, node string) micro.Info {
s, _ := catalogue.SeatNamed(seat)
var endpoints []micro.EndpointInfo
for _, v := range s.Serves {
schema, _ := json.Marshal(v.Input)
endpoints = append(endpoints, micro.EndpointInfo{
Name: seat + "__" + v.Name,
Subject: link.NodeSeatToolSubject(seat, v.Name, node),
// The queue group the verbs are served in, as every runtime announces its own.
QueueGroup: "seat." + seat,
Metadata: map[string]string{
"kind": "seat", "module": module, "tool": v.Name, "seat": seat, "scope": "node",
"node": node, "interchangeable": "false", "description": v.Description, "schema": string(schema),
},
})
}
return micro.Info{
ServiceIdentity: micro.ServiceIdentity{Name: seat, ID: node, Version: "0.1.0",
Metadata: map[string]string{"seat": seat, "scope": "node", "node": node, "module": module}},
Description: "what the build running on " + node + " is, and ending, pausing and resuming it (novox/hq ADR 0219)",
Endpoints: endpoints,
}
}
// moduleOf is the module a credential was issued for: its user is `<node>.<module>`.
func moduleOf(user string) string {
if _, module, ok := strings.Cut(user, "."); ok && module != "" {
return module
}
return "build-agent"
}
+172
View File
@@ -0,0 +1,172 @@
package main
import (
"context"
"encoding/json"
"errors"
"strings"
"testing"
"github.com/novox/mesh-controller/internal/link"
)
// A holder paused stays paused across its restart: the flag is kept in its workspace (novox/hq ADR
// 0219), and what it says about itself follows.
func TestAPausedHolderStaysPausedAcrossARestart(t *testing.T) {
workspace := t.TempDir()
var said []link.HolderState
h := newHolder("ace", link.TheBuildMachine, workspace, func(s link.HolderState) error {
said = append(said, s)
return nil
})
if h.Paused() {
t.Fatal("a new holder starts paused")
}
if _, err := h.handlers()["pause"](context.Background(), json.RawMessage(`{}`)); err != nil {
t.Fatal(err)
}
if !h.Paused() || len(said) != 1 || !said[0].Paused || said[0].On != "ace" {
t.Fatalf("paused: %v, said %+v", h.Paused(), said)
}
again := newHolder("ace", link.TheBuildMachine, workspace, nil)
if !again.Paused() {
t.Fatal("restarted, the holder forgot it was paused")
}
if _, err := again.handlers()["resume"](context.Background(), json.RawMessage(`{}`)); err != nil {
t.Fatal(err)
}
if newHolder("ace", link.TheBuildMachine, workspace, nil).Paused() {
t.Fatal("resumed, the holder came back paused")
}
}
// kill ends the build with that id — its context, which ends its commands — removes what it left by
// label, and refuses an id it is not building.
func TestKillEndsTheBuildRunningHereAndNoOther(t *testing.T) {
h := newHolder("ace", link.TheBuildMachine, t.TempDir(), nil)
var removed []string
h.remove = func(_ context.Context, _ string, name string, args ...string) (string, error) {
removed = append(removed, name+" "+strings.Join(args, " "))
if args[0] == "ps" {
return "c1\n", nil
}
return "", nil
}
kill := h.handlers()["kill"]
if _, err := kill(context.Background(), json.RawMessage(`{"id":"build-1"}`)); err == nil {
t.Fatal("killed a build while none ran")
}
building, cancel := context.WithCancel(context.Background())
r := h.begin(link.BuildRequest{ID: "build-1", Repository: "novox/a"}, cancel)
h.stepped(r, "image")
if c := h.current(); c.Running == nil || c.Running.ID != "build-1" || c.Running.Step != "image" {
t.Fatalf("current says %+v", c)
}
if _, err := kill(context.Background(), json.RawMessage(`{"id":"build-2"}`)); err == nil ||
!strings.Contains(err.Error(), "build-1") {
t.Fatalf("killed another id: %v", err)
}
// The build's own goroutine: it ends when its context does, as a build's commands do, and
// announces what came of it.
go func() {
<-building.Done()
if h.returned(r) {
h.said(r, link.KilledByHand, true)
}
h.end(r)
}()
answer, err := kill(context.Background(), json.RawMessage(`{"id":"build-1"}`))
if err != nil {
t.Fatal(err)
}
if building.Err() == nil {
t.Fatal("the build's context was not cancelled")
}
if !r.killed {
t.Error("the build is not marked killed, so it would be announced as an ordinary failure")
}
said := answer.(map[string]any)["said"].(string)
if !strings.Contains(said, "2 container(s)") || !strings.Contains(said, "announced as failed") || !strings.Contains(said, link.KilledByHand) {
t.Errorf("kill said %q", said)
}
// Removed at the kill and again once the build had ended: a container made in between is caught.
if len(removed) != 4 || !strings.Contains(removed[0], "label=mesh.build=build-1") || !strings.Contains(removed[2], "label=mesh.build=build-1") {
t.Errorf("removed %v", removed)
}
if c := h.current(); c.Running != nil {
t.Errorf("after the kill current says %+v", c)
}
}
// A holder announces this machine's verbs of the seat as the console reads a seat's verb, and no more.
func TestAHolderAnnouncesItsMachinesVerbsForTheConsole(t *testing.T) {
info := announcement(link.TheBuildMachine, moduleOf("ace.build-agent"), "ace")
if info.Name != "node-build-agent" || info.ID != "ace" || len(info.Endpoints) != 4 {
t.Fatalf("announced %s/%s with %d endpoints", info.Name, info.ID, len(info.Endpoints))
}
kill := info.Endpoints[1]
md := kill.Metadata
if kill.Subject != "mesh.seat.node-build-agent.tool.kill.ace" || kill.QueueGroup != "seat.node-build-agent" || md["kind"] != "seat" || md["seat"] != "node-build-agent" ||
md["scope"] != "node" || md["node"] != "ace" || md["tool"] != "kill" || md["module"] != "build-agent" ||
!strings.Contains(md["description"], "killed by hand") || !strings.Contains(md["schema"], `"id"`) {
t.Fatalf("kill is announced as %+v", kill)
}
body, err := json.Marshal(info)
if err != nil || len(body) > 8*1024 {
t.Fatalf("the answer is %d bytes (%v)", len(body), err)
}
}
// A build that ended on its own as the kill arrived says what it did: the kill is refused, and its
// own error is its outcome. One whose outcome could not be announced is not said to be settled.
func TestAKillArrivingAfterTheBuildEndedIsRefusedAndAnUnannouncedKillSaysSo(t *testing.T) {
h := newHolder("ace", link.TheBuildMachine, t.TempDir(), nil)
h.remove = func(context.Context, string, string, ...string) (string, error) { return "", nil }
_, cancel := context.WithCancel(context.Background())
r := h.begin(link.BuildRequest{ID: "build-1"}, cancel)
if killed := h.returned(r); killed {
t.Fatal("a build nobody killed reads as killed")
}
if _, err := h.kill("build-1"); err == nil || !strings.Contains(err.Error(), "ended on its own") {
t.Fatalf("killed a build that had ended: %v", err)
}
h.end(r)
building, cancel := context.WithCancel(context.Background())
r = h.begin(link.BuildRequest{ID: "build-2"}, cancel)
go func() {
<-building.Done()
if h.returned(r) {
h.said(r, link.KilledByHand, false)
}
h.end(r)
}()
said, err := h.kill("build-2")
if err != nil || strings.Contains(said, "announced as failed") || !strings.Contains(said, "could not be announced") {
t.Fatalf("kill said %q (%v)", said, err)
}
}
// **Issue 306**: a holder starting removes the containers earlier builds left on its machine, through
// the runner a kill uses, and a runtime that cannot be asked does not stop it.
func TestAHolderStartingRemovesWhatEarlierBuildsLeftHere(t *testing.T) {
h := newHolder("novox", link.TheBuildMachine, t.TempDir(), nil)
var removed []string
h.remove = func(_ context.Context, _ string, name string, args ...string) (string, error) {
removed = append(removed, name+" "+strings.Join(args, " "))
if args[0] == "ps" {
return "c1 build-1\nc2 build-1\nc3 check-here-2\n", nil
}
return "", nil
}
if n := h.removeLeftBehind(); n != 2 || len(removed) != 2 || removed[1] != "docker rm -f c1 c2" {
t.Fatalf("removed %d: %v", n, removed)
}
h.remove = func(context.Context, string, string, ...string) (string, error) {
return "", errors.New("no runtime")
}
if n := h.removeLeftBehind(); n != 0 {
t.Errorf("a runtime that cannot be asked removed %d", n)
}
}
+936
View File
@@ -0,0 +1,936 @@
// mesh-builder — the thing a build machine runs.
//
// It takes work from the mesh, turns a repository into artifacts, publishes them, and says what
// came out. It is **not** the control plane and it is **not** the host:
//
// - the control plane decides and never touches a machine. Building runs commands on one, and
// what the control plane may send a machine is bounded by the declaration language
// (novox/hq ADR 0005). "Run this build" is not in it, and widening the language so it could
// be would make the control plane able to run anything anywhere.
// - the host applies declarations and holds no opinion about what they contain. A host that
// also built things would need a container runtime and git, on every machine, to do something
// almost none of them will ever do.
//
// So it is a module: a program a machine runs because the mesh told it to, holding its own broker
// credential and nothing else. Compromise of a build machine is compromise of a build machine.
package main
import (
"context"
"encoding/json"
"errors"
"fmt"
"log"
"net/url"
"os"
"os/signal"
"sort"
"strings"
"sync"
"sync/atomic"
"syscall"
"time"
"github.com/nats-io/nats.go"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/builder"
"github.com/novox/mesh-controller/internal/link"
)
// version is set at build time.
var version = "development"
func main() {
if err := run(); err != nil {
fmt.Fprintf(os.Stderr, "mesh-builder: %v\n", err)
os.Exit(1)
}
}
const usage = `mesh-builder — builds modules for the mesh
It consumes build requests and answers with what it made. Nothing is listened on and nothing
is dialled except the broker.
MESH_BROKER_FILE a file the mesh sealed to this machine holding the same
MESH_REGISTRY host:port to publish artifacts to, when the mesh has not said
MESH_BINDING a file the mesh wrote saying where the artifact store is
MESH_PACKAGE_BINDING a file the mesh wrote saying where the package registry is
MESH_NPM_TOKEN_FILE a file the mesh sealed holding the token for it
MESH_GENESIS_PACKAGE_REGISTRY the document the installer sealed at genesis, read while no
binding is present (novox/hq ADR 0336)
MESH_NPM_REGISTRY a package registry URL, when the mesh has not said (a person, the bootstrap)
MESH_NPM_TOKEN the token for it, likewise
MESH_NPM_SCOPE the scope it answers for (default: @novox)
MESH_WORKSPACE where to clone and build (default: a temporary directory)
It also builds one module and stops, which is how a mesh is raised — before there is a
broker to take work from or a registry to publish into:
mesh-builder build <repository> [--path P] [--ref COMMIT] [--registry HOST:PORT]
Without --registry the artifacts stay in this machine's container runtime, named by the
digest of their own configuration. The result is printed as JSON.
mesh-builder ready
exits 0 once the agent running in this container is bound to its build work and pulls it, and
1 before: its container's health check.
mesh-builder busy
exits 0 while the agent running in this container has a build in hand, and 1 when it has none;
from the first ask the agent takes no new build: its container's busy command, asked before the
container is replaced (novox/hq ADR 0335).
`
func run() error {
if len(os.Args) > 1 {
switch os.Args[1] {
case "version":
fmt.Println(version)
return nil
case "build":
return buildOnce(context.Background(), os.Args[2:])
case "ready":
// The container's health check (novox/hq issue 544): exits non-zero until the agent takes work, and
// once it has been drained longer than its node-engine would wait to replace it (ADR 0335).
if err := ready(); err != nil {
return err
}
return drainedTooLong(time.Now())
case "busy":
// The container's busy command (novox/hq ADR 0335): drains the agent, and exits non-zero once
// nothing is in hand.
said, err := busy()
if err != nil {
return err
}
fmt.Println(said)
return nil
case "help", "-h", "--help":
fmt.Print(usage)
return nil
default:
// **An ask this build does not know fails** (novox/hq ADR 0335): an image older than `busy`, asked
// it, must read as idle — exiting 0 would hold its container for the node-engine's whole bound.
fmt.Fprint(os.Stderr, usage)
return fmt.Errorf("%q is not a command this mesh-builder knows", os.Args[1])
}
}
if err := notReady(); err != nil {
return err
}
// **The builder's own version is part of every reuse key** (novox/hq ADR 0350 §1): a builder that builds
// differently never takes a build another one made.
builder.Version = builder.SelfVersion(version)
if err := undrained(); err != nil {
return err
}
credential, err := brokerFrom()
if err != nil {
return err
}
registry, err := whereToPublish()
if err != nil {
return err
}
workspace := os.Getenv("MESH_WORKSPACE")
if workspace == "" {
workspace = os.TempDir() + "/mesh-builder"
}
// **The mesh's name for this machine, not the container's.** A build is reported to the rest
// of the mesh, and a report whose origin reads `104cb10e105b` names something no other module
// can look up. The mesh already knows the answer and has a way to say it — `${machine:name}`
// in the environment file this module is handed — so the hostname is only what is left when
// nobody said.
on := os.Getenv("MESH_NODE")
if on == "" {
hostname, err := os.Hostname()
if err != nil {
return fmt.Errorf("this build machine has no name: nothing said MESH_NODE and the host would not say either: %w", err)
}
on = hostname
}
ctx, stop := signal.NotifyContext(context.Background(), syscall.SIGINT, syscall.SIGTERM)
defer stop()
js, seat, err := dialFor(credential)
if err != nil {
return err
}
defer js.Close()
// **This machine's holder: paused or not, and the build it is running** (novox/hq ADR 0219). The
// paused flag is read from the workspace before anything is taken, so a holder restarted while
// paused takes nothing.
h := newHolder(on, seat, workspace, func(state link.HolderState) error {
body, err := json.Marshal(state)
if err != nil {
return err
}
_, err = js.Context().Publish(link.BuildPausedOf(seat, on), body)
return err
})
if h.Paused() {
fmt.Fprintf(os.Stderr, "paused (kept in %s): taking no build until resumed\n", pausedFile(workspace))
}
// **What an earlier holder here left goes before anything is taken** (novox/hq issue 306): a holder
// recreated mid-check left the check's store and bus running, and the ask went to another machine.
h.removeLeftBehind()
// Paused or drained, it takes nothing new (novox/hq ADR 0219, ADR 0335): a drained agent finishes the build
// in hand and then says idle, and its container is replaced between builds.
machine := link.MachineOverNATSWith(js, on, seat, link.MachineOptions{
Paused: func() bool { return h.Paused() || drained() },
Taking: func() { taking(seat) },
Claim: claim,
})
defer machine.Close()
// The seat's verbs, on this machine's subjects. Only the seat that declares them: the retired
// one serves none, and a subscription its holder has no grant for would be refused for ever.
if seat == link.TheBuildMachine {
stopServing, err := link.OverNATS{Conn: js.Conn()}.ServeNodeSeatTools(seat, on, h.handlers(),
log.New(os.Stderr, "", 0))
if err != nil {
return err
}
defer stopServing()
// And says so, for the console to find (novox/hq ADR 0197).
stopAnnouncing, err := link.OverNATS{Conn: js.Conn()}.Announce(
announcement(seat, moduleOf(credential.User), on), log.New(os.Stderr, "", 0))
if err != nil {
return err
}
defer stopAnnouncing()
go h.stateWhilePaused(ctx, time.Hour)
}
fmt.Fprintf(os.Stderr, "building for the mesh, publishing to %s\n", registry)
publisher := builder.Registry{Address: registry, Run: builder.Command}
// Check builds are let go of after a week (novox/hq ADR 0350 §5).
h.scratch = publisher
go h.expireChecks(ctx, time.Hour, time.Now)
return machine.Take(ctx, func(ctx context.Context, work link.Build) {
defer released()
holding(work.Request().Repository + " (" + work.Request().ID + ")")
answer(ctx, publisher, on, workspace, work, h)
})
}
// dialFor opens this machine's link to whichever bus the mesh is on, and says which build seat it
// holds.
//
// **One place chooses**, as everywhere else the bus change went (novox/hq ADR 0116 step 5): a build
// machine told about both would take work from one and answer on the other, and every log line would
// say it was fine.
func dialFor(credential Credential) (*broker.JetStream, string, error) {
// **The credential names the bus, and there is one** (novox/hq ADR 0131, design 28 task 5.5).
// A credential for the mesh's bus carries user, password and fingerprint beside the address,
// and that is enough to dial it, pinned.
if !credential.onTheNewBus() {
return nil, "", fmt.Errorf("the credential at hand names %q, which is not the mesh's bus", credential.URL)
}
// Named for what it is, not the controller whose code dials it (novox/hq issue 327).
host, _ := os.Hostname()
js, err := broker.DialPinned(credential.natsURL(), credential.Fingerprint, nats.Name("build agent on "+host))
if err != nil {
return nil, "", err
}
// **The seat this machine serves is the one its credential claims** (novox/hq ADR 0190, the
// handover): the mesh issues a build machine's credential naming the seat its module claims,
// and one binary serves the old role as `builder` and the new as `build-agent` from that alone.
seat := link.BuildSeatClaimed(credential.seatsClaimed())
fmt.Fprintf(os.Stderr, "taking build work as a holder of %s\n", seat)
return js, seat, nil
}
// answer does one build and says what happened, whichever way it went.
func answer(ctx context.Context, publisher builder.Publisher, on, workspace string, work link.Build, h *holder) {
request := work.Request()
// **Its own context, so it can be killed alone** (novox/hq ADR 0219): cancelled by `kill`, it ends
// this build's commands and nothing else; the machine's own context ending — a SIGTERM — still
// reaches it through the parent, and that keeps today's meaning below.
building, cancel := context.WithCancel(ctx)
defer cancel()
var mine *running
if h != nil {
mine = h.begin(request, cancel)
defer h.end(mine)
}
// **First thing, and to stdout.** A build request that arrives and produces no visible line until
// it either finishes or fails is indistinguishable from one that never arrived — which cost a long
// diagnosis against a running mesh, chasing "the handler never fired" when the truth was only that
// the handler said nothing until the end.
fmt.Fprintf(os.Stderr, "a build request arrived for %s (%s)\n", request.Repository, request.ID)
// **Everything a build says goes two ways**: to stderr, as always, and onto the bus as the
// role's own events under the build's id (novox/hq ADR 0157) — so whoever asked, and anybody
// watching, reads the same lines this container's log holds, live, and after the fact from the
// stream. Said first, before anything runs, so a build that hangs is one that visibly started.
// One line at a time: a check's builds for its merge say their lines while its tests say theirs (novox/hq
// issue 716).
var saying sync.Mutex
say := func(step, message string) {
saying.Lock()
defer saying.Unlock()
fmt.Fprintf(os.Stderr, " [%s] %s\n", step, message)
work.Say(step, message)
if mine != nil && step != "run" && step != "output" {
h.stepped(mine, step)
}
}
builder.Said = say
defer func() { builder.Said = nil }()
if err := work.Began(ctx); err != nil {
fmt.Fprintf(os.Stderr, "cannot say a build started: %v\n", err)
}
// **A running check watches its pull request's key** (novox/hq ADR 0406 §4, issue 974): when the head or the trunk
// commit in it moves, the check is for a state nobody wants any more, and ends itself — nothing is asked of the forge
// (issue 726). A group's check, a try and a build watch nothing.
var superseded atomic.Bool
if w, ok := work.(interface {
WatchPull(context.Context, func(link.PullKey)) (func(), error)
}); ok && request.Check != nil {
stopWatching, err := w.WatchPull(building, func(k link.PullKey) {
superseded.Store(true)
say("superseded", fmt.Sprintf("the pull request's key moved to %.12s on %.12s while this checked %.12s on %.12s: stopped",
k.Head, k.Trunk, request.Ref, request.Check.Trunk))
cancel()
})
if err != nil {
fmt.Fprintf(os.Stderr, "the key of %s/%s#%d could not be watched, so its check runs on if the key moves: %v\n",
request.Check.Owner, request.Check.Repo, request.Check.Number, err)
} else {
defer stopWatching()
}
}
result := link.BuildResult{
ID: request.ID, Repository: request.Repository, Path: request.Path,
Ref: request.Ref, On: on, Source: request.Source, DryRun: request.DryRun,
}
what := "building " + request.Repository
if request.Path != "" {
what += " at " + request.Path
}
if request.Ref != "" {
what += " on " + request.Ref
}
say("build", what)
npmrc, err := packagesFrom(on)
if err == nil {
say("packages", saysRegistry(npmrc))
}
registryHere, isRegistry := publisher.(builder.Registry)
var built builder.Result
if request.Check != nil {
// **A pull request's merge check, not a build** (novox/hq to-be 45 §9): nothing is built,
// published or registered; the verdict is the outcome.
result.Checked = request.Check
registry := ""
if r, ok := publisher.(builder.Registry); ok {
registry = r.Address
}
spec := checkSpecOf(request)
// The judge cache the build agent declares as its own data, never in the workspace: only judge builds and
// mesh-lab's replays mount it, never a check's containers, so no pull request's script can write what a judge
// is built with or kept as, or what the replays run (novox/hq ADR 0358 §6, issue 709).
spec.JudgeCache = judgeCache()
// What the merge would move, built by the check for the walk to take (novox/hq ADR 0350 §1): into the
// scratch namespace. **Never with the package registry's credential**: this is code nobody has approved
// yet, and a recipe could read the credential it is handed (issue 462). A module that resolves packages
// at build time has no reuse key anyway, so nothing is lost by refusing it here.
if isRegistry && len(spec.Builds) > 0 {
spec.BuildWith = &builder.CheckBuilding{Publish: publisher, Scratch: registryHere, Held: request.Held,
Running: request.Check.Running, Seats: request.Seats, Named: namedOf(request.Check.Named)}
}
var v builder.CheckVerdict
v, err = builder.Check(building, builder.Command, spec, workspace, registry, forgeFrom(), say)
if err == nil {
result.Check = &link.CheckOutcome{Verdict: v.Verdict, Summary: v.Summary, Report: v.Report,
Took: v.Took.Round(time.Second).String(), Gate: layerOf(v.Gate), RepoCheck: layerOf(v.Repo),
Head: v.Head, Trunk: v.Trunk, Beside: v.Beside, Built: checkBuiltOf(v.Built),
Validator: v.Validator, Siblings: siblingsOut(v.Siblings), SiblingsUnknown: v.SiblingsUnknown, JudgeFrom: v.JudgeFrom}
// What each of a group's heads was judged as (novox/hq ADR 0323 §6).
for _, m := range v.Members {
result.Check.Members = append(result.Check.Members, link.CheckedMember{Owner: m.Owner, Repo: m.Repo,
Commit: m.Head, Trunk: m.Trunk})
}
// A branch tried before its pull request reads its outcome from this log, whole (novox/hq issue 686).
if request.Check.Try {
if line, err := json.Marshal(triedOf(v)); err == nil {
say(link.StepTried, string(line))
}
}
}
} else if err == nil {
// The package-registry credential is a build input, so it is resolved before the clone: a
// build that could not have resolved its dependencies is refused in front of the reason, not
// after a clone that then fails at npm ci.
// Every container it starts is labelled with its id, so a kill finds what outlived the
// docker client (ADR 0219).
if request.Reuse != nil && isRegistry && !request.DryRun {
// **A walk's build takes the check's build of the same key** (novox/hq ADR 0350 §2).
reuse := builder.Reuse{Scratch: registryHere}
for _, p := range request.Reuse.Pulls {
reuse.Pulls = append(reuse.Pulls, builder.PullRef{Owner: p.Owner, Repo: p.Repo, Number: p.Number})
}
// Only what a verdict names is taken (ADR 0350 §2).
reuse.Named = namedOf(request.Reuse.Builds)
built, err = builder.BuildReusing(building, builder.Labelled(builder.Command, request.ID), publisher, reuse,
request.Repository, request.Path, request.Ref, workspace, request.Held, npmrc,
forgeFrom(), say, request.Seats)
} else {
built, err = builder.Build(building, builder.Labelled(builder.Command, request.ID), publisher,
request.Repository, request.Path, request.Ref, workspace, request.Held, npmrc,
forgeFrom(), say, request.Seats)
}
}
// What it copied into the store, whatever became of the build (novox/hq ADR 0257).
result.Mirrored = built.Mirrored
// Only a build the kill ended: the kill came before its work did. One that finished — built, or
// failed on its own — in the moment the kill arrived says what it did, and the kill is refused.
killed := false
if mine != nil {
killed = h.returned(mine) && err != nil
}
if superseded.Load() && err != nil {
// Ended by its own key moving on (ADR 0406 §4): said as the check's failure, as a kill is.
result.Failed = link.SupersededByKey
say("failed", link.SupersededByKey)
} else if killed {
// **Killed by hand is the outcome, whatever the build was doing** (novox/hq ADR 0219): the
// error it ended with is the kill's consequence, not a fault of the source.
result.Failed = link.KilledByHand
say("failed", link.KilledByHand)
} else if err != nil {
// A failure is a result. A build that fails and says nothing is indistinguishable from a
// builder that is not running, and those want completely different responses.
result.Failed = err.Error()
say("failed", err.Error())
} else if request.Check != nil {
say("checked", result.Check.Verdict+": "+result.Check.Summary)
} else {
manifest, marshalErr := json.Marshal(built.Manifest)
if marshalErr != nil {
result.Failed = marshalErr.Error()
} else {
result.Commit = built.Commit
result.Manifest = manifest
for _, made := range built.Built {
result.Made = append(result.Made, link.MadeArtifact{
Name: made.Name, Kind: made.Kind, Reference: made.Reference,
})
}
result.Against = built.Against
result.SourceFingerprint = built.Source
result.SourceInputs = built.SourceInputs
// What the walk did instead of building, or why it built (novox/hq ADR 0350 §3, §4).
result.ReuseKey = built.ReuseKey
if built.Reused != nil {
f := built.Reused.From
result.Reused = built.Reused.Words()
result.ReusedFrom = &link.ReusedFrom{Pull: link.PullRef{Owner: f.Pull.Owner, Repo: f.Pull.Repo,
Number: f.Pull.Number}, Head: f.Head, Trunk: f.Trunk, Check: f.Check, Key: f.Key}
} else if request.Reuse != nil {
result.BuiltWhy = built.Why
if result.BuiltWhy == "" {
result.BuiltWhy = "it has no artifact to build"
}
}
result.Trunk, result.OnTrunk, result.Branches = built.Trunk, built.OnTrunk, built.Branches
for _, r := range built.Read {
result.Read = append(result.Read, link.ReadRepository{Repository: r.Repository, Ref: r.Ref})
}
for _, s := range built.Sources {
result.Sources = append(result.Sources, link.BuildSource{Repository: s.Repository, Ref: s.Ref, Paths: s.Paths})
}
// The list of predefined keys the manifest was judged against, for the controller to publish
// (novox/hq ADR 0348 §7).
if p := built.Predefined; p != nil {
result.Predefined = &link.PredefinedList{Keys: p.Keys, Generation: p.Generation}
}
say("built", built.Manifest.Module+" from "+short(built.Commit))
}
}
// A killed build is announced on a context of its own: the build's was the one cancelled, and the
// outcome must go out and the ask be settled — acknowledged, never redelivered to another machine
// to be built again. A machine being stopped is the other case and keeps its meaning: the
// parent's context is gone, nothing is announced or settled, and the ask is redelivered.
announcing := ctx
if killed {
fresh, stop := context.WithTimeout(context.Background(), 30*time.Second)
defer stop()
announcing = fresh
}
announceErr := work.Announce(announcing, result)
if mine != nil {
h.said(mine, result.Failed, announceErr == nil)
}
if err := announceErr; err != nil {
// Said, not fatal: the build happened. A build reported as failed because announcing it
// failed is a lie about work that was done — and the request stays unsettled below only if
// nothing was said at all, so another machine can try.
fmt.Fprintf(os.Stderr, "cannot say what came of a build: %v\n", err)
return
}
// Settled only once the outcome is away, so a machine that dies before answering leaves the work
// for another rather than losing it.
if err := work.Done(); err != nil {
fmt.Fprintf(os.Stderr, "the outcome is away and the request could not be settled: %v\n", err)
}
}
// checkSpecOf is a check request as the builder runs it.
func checkSpecOf(request link.BuildRequest) builder.CheckSpec {
c := request.Check
spec := builder.CheckSpec{ID: request.ID, Repository: request.Repository, Ref: request.Ref,
Owner: c.Owner, Repo: c.Repo, Number: c.Number, Paths: c.Paths, Beside: map[string]builder.Beside{},
Modules: c.Modules, New: c.New, Manifests: c.Manifests, Judge: c.Judge, Base: c.Base,
Toolchain: builder.ToolchainOf(request.Held), Toolchains: builder.ToolchainsOf(request.Held),
Tests: c.Tests, StoreOnly: c.StoreOnly, SiblingsUnknown: c.SiblingsUnknown}
for _, s := range c.Siblings {
spec.Siblings = append(spec.Siblings, builder.Sibling{Repository: s.Repository, Number: s.Number, Head: s.Head,
Base: s.Base, Dir: s.Dir})
}
for dir, b := range c.Beside {
spec.Beside[dir] = builder.Beside{Repository: b.Repository, Ref: b.Ref}
}
for _, b := range c.Builds {
spec.Builds = append(spec.Builds, builder.CheckBuildSpec{Module: b.Module, Repository: b.Repository, Path: b.Path,
Ref: b.Ref, Merged: b.Merged, Tier: b.Tier, Skip: b.Skip})
}
for _, m := range c.Members {
spec.Group = append(spec.Group, builder.GroupHead{Owner: m.Owner, Repo: m.Repo, Repository: m.Repository,
Ref: m.Ref, Paths: m.Paths, Base: m.Base})
}
return spec
}
// siblingsOut is the siblings a check read as the outcome carries them (novox/hq ADR 0382 §4).
func siblingsOut(read []builder.JudgedSibling) []link.CheckSibling {
var out []link.CheckSibling
for _, s := range read {
out = append(out, link.CheckSibling{Repository: s.Repository, Number: s.Number, Head: s.Head, Base: s.Base,
BaseCommit: s.BaseCommit})
}
return out
}
// judgeCache is the judge cache the build agent's manifest declares (MESH_JUDGE_CACHE, novox/hq ADR 0358 §6): a
// directory mounted at the same path on the host, so a judge build's mounts name what the agent sees. Empty when the
// agent declares none, and every judge is then built cold under its check's root.
func judgeCache() string { return os.Getenv("MESH_JUDGE_CACHE") }
// triedOf is a try's outcome as its log line says it (novox/hq issue 686): the repository's own check's verdict,
// or the check's where that did not run — a head that does not merge — and the lines naming what failed.
func triedOf(v builder.CheckVerdict) link.Tried {
t := link.Tried{Verdict: v.Verdict, Summary: v.Summary, Took: v.Took.Round(time.Second).String(),
Head: v.Head, Trunk: v.Trunk}
if v.Repo != nil && (v.Gate == nil || v.Gate.Verdict == "pass") {
t.Verdict, t.Summary = v.Repo.Verdict, v.Repo.Summary
}
if v.Repo != nil && v.Repo.Failed != "" {
t.Failed = strings.Split(v.Repo.Failed, "\n")
}
if t.Verdict == "" {
t.Verdict = "error"
}
return t
}
// namedOf are the builds the seat's verdicts name, as the builder believes a scratch record by them (novox/hq ADR
// 0350 §2, issue 703): each module's key with every artifact's digest.
func namedOf(builds []link.CheckedBuild) []builder.NamedBuild {
var out []builder.NamedBuild
for _, b := range builds {
named := builder.NamedBuild{Pull: builder.PullRef{Owner: b.Pull.Owner, Repo: b.Pull.Repo,
Number: b.Pull.Number}, Check: b.Check, Module: b.Built.Module, Key: b.Built.Key,
Digests: map[string]string{}}
for _, m := range b.Built.Made {
named.Digests[m.Name] = builder.DigestOf(m.Reference)
}
out = append(out, named)
}
return out
}
// checkBuiltOf is a check's builds for its merge as the outcome carries them (novox/hq ADR 0350 §1).
func checkBuiltOf(built []builder.CheckBuilt) []link.CheckBuilt {
var out []link.CheckBuilt
for _, b := range built {
c := link.CheckBuilt{Module: b.Module, Key: b.Key, NotBuilt: b.NotBuilt, Took: b.Took.Round(time.Second).String(),
Kept: b.Kept, Same: b.Same, SameAs: b.SameAs}
for _, m := range b.Made {
c.Made = append(c.Made, link.MadeArtifact{Name: m.Name, Kind: m.Kind, Reference: m.Reference})
}
out = append(out, c)
}
return out
}
// layerOf is one layer of a check as the outcome carries it.
func layerOf(l *builder.Layer) *link.CheckLayer {
if l == nil {
return nil
}
return &link.CheckLayer{Verdict: l.Verdict, Summary: l.Summary, Modules: l.Modules}
}
// packagesFrom is where a build resolves the mesh's own published packages — the SDK above all
// (novox/hq ADR 0076, issue 053) — on this machine.
//
// **The binding wins, then the genesis document** (novox/hq ADR 0336 §3, §4). The build agent's
// npm-package-registry requirement is optional: on genesis's node, before the gitea module is assigned,
// nothing provides it, and the installer sealed the forge it raised to this agent as one document
// instead. Once the requirement binds, the binding names the endpoint the way the artifact store's
// does and a sealed token file the credential, and the genesis document is left unread. A binding
// file named and absent is a requirement not bound here, never a fault: the environment names the
// path whether or not the mesh wrote it.
//
// The environment variables remain for a builder run by a person. With none of the three the result
// is disabled and says this machine, so a build that needs the registry is refused naming it (§6),
// and a build that needs none builds anyway.
func packagesFrom(machine string) (builder.Npmrc, error) {
scope := strings.TrimSpace(os.Getenv("MESH_NPM_SCOPE"))
if scope == "" {
scope = "@novox"
}
none := builder.Npmrc{Machine: machine}
if path := strings.TrimSpace(os.Getenv("MESH_PACKAGE_BINDING")); path != "" {
raw, err := os.ReadFile(path)
switch {
case errors.Is(err, os.ErrNotExist):
// Not bound on this machine: the genesis document, or nothing.
case err != nil:
return none, fmt.Errorf("cannot read what the mesh said about the package registry: %w", err)
default:
return fromBinding(path, raw, scope, machine)
}
}
if path := strings.TrimSpace(os.Getenv("MESH_GENESIS_PACKAGE_REGISTRY")); path != "" {
raw, err := os.ReadFile(path)
switch {
case errors.Is(err, os.ErrNotExist):
// Not this mesh's genesis node, or the installer has not sealed it yet.
case err != nil:
return none, fmt.Errorf("cannot read the genesis document for the package registry: %w", err)
default:
return fromGenesis(path, raw, scope, machine)
}
}
// A builder run by a person.
registry := strings.TrimSpace(os.Getenv("MESH_NPM_REGISTRY"))
secret := strings.TrimSpace(os.Getenv("MESH_NPM_TOKEN"))
username := strings.TrimSpace(os.Getenv("MESH_NPM_USER"))
if registry == "" && secret == "" {
return none, nil
}
n := builder.Npmrc{Scope: scope, Registry: registry, From: builder.FromEnvironment, Machine: machine}
if username != "" {
n.Username, n.Password = username, secret
} else {
n.Token = secret
}
return n, nil
}
// fromBinding composes the registry from the npm-package-registry binding the mesh wrote and the
// credential it sealed beside it. The mesh authenticates the ordinary way — a generated password the
// provider only applies (novox/hq ADR 0048) — so with a username the secret is a password (basic
// auth); without one it is a bearer token a provider minted.
func fromBinding(path string, raw []byte, scope, machine string) (builder.Npmrc, error) {
none := builder.Npmrc{Machine: machine}
var told struct {
From string `json:"from"`
At string `json:"at"`
As string `json:"as"`
Serves map[string]any `json:"serves"`
}
if err := json.Unmarshal(raw, &told); err != nil {
return none, fmt.Errorf("%s is not a binding: %w", path, err)
}
if told.At == "" {
return none, fmt.Errorf(
"%s says the package registry is on %q and gives no address for it", path, told.From)
}
// Composed from what the provider serves, so nothing here knows gitea's URL shape from
// another registry's: it states its port, the path its registry answers on, and the scheme.
scheme := "https"
if s, ok := told.Serves["scheme"]; ok {
scheme = fmt.Sprintf("%v", s)
}
port, ok := told.Serves["port"]
if !ok {
return none, fmt.Errorf("%s says nothing about which port the package registry answers on", path)
}
npmPath, ok := told.Serves["npm-path"]
if !ok {
return none, fmt.Errorf("%s says nothing about the path the package registry answers on", path)
}
secret := strings.TrimSpace(os.Getenv("MESH_NPM_TOKEN"))
if file := strings.TrimSpace(os.Getenv("MESH_NPM_TOKEN_FILE")); file != "" {
raw, err := os.ReadFile(file)
if err != nil {
return none, fmt.Errorf("cannot read this builder's package-registry credential: %w", err)
}
secret = strings.TrimSpace(string(raw))
}
n := builder.Npmrc{Scope: scope, Registry: fmt.Sprintf("%s://%s:%v%v", scheme, told.At, port, npmPath),
From: builder.FromBinding, Machine: machine}
if told.As != "" {
n.Username, n.Password = told.As, secret
} else {
n.Token = secret
}
return n, nil
}
// GenesisRegistry is the genesis document (novox/hq ADR 0336 §3): what the installer sealed to the
// build agent on its node at genesis's packages step, as the `genesis-package-registry` own-secret —
// the forge it raised, the port this node gave the package registry, its npm path, and the account it
// made with that account's password. mesh-host writes it; the field names are the two repositories'
// agreement.
type GenesisRegistry struct {
Scheme string `json:"scheme"`
At string `json:"at"`
Port int `json:"port"`
NpmPath string `json:"npm-path"`
As string `json:"as"`
Password string `json:"password"`
}
// fromGenesis composes the registry from the genesis document.
func fromGenesis(path string, raw []byte, scope, machine string) (builder.Npmrc, error) {
none := builder.Npmrc{Machine: machine}
var g GenesisRegistry
if err := json.Unmarshal(raw, &g); err != nil {
return none, fmt.Errorf("%s is not the genesis document for the package registry: %w", path, err)
}
var missing []string
for name, value := range map[string]string{"at": g.At, "npm-path": g.NpmPath, "as": g.As, "password": g.Password} {
if strings.TrimSpace(value) == "" {
missing = append(missing, name)
}
}
if g.Port <= 0 {
missing = append(missing, "port")
}
if len(missing) > 0 {
sort.Strings(missing)
return none, fmt.Errorf("%s, the genesis document for the package registry, says nothing of %s",
path, strings.Join(missing, ", "))
}
scheme := g.Scheme
if scheme == "" {
scheme = "http"
}
return builder.Npmrc{Scope: scope, Registry: fmt.Sprintf("%s://%s:%d%s", scheme, g.At, g.Port, g.NpmPath),
Username: g.As, Password: g.Password, From: builder.FromGenesis, Machine: machine}, nil
}
// saysRegistry is the line each build's log carries about the package registry (novox/hq ADR 0336
// §4): which of the binding and the genesis document it resolved from, or that it has neither here.
func saysRegistry(n builder.Npmrc) string {
if n.Enabled() {
return "the package registry is resolved from " + n.From
}
machine := n.Machine
if machine == "" {
machine = "this machine"
}
return fmt.Sprintf("no package registry: %s is unbound on %s and no genesis document is here; "+
"a build needing it is refused", builder.RegistryRequirement, machine)
}
// forgeFrom is the git credential this builder may offer a clone, composed from the same binding
// and sealed secret its package-registry half already reads: the forge that answers npm is the
// forge that hosts the repositories, and its provisioner applies one password to one user for
// both. Anything missing means no credential, and every clone stays anonymous — which is all a
// mesh of public repositories ever needs.
//
// The URL names the binding's own address — the machine the mesh says the forge is on — so a
// private repository is registered and built by that address, and a clone of anything else is
// never shown this credential (git's credential store matches the whole origin).
func forgeFrom() builder.GitCredential {
path := strings.TrimSpace(os.Getenv("MESH_PACKAGE_BINDING"))
if path == "" {
return builder.GitCredential{}
}
raw, err := os.ReadFile(path)
if err != nil {
return builder.GitCredential{}
}
var told struct {
At string `json:"at"`
As string `json:"as"`
Serves map[string]any `json:"serves"`
}
if err := json.Unmarshal(raw, &told); err != nil || told.At == "" || told.As == "" {
return builder.GitCredential{}
}
secret := strings.TrimSpace(os.Getenv("MESH_NPM_TOKEN"))
if file := strings.TrimSpace(os.Getenv("MESH_NPM_TOKEN_FILE")); file != "" {
if raw, err := os.ReadFile(file); err == nil {
secret = strings.TrimSpace(string(raw))
}
}
if secret == "" {
return builder.GitCredential{}
}
scheme := "https"
if s, ok := told.Serves["scheme"]; ok {
scheme = fmt.Sprintf("%v", s)
}
host := told.At
if port, ok := told.Serves["port"]; ok {
host = fmt.Sprintf("%s:%v", told.At, port)
}
made := url.URL{Scheme: scheme, User: url.UserPassword(told.As, secret), Host: host}
return builder.GitCredential{URL: made.String()}
}
func short(commit string) string {
if len(commit) > 8 {
return commit[:8]
}
return commit
}
// whereToPublish is the artifact store this builder uses.
//
// **Preferably from the mesh.** A builder that is a module requires an artifact store, and the
// mesh writes it a file saying which machine answers that and on what port — the same binding any
// consumer of any provision gets. Reading it means the address is not a setting somebody keeps in
// step by hand, and moving the store is an ordinary reassignment rather than an edit on every
// build machine.
//
// The environment variable remains for a builder run by a person, which is how this started and
// how it is still run while being developed.
func whereToPublish() (string, error) {
binding := strings.TrimSpace(os.Getenv("MESH_BINDING"))
if binding == "" {
registry := strings.TrimSpace(os.Getenv("MESH_REGISTRY"))
if registry == "" {
return "", fmt.Errorf("neither MESH_BINDING nor MESH_REGISTRY: a built artifact " +
"nobody can fetch is not built")
}
return registry, nil
}
raw, err := os.ReadFile(binding)
if err != nil {
return "", fmt.Errorf("cannot read what the mesh said about the artifact store: %w", err)
}
var told struct {
From string `json:"from"`
At string `json:"at"`
Serves map[string]any `json:"serves"`
}
if err := json.Unmarshal(raw, &told); err != nil {
return "", fmt.Errorf("%s is not a binding: %w", binding, err)
}
if told.At == "" {
// The provider is not on the private network, so there is no name to reach it by. Said
// rather than falling back to the machine's own name, which would publish to a store on
// the wrong machine and be found out much later.
return "", fmt.Errorf(
"%s says the artifact store is on %q and gives no address for it", binding, told.From)
}
port, ok := told.Serves["port"]
if !ok {
return "", fmt.Errorf("%s says nothing about which port the artifact store answers on",
binding)
}
return fmt.Sprintf("%s:%v", told.At, port), nil
}
// brokerFrom is where this builder connects, and with what.
//
// **Preferably from a file the mesh sealed to this machine.** A builder that is a module is given
// its credential the way every other module is given one: generated or accepted centrally, sealed
// to the machine, written by the host. Putting it in an environment variable instead would mean
// the one copy that matters passing through a terminal and a process listing.
//
// The variable remains for a builder run by a person.
func brokerFrom() (Credential, error) {
if path := strings.TrimSpace(os.Getenv("MESH_BROKER_FILE")); path != "" {
raw, err := os.ReadFile(path)
if err != nil {
return Credential{}, fmt.Errorf("cannot read this builder's credential: %w", err)
}
said := strings.TrimSpace(string(raw))
if said == "" {
// An empty credential file is a machine that will connect as nobody and be refused,
// with the reason three layers away.
return Credential{}, fmt.Errorf("%s is empty, so this builder has no credential", path)
}
var held Credential
if err := json.Unmarshal([]byte(said), &held); err == nil && held.URL != "" {
return held, nil
}
// A file holding only a URL, which is what a person writing one by hand produces. The
// broker is then verified against whatever this machine already trusts.
return Credential{URL: said}, nil
}
return Credential{}, fmt.Errorf(
"no MESH_BROKER_FILE: a build machine with no credential for the bus has nothing to build")
}
// Credential is what a build machine is given so it can reach the broker.
//
// Two things, because reaching a broker over TLS needs both: who to connect as, and what to check
// the certificate against. A mesh's broker presents a certificate of the mesh's own, which is in
// no public trust store, so a URL alone can only connect to a broker somebody else vouches for.
//
// **The same shape a node gets, for the same reason** (novox/hq ADR 0004): the fingerprint travels
// out of band — here, sealed with the credential — and the endpoint is verified once at connect.
type Credential struct {
URL string `json:"url"`
Fingerprint string `json:"fingerprint,omitempty"`
// User and Password ride beside the address on the bus being built (design 25): a credential
// embedded in a URL leaks into every log line that prints a connection, so the mesh seals them
// as two fields and this machine joins them once, here, to dial.
User string `json:"user,omitempty"`
Password string `json:"password,omitempty"`
// Claims are the seats the module this credential was issued for claims, as the mesh writes
// them beside the credential (novox/hq ADR 0159). The first is the build role this machine
// serves; a credential naming none is from before claims travelled in it.
Claims []struct {
Seat string `json:"seat"`
} `json:"claims,omitempty"`
}
// seatsClaimed is the seats the credential names, in order.
func (c Credential) seatsClaimed() []string {
out := make([]string, 0, len(c.Claims))
for _, claim := range c.Claims {
out = append(out, claim.Seat)
}
return out
}
// onTheNewBus is whether a credential is for the bus being built: its address says so, and the
// mesh only ever seals such a credential with the user and password beside it.
func (c Credential) onTheNewBus() bool { return strings.HasPrefix(strings.TrimSpace(c.URL), "nats://") }
// natsURL is the address with this machine's credential in it, for the one dial that needs it.
func (c Credential) natsURL() string {
rest := strings.TrimPrefix(strings.TrimSpace(c.URL), "nats://")
if c.User == "" {
return "nats://" + rest
}
return "nats://" + c.User + ":" + c.Password + "@" + rest
}
+188
View File
@@ -0,0 +1,188 @@
package main
import (
"context"
"encoding/json"
"errors"
"flag"
"fmt"
"os"
"strings"
"github.com/novox/mesh-controller/internal/builder"
)
// buildOnce is the builder doing one build and stopping, with no broker and no mesh.
//
// **This is how a mesh is raised** (novox/hq ADR 0073). The installer carries this program and runs
// it once, before anything else exists, to produce the control plane from the same repository and
// path that every later rebuild of it will use. What raises the mesh is therefore the same thing
// that maintains it — not a second mechanism that has to be kept in step with the first and is
// exercised once per new mesh, which is how often enough to rot.
//
// It takes no work from a queue and answers nobody: there is no broker yet, and the only thing
// waiting for the answer is the installer that started it. So the result goes to standard output as
// JSON, which is what the installer reads.
//
// With no --registry it publishes nowhere and the image stays in this machine's container runtime,
// named by the digest of its own configuration (see builder.Local). That is the genesis case. With
// one, it publishes as it always does — the same command is how an operator builds a module by hand
// on a mesh that already exists.
func buildOnce(ctx context.Context, args []string) error {
set := flag.NewFlagSet("build", flag.ContinueOnError)
path := set.String("path", "", "the module's directory inside the repository (default: its root)")
ref := set.String("ref", "", "the commit to build; a branch is a moving target somebody else controls")
registry := set.String("registry", "",
"host:port to publish to. Without it the artifacts stay in this machine's container runtime, which is the genesis case")
workspace := set.String("workspace", "", "where to clone and build (default: a temporary directory)")
on := multiple(set, "on",
"a base this build stands on, as <module>/<artifact>=<reference>. Repeatable")
positionals, err := parseAround(set, args)
if err != nil {
return err
}
if len(positionals) != 1 {
return errors.New("mesh-builder build <repository> [--path P] [--ref COMMIT] [--registry HOST:PORT]")
}
repository := positionals[0]
where := *workspace
if where == "" {
where = os.TempDir() + "/mesh-builder-once"
}
// What this build may stand on, said on the command line because there is no mesh to ask.
//
// **This is the genesis case and it is meant to be awkward.** On a running mesh the control
// plane answers this, because only it knows what this mesh holds. Here nothing has been built
// yet, so whoever runs this says what to use — and for the control plane, which is the one
// module raised before anything else exists, the answer is ordinarily nothing at all.
bases := map[string]string{}
for _, pair := range *on {
key, reference, found := strings.Cut(pair, "=")
if !found || key == "" || reference == "" {
return fmt.Errorf(
"--on takes <module>/<artifact>=<reference>, and %q is not that", pair)
}
bases[key] = reference
}
// Local unless told otherwise, because the moment this exists for has nowhere to publish. A
// default pointing at a registry would mean genesis failing at a push to something that is not
// there yet, one step away from the thing that could explain it.
var publisher builder.Publisher = builder.Local{Run: builder.Command}
if *registry != "" {
publisher = builder.Registry{Address: *registry, Run: builder.Command}
}
fmt.Fprintf(os.Stderr, "building %s", repository)
if *path != "" {
fmt.Fprintf(os.Stderr, " at %s", *path)
}
if *ref != "" {
fmt.Fprintf(os.Stderr, " at %s", *ref)
}
fmt.Fprintln(os.Stderr)
machine := os.Getenv("MESH_NODE")
if machine == "" {
machine, _ = os.Hostname()
}
npmrc, err := packagesFrom(machine)
if err != nil {
return err
}
fmt.Fprintf(os.Stderr, " [packages] %s\n", saysRegistry(npmrc))
built, buildErr := builder.Build(ctx, builder.Command, publisher, repository, *path, *ref, where, bases, npmrc,
forgeFrom(),
func(step, message string) { fmt.Fprintf(os.Stderr, " [%s] %s\n", step, message) })
if buildErr != nil {
return buildErr
}
// To standard output, and everything else to standard error, so the caller can read this
// without having to separate it from progress.
out := onceResult{
Module: built.Manifest.Module,
Commit: built.Commit,
Repository: repository,
Path: *path,
Ref: *ref,
Manifest: built.Manifest,
Against: built.Against,
Source: built.Source,
}
for _, r := range built.Read {
out.Read = append(out.Read, readRepository{Repository: r.Repository, Ref: r.Ref})
}
for _, made := range built.Built {
out.Made = append(out.Made, madeArtifact{Name: made.Name, Kind: made.Kind, Reference: made.Reference})
}
body, err := json.MarshalIndent(out, "", " ")
if err != nil {
return err
}
fmt.Println(string(body))
fmt.Fprintf(os.Stderr, " built %s from %s\n", built.Manifest.Module, short(built.Commit))
return nil
}
// onceResult is what one build reports to whoever started it.
//
// The same fields the mesh records for a build, so a reader comparing a genesis build against an
// ordinary one is comparing the same thing said the same way.
type onceResult struct {
Module string `json:"module"`
Commit string `json:"commit"`
Repository string `json:"repository"`
Path string `json:"path,omitempty"`
Ref string `json:"ref,omitempty"`
Manifest any `json:"manifest"`
Made []madeArtifact `json:"made"`
Against []string `json:"against,omitempty"`
Read []readRepository `json:"read,omitempty"`
Source string `json:"source-fingerprint,omitempty"`
}
// readRepository is a repository this build read source from besides the module's own.
type readRepository struct {
Repository string `json:"repository"`
Ref string `json:"ref,omitempty"`
}
type madeArtifact struct {
Name string `json:"name"`
Kind string `json:"kind"`
Reference string `json:"reference"`
}
// parseAround lets flags appear on either side of the repository, because a person writing this by
// hand will put them wherever reads best and the standard parser stops at the first thing that is
// not a flag. The same helper the control plane's commands use, for the same reason.
func parseAround(set *flag.FlagSet, args []string) ([]string, error) {
var positionals []string
rest := args
for {
if err := set.Parse(rest); err != nil {
return nil, err
}
rest = set.Args()
if len(rest) == 0 {
return positionals, nil
}
positionals = append(positionals, rest[0])
rest = rest[1:]
}
}
// multiple is a flag that may be given more than once.
type repeated []string
func (r *repeated) String() string { return strings.Join(*r, ", ") }
func (r *repeated) Set(v string) error { *r = append(*r, v); return nil }
func multiple(set *flag.FlagSet, name, usage string) *[]string {
var values repeated
set.Var(&values, name, usage)
return (*[]string)(&values)
}
+102
View File
@@ -0,0 +1,102 @@
package main
import (
"os"
"path/filepath"
"strings"
"testing"
"github.com/novox/mesh-controller/internal/builder"
)
// The build agent's package registry (novox/hq ADR 0336, issue 121): its npm-package-registry
// requirement is optional, so the binding may be absent, and on genesis's node the installer seals
// the forge it raised as one genesis document instead.
// registryFiles names the three paths the build agent's environment always names, writing only those
// given.
func registryFiles(t *testing.T, binding, token, genesis string) {
t.Helper()
dir := t.TempDir()
for env, content := range map[string]string{
"MESH_PACKAGE_BINDING": binding,
"MESH_NPM_TOKEN_FILE": token,
"MESH_GENESIS_PACKAGE_REGISTRY": genesis,
} {
path := filepath.Join(dir, env)
t.Setenv(env, path)
if content != "" {
if err := os.WriteFile(path, []byte(content), 0o600); err != nil {
t.Fatal(err)
}
}
}
for _, env := range []string{"MESH_NPM_REGISTRY", "MESH_NPM_TOKEN", "MESH_NPM_USER", "MESH_NPM_SCOPE"} {
t.Setenv(env, "")
}
}
const (
aBinding = `{"from":"gitea","at":"anchor.mesh","as":"anchor-agent",
"serves":{"scheme":"http","port":3000,"npm-path":"/api/packages/novox/npm/"}}`
aGenesisDocument = `{"scheme":"http","at":"127.0.0.1","port":3100,"npm-path":"/api/packages/novox/npm/",
"as":"mesh-builder","password":"genesis-secret"}`
)
func TestUntilTheRequirementBindsTheGenesisDocumentIsRead(t *testing.T) {
registryFiles(t, "", "", aGenesisDocument)
n, err := packagesFrom("anchor")
if err != nil {
t.Fatalf("an absent binding beside a genesis document was refused: %v", err)
}
if n.From != builder.FromGenesis || n.Registry != "http://127.0.0.1:3100/api/packages/novox/npm/" ||
n.Username != "mesh-builder" || n.Password != "genesis-secret" || n.Scope != "@novox" {
t.Fatalf("the genesis document was not what the registry came from: %+v", n)
}
file, err := n.File()
if err != nil || !strings.Contains(file, "//127.0.0.1:3100/api/packages/novox/npm/:username=mesh-builder") {
t.Fatalf("the .npmrc is not written from the genesis document: %v\n%s", err, file)
}
if got := saysRegistry(n); !strings.Contains(got, "genesis document") {
t.Errorf("the log does not say the genesis document: %s", got)
}
}
func TestARealBindingWinsOverTheGenesisDocument(t *testing.T) {
registryFiles(t, aBinding, "bound-secret\n", aGenesisDocument)
n, err := packagesFrom("anchor")
if err != nil {
t.Fatal(err)
}
if n.From != builder.FromBinding || n.Registry != "http://anchor.mesh:3000/api/packages/novox/npm/" ||
n.Username != "anchor-agent" || n.Password != "bound-secret" {
t.Fatalf("the binding did not win over the genesis document: %+v", n)
}
if got := saysRegistry(n); !strings.Contains(got, "binding") {
t.Errorf("the log does not say the binding: %s", got)
}
}
func TestWithNeitherTheAgentHasNoRegistryAndSaysWhere(t *testing.T) {
registryFiles(t, "", "", "")
n, err := packagesFrom("workstation")
if err != nil {
t.Fatalf("an unbound optional requirement was a fault: %v", err)
}
if n.Enabled() || n.Machine != "workstation" {
t.Fatalf("a registry came from nowhere: %+v", n)
}
got := saysRegistry(n)
for _, want := range []string{"npm-package-registry", "unbound", "workstation"} {
if !strings.Contains(got, want) {
t.Errorf("the log does not say %q: %s", want, got)
}
}
}
func TestAGenesisDocumentMissingAFieldIsRefusedByName(t *testing.T) {
registryFiles(t, "", "", `{"at":"127.0.0.1","port":3000,"npm-path":"/api/packages/novox/npm/","as":"mesh-builder"}`)
if _, err := packagesFrom("anchor"); err == nil || !strings.Contains(err.Error(), "password") {
t.Fatalf("a genesis document with no password was taken: %v", err)
}
}
+44
View File
@@ -0,0 +1,44 @@
package main
import (
"errors"
"fmt"
"os"
"path/filepath"
)
// **The build agent says it is ready** (novox/hq issue 544). It listens on nothing, so no http or tcp
// check can see it work; it writes a mark once it is bound to its worker and pulls build work, and its
// container's health check runs `mesh-builder ready`, which passes only while the mark is there. Without
// it the node-engine could judge the agent only alive, and every walk that moved it waited out the whole
// grace on each machine.
//
// The mark lives in the container's own temporary directory, never in the workspace the machine keeps:
// a new container starts without it, and a restarted one removes it before it binds again.
func readyMark() string { return filepath.Join(os.TempDir(), "mesh-builder.ready") }
// notReady removes the mark: a holder starting is not taking work until it says so again.
func notReady() error {
if err := os.Remove(readyMark()); err != nil && !errors.Is(err, os.ErrNotExist) {
return fmt.Errorf("cannot remove the mark that this agent takes build work (%s): %w", readyMark(), err)
}
return nil
}
// taking writes the mark. A mark that cannot be written is said, and the agent goes on building: its
// check then says it is not ready, which is the truth the gate should hear.
func taking(seat string) {
if err := os.WriteFile(readyMark(), []byte(seat+"\n"), 0o644); err != nil {
fmt.Fprintf(os.Stderr, "cannot mark that this agent takes build work (%s): %v\n", readyMark(), err)
return
}
fmt.Fprintf(os.Stderr, "taking build work from %s\n", seat)
}
// ready is the health check: nil when this container's agent is bound and pulling build work.
func ready() error {
if _, err := os.Stat(readyMark()); err != nil {
return fmt.Errorf("not taking build work yet: %s is not there", readyMark())
}
return nil
}
+32
View File
@@ -0,0 +1,32 @@
package main
import (
"os"
"testing"
)
// The health check passes only between the agent saying it takes work and its next start (novox/hq
// issue 544).
func TestTheAgentIsReadyOnlyOnceItTakesWork(t *testing.T) {
t.Setenv("TMPDIR", t.TempDir())
if err := ready(); err == nil {
t.Fatal("an agent that never took work is ready")
}
taking("mesh-build-machine")
if err := ready(); err != nil {
t.Fatalf("an agent taking work is not ready: %v", err)
}
// Started again, it is not ready until it binds again.
if err := notReady(); err != nil {
t.Fatal(err)
}
if err := ready(); err == nil {
t.Fatal("an agent starting again is ready before it takes work")
}
if err := notReady(); err != nil {
t.Fatalf("removing a mark that is not there failed: %v", err)
}
if _, err := os.Stat(readyMark()); err == nil {
t.Fatal("the mark is still there")
}
}
+27
View File
@@ -0,0 +1,27 @@
package main
import (
"encoding/json"
"testing"
"github.com/novox/mesh-controller/internal/link"
)
// The seat a build machine serves comes from its credential (novox/hq ADR 0190 handover).
func TestTheCredentialSaysWhichBuildRoleThisMachineServes(t *testing.T) {
var held Credential
if err := json.Unmarshal([]byte(`{"url":"nats://bus:4222","user":"anchor.builder","password":"x",
"claims":[{"seat":"mesh-build-machine","scope":"mesh","serves":[]}]}`), &held); err != nil {
t.Fatal(err)
}
if got := link.BuildSeatClaimed(held.seatsClaimed()); got != "mesh-build-machine" {
t.Errorf("the old builder's credential serves %q", got)
}
var bare Credential
if err := json.Unmarshal([]byte(`{"url":"nats://bus:4222","user":"anchor.build-agent","password":"x"}`), &bare); err != nil {
t.Fatal(err)
}
if got := link.BuildSeatClaimed(bare.seatsClaimed()); got != link.TheBuildMachine {
t.Errorf("a credential without claims serves %q, want %s", got, link.TheBuildMachine)
}
}
+45
View File
@@ -0,0 +1,45 @@
package main
import (
"os"
"strings"
"testing"
)
// **The genesis path writes its result to stdout and nothing else there.** The installer captures
// stdout alone and parses it as JSON; one stray log line makes that fail with "invalid character"
// — exactly what a builder fmt.Printf caused, breaking a build that had worked. This keeps
// diagnostics on stderr so a future print cannot repeat it silently.
func TestBuilderDiagnosticsStayOffStdout(t *testing.T) {
allowed := map[string]bool{
"string(body)": true, // once.go: the result JSON, which IS stdout
"version)": true, // --version
`"stopping")`: true, // the loop.s shutdown line
"usage)": true, // --help text, for a human
"said)": true, // busy: the answer its node-engine reads (novox/hq ADR 0335)
}
for _, file := range []string{"once.go", "main.go"} {
src, err := os.ReadFile(file)
if err != nil {
t.Fatal(err)
}
for i, raw := range strings.Split(string(src), "\n") {
line := strings.TrimSpace(raw)
if !strings.HasPrefix(line, "fmt.Printf(") &&
!strings.HasPrefix(line, "fmt.Println(") &&
!strings.HasPrefix(line, "fmt.Print(") {
continue
}
ok := false
for token := range allowed {
if strings.Contains(line, token) {
ok = true
}
}
if !ok {
t.Errorf("%s:%d writes a diagnostic to stdout, which the installer parses as JSON:\n %s",
file, i+1, line)
}
}
}
}
+31
View File
@@ -0,0 +1,31 @@
package main
import (
"testing"
"time"
"github.com/novox/mesh-controller/internal/builder"
)
// A try's outcome, as the build seat says it in the build's log (novox/hq issue 686): the repository's own
// check's verdict, how long it took, and each line naming what failed.
func TestATrysOutcomeNamesWhatFailed(t *testing.T) {
tried := triedOf(builder.CheckVerdict{Verdict: "pass", Summary: "the change touches no module", Took: 303 * time.Second,
Head: "c0ffee", Trunk: "ca3f43c", Gate: &builder.Layer{Verdict: "pass"},
Repo: &builder.Layer{Verdict: "fail", Summary: "its merge-check.sh failed: TestX",
Failed: "--- FAIL: TestX (0.01s)\n x_test.go:12: got 1, want 2"}})
if tried.Verdict != "fail" || tried.Took != "5m3s" || len(tried.Failed) != 2 || tried.Trunk != "ca3f43c" ||
tried.Summary != "its merge-check.sh failed: TestX" {
t.Fatalf("a failed try: %+v", tried)
}
// A head that does not merge fails both layers: the check's own word.
conflict := triedOf(builder.CheckVerdict{Verdict: "fail", Summary: "does not merge", Gate: &builder.Layer{Verdict: "fail"},
Repo: &builder.Layer{Verdict: "fail", Summary: "does not merge"}})
if conflict.Verdict != "fail" || conflict.Summary != "does not merge" {
t.Fatalf("a try that does not merge: %+v", conflict)
}
// No verdict is an error, never a pass.
if none := triedOf(builder.CheckVerdict{}); none.Verdict != "error" {
t.Fatalf("a try with no verdict: %+v", none)
}
}
+225
View File
@@ -0,0 +1,225 @@
package main
import (
"crypto/ed25519"
"crypto/rand"
"crypto/sha256"
"crypto/tls"
"crypto/x509"
"crypto/x509/pkix"
"encoding/hex"
"math/big"
"net"
"os"
"path/filepath"
"strings"
"testing"
"time"
"github.com/novox/mesh-controller/internal/broker"
)
// Where a builder publishes.
//
// Preferably from the mesh: a builder that is a module requires an artifact store, and the mesh
// writes it a binding saying which machine answers and on what port. Reading it means the address
// is not a setting somebody keeps in step by hand.
func binding(t *testing.T, body string) string {
t.Helper()
path := filepath.Join(t.TempDir(), "artifact-store.json")
if err := os.WriteFile(path, []byte(body), 0o644); err != nil {
t.Fatal(err)
}
return path
}
func TestTheMeshSaysWhereToPublish(t *testing.T) {
t.Setenv("MESH_BINDING", binding(t, `{"binding":1,"provision":"artifact-store",
"from":"anchor","at":"anchor.internal","serves":{"port":5000,"scheme":"http"}}`))
where, err := whereToPublish()
if err != nil {
t.Fatal(err)
}
if where != "anchor.internal:5000" {
t.Fatalf("got %q", where)
}
}
func TestABindingWithNoAddressIsRefused(t *testing.T) {
// The provider is not on the private network, so there is no name to reach it by. Falling
// back to anything would publish to a store on the wrong machine and be found out much later.
t.Setenv("MESH_BINDING", binding(t, `{"binding":1,"from":"anchor","serves":{"port":5000}}`))
_, err := whereToPublish()
if err == nil {
t.Fatal("a binding with nowhere to reach was accepted")
}
if !strings.Contains(err.Error(), "anchor") {
t.Fatalf("the failure does not name the machine: %v", err)
}
}
func TestABindingWithNoPortIsRefused(t *testing.T) {
t.Setenv("MESH_BINDING", binding(t, `{"binding":1,"from":"a","at":"a.internal","serves":{}}`))
if _, err := whereToPublish(); err == nil {
t.Fatal("a binding saying nothing about a port was accepted")
}
}
func TestTheVariableStillWorksForABuilderRunByAPerson(t *testing.T) {
// Which is how this started and how it is still run while being developed.
t.Setenv("MESH_BINDING", "")
t.Setenv("MESH_REGISTRY", "127.0.0.1:5000")
where, err := whereToPublish()
if err != nil {
t.Fatal(err)
}
if where != "127.0.0.1:5000" {
t.Fatalf("got %q", where)
}
}
func TestNeitherIsRefusedRatherThanGuessed(t *testing.T) {
t.Setenv("MESH_BINDING", "")
t.Setenv("MESH_REGISTRY", "")
if _, err := whereToPublish(); err == nil {
t.Fatal("a builder with nowhere to publish reported somewhere")
}
}
func TestTheCredentialComesFromAFileTheMeshSealed(t *testing.T) {
// A builder that is a module is given its credential the way every module is: sealed to the
// machine and written by the host. An environment variable instead would put the one copy
// that matters through a terminal and a process listing.
path := filepath.Join(t.TempDir(), "broker")
if err := os.WriteFile(path, []byte("amqps://a-builder:secret@broker.internal:5671/\n"), 0o600); err != nil {
t.Fatal(err)
}
t.Setenv("MESH_BROKER_FILE", path)
t.Setenv("MESH_BROKER_AMQP", "amqp://should-not-be-used@nowhere/")
got, err := brokerFrom()
if err != nil {
t.Fatal(err)
}
if got.URL != "amqps://a-builder:secret@broker.internal:5671/" {
t.Fatalf("got %q", got.URL)
}
}
// The credential the mesh seals carries what to check the broker's certificate against, because a
// mesh's broker presents a certificate of the mesh's own and no public trust store has it. A URL
// alone can only reach a broker somebody else vouches for.
func TestTheSealedCredentialCarriesWhatVerifiesTheBroker(t *testing.T) {
path := filepath.Join(t.TempDir(), "broker")
if err := os.WriteFile(path, []byte(
`{"url":"amqps://a-builder:secret@broker.internal:5671/","fingerprint":"abc123"}`),
0o600); err != nil {
t.Fatal(err)
}
t.Setenv("MESH_BROKER_FILE", path)
t.Setenv("MESH_BROKER_AMQP", "")
got, err := brokerFrom()
if err != nil {
t.Fatal(err)
}
if got.URL != "amqps://a-builder:secret@broker.internal:5671/" {
t.Fatalf("the url was lost: %q", got.URL)
}
if got.Fingerprint != "abc123" {
t.Fatal("the builder was given nothing to check the broker against, so it can only " +
"connect to a broker some public authority vouches for")
}
}
func TestAnEmptyCredentialFileIsRefused(t *testing.T) {
// Otherwise the builder connects as nobody and is refused, with the reason three layers away.
path := filepath.Join(t.TempDir(), "broker")
if err := os.WriteFile(path, []byte("\n"), 0o600); err != nil {
t.Fatal(err)
}
t.Setenv("MESH_BROKER_FILE", path)
t.Setenv("MESH_BROKER_AMQP", "")
if _, err := brokerFrom(); err == nil {
t.Fatal("an empty credential was accepted")
}
}
func TestABuilderWithNoCredentialAtAllSaysSo(t *testing.T) {
t.Setenv("MESH_BROKER_FILE", "")
t.Setenv("MESH_BROKER_AMQP", "")
if _, err := brokerFrom(); err == nil {
t.Fatal("a builder with no broker reported one")
}
}
// The pin is compared in the spelling the mesh writes it.
//
// A bare digest against a written fingerprint never matches, and the failure is indistinguishable
// from being pointed at the wrong broker — which is the one thing this check exists to report
// truthfully. It cost a lab run.
func TestThePinIsComparedInTheSpellingTheMeshWritesIt(t *testing.T) {
certificate, key := aServerCertificate(t)
der := certificate.Certificate[0]
sum := sha256.Sum256(der)
written := "sha256:" + hex.EncodeToString(sum[:])
listener, err := tls.Listen("tcp", "127.0.0.1:0", &tls.Config{
Certificates: []tls.Certificate{certificate}, MinVersion: tls.VersionTLS12,
})
if err != nil {
t.Fatal(err)
}
defer listener.Close()
go func() {
for {
conn, err := listener.Accept()
if err != nil {
return
}
_ = conn.(*tls.Conn).Handshake()
conn.Close()
}
}()
_ = key
// The pin the mesh wrote must be accepted.
if err := handshakeWith(listener.Addr().String(), written); err != nil {
t.Fatalf("the broker this builder was told about was refused: %v", err)
}
// And a different one refused, or the check reports nothing.
other := "sha256:" + strings.Repeat("ab", 32)
if err := handshakeWith(listener.Addr().String(), other); err == nil {
t.Fatal("a broker this builder was not told about was accepted")
}
}
// handshakeWith runs the pin check the builder dials with against an address.
func handshakeWith(address, pin string) error {
conn, err := tls.Dial("tcp", address, broker.PinnedToFingerprint(pin))
if err != nil {
return err
}
return conn.Close()
}
func aServerCertificate(t *testing.T) (tls.Certificate, ed25519.PrivateKey) {
t.Helper()
public, private, err := ed25519.GenerateKey(rand.Reader)
if err != nil {
t.Fatal(err)
}
template := &x509.Certificate{
SerialNumber: big.NewInt(1),
Subject: pkix.Name{CommonName: "a broker"},
NotBefore: time.Now().Add(-time.Hour),
NotAfter: time.Now().Add(time.Hour),
IPAddresses: []net.IP{net.ParseIP("127.0.0.1")},
}
der, err := x509.CreateCertificate(rand.Reader, template, template, public, private)
if err != nil {
t.Fatal(err)
}
return tls.Certificate{Certificate: [][]byte{der}, PrivateKey: private}, private
}
+2 -21
View File
@@ -408,21 +408,12 @@ func issueOnAssign(ctx context.Context, open *stores, node, module string) strin
return ""
}
user := broker.Principal{Kind: broker.KindModule, Node: node, Module: module}.Username()
_, reads := m.OwnSecrets["broker"]
if !reads && !runsTheProvisionerClient(m) {
// A module whose bus accounts are its processes' alone has none of its own: only the processes that
// are not issued yet are issued (issue 1226).
return processAccountsOnAssign(ctx, inv, m, node)
}
if !reads {
if _, reads := m.OwnSecrets["broker"]; !reads {
// A module running the provisioner client alone is minted the client's account (issue 917).
user = broker.Principal{Kind: broker.KindProvisioner, Node: node, Module: module}.Username()
}
if _, minted, err := inv.BusUserHash(ctx, user); err != nil {
if _, minted, err := inv.BusUserHash(ctx, user); err != nil || minted {
return ""
} else if minted {
// Its own account stands and is kept; a process it gained is not issued yet (issue 1226).
return processAccountsOnAssign(ctx, inv, m, node)
}
busAddress, err := broker.BusAddress()
if err == nil {
@@ -435,16 +426,6 @@ func issueOnAssign(ctx context.Context, open *stores, node, module string) strin
return fmt.Sprintf("its bus credential is issued and sealed to %s, and arrives with the push", node)
}
// processAccountsOnAssign issues the process accounts of an assigned module not issued yet, and says what it did.
func processAccountsOnAssign(ctx context.Context, inv *inventory.Inventory, m catalogue.Manifest, node string) string {
line, err := issueMissingProcessAccounts(ctx, inv, m, node)
if err != nil {
return fmt.Sprintf("its process accounts are not issued (%v): `module issue %s --node %s` first — "+
"`push %s` refuses to send %s until it is", err, m.Module, node, node, m.Module)
}
return line
}
// refusedMovingTheController refuses assigning or unassigning the controller's module through a verb (review of
// novox/hq ADR 0272): the node it is assigned to is the control-node, and the control-node's operator account is
// the controller's terminal, so whoever moves the module chooses the terminal. At the terminal alone, as every
@@ -221,4 +221,8 @@ func TestOnANodeWithNoNetworkTheStoreIsReachedByLoopback(t *testing.T) {
if image != "127.0.0.1:5100/postgres/runtime@"+aDigest {
t.Fatalf("on the store's own node, off any network, the image is fetched as %v", image)
}
held := heldBy(ctx)
if got := held["postgres/runtime"]; got != "127.0.0.1:5100/postgres/runtime@"+aDigest {
t.Fatalf("a builder beside the store is handed the base %q", got)
}
}
+59
View File
@@ -1,9 +1,12 @@
package main
import (
"encoding/json"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/link"
)
// An address is read from the node's settings where it is used, never recorded with a port
@@ -11,6 +14,62 @@ import (
var aDigest = "sha256:" + strings.Repeat("e", 64)
// **A build is recorded by digest and path**, whatever address the builder pushed to — what it
// made and what it stood on both; an image the module runs from elsewhere is left where it says.
func TestABuildIsRecordedWithoutTheStoresAddress(t *testing.T) {
manifest, _ := json.Marshal(map[string]any{
"module": "gitea", "version": "1",
"resources": []map[string]any{
{"id": "server", "type": "container", "name": "mesh-gitea",
"image": "anchor.internal:5100/gitea/server@" + aDigest},
{"id": "config", "type": "archive", "path": "/etc/gitea", "digest": aDigest,
"source": "http://anchor.internal:5100/v2/gitea/config/blobs/" + aDigest},
{"id": "cache", "type": "container", "name": "mesh-gitea-cache",
"image": "valkey/valkey@" + aDigest},
},
})
kept := buildFrom(link.BuildResult{
ID: "b1", Repository: "https://forge.example/gitea.git", Commit: "abc", On: "laptop",
Manifest: manifest,
Made: []link.MadeArtifact{
{Name: "server", Kind: "image", Reference: "anchor.internal:5100/gitea/server@" + aDigest},
{Name: "config", Kind: "archive", Reference: "http://anchor.internal:5100/v2/gitea/config/blobs/" + aDigest},
},
Against: []string{"anchor.internal:5100/mesh-tools/runtime@" + aDigest},
})
if kept.Module != "gitea" {
t.Fatalf("the module was not read from the recorded manifest: %q", kept.Module)
}
if kept.Made[0].Reference != catalogue.ArtifactStoreScheme+"gitea/server@"+aDigest {
t.Errorf("the image is recorded as %q, address and all", kept.Made[0].Reference)
}
if kept.Made[1].Reference != catalogue.ArtifactStoreScheme+"gitea/config/blobs/"+aDigest {
t.Errorf("the archive is recorded as %q, address and all", kept.Made[1].Reference)
}
recorded, err := catalogue.ParseManifest(kept.Manifest)
if err != nil {
t.Fatal(err)
}
if got := recorded.Resources[0]["image"]; got != catalogue.ArtifactStoreScheme+"gitea/server@"+aDigest {
t.Errorf("the recorded manifest's image is %v", got)
}
if got := recorded.Resources[1]["source"]; got != catalogue.ArtifactStoreScheme+"gitea/config/blobs/"+aDigest {
t.Errorf("the recorded manifest's archive is %v", got)
}
if got := recorded.Resources[2]["image"]; got != "valkey/valkey@"+aDigest {
t.Errorf("an image the build did not make was rewritten: %v", got)
}
if strings.Contains(string(kept.Manifest), "anchor.internal:5100") {
t.Errorf("the recorded manifest still carries the store's address:\n%s", kept.Manifest)
}
// What the build stood on is an edge to another module's artifact, and it is recorded the way
// that artifact is: by path in the store, so the edge still names the same thing when the
// store answers at another address.
if kept.Against[0] != catalogue.ArtifactStoreScheme+"mesh-tools/runtime@"+aDigest {
t.Errorf("what the build stood on was recorded by address: %v", kept.Against)
}
}
// aStore is a module offering the artifact store on 5000, published the long way as the
// distribution module does, so a node may be given another number for it.
func aStore() catalogue.Manifest {
@@ -0,0 +1,134 @@
//go:build integration
package main
import (
"context"
"fmt"
"strings"
"testing"
"time"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
)
// Issue 1334, the look: when the bus did not answer an ask of any module of a tier, no build is in flight and no event follows,
// so the look itself comes back (it fails, and the looker retries it), asks the modules again, and past the bound fails the plan.
func TestALookComesBackForATierTheBusRefusedWhole(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
inv := open.inventory
for _, name := range []string{"a", "b"} {
if err := inv.RegisterModule(ctx, catalogue.Manifest{Module: name, Version: "1"},
inventory.Source{Repository: "novox/" + name, Seat: "git", Ref: "main", BuiltFrom: "c0", Head: "c0"}); err != nil {
t.Fatal(err)
}
}
p := inventory.Plan{ID: "plan-bus", Repository: "novox/a", Branch: "main", Commit: "c1", Created: time.Now().UTC(),
State: inventory.PlanBuilding, Tiers: [][]string{{"a", "b"}}, Modules: map[string]*inventory.PlanModule{"a": {}, "b": {}}}
if err := inv.SavePlan(ctx, &p); err != nil {
t.Fatal(err)
}
bus := false
was := askABuild
askABuild = func(_ context.Context, source buildSource, _, _ string) (string, error) {
if !bus {
return "", fmt.Errorf("%w: nats: no responders available for request", inventory.ErrNoChain)
}
return "build-" + source.Repository, nil
}
t.Cleanup(func() { askABuild = was })
look := func() bool {
t.Helper()
release, err := inv.HoldPlans(ctx, true)
if err != nil {
t.Fatal(err)
}
defer release()
return advanceHeld(ctx, open)
}
read := func() inventory.Plan {
t.Helper()
got, err := inv.PlanByID(ctx, "plan-bus")
if err != nil {
t.Fatal(err)
}
return got
}
// Every module refused: the look says it failed, which has the looker come back; the plan is building and says why.
if !look() {
t.Fatal("a look that left the whole tier unasked for the bus did not fail, so nothing would come back to it")
}
if got := read(); got.State != inventory.PlanBuilding || !strings.Contains(got.Note, "bus did not answer") ||
got.Modules["a"].State != "" || got.Modules["b"].State != "" {
t.Fatalf("the plan after a refused tier: %s %q %+v %+v", got.State, got.Note, got.Modules["a"], got.Modules["b"])
}
// Again, still refused: the same fact is not saved again, and the look still comes back.
before := read().Revision
if !look() || read().Revision != before {
t.Fatalf("a second refused look failed to come back or saved the plan again (revision %d, was %d)", read().Revision, before)
}
// The bus answers at the next look: both are asked, and nothing waits any more.
bus = true
if look() {
t.Fatal("a look that asked the whole tier failed")
}
if got := read(); got.Modules["a"].State != "asked" || got.Modules["b"].State != "asked" || got.Modules["a"].BusWaitFrom != nil {
t.Fatalf("the tier was not asked once the bus answered: %+v %+v", got.Modules["a"], got.Modules["b"])
}
}
func TestALookFailsTheTierTheBusRefusedPastTheBound(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
inv := open.inventory
if err := inv.RegisterModule(ctx, catalogue.Manifest{Module: "a", Version: "1"},
inventory.Source{Repository: "novox/a", Seat: "git", Ref: "main", BuiltFrom: "c0", Head: "c0"}); err != nil {
t.Fatal(err)
}
long := time.Now().UTC().Add(-busWaitBound - time.Minute)
p := inventory.Plan{ID: "plan-bus", Repository: "novox/a", Branch: "main", Commit: "c1", Created: time.Now().UTC(),
State: inventory.PlanBuilding, Tiers: [][]string{{"a"}}, Modules: map[string]*inventory.PlanModule{"a": {BusWaitFrom: &long}}}
if err := inv.SavePlan(ctx, &p); err != nil {
t.Fatal(err)
}
was := askABuild
askABuild = func(context.Context, buildSource, string, string) (string, error) {
return "", fmt.Errorf("%w: nats: no responders available for request", inventory.ErrNoChain)
}
t.Cleanup(func() { askABuild = was })
release, err := inv.HoldPlans(ctx, true)
if err != nil {
t.Fatal(err)
}
advanceHeld(ctx, open)
release()
got, err := inv.PlanByID(ctx, "plan-bus")
if err != nil {
t.Fatal(err)
}
if got.State != inventory.PlanFailed || !strings.Contains(got.Note, "the bus did not answer an ask for 10m0s (since ") {
t.Fatalf("a bus silent past its bound did not fail the plan in words: %s %q", got.State, got.Note)
}
// Rebuilt by hand while the bus still reloads: said as waiting, not as a failure and not with an empty reason.
joined, said, err := joinAPlan(ctx, open, "a")
if err != nil || !joined || !strings.Contains(said, "not asked yet: the bus did not answer") {
t.Fatalf("a rebuild left for the bus is said as %v %q %v", joined, said, err)
}
// And retried by hand: the module is named as waiting, not as asked again.
got, _ = inv.PlanByID(ctx, "plan-bus")
if got.State == inventory.PlanFailed {
got.Modules["a"].State, got.Modules["a"].Why = "failed", "x"
if err := inv.SavePlan(ctx, &got); err != nil {
t.Fatal(err)
}
said, err := retryPlan(ctx, open, "plan-bus")
if err != nil || !strings.Contains(said, "not asked yet: the bus did not answer") {
t.Fatalf("a retry left for the bus is said as %q %v", said, err)
}
}
}
+139
View File
@@ -0,0 +1,139 @@
package main
import (
"context"
"errors"
"fmt"
"strings"
"testing"
"time"
"github.com/novox/mesh-controller/internal/inventory"
)
// Issue 1334: a bus that does not answer for a moment (it reloads its user list, and JetStream's API has no responder for a few
// seconds) is not a failed build. The module is left unasked and the plan building, and the next look asks it again; only a
// module's own fault, or a bus silent for ten minutes, fails the plan.
func aTierOfThree() (*inventory.Plan, map[string]inventory.Entry) {
p := &inventory.Plan{ID: "plan-1", State: inventory.PlanBuilding, Tiers: [][]string{{"a", "b", "c"}},
Modules: map[string]*inventory.PlanModule{}}
byName := map[string]inventory.Entry{}
for _, m := range []string{"a", "b", "c"} {
byName[m] = inventory.Entry{Source: inventory.Source{Repository: "novox/" + m, Seat: "git", Ref: "main"}}
}
return p, byName
}
func TestABusThatDoesNotAnswerForAMomentDoesNotFailTheTier(t *testing.T) {
p, byName := aTierOfThree()
bus := false
was := askABuild
askABuild = func(_ context.Context, source buildSource, _, _ string) (string, error) {
if source.Repository == "novox/b" && !bus {
return "", fmt.Errorf("%w: the bucket mesh-controller_builds is not on the bus: nats: no responders available for request",
inventory.ErrNoChain)
}
return "build-" + source.Repository, nil
}
t.Cleanup(func() { askABuild = was })
for _, m := range []string{"a", "b", "c"} {
askModule(t.Context(), p, m, byName)
}
if p.State != inventory.PlanBuilding {
t.Fatalf("a bus that did not answer for a moment failed the plan: %s, %q", p.State, p.Note)
}
if p.Modules["a"].State != "asked" || p.Modules["c"].State != "asked" {
t.Fatalf("the modules the bus did answer for were not asked: %+v %+v", p.Modules["a"], p.Modules["c"])
}
if s := p.Modules["b"]; s.State != "" {
t.Fatalf("the module the bus did not answer for is %q, not left unasked", s.State)
}
if got := unaskedOf(*p); len(got) != 1 || got[0] != "b" {
t.Fatalf("the next look would ask %v, not b", got)
}
if !strings.Contains(p.Note, "bus did not answer") {
t.Fatalf("the plan says nothing of the wait: %q", p.Note)
}
// The next look, the bus answering: b is asked and the plan goes on.
bus = true
for _, m := range unaskedOf(*p) {
askModule(t.Context(), p, m, byName)
}
if p.State != inventory.PlanBuilding || p.Modules["b"].State != "asked" || p.Modules["b"].Build == "" {
t.Fatalf("b was not asked once the bus answered: %s %+v", p.State, p.Modules["b"])
}
}
func TestABusSilentForTenMinutesFailsTheTierAsBefore(t *testing.T) {
p, byName := aTierOfThree()
was := askABuild
askABuild = func(context.Context, buildSource, string, string) (string, error) {
return "", fmt.Errorf("%w: nats: no responders available for request", inventory.ErrNoChain)
}
t.Cleanup(func() { askABuild = was })
askModule(t.Context(), p, "a", byName)
if p.State != inventory.PlanBuilding {
t.Fatalf("the first refusal failed the plan: %q", p.Note)
}
long := time.Now().Add(-busWaitBound - time.Minute)
p.Modules["a"].BusWaitFrom = &long
askModule(t.Context(), p, "a", byName)
if p.State != inventory.PlanFailed || p.Modules["a"].State != "failed" {
t.Fatalf("a bus silent past its bound did not fail the plan: %s %+v", p.State, p.Modules["a"])
}
// Said in plain words: the bus, the wait, and since when.
want := "a could not be asked for: the bus did not answer an ask for 10m0s (since " + long.UTC().Format(time.RFC3339) +
"): this kind of record lives on the bus"
if !strings.HasPrefix(p.Note, want) || p.Modules["a"].Why != strings.TrimPrefix(want, "a could not be asked for: ")+p.Note[len(want):] {
t.Fatalf("the plan says %q, not words starting %q", p.Note, want)
}
// The wait has ended: a Retry made while the bus still reloads waits again, and does not fail at its first refusal.
if p.Modules["a"].BusWaitFrom != nil {
t.Fatalf("a failed module still holds the start of its wait: %v", p.Modules["a"].BusWaitFrom)
}
p.State = inventory.PlanBuilding
askModule(t.Context(), p, "a", byName)
if p.State != inventory.PlanBuilding || p.Modules["a"].State != "" || p.Modules["a"].BusWaitFrom == nil {
t.Fatalf("a retry made while the bus is silent did not wait afresh: %s %+v", p.State, p.Modules["a"])
}
}
func TestAPlanWithAModuleWaitingForTheBusIsOneTheLookComesBackTo(t *testing.T) {
p, _ := aTierOfThree()
if busWaiting(*p) {
t.Fatal("a plan whose modules are not asked yet waits for the bus")
}
now := time.Now()
p.Modules["b"] = &inventory.PlanModule{BusWaitFrom: &now}
p.Modules["a"] = &inventory.PlanModule{State: "asked"}
if got := busWaitingIn(*p); len(got) != 1 || got[0] != "b" {
t.Fatalf("the modules waiting for the bus are %v, not b", got)
}
p.State = inventory.PlanFailed
if busWaiting(*p) {
t.Fatal("a plan that is not open is waited for")
}
}
func TestAModulesOwnFaultStillFailsThePlan(t *testing.T) {
p, byName := aTierOfThree()
delete(byName, "a")
askModule(t.Context(), p, "a", byName)
if p.State != inventory.PlanFailed || !strings.Contains(p.Note, "no longer in the catalogue") {
t.Fatalf("a module gone from the catalogue did not fail the plan: %s %q", p.State, p.Note)
}
p, byName = aTierOfThree()
was := askABuild
askABuild = func(context.Context, buildSource, string, string) (string, error) {
return "", errors.New("git.example/novox/a is no repository the catalogue builds a module from")
}
t.Cleanup(func() { askABuild = was })
askModule(t.Context(), p, "a", byName)
if p.State != inventory.PlanFailed || p.Modules["a"].State != "failed" {
t.Fatalf("a module's own refusal did not fail the plan: %s %+v", p.State, p.Modules["a"])
}
}
+35
View File
@@ -1,12 +1,16 @@
package main
import (
"errors"
"fmt"
"slices"
"sort"
"strings"
"time"
"github.com/nats-io/nats.go"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/inventory"
)
@@ -14,6 +18,37 @@ import (
// (issue 1334).
const busWaitBound = 10 * time.Minute
// busDidNotAnswer says an ask failed because the bus's JetStream had no responder at that moment, as it has for a few seconds
// when the bus reloads its user list (novox/hq issue 353), and not because of the module: the same ask is made again.
func busDidNotAnswer(err error) bool {
if err == nil {
return false
}
if errors.Is(err, nats.ErrNoResponders) || errors.Is(err, nats.ErrJetStreamNotEnabled) ||
errors.Is(err, jetstream.ErrJetStreamNotEnabled) || errors.Is(err, inventory.ErrNoChain) {
return true
}
// An error that crossed a boundary as text only.
text := err.Error()
return strings.Contains(text, "no responders available") || strings.Contains(text, "jetstream not enabled")
}
// waitForTheBus leaves a module whose ask the bus did not answer unasked, the plan building, and says so in the plan's note: the
// next look asks the tier's unasked modules again. It says false, so the caller fails the plan as for any refusal, once the bus
// has been silent for busWaitBound.
func waitForTheBus(p *inventory.Plan, state *inventory.PlanModule, name string, err error, now time.Time) bool {
if state.BusWaitFrom == nil {
state.BusWaitFrom = &now
}
if now.Sub(*state.BusWaitFrom) > busWaitBound {
return false
}
state.State, state.Why = "", ""
p.Note = fmt.Sprintf("tier %d: %s not asked: the bus did not answer (%v); asked again at the next look, failing after %s",
p.Tier, name, firstLine(err.Error()), busWaitBound)
return true
}
// busWaiting says an open plan has a module of its current tier that the bus did not answer an ask of: the look must come back
// for it, as nothing else will when no build of the tier is in flight (issue 1334).
func busWaiting(p inventory.Plan) bool {
-23
View File
@@ -1,23 +0,0 @@
package main
import (
"strings"
"testing"
"time"
"github.com/novox/mesh-controller/internal/inventory"
)
// A retry of a plan failed past the bus's bound, whose module the bus still does not answer, names that module as waiting:
// read while the plan still says failed, it was named nowhere, and the retry said "asked" with nothing after it (issue
// 1403, met again once the hand rebuild no longer resumed the plan first).
func TestARetryNamesTheModulesStillWaitingForTheBus(t *testing.T) {
since := time.Now().UTC()
p := inventory.Plan{ID: "plan-bus", State: inventory.PlanFailed, Tiers: [][]string{{"a"}},
Modules: map[string]*inventory.PlanModule{"a": {BusWaitFrom: &since}}}
waiting := busWaitingOnRetry(p)
_, said := retryWords(p, []string{"a"}, nil, waiting)
if !strings.Contains(said, "a not asked yet: the bus did not answer") || strings.Contains(said, "asked ;") {
t.Fatalf("a retry left for the bus is said as %q", said)
}
}
-4
View File
@@ -645,10 +645,6 @@ func doesWords(act conditions.Action) string {
return "the delivery is stopped"
case act.Verb == broker.ControllerSeat+".plans" && act.Arguments["retry"] != "":
return "the failed builds are asked again"
case act.Verb == broker.ControllerSeat+".push" && act.Arguments["node"] != "":
return "the held builds are sent to " + act.Arguments["node"]
case act.Verb == broker.ControllerSeat+".upgrade" && act.Arguments["release-backlog"] != "":
return "the builds held after the failed delivery are released"
case strings.HasSuffix(act.Verb, ".restart"):
return "its service is restarted on " + act.Machine
}
@@ -23,7 +23,7 @@ import (
func TestAWarrantForRetryRetriesAFailedPlanInTheStoreAndEndsItsCondition(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
asksRecorded(t)
asked := asksRecorded(t)
planMovedHeard(t) // the bus the verb says its moves on, and the hand-act log
twoTiers(t, open)
before := time.Now().UTC().Add(-time.Hour)
@@ -33,7 +33,7 @@ func TestAWarrantForRetryRetriesAFailedPlanInTheStoreAndEndsItsCondition(t *test
Note: "a failed to build in tier 0",
Delivery: &inventory.PlanDelivery{},
Modules: map[string]*inventory.PlanModule{"a": {State: "failed", AskedAt: &before, Build: "build-1",
Why: "killed by hand"}},
Why: link.KilledByHand}},
Times: &inventory.PlanTimes{Cut: &before, Started: &before, Ended: &ended, Class: inventory.ClassLeaf}}
if err := open.inventory.SavePlan(ctx, &failed); err != nil {
t.Fatal(err)
@@ -123,9 +123,12 @@ func TestAWarrantForRetryRetriesAFailedPlanInTheStoreAndEndsItsCondition(t *test
t.Fatal(err)
}
m := p.Modules["a"]
if p.State != inventory.PlanBuilding || p.Tier != 0 || m.State != "asked" || !m.Key || m.Why != "waits for the builder's key" || m.Build == "build-1" {
if p.State != inventory.PlanBuilding || p.Tier != 0 || m.State != "asked" || m.Build == "" || m.Build == "build-1" {
t.Fatalf("after the warrant the plan is %s at tier %d and a is %+v", p.State, p.Tier, m)
}
if len(*asked) != 1 {
t.Fatalf("the build was asked %d times: %v", len(*asked), *asked)
}
if got := store[w.Ask]; got.Acted != "done" {
t.Fatalf("the ask ended %q", got.Acted)
}
+16 -51
View File
@@ -8,7 +8,6 @@ import (
"fmt"
"strings"
"sync"
"sync/atomic"
"testing"
"time"
@@ -16,7 +15,6 @@ import (
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/builderkeys"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
"github.com/novox/mesh-controller/internal/testbus"
@@ -68,18 +66,17 @@ func TestAPlanGoesOnWithoutAMachineAwayWhenItsTierIsSent(t *testing.T) {
// The build machine and the machines, as far as the plan meets them: an ask is counted, a send is recorded and kept.
var mu sync.Mutex
var asks []string
var sends [][]string
everyKeyDone(t) // the builds of the tiers: each module is taken from a done key as soon as the walk reaches it
// Except tier 1's module b, whose key is not built until the test says so: a walk that takes it in the same look
// goes on through tier 1's send and overwrites the note of the move before the test reads it (the old build ask
// was where the walk paused).
wholeKeys := openPlanKeys
var holdB atomic.Bool
holdB.Store(true)
openPlanKeys = func() (planKeyReads, func(), error) {
all, done, err := wholeKeys()
return keysHolding{all, "b", &holdB, ""}, done, err
wasAsk := askABuild
askABuild = func(_ context.Context, source buildSource, _, _ string) (string, error) {
mu.Lock()
defer mu.Unlock()
id := link.NewBuildID(time.Now().Add(time.Duration(len(asks)) * time.Millisecond))
asks = append(asks, source.Repository+"#"+id)
return id, nil
}
t.Cleanup(func() { askABuild = wasAsk })
digests := 0
wasSend := sendRollout
sendRollout = func(ctx context.Context, open *stores, names []string) ([]string, error) {
@@ -201,9 +198,13 @@ func TestAPlanGoesOnWithoutAMachineAwayWhenItsTierIsSent(t *testing.T) {
if !strings.Contains(moved, "laptop was away: sent when it is heard again") || strings.Contains(moved, "anchor") {
t.Errorf("1161/AC2: the plan says %q; want laptop named away and not anchor", moved)
}
// Tier 1's module is taken from its key once that is built, which the walk's look does by itself.
holdB.Store(false)
askToLook()
// The ask for tier 1's build is the next step, and records its build id on the module.
eventually(t, "tier 1 asked", func() bool { m := plan().Modules["b"]; return m != nil && m.Build != "" })
b := plan().Modules["b"]
askedAt, _ := link.BuildAskedAt(b.Build)
record(b.Build, "b", "c2", askedAt)
register("b", "c2", askedAt)
planBuilt(ctx, open, "b", "c2", "", askedAt, b.Build)
eventually(t, "tier 1 sent to anchor alone", func() bool {
mu.Lock()
defer mu.Unlock()
@@ -225,39 +226,3 @@ func TestAPlanGoesOnWithoutAMachineAwayWhenItsTierIsSent(t *testing.T) {
t.Errorf("1161/AC2: the plan keeps %v away; want laptop", got)
}
}
// keysHolding is the stand-in builder's keys but for one module, whose key is not built while held; only the key of
// the fingerprint fp when that is given.
type keysHolding struct {
planKeyReads
module string
held *atomic.Bool
fp string
}
func (k keysHolding) Build(ctx context.Context, module, fp string) (builderkeys.Key, error) {
if module == k.module && k.held.Load() && (k.fp == "" || fp == k.fp) {
return builderkeys.Key{Module: module, Fingerprint: fp}, nil
}
return k.planKeyReads.Build(ctx, module, fp)
}
// holdKeys has the stand-in builder's keys of these modules (at any commit) unbuilt until the returned flag is cleared,
// so a walk that took them stays on its tier as it did while its build ran. It restores what it replaced at the end of
// the test, and the flag is set until the test lowers it.
func holdKeys(t *testing.T, modules ...string) *atomic.Bool {
t.Helper()
held := &atomic.Bool{}
held.Store(true)
was := openPlanKeys
openPlanKeys = func() (planKeyReads, func(), error) {
all, done, err := was()
var keys planKeyReads = all
for _, m := range modules {
keys = keysHolding{keys, m, held, ""}
}
return keys, done, err
}
t.Cleanup(func() { openPlanKeys = was })
return held
}
@@ -211,12 +211,8 @@ func TestADeliveryThatMovesABackedUpModuleWaitsForItsBackupAndAFailedOneHoldsItA
}
seats["laptop"].answers("notes", true, "")
advancePlans(ctx, open)
// One key taken, not two: plan-one and plan-two carry the same module at the same commit, so they share one key
// build id (key_takein.go keyBuildID), and a walk finds a key already taken (keyTakes.taken, plan_keys.go) and takes
// it in no more. So that the smaller count hides no plan that stopped, plan-two's module must be built from that key.
if got := read("plan-two"); got.Backups.State != inventory.BackupDone || len(*asked) != 1 ||
got.Modules["notes"].Build == "" || got.Modules["notes"].State == "failed" || got.Modules["notes"].Key && got.Modules["notes"].State == "asked" {
t.Fatalf("the plan did not go on: %+v module=%+v builds=%v", got.Backups, got.Modules["notes"], *asked)
if got := read("plan-two"); got.Backups.State != inventory.BackupDone || len(*asked) != 2 {
t.Fatalf("the plan did not go on: %+v builds=%v", got.Backups, *asked)
}
if said, _ := probeBackupHeld(ctx, &doctor{open: open}); len(said) != 0 {
t.Fatalf("the condition stays: %+v", said)
-5
View File
@@ -232,8 +232,6 @@ func (f following) hearMerge(ctx context.Context, m link.SourceMoved, now time.T
fmt.Print(nothingReadsIt(repository, m))
return nil
}
// A module the merge adds is registered from the builder's first done key of it, which the builder's write wakes
// (novox/hq ADR 0501 §6): nothing is asked here.
// **Its delivery group's word, before the plans are held** (ADR 0306 decision 2): a slow answer holds up no walk.
own := ownPath(touches)
waits, holding, err := groupWaitHolding(ctx, inv, m, deliverySeatHeld(entries))
@@ -800,9 +798,6 @@ type keptMerge struct {
// FirstBuild says the merge is a first merge into a repository the mesh builds nothing from, kept to ask the first
// build of its modules once mesh-delivery's word on it is kept (novox/hq ADR 0432, issue 1146).
FirstBuild bool `json:"first_build,omitempty"`
// Held says which first builds of the merge were refused for a word the running build-agent does not read: the merge
// waits for a build-agent that reads them, and is never failed (novox/hq ADR 0482 §5, issue 1173).
Held []heldFirstBuild `json:"held,omitempty"`
}
// announcedOf is a kept merge as the forge announced it, and what it moved; empty where the record says none.
+14 -43
View File
@@ -8,7 +8,6 @@ import (
"os"
"slices"
"strings"
"sync/atomic"
"testing"
"time"
@@ -66,7 +65,7 @@ func cutAt(t *testing.T, open *stores, now time.Time) {
func TestTwoMergesSecondsApartAreOneWalkAtTheLaterCommit(t *testing.T) {
open := windowed(t)
everyHeadWaits(t)
asksWithPaths(t)
asked := asksWithPaths(t)
claude := catalogueMerge("553b7191claude", "app", t0)
dunst := catalogueMerge("48bda475dunst", "notes", t0.Add(17*time.Second))
hear(t, open, claude, t0.Add(time.Second))
@@ -97,12 +96,8 @@ func TestTwoMergesSecondsApartAreOneWalkAtTheLaterCommit(t *testing.T) {
if w.Delivery == nil || !slices.EqualFunc(w.Delivery.Merges, want, sameMerge) {
t.Fatalf("the walk answers %+v, want %+v", w.Delivery, want)
}
// The cut takes its first tier from the builder's keys (a look takes them in later; none runs here), and says at
// which commit: the one the branch holds, which is the later merge the walk carries.
for _, m := range w.Tiers[w.Tier] {
if s := w.Modules[m]; s == nil || !s.Key || s.State != "asked" || s.KeyAt != dunst.Commit {
t.Fatalf("the walk did not take %s from its key at the commit it carries (%s): %+v", m, dunst.Commit, s)
}
if len(*asked) != 2 || (*asked)[0][2] != "main" || (*asked)[1][2] != "main" {
t.Fatalf("the walk did not ask its modules at the branch, which holds the commit it carries: %v", *asked)
}
// Heard again, as the bus may hand it over twice: never a second merge, nor a second walk.
hear(t, open, claude, t0.Add(5*time.Minute))
@@ -136,10 +131,11 @@ func TestThreeRepositoriesInOneWindowAreOneWalk(t *testing.T) {
t.Fatalf("the walk does not build %s: %v", r, w.Modules)
}
}
// Each module of the first tier is taken from its key at its own repository's commit (the cut asks no build).
for _, r := range w.Tiers[w.Tier] {
if s := w.Modules[r]; s == nil || !s.Key || s.State != "asked" || s.KeyAt != "c-"+r {
t.Fatalf("%s was not taken from its key at its own commit: %+v", r, s)
// Each module's ask is recorded at its own repository's commit.
for _, r := range []string{"one", "two", "three"} {
q, found, err := open.inventory.BuildRequestByID(t.Context(), w.Modules[r].Build)
if err != nil || !found || q.Commit != "c-"+r {
t.Fatalf("%s's build was not recorded at its own commit: %+v %v %v", r, q, found, err)
}
}
}
@@ -344,7 +340,7 @@ func TestALateMergeIsAnsweredByTheWalkOfTheLaterOne(t *testing.T) {
func TestAFailedWalkWalksItsEarlierMergesAlone(t *testing.T) {
open := windowed(t)
everyHeadWaits(t)
asksWithPaths(t)
asked := asksWithPaths(t)
ctx := t.Context()
// Made before the cut marks their modules seen, as merges are: walked again, they are news all the same.
oldest := catalogueMerge("aaaa0001", "app", t0.Add(-3*time.Minute))
@@ -371,10 +367,9 @@ func TestAFailedWalkWalksItsEarlierMergesAlone(t *testing.T) {
!sameMerge(alone.Delivery.Merges[0], inventory.PlanMerge{Repository: "novox/mesh-catalog", Commit: middle.Commit}) {
t.Fatalf("the newest earlier merge was not walked alone on its own commit: %+v", alone)
}
// The cut takes the first tier from the builder's keys and asks no build, so nothing is recorded as asked: the
// module is taken from its key at the commit the merge was walked on.
if app := alone.Modules["app"]; app == nil || !app.Key || app.State != "asked" || app.KeyAt != middle.Commit {
t.Fatalf("the merge walked alone does not build app at its own commit: %v", alone.Modules)
if _, in := alone.Modules["app"]; !in || (*asked)[len(*asked)-1] != [3]string{"novox/mesh-catalog", "modules/app",
middle.Commit} {
t.Fatalf("the merge walked alone does not build app at its own commit: %v, asked %v", alone.Modules, *asked)
}
// Retried, the failed walk would name merges the search answers now.
if _, err := retryPlan(ctx, open, failed.ID); err == nil || !strings.Contains(err.Error(), "walked alone") {
@@ -513,14 +508,6 @@ func TestALateMergeOfAOneModuleRepositoryIsNamed(t *testing.T) {
func TestAWalkLetGoBesideAStartedOneStartsOnceItEnded(t *testing.T) {
open := windowed(t)
asked := asksWithPaths(t)
// The started walk's key (one at c1) is not built, so that walk stays started as it did while its build ran: with
// the stand-in builder's key done at once, one step would carry it to done and nothing would be beside the other.
wholeKeys, held := openPlanKeys, &atomic.Bool{}
held.Store(true)
openPlanKeys = func() (planKeyReads, func(), error) {
all, done, err := wholeKeys()
return keysHolding{all, "one", held, fpOfModuleAt("one", "c1")}, done, err
}
ctx := t.Context()
hear(t, open, repoMerge("one", "c1", t0), t0)
cutAt(t, open, t0.Add(2*time.Minute))
@@ -554,11 +541,7 @@ func TestAWalkLetGoBesideAStartedOneStartsOnceItEnded(t *testing.T) {
t.Fatalf("a deferred walk is watched as a tier running late: %+v", f)
}
}
// Read again: the steps since wrote the started walk (what its key waits for), so the copy read at its cut is stale.
done, err := open.inventory.PlanByID(ctx, started[0].ID)
if err != nil {
t.Fatal(err)
}
done := started[0]
done.State = inventory.PlanDone
if err := open.inventory.SavePlan(ctx, &done); err != nil {
t.Fatal(err)
@@ -641,14 +624,6 @@ func TestAMergeOnTheControllersPathNeverSharesABatch(t *testing.T) {
// A catalogue walk waiting for its word when an own-path batch is cut keeps waiting beside the own-path
// walk, is not folded into it, and its word is taken meanwhile: it starts once the own-path walk ended.
catalogueWalk := ws[0]
// The controller's second walk keeps its key unbuilt, so it stays started as it did while its build ran; with the
// stand-in builder's key done at once one step would carry it to done and nothing would be beside the catalogue walk.
wholeKeys, held := openPlanKeys, &atomic.Bool{}
held.Store(true)
openPlanKeys = func() (planKeyReads, func(), error) {
all, done, err := wholeKeys()
return keysHolding{all, "mesh-controller", held, fpOfModuleAt("mesh-controller", "k2")}, done, err
}
hear(t, open, repoMerge("mesh-controller", "k2", t0.Add(4*time.Minute)), t0.Add(4*time.Minute))
cutAt(t, open, t0.Add(6*time.Minute))
ws, _ = walks(t, open)
@@ -667,11 +642,7 @@ func TestAMergeOnTheControllersPathNeverSharesABatch(t *testing.T) {
if len(*asked) != before {
t.Fatalf("the catalogue walk started beside the controller's: asked %v", (*asked)[before:])
}
// Read again: the step above wrote the walk (what its key waits for), so the copy read at its cut is stale.
own2, err := open.inventory.PlanByID(ctx, ws[0].ID)
if err != nil {
t.Fatal(err)
}
own2 := ws[0]
own2.State = inventory.PlanDone
if err := open.inventory.SavePlan(ctx, &own2); err != nil {
t.Fatal(err)
+1 -1
View File
@@ -264,7 +264,7 @@ func TestAPlanLineNamesWhatItMoves(t *testing.T) {
}
return nil
}
if got, want := planLineOn(p, now, tierAtLeast, running),
if got, want := planLineOn(p, now, pauseView{}, tierAtLeast, running),
"mesh-catalog #175 a tap shows its outcome · messenger, telegram → novox · tier 1 of 1, building for 2m0s"; got != want {
t.Fatalf("the line reads %q, want %q", got, want)
}
-9
View File
@@ -4,7 +4,6 @@ package main
import (
"strings"
"sync/atomic"
"testing"
"time"
@@ -16,14 +15,6 @@ import (
func TestAnUnrelatedBatchWalksBesideAnOpenWalk(t *testing.T) {
open := windowed(t)
asksWithPaths(t)
// Both modules' keys are not built, so each walk stays open as it did while its build ran: with the stand-in
// builder's key done at once, one step would carry a one-module walk to done.
wholeKeys, held := openPlanKeys, &atomic.Bool{}
held.Store(true)
openPlanKeys = func() (planKeyReads, func(), error) {
all, done, err := wholeKeys()
return keysHolding{keysHolding{all, "one", held, ""}, "two", held, ""}, done, err
}
ctx := t.Context()
hear(t, open, repoMerge("one", "c1", t0), t0)
cutAt(t, open, t0.Add(90*time.Second))
File diff suppressed because it is too large Load Diff
+43
View File
@@ -0,0 +1,43 @@
package main
import (
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
func claiming(module, seat string, on ...string) inventory.Entry {
return inventory.Entry{
Manifest: catalogue.Manifest{Module: module, Claims: []catalogue.Claim{{Name: seat}}},
On: on,
}
}
// The controller asks the build role that has a holder (novox/hq ADR 0190 handover): the retired
// one while only the builder is assigned, the current one from the first build-agent on, and the
// current one when nothing holds either — where every ask goes once the handover is done.
func TestTheControllerAsksTheBuildRoleThatHasAHolder(t *testing.T) {
onlyTheBuilder := []inventory.Entry{
claiming("builder", link.TheBuildMachineBefore, "anchor"),
claiming("build-agent", link.TheBuildMachine), // registered, assigned nowhere yet
}
if got := buildSeatAmong(onlyTheBuilder); got != link.TheBuildMachineBefore {
t.Errorf("with only the builder assigned, asked %q", got)
}
bothHeld := []inventory.Entry{
claiming("builder", link.TheBuildMachineBefore, "anchor"),
claiming("build-agent", link.TheBuildMachine, "home-server"),
}
if got := buildSeatAmong(bothHeld); got != link.TheBuildMachine {
t.Errorf("with a build-agent assigned anywhere, asked %q", got)
}
neither := []inventory.Entry{claiming("builder", link.TheBuildMachineBefore)}
if got := buildSeatAmong(neither); got != link.TheBuildMachine {
t.Errorf("with no holder of either, asked %q, want the current role", got)
}
if got := buildSeatAmong(nil); got != link.TheBuildMachine {
t.Errorf("an empty catalogue asks %q", got)
}
}
+105 -86
View File
@@ -7,7 +7,6 @@ import (
"fmt"
"os"
"regexp"
"slices"
"strings"
"time"
@@ -189,13 +188,25 @@ func (f *sourceForms) same(aRepository, aSeat, bRepository, bSeat string) bool {
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, a merge whitelist naming
// the tools' account and no agent's account, and no administrator merging past the rule (novox/hq ADR 0534 §6; no
// required status, which the forge no longer asks, §1). Why says what is missing when it is not.
// 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
}
@@ -350,68 +361,46 @@ func askTheForgeOnce(ctx context.Context, owner, repo, branch string) (forgeFact
return facts, nil
}
// forgeToolsAccount is the forge tools' account, which main's merge whitelist names for mesh-delivery (novox/hq ADR 0398
// §2, the gitea module's tools-account.ts TOOLS_ACCOUNT); beside it the whitelist names only people's own logins, the
// operator's (ADR 0534 §2), never an agent's account.
const forgeToolsAccount = "mesh-tools"
// isAgentAccount is whether a forge login is a node's agent's account, agent-<node> (novox/hq ADR 0398 §1): one that
// may never merge into a module's trunk by itself.
func isAgentAccount(login string) bool { return strings.HasPrefix(login, "agent-") }
// 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"`
// MergeWhitelist is who may merge into the branch; nil when the rule keeps no merge whitelist (anyone who writes).
MergeWhitelist []string `json:"merge_whitelist"`
}
// 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. A trunk is trusted for who can land a commit on
// it, not for a status it requires (novox/hq ADR 0534 §6): no push, a merge whitelist naming the tools' account and
// beside it only people's own logins, never an agent's, and no administrator merging past the rule; then a commit lands
// there through mesh-delivery, on a passing mesh/check and the operator's Approve (ADR 0534 §4), or by the operator's
// own hand at the forge (§2).
// later rule that happens to be stronger: the forge does not read it.
func judgedRule(rules []protectionRule, branch string) forgeFacts {
switch r := appliedRule(rules, branch); {
var applied *protectionRule
for i := range rules {
if rules[i].Rule == branch {
applied = &rules[i]
break
}
}
if applied == nil {
for i := range rules {
if ruleCovers(rules[i].Rule, branch) {
applied = &rules[i]
break
}
}
}
switch r := applied; {
case r == nil:
return forgeFacts{Why: fmt.Sprintf("no protection rule covers %s", branch)}
case r.Push:
return forgeFacts{Why: fmt.Sprintf("the rule %s lets a person push to %s directly", r.Rule, branch)}
case r.MergeWhitelist == nil || len(r.MergeWhitelist) == 0:
return forgeFacts{Why: fmt.Sprintf("the rule %s keeps no merge whitelist naming %s alone for %s", r.Rule, forgeToolsAccount, branch)}
case !slices.Contains(r.MergeWhitelist, forgeToolsAccount):
return forgeFacts{Why: fmt.Sprintf("the rule %s lets %s merge into %s, and not %s", r.Rule,
strings.Join(r.MergeWhitelist, ", "), branch, forgeToolsAccount)}
case slices.ContainsFunc(r.MergeWhitelist, isAgentAccount):
return forgeFacts{Why: fmt.Sprintf("the rule %s lets an agent's account merge into %s: %s", r.Rule, branch,
strings.Join(r.MergeWhitelist, ", "))}
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 it", r.Rule, branch)}
return forgeFacts{Why: fmt.Sprintf("the rule %s lets an administrator merge into %s past a status", r.Rule, branch)}
}
return forgeFacts{Guarded: true}
}
// appliedRule is the one rule the forge applies to a branch: the rule named for it, else the first glob rule, in the
// forge's order, that covers it; nil when none does.
func appliedRule(rules []protectionRule, branch string) *protectionRule {
for i := range rules {
if rules[i].Rule == branch {
return &rules[i]
}
}
for i := range rules {
if ruleCovers(rules[i].Rule, branch) {
return &rules[i]
}
}
return nil
}
// ruleCovers says a protection rule's name — a branch, or a glob of them — covers a branch, as the forge reads it.
func ruleCovers(rule, branch string) bool {
if rule == branch {
@@ -474,42 +463,14 @@ func (f *sourceForms) onTheForge(repository, seat string) (owner, name string, o
// - 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
// a merge whitelist naming the tools' account and no agent's, no administrator merging past the rule (ADR 0534 §6).
// 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; or, for the build of a merge, when mesh-delivery announced it on the operator's Approve of
// that head (novox/hq ADR 0420 §7 for a moved module, ADR 0432 for a new one), the forge's checks still holding. path is the module's directory as built; branch the trunk it is registered from.
func mayRegisterFrom(ctx context.Context, inv *inventory.Inventory, module string, built inventory.Source,
ask registrationAsk, path, branch string) (int64, error) {
return mayRegisterIn(ctx, inv, newSourceForms(ctx, inv), module, built, ask, path, branch)
}
// registrationAsk is what a registration's warrant is read from (novox/hq ADR 0501 §3): the build an outcome answers,
// whose kept request says whether it was asked at the terminal or for a merge (build-agent's outcome, the builder's own
// module until R4g); or, for a take-in from the builder's done key, the merge the indexer's trunk value names and
// whether the operator asked the registration at the controller's terminal (ADR 0501 §4). A key asks no build, so it
// keeps no request.
type registrationAsk struct {
// Build is the build whose kept request is read; empty for a take-in from a key.
Build string
// Merge is the commit of the merge a key's module is registered from (the indexer's trunk value); empty for an
// outcome, whose request names it.
Merge string
// Terminal is a key registration the operator asked at the controller's terminal for this repository and path.
Terminal bool
}
// String is the ask as the registration's words name it.
func (a registrationAsk) String() string {
switch {
case a.Build != "":
return a.Build
case a.Terminal:
return "the registration asked at the terminal"
case a.Merge != "":
return fmt.Sprintf("the merge %.8s", a.Merge)
}
return "a key"
buildID, path, branch string) (int64, error) {
return mayRegisterIn(ctx, inv, newSourceForms(ctx, inv), module, built, buildID, path, branch)
}
// catalogueReader is the catalogue's manifests, as the router's records are found in them.
@@ -529,19 +490,15 @@ type registrations interface {
// mayRegisterIn is mayRegisterFrom on what it reads of the mesh handed in.
func mayRegisterIn(ctx context.Context, inv registrations, forms *sourceForms, module string, built inventory.Source,
ask registrationAsk, path, branch string) (int64, error) {
buildID := ask.String()
buildID, path, branch string) (int64, error) {
was, err := inv.SourceOf(ctx, module)
isNew := errors.Is(err, inventory.ErrNoSuchModule)
if err != nil && !isNew {
return 0, err
}
terminal := ask.Terminal
if ask.Build != "" {
var err error
if terminal, err = askedHereFor(ctx, inv, forms, ask.Build, built, path); err != nil {
return 0, err
}
terminal, err := askedHereFor(ctx, inv, forms, buildID, built, path)
if err != nil {
return 0, err
}
refuse := func(why string) (int64, error) {
return 0, fmt.Errorf("%w: %s. A module is registered from such a source only by a build asked at the "+
@@ -585,7 +542,7 @@ func mayRegisterIn(ctx context.Context, inv registrations, forms *sourceForms, m
// The forge's protection and the repository's identity are checked below as for every registration.
movedWhy := ""
if why != "" && !terminal {
accepted, err := approvedMove(ctx, inv, forms, ask, built)
accepted, err := approvedMove(ctx, inv, forms, buildID, built)
switch {
case err == nil:
movedWhy, why = accepted, ""
@@ -704,3 +661,65 @@ func sourceWords(repository, seat string) string {
}
return buildSource{Repository: repository, Seat: seat}.String()
}
// firstBuildWarrant is what the controller checked before it asks the first build of the modules a merge adds to a
// repository nothing is built from: the merge was made on the operator's Approve of its head (approvedAnnounced). It
// travels in the context of that ask alone, in this process, so no verb carries one.
type firstBuildWarrant struct {
repository string // owner/name on the git seat
paths map[string]bool
why string
}
type firstBuildWarrantKey struct{}
// withFirstBuildWarrant is ctx carrying the warrant for asking the first builds of paths in repository.
func withFirstBuildWarrant(ctx context.Context, repository string, paths []string, why string) context.Context {
w := &firstBuildWarrant{repository: repository, paths: map[string]bool{}, why: why}
for _, p := range paths {
w.paths[strings.Trim(p, "/")] = true
}
return context.WithValue(ctx, firstBuildWarrantKey{}, w)
}
func firstBuildWarrantIn(ctx context.Context) *firstBuildWarrant {
w, _ := ctx.Value(firstBuildWarrantKey{}).(*firstBuildWarrant)
return w
}
// covers says the warrant is for exactly this ask: the repository on the git seat, and a module path the merge holds.
func (w *firstBuildWarrant) covers(source buildSource, path string) bool {
return w != nil && source.Seat == gitSeat && strings.EqualFold(strings.Trim(source.Repository, "/"), w.repository) &&
w.paths[strings.Trim(path, "/")]
}
// verbMayAsk refuses, for a call through a verb, a build of a repository the catalogue builds no module from
// (novox/hq ADR 0266): the build node would run what an agent wrote, and its outcome could never be
// registered anyway. url is the repository as it is cloned. At the terminal anything may be asked.
func verbMayAsk(ctx context.Context, source buildSource, url, path string) error {
if startedAtTheTerminal() {
return nil
}
// **Or the first build of a module a merge adds on the operator's Approve** (novox/hq ADR 0432, issue 1146): the
// controller's own ask, carrying the warrant it checked, for that repository and the module paths the merge holds.
if firstBuildWarrantIn(ctx).covers(source, path) {
return nil
}
open, err := openStores(ctx)
if err != nil {
return err
}
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)
}
@@ -129,6 +129,27 @@ func TestADeliveryFromTheRegisteredRepositoryIsRegistered(t *testing.T) {
}
}
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)
}
}
func TestARollbackNeverPutsBackABuildFromAnotherRepository(t *testing.T) {
open := theCatalogue(t)
ctx := t.Context()
+4 -38
View File
@@ -102,17 +102,16 @@ func init() {
// 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) {
tools := []string{forgeToolsAccount}
guarded := protectionRule{Rule: "*", MergeWhitelist: tools}
open := protectionRule{Rule: "main", Push: true, MergeWhitelist: tools}
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*", MergeWhitelist: nil}, {Rule: "*", MergeWhitelist: tools}},
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", MergeWhitelist: tools}}, "main"); !f.Guarded {
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 {
@@ -120,39 +119,6 @@ func TestTheRuleJudgedIsTheOneTheForgeApplies(t *testing.T) {
}
}
// **A module's trunk is trusted when it takes no push, only the tools' account and people's own logins merge into it,
// and no administrator merges past its rule** (novox/hq ADR 0534 §2, §6): the forge requires no status to merge (§1), so
// a rule requiring none is trusted; one with no merge whitelist, no tools' account in it, an agent's account in it, a
// push allowed or an administrator's override is not.
func TestATrunkIsTrustedForItsMergersNotForAStatus(t *testing.T) {
tools := []string{forgeToolsAccount}
for _, c := range []struct {
name string
rule protectionRule
want bool
why string
}{
{"no status, no push, the tools alone merge", protectionRule{Rule: "main", MergeWhitelist: tools}, true, ""},
{"a status required as well", protectionRule{Rule: "main", MergeWhitelist: tools, RequiredStatuses: []string{"mesh/check"}}, true, ""},
{"no merge whitelist", protectionRule{Rule: "main"}, false, "merge whitelist"},
{"an empty merge whitelist", protectionRule{Rule: "main", MergeWhitelist: []string{}}, false, "merge whitelist"},
{"the operator's own login beside the tools", protectionRule{Rule: "main", MergeWhitelist: []string{forgeToolsAccount, "jschoubben"}}, true, ""},
{"an agent's account in the whitelist", protectionRule{Rule: "main", MergeWhitelist: []string{forgeToolsAccount, "agent-ace"}}, false, "agent-ace"},
{"an agent's account alone", protectionRule{Rule: "main", MergeWhitelist: []string{"agent-ace"}}, false, "agent-ace"},
{"a person's login without the tools", protectionRule{Rule: "main", MergeWhitelist: []string{"jschoubben"}}, false, forgeToolsAccount},
{"a push allowed", protectionRule{Rule: "main", Push: true, MergeWhitelist: tools}, false, "push"},
{"an administrator's override", protectionRule{Rule: "main", MergeWhitelist: tools, AdminMayOverride: true}, false, "administrator"},
} {
f := judgedRule([]protectionRule{c.rule}, "main")
if f.Guarded != c.want || !strings.Contains(f.Why, c.why) {
t.Errorf("%s: %+v, want guarded %v saying %q", c.name, f, c.want, c.why)
}
if strings.Contains(f.Why, "requires no status") {
t.Errorf("%s: a trunk is still judged on a required status: %s", c.name, f.Why)
}
}
}
// An id the forge could not give keeps the id already recorded.
// **The operator's Approve of a reviewed group registers a moved module from its new repository** (novox/hq ADR 0420 §7,
+11
View File
@@ -2,10 +2,12 @@ package main
import (
"bytes"
"reflect"
"strings"
"testing"
"time"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
@@ -66,3 +68,12 @@ func TestBuildsSaysTheSourceFingerprint(t *testing.T) {
t.Errorf("the other lines changed:\n%s\nwant\n%s", plain.String(), want)
}
}
// 1215/AC4: what the builder said of the inputs is what the controller keeps of the build.
func TestTheControllerKeepsTheInputsABuilderSaid(t *testing.T) {
in := &sdkdelivery.SourceInputs{Module: "mailu", Tree: "t", Toolchains: map[string]string{"code": "ts sha256:aa 01"}}
kept := buildFrom(link.BuildResult{ID: "build-1", Repository: "r", Commit: "c", SourceFingerprint: "src1:x", SourceInputs: in})
if !reflect.DeepEqual(kept.SourceInputs, in) {
t.Errorf("the controller keeps %+v, the builder said %+v", kept.SourceInputs, in)
}
}
@@ -1,109 +0,0 @@
//go:build integration
package main
import (
"context"
"testing"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/testbus"
)
// builderOnTheBus is the builder's and the indexer's buckets on a test's own bus, which the controller's connection for
// the moment (servingBus) reaches: what a take-in from a key reads (R4e, issue 1851).
type builderOnTheBus struct {
js *broker.JetStream
builds, modules, manifests jetstream.KeyValue
}
func aBuilderOnTheBus(t *testing.T) builderOnTheBus {
t.Helper()
js, err := broker.Dial(testbus.URL(t))
if err != nil {
t.Fatal(err)
}
t.Cleanup(js.Close)
api, err := jetstream.New(js.Conn())
if err != nil {
t.Fatal(err)
}
ctx := t.Context()
b := builderOnTheBus{js: js}
for _, c := range []struct {
into *jetstream.KeyValue
bucket string
history uint8
}{
{&b.builds, sdkdelivery.BuilderBuildsBucket, sdkdelivery.BuilderBuildsHistory},
{&b.modules, sdkdelivery.ModulesBucket, sdkdelivery.ModulesHistory},
{&b.manifests, sdkdelivery.ManifestsBucket, 1},
} {
kv, err := api.CreateKeyValue(ctx, jetstream.KeyValueConfig{Bucket: c.bucket, History: c.history})
if err != nil {
t.Fatal(err)
}
*c.into = kv
}
servingBus.Store(js)
t.Cleanup(func() { servingBus.Store(nil) })
withTrunkList(t, true)
// The indexer on this bus is read, not the "not held" a mesh with no bus reads (aMeshOnTheStore).
wasIndexed := indexedModule
indexedModule = theIndexedModule
t.Cleanup(func() { indexedModule = wasIndexed })
return b
}
// indexed writes the indexer's trunk value of module at commit, naming fp, and its manifest: a module of one archive.
func (b builderOnTheBus) indexed(t *testing.T, module, version, repository, path, commit, fp string) string {
t.Helper()
keys, trunk := aKeyFor(module, version, fp, repository, path, commit)
raw := keys.manifests[trunk.Manifest]
value, err := sdkdelivery.EncodeManifest(sdkdelivery.Manifest{Module: module, JSON: raw})
if err != nil {
t.Fatal(err)
}
if _, err := b.manifests.Put(context.Background(), trunk.Manifest, value); err != nil {
t.Fatal(err)
}
trunk.FP = "src1:" + fp[len("src2:"):]
encoded, err := sdkdelivery.EncodeModule(trunk)
if err != nil {
t.Fatal(err)
}
if _, err := b.modules.Put(context.Background(), module, encoded); err != nil {
t.Fatal(err)
}
return trunk.Manifest
}
// built writes the builder's key of module at fp: done, its one archive named with a reference, or in another state.
func (b builderOnTheBus) built(t *testing.T, module, version, fp, manifest, state string) {
t.Helper()
keys, _ := aKeyFor(module, version, fp, "", "", "")
v := keys.keys[module].Build
v.Manifest = manifest
if state != sdkdelivery.BuildDone {
v = sdkdelivery.BuilderBuild{State: state, By: "ace"}
if state == sdkdelivery.BuildFailed {
v.Error = "digests differ: ace sha256:1, g14 sha256:2"
}
}
value, err := sdkdelivery.EncodeBuilderBuild(v)
if err != nil {
t.Fatal(err)
}
key, err := sdkdelivery.BuilderBuildKey(module, fp)
if err != nil {
t.Fatal(err)
}
if _, err := b.builds.Put(context.Background(), key, value); err != nil {
t.Fatal(err)
}
}
// theIndexedModule is the look at the indexer as the controller makes it, kept before any test replaces it.
var theIndexedModule = indexedModule
-7
View File
@@ -32,13 +32,6 @@ func assertBusObjects(ctx context.Context, inv *inventory.Inventory, r broker.Ra
if err := broker.Raise(r, names); err != nil {
return nil, err
}
// And each node's bucket for the files of its declaration that travel by digest (novox/hq ADR 0486 §4), so a node finds its own
// whenever the bus is up.
if files, ok := r.(interface{ EnsureDeclarationFiles(nodes []string) error }); ok {
if err := files.EnsureDeclarationFiles(names); err != nil {
return nil, err
}
}
// The work queues of the mesh's own roles (novox/hq ADR 0121). The queue before the holder,
// deliberately: work queues until somebody arrives to do it, so assigning a build machine a week
// after something started asking for builds flushes the backlog instead of having lost it.
+9 -17
View File
@@ -46,33 +46,25 @@ func aCarriedConsumer(t *testing.T, open *stores) broker.Consumer {
return broker.Consumer{}
}
// aLateHolder is a module holding a build seat (node-build-go) on anchor, assigned after the controller started, and
// the worker its seat's queue should have for it: a build seat is pulled by kind, so the worker is the one for the
// seat's first accepted kind (the build agent's seat, which had one worker for all, is gone with the builder).
// aLateHolder is a module holding the build agent's seat on anchor, assigned after the controller
// started, and the worker its seat's queue should have for it.
func aLateHolder(t *testing.T, open *stores) broker.Consumer {
t.Helper()
// With an account to read, as a holder has: a module with none is no bus user, so its seat's workers are not derived.
register(t, open, catalogue.Manifest{Module: "late-builder", Version: "1",
OwnSecrets: catalogue.OwnSecrets{"broker": {Path: "/var/lib/mesh/late-builder/broker"}},
Claims: []catalogue.Claim{{Name: "node-build-go", Scope: catalogue.ScopeNode}}})
Claims: []catalogue.Claim{{Name: "node-build-agent", Scope: catalogue.ScopeNode}}})
if _, err := open.inventory.Assign(t.Context(), "anchor", "late-builder"); err != nil {
t.Fatal(err)
}
_, workers, err := seatTrafficObjects(t.Context(), open.inventory)
if err != nil {
t.Fatal(err)
}
for _, s := range inventory.MeshSeats() {
if s.Name != "node-build-go" || len(s.Accepts) == 0 {
continue
}
for _, w := range workers {
if w.Name == broker.WorkerName(s.Name, s.Accepts[0]) {
return w
if s.Name == "node-build-agent" {
c, needed := broker.HolderConsumerFor("anchor", "late-builder", s)
if !needed {
t.Fatal("the build agent's seat needs no worker")
}
return c
}
}
t.Fatal("the build seat's holder is derived no worker")
t.Fatal("the mesh declares no build agent's seat")
return broker.Consumer{}
}
+7 -3
View File
@@ -241,9 +241,10 @@ func mergeAskPrefix(number int, head string) string {
return fmt.Sprintf("merge-%d-%s-", number, head)
}
// askersOfMerges are the askers a merge's warrant is read under: the teller alone (ADR 0437 §6), since R4e stopped taking
// mesh-delivery's (ADR 0450 §8; issue 1318).
func askersOfMerges() []string { return []string{sdkdelivery.TellerModule} }
// askersOfMerges are the askers a merge's warrant is read under: the teller, and mesh-delivery for a merge before release 3e
// (ADR 0437 §6). Release 4's controller work stops accepting mesh-delivery (ADR 0450 §8); a merge asked under it earlier is
// still found here until then.
func askersOfMerges() []string { return []string{sdkdelivery.TellerModule, theDelivery} }
// caughtUpWarrant judges a chosen merge ask of the router as the warrant of a merge: the ask is the operator's, the warrant
// answers that very ask as it was published, and it is the operator's Approve proved by the channel it came through (ADR 0437
@@ -390,6 +391,9 @@ var findMergeWarrant = func(ctx context.Context, inv catalogueReader, owner, rep
if asker == sdkdelivery.TellerModule && !strings.HasPrefix(id, sdkdelivery.AskKindMerge+"-") {
continue
}
if asker == theDelivery && !named {
continue
}
ask, w, err := closedRecordOf(ctx, inv, asker, id)
if err != nil {
return "", false, "", err
+34 -6
View File
@@ -69,12 +69,12 @@ func TestAForgeThatDoesNotAnswerIsAnErrorNotAnEmptyTrunk(t *testing.T) {
func TestACaughtUpMergeIsAcceptedOnTheSameWarrantAsAnAnnouncedOne(t *testing.T) {
ok := approvedMerge(t)
if _, err := caughtUpWarrant(sdkdelivery.TellerModule, &ok.ask, &ok.warrant); err != nil {
t.Errorf("the operator's Approve under the teller was refused: %v", err)
}
// 1318 (ADR 0450 §8): mesh-delivery's warrant of before the gate is no longer taken.
if _, err := caughtUpWarrant(theDelivery, &ok.ask, &ok.warrant); err == nil {
t.Error("a warrant under mesh-delivery was accepted")
for _, asker := range []string{sdkdelivery.TellerModule, theDelivery} {
w := ok.warrant
w.Asker = asker
if _, err := caughtUpWarrant(asker, &ok.ask, &w); err != nil {
t.Errorf("the operator's Approve under %s was refused: %v", asker, err)
}
}
if _, err := caughtUpWarrant("keycloak", &ok.ask, &ok.warrant); err == nil {
t.Error("a warrant of an asker the controller takes none from was accepted")
@@ -229,6 +229,34 @@ func TestAGroupsWarrantIsFoundByTheAsksWordsForThePullRequestAndItsHead(t *testi
}
}
// Keycloak's merge was asked as mesh-delivery.merge-378-b9104454f9a5-1, before release 3e: the id names the pull request and
// the first twelve digits of its head, and the warrant is found by them under mesh-delivery.
func TestAWarrantAskedUnderMeshDeliveryIsFoundByThePullRequestAndItsHead(t *testing.T) {
ok := approvedMerge(t)
head := "b9104454f9a5e0c4d3b2a1f0e9d8c7b6a5948372"
askID := "merge-378-b9104454f9a5-1"
ask, w := ok.ask, ok.warrant
ask.ID, w.Ask, w.Asker = askID, askID, theDelivery
w.AskDigest = ask.Digest()
withRouter(t, map[string]map[string]aRouterRecord{theDelivery: {
"merge-377-aaaaaaaaaaaa-1": {},
askID: {ask, w},
}})
why, found, reason, err := findMergeWarrant(t.Context(), nil, "novox", "mesh-catalog", 378, head)
if err != nil || !found || !strings.Contains(why, askID) {
t.Fatalf("the warrant was not found: %q %v %q %v", why, found, reason, err)
}
// Another head names another ask, and none is found.
if _, found, _, _ := findMergeWarrant(t.Context(), nil, "novox", "mesh-catalog", 378, "ffffffffffff0000"); found {
t.Fatal("another head found a warrant")
}
// A router that cannot be read is an error, never "no warrant".
routerAskIDs = func(context.Context, catalogueReader, string) ([]string, error) { return nil, errors.New("no answer") }
if _, found, _, err := findMergeWarrant(t.Context(), nil, "novox", "mesh-catalog", 378, head); found || err == nil {
t.Fatalf("an unreadable router read as no warrant: %v %v", found, err)
}
}
// A trunk shorter than a page, or with none, is read without a fault: the pages end when the forge's do.
func TestAShortTrunkIsReadWholeAndAnEmptyOneIsNoMerges(t *testing.T) {
was := trunkCommitsPage
+44 -2
View File
@@ -51,13 +51,55 @@ func attachTheChain(ctx context.Context, inv *inventory.Inventory, server *link.
fmt.Printf("chain: the builds are not laid out under their modules yet, and none is read until they are: %v\n", err)
go layBuildsOutSoon(ctx, inv)
}
// The requests the move of delivery 1 found never handed over are not asked again: the controller asks no build
// (novox/hq ADR 0501; issue 1851, A10), and the builder builds every fingerprint the indexer writes.
// And the requests the move of delivery 1 found never handed over, asked again once (askAgain), once it is serving.
go askAgainSoon(ctx, inv)
// And what delivery 1 recorded of what each machine was sent, copied onto the keys of the assignments (CopySentBuilds).
go copySentBuildsSoon(ctx, inv)
return nil
}
// askAgain asks once more for each build request the move found never handed over (novox/hq issue 981 note 7 Q5): the
// requests a merge could not ask for, which a person's hand `build` would otherwise have to repeat. **At most once**: a
// request is claimed before it is asked and settled after, so a crash between loses the ask, which is where it stood.
func askAgain(ctx context.Context, inv *inventory.Inventory) {
pending, err := inv.RequestsToAskAgain(ctx)
if err != nil {
fmt.Printf("chain: the requests never handed over cannot be read: %v\n", err)
return
}
for _, r := range pending {
if claimed, err := inv.ClaimToAskAgain(ctx, r.ID); err != nil || !claimed {
if err != nil {
fmt.Printf("chain: %s cannot be claimed to be asked again: %v\n", r.ID, err)
}
continue
}
id, err := askABuild(ctx, buildSource{Repository: r.Repository, Seat: r.Seat}, r.Path, r.Ref)
if err != nil {
if serr := inv.AskedAgainAs(ctx, r.ID, "not asked again: "+err.Error()); serr != nil {
fmt.Printf("chain: %s was not asked again, and that could not be kept: %v\n", r.ID, serr)
}
fmt.Printf("chain: %s was never handed over and cannot be asked again: %v\n", r.ID, err)
continue
}
recordBuildRequest(ctx, inv, inventory.BuildRequest{ID: id, Repository: r.Repository, Seat: r.Seat, Path: r.Path,
Ref: r.Ref, Commit: r.Commit, For: r.For})
if err := inv.AskedAgainAs(ctx, r.ID, id); err != nil {
fmt.Printf("chain: %s was asked again as %s, and that could not be kept: %v\n", r.ID, id, err)
}
fmt.Printf("chain: %s, never handed over, is asked again as %s\n", r.ID, id)
}
}
// askAgainSoon waits for the controller to be serving, which an ask needs, and then asks again.
func askAgainSoon(ctx context.Context, inv *inventory.Inventory) {
select {
case <-ctx.Done():
case <-time.After(time.Minute):
askAgain(ctx, inv)
}
}
// kindSentBuildsNotCopied is the condition of the one-time copy of node.sent_builds onto the assignments' keys that has not
// succeeded: until it has, a module an assignment pulled in may be missing from what a machine is read to run.
const kindSentBuildsNotCopied = "sent-builds-not-copied"
+17 -4
View File
@@ -14,6 +14,7 @@ import (
"github.com/novox/mesh-controller/internal/chain"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/lease"
"github.com/novox/mesh-controller/internal/link"
)
// **Acceptance 10 of design 47: one read of an assignment names the commit, the build and the rollout that put it there,
@@ -79,7 +80,7 @@ func TestOneReadOfAnAssignmentNamesItsCommitBuildAndRolloutAndOneReadOfTheRollou
if err != nil {
t.Fatal(err)
}
for try := 0; try < 6 && (asking.Modules["notes"] == nil || asking.Modules["notes"].Build == "" || asking.Modules["notes"].State == "asked"); try++ {
for try := 0; try < 3 && (asking.Modules["notes"] == nil || asking.Modules["notes"].Build == ""); try++ {
advancePlans(ctx, open)
if asking, err = inv.PlanByID(ctx, rollout); err != nil {
t.Fatal(err)
@@ -89,6 +90,13 @@ func TestOneReadOfAnAssignmentNamesItsCommitBuildAndRolloutAndOneReadOfTheRollou
if buildID == "" || len(*asked) != 1 || !strings.HasSuffix((*asked)[0], "#"+buildID) {
t.Fatalf("the walk asked %v, and notes' build is %q", *asked, buildID)
}
digest := "sha256:" + strings.Repeat("a", 64)
result := outcome(buildID, "modules/notes", "c-notes", "notes", "")
result.ReuseKey = "rk-notes"
result.Made = []link.MadeArtifact{{Name: "notes", Kind: "image", Reference: "artifact-store://notes/notes@" + digest}}
if err := (builds{inv: inv, open: open}).Built(ctx, result); err != nil {
t.Fatal(err)
}
// The walk sends the tier to its first machine.
for try := 0; try < 3 && len(sent) == 0; try++ {
@@ -108,9 +116,9 @@ func TestOneReadOfAnAssignmentNamesItsCommitBuildAndRolloutAndOneReadOfTheRollou
if wireguard, pulled := assignmentOf(on, "mesh-wireguard"); pulled && (wireguard.Build != "" || wireguard.Rollout != "") {
t.Fatalf("a module this rollout did not send names a build or a rollout: %+v", wireguard)
}
if a.Module != "notes" || a.Commit != "c-notes" || a.Build != buildID || a.Rollout != rollout ||
if a.Module != "notes" || a.Commit != "c-notes" || a.Build != buildID || a.Rollout != rollout || a.Artifact != digest ||
a.Since.IsZero() || a.Assigned.IsZero() || !a.Stands() {
t.Fatalf("the assignment names %+v, not commit c-notes, build %s, rollout %s", a, buildID, rollout)
t.Fatalf("the assignment names %+v, not commit c-notes, build %s, artifact %s, rollout %s", a, buildID, digest, rollout)
}
// One read of the rollout.
p, err := inv.PlanByID(ctx, rollout)
@@ -122,11 +130,16 @@ func TestOneReadOfAnAssignmentNamesItsCommitBuildAndRolloutAndOneReadOfTheRollou
p.Steps[0].Node != "anchor" || p.Steps[0].Sent == nil {
t.Fatalf("the rollout lists changed %+v, builds %v, steps %+v; want %+v", p.Changed, p.Builds, p.Steps, wantChange)
}
// And one read of the build names the same commit.
// And one read of the build names the same commit, with what it published and its reuse key.
built, found, err := inv.BuildByID(ctx, buildID)
if err != nil || !found || built.Commit != "c-notes" || built.Module != "notes" {
t.Fatalf("the build: %+v %v %v", built, found, err)
}
lately, err := inv.Builds(ctx, "notes", 1)
if err != nil || len(lately) != 1 || lately[0].ReuseKey != "rk-notes" || lately[0].Published == nil ||
lately[0].Published.Artifact != digest {
t.Fatalf("what the build published: %+v %v", lately, err)
}
// The verbs say the links.
if lines := changedLines(p); len(lines) != 2 || !strings.Contains(lines[1], "anchor/notes") ||
!strings.Contains(lines[1], buildID) {
@@ -6,6 +6,8 @@ import (
"strings"
"testing"
"time"
"github.com/novox/mesh-controller/internal/link"
)
// **Acceptance 12 of design 47: no table in the controller's store holds builds, rollouts or assignments** (novox/hq ADR
@@ -32,7 +34,7 @@ func TestNoTableOfTheControllersStoreHoldsBuildsRolloutsOrAssignments(t *testing
if err != nil {
t.Fatal(err)
}
for try := 0; try < 6 && (asking.Modules["notes"] == nil || asking.Modules["notes"].Build == "" || asking.Modules["notes"].State == "asked"); try++ {
for try := 0; try < 3 && (asking.Modules["notes"] == nil || asking.Modules["notes"].Build == ""); try++ {
advancePlans(ctx, open)
if asking, err = inv.PlanByID(ctx, rollout); err != nil {
t.Fatal(err)
@@ -42,6 +44,12 @@ func TestNoTableOfTheControllersStoreHoldsBuildsRolloutsOrAssignments(t *testing
if buildID == "" || len(*asked) != 1 {
t.Fatalf("the walk asked %v, and notes' build is %q", *asked, buildID)
}
digest := "sha256:" + strings.Repeat("b", 64)
result := outcome(buildID, "modules/notes", "c-notes", "notes", "")
result.Made = []link.MadeArtifact{{Name: "notes", Kind: "image", Reference: "artifact-store://notes/notes@" + digest}}
if err := (builds{inv: inv, open: open}).Built(ctx, result); err != nil {
t.Fatal(err)
}
// What the live path made is in the chain.
if built, found, err := inv.BuildByID(ctx, buildID); err != nil || !found || built.Commit != "c-notes" {
+29
View File
@@ -190,3 +190,32 @@ func summaryOf(p link.ChangePlan) string {
}
return strings.Join(parts, "; ")
}
// planText is a change plan as a person reads it, for the pull request's comment.
func planText(p link.ChangePlan) string {
var b strings.Builder
fmt.Fprintf(&b, "change plan of %s at %.8s into %s: %s\n", p.Repository, p.Head, p.Base, p.Summary)
for i, tier := range p.Tiers {
fmt.Fprintf(&b, " tier %d: %s\n", i, strings.Join(tier, ", "))
}
for _, m := range p.Machines {
line := " " + m.Machine + ": "
if len(m.Receives) > 0 {
line += "receives " + strings.Join(m.Receives, ", ")
}
if len(m.Waits) > 0 {
if len(m.Receives) > 0 {
line += "; "
}
line += "waits for a person: " + strings.Join(m.Waits, ", ")
}
b.WriteString(line + "\n")
}
for _, s := range p.Steps {
b.WriteString(" - " + s + "\n")
}
if len(p.Unread) > 0 {
fmt.Fprintf(&b, " read by no module's build: %s\n", strings.Join(p.Unread, ", "))
}
return b.String()
}
+3
View File
@@ -72,4 +72,7 @@ func TestAChangePlanSaysWhatEachMachineReceives(t *testing.T) {
if !strings.HasPrefix(p.Summary, "builds nothing") || len(p.Machines) != 0 || strings.Join(p.Unread, ",") != "merge-check.sh" {
t.Errorf("a root file's plan reads %+v", p)
}
if !strings.Contains(planText(p), "read by no module's build: merge-check.sh") {
t.Errorf("the plan's text does not say why nothing is built:\n%s", planText(p))
}
}
+87
View File
@@ -0,0 +1,87 @@
package main
import (
"context"
"errors"
"fmt"
"os"
"strconv"
"github.com/novox/mesh-controller/internal/builder"
"github.com/novox/mesh-controller/internal/inventory"
)
// A pull request's check builds go when it closes (novox/hq ADR 0350 §5): the build seat keeps them in the artifact
// store's scratch namespace for the walk of its merge to take. **Let go of where the moment is heard**: the
// controller ends the walk answering a merge — after that walk took what it reuses — and the delivery's owner, which
// hears the forge say a pull request closed unmerged, asks the `forget-checks` verb; both let go through the store,
// as the controller's own collection does. The build seat's holders let go of anything a week old on their own,
// so a word lost here leaves a week's builds, never more.
// checkBuildsIn is the scratch namespace as the controller reaches the artifact store.
func checkBuildsIn(ctx context.Context, inv *inventory.Inventory) (builder.Scratch, error) {
shelf, err := inv.Catalogue(ctx)
if err != nil {
return nil, err
}
address, err := artifactStoreAddress(ctx, inv, shelf, "")
if err != nil {
return nil, err
}
if address == "" {
return nil, errors.New("the artifact store is not on the mesh's network")
}
return builder.Registry{Address: address}, nil
}
// forgetChecks lets go of a pull request's check builds and answers what it did.
var forgetChecks = func(ctx context.Context, inv *inventory.Inventory, pull builder.PullRef) (string, error) {
scratch, err := checkBuildsIn(ctx, inv)
if err != nil {
return "", err
}
n, err := scratch.Forget(ctx, pull)
if err != nil {
return "", fmt.Errorf("%s: %d check(s)' builds let go of, and not all: %w", pull, n, err)
}
return fmt.Sprintf("%s: %d check(s)' builds let go of", pull, n), nil
}
// forgetChecksOfWalk lets go, on the save that ends a walk, of the check builds of every pull request whose merge
// it answers: the walk took what it reuses. Said, never fatal.
func forgetChecksOfWalk(ctx context.Context, inv *inventory.Inventory, p inventory.Plan) {
if p.WalkSaid != inventory.WalkEnded || p.Delivery == nil {
return
}
for _, pull := range reuseAskOf(&p).Pulls {
said, err := forgetChecks(ctx, inv, builder.PullRef{Owner: pull.Owner, Repo: pull.Repo, Number: pull.Number})
if err != nil {
fmt.Fprintf(os.Stderr, "%s: its check builds stay until the week's sweep: %v\n", p.ID, err)
continue
}
fmt.Printf("%s: %s\n", p.ID, said)
}
}
// forgetChecksCommand is `forget-checks <owner> <repo> <number>`: the verb the delivery's owner asks when the forge
// says a pull request closed unmerged (novox/hq ADR 0350 §5).
func forgetChecksCommand(ctx context.Context, args []string) error {
if len(args) != 3 {
return errors.New("forget-checks <owner> <repo> <number>")
}
number, err := strconv.Atoi(args[2])
if err != nil || number <= 0 {
return fmt.Errorf("%q is not a pull request's number", args[2])
}
open, err := openStores(ctx)
if err != nil {
return err
}
defer open.Close()
said, err := forgetChecks(ctx, open.inventory, builder.PullRef{Owner: args[0], Repo: args[1], Number: number})
if err != nil {
return err
}
fmt.Println(said)
return nil
}
@@ -0,0 +1,102 @@
//go:build integration
package main
import (
"encoding/json"
"reflect"
"strings"
"testing"
"time"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
func TestOnlyAWalksAskTakesAChecksBuildAndADryRunNever(t *testing.T) {
p := &inventory.Plan{Delivery: &inventory.PlanDelivery{Merges: []inventory.PlanMerge{
{Repository: "novox/shop", Number: 7}, {Repository: "http://forge.invalid/novox/web.git", Number: 3},
{Repository: "novox/shop", Number: 7}, {Repository: "novox/shop"}}}}
asked := reuseAskOf(p)
if want := []link.PullRef{{Owner: "novox", Repo: "shop", Number: 7}, {Owner: "novox", Repo: "web", Number: 3}}; !reflect.DeepEqual(asked.Pulls, want) {
t.Fatalf("the walk names %+v", asked.Pulls)
}
walk := withReuse(t.Context(), asked)
if reuseAsked(walk, false) != asked {
t.Fatal("a walk's ask does not carry its leave")
}
if reuseAsked(walk, true) != nil {
t.Fatal("a dry run carries a leave to take a check's build")
}
if reuseAsked(t.Context(), false) != nil {
t.Fatal("an ask that is not a walk's carries a leave")
}
}
func TestAReusedCheckBuildIsRegisteredByItsDigestAndGatedAsABuiltOne(t *testing.T) {
withTheGate(t) // the first-machine gate's own path (novox/hq ADR 0360 keeps it for earlier walks)
g := aGateMesh(t)
ctx := t.Context()
inv := g.open.inventory
gateEvery = 0
digest := "sha256:" + strings.Repeat("c", 64)
reused := func(id, reference string) link.BuildResult {
manifest := mustJSON(t, catalogue.Manifest{Module: "app", Version: "c3", Resources: []map[string]any{
{"id": "svc", "type": "container", "name": "app", "image": reference}}})
return link.BuildResult{ID: id, Repository: "novox/mesh-catalog", Path: "modules/app", Ref: "main", On: "anchor",
Commit: "c3", Manifest: manifest, Trunk: "main", OnTrunk: true, Branches: []string{"main"},
Made: []link.MadeArtifact{{Name: "server", Kind: "image", Reference: reference}},
Source: &link.SourceOnSeat{Seat: "git", Repository: "novox/mesh-catalog"},
Reused: "reused from the check of novox/mesh-catalog#9 at head abcdefabcdef (merged into trunk 123456123456): server " + digest,
ReusedFrom: &link.ReusedFrom{Pull: link.PullRef{Owner: "novox", Repo: "mesh-catalog", Number: 9},
Head: "abcdefabcdef", Trunk: "123456123456", Key: "r1:" + strings.Repeat("1", 64)}}
}
if _, _, err := takeIn(ctx, inv, reused(link.NewBuildID(time.Now()), "store.internal:5000/scratch/app/server@"+digest)); err == nil ||
!strings.Contains(err.Error(), "scratch") {
t.Fatalf("a build in the scratch namespace was registered: %v", err)
}
id := link.NewBuildID(time.Now().Add(time.Second))
result := reused(id, "store.internal:5000/app/server@"+digest)
if _, _, err := takeIn(ctx, inv, result); err != nil {
t.Fatal(err)
}
current, err := inv.CurrentBuilds(ctx)
if err != nil || current["app"].Commit != "c3" {
t.Fatalf("the reused build is not what app is registered at: %+v %v", current["app"], err)
}
shelf, err := inv.Catalogue(ctx)
if err != nil {
t.Fatal(err)
}
raw, _ := json.Marshal(shelf["app"].Resources)
if !strings.Contains(string(raw), "app/server@"+digest) {
t.Fatalf("app is registered as %s, not by the check's digest in its own namespace", raw)
}
// The walk asked it: the walk keeps what its build did, and its gate judges the registered build.
asked := time.Now().UTC()
p := g.plan(t)
p.Commit = "c3"
p.Modules["app"] = &inventory.PlanModule{State: "asked", AskedAt: &asked, Build: id}
if err := inv.SavePlan(ctx, &p); err != nil {
t.Fatal(err)
}
planBuiltSaying(ctx, g.open, "app", "c3", "", asked, id, result.MadeWords())
p = g.plan(t)
if m := p.Modules["app"]; m.State != "built" || !strings.HasPrefix(m.Made, "reused from the check of novox/mesh-catalog#9") {
t.Fatalf("the walk keeps %+v", m)
}
for i := 0; i < 6; i++ {
advancePlans(ctx, g.open)
}
p = g.plan(t)
gate := p.Modules["app"].Gate
if gate == nil || gate.To != "c3" || gate.Verdict != inventory.GatePassed {
t.Fatalf("the gate judged %+v, not the reused build", gate)
}
if sent, _, _ := inv.SentBuilds(ctx, "anchor"); sent["app"] != "c3" {
t.Fatalf("the first machine was sent %v, not the reused build", sent)
}
}
+193
View File
@@ -0,0 +1,193 @@
package main
import (
"context"
"errors"
"reflect"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/builder"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// One build per change (novox/hq ADR 0350), as the controller asks it and takes it in.
// The check of a change is asked to build what a merge of it would move, in the walk's order: a module of the
// repository checked from the merge, one standing on it from its own repository as the walk builds it, and one
// that reads the repository through a context not at all. A change that moves no module builds nothing.
func TestACheckIsAskedToBuildWhatTheMergeWouldMove(t *testing.T) {
entries := []inventory.Entry{
{Manifest: catalogue.Manifest{Module: "app"}, Source: inventory.Source{Repository: "novox/shop", Seat: "git", Path: "modules/app"}},
{Manifest: catalogue.Manifest{Module: "web"}, Source: inventory.Source{Repository: "novox/web", Seat: "git", Ref: "main"}},
{Manifest: catalogue.Manifest{Module: "docs"}, Source: inventory.Source{Repository: "novox/docs", Seat: "git"}},
}
p := link.PullUpdated{Owner: "novox", Repo: "shop", Number: 7, Base: "main", Commit: "abc"}
scope := checkScope{Reach: mergeReach{Packaging: entries[2:], Plan: inventory.Plan{Tiers: [][]string{{"app", "docs"}, {"web"}}}}}
clone := func(s inventory.Source) (string, error) { return "http://forge.invalid/" + s.Repository + ".git", nil }
got := checkBuildsOf(p, scope, entries, clone)
want := []link.CheckBuild{
{Module: "app", Path: "modules/app", Merged: true, Tier: 0},
{Module: "docs", Tier: 0, Skip: "its build reads novox/shop through a context, which a check cannot hand the merge"},
{Module: "web", Repository: "http://forge.invalid/novox/web.git", Ref: "main", Tier: 1},
}
if !reflect.DeepEqual(got, want) {
t.Fatalf("the check is asked to build\n%+v\nwant\n%+v", got, want)
}
if got := checkBuildsOf(p, checkScope{}, entries, clone); got != nil {
t.Fatalf("a change that moves no module asks for builds: %+v", got)
}
}
// A walk's ask carries its leave to take a check's build, naming the pull requests of the merges it answers; a dry
// run never does, and neither does an ask that is not a walk's.
// A walk ended lets go of its merges' check builds — only once it ended, after it took what it reuses.
func TestAWalkEndedLetsGoOfItsMergesCheckBuilds(t *testing.T) {
var forgot []builder.PullRef
was := forgetChecks
forgetChecks = func(_ context.Context, _ *inventory.Inventory, pull builder.PullRef) (string, error) {
forgot = append(forgot, pull)
return "", nil
}
t.Cleanup(func() { forgetChecks = was })
p := inventory.Plan{ID: "plan-1", Delivery: &inventory.PlanDelivery{Merges: []inventory.PlanMerge{
{Repository: "novox/shop", Number: 7}}}}
p.WalkSaid = inventory.WalkStarted
forgetChecksOfWalk(t.Context(), nil, p)
p.WalkSaid = ""
forgetChecksOfWalk(t.Context(), nil, p)
if len(forgot) != 0 {
t.Fatalf("a walk not ended let go of %v, which it may still take", forgot)
}
p.WalkSaid = inventory.WalkEnded
forgetChecksOfWalk(t.Context(), nil, p)
if want := []builder.PullRef{{Owner: "novox", Repo: "shop", Number: 7}}; !reflect.DeepEqual(forgot, want) {
t.Fatalf("the walk ended let go of %v", forgot)
}
}
// A build in the scratch namespace is never registered: only the walk's copy in the module's own namespace is.
func TestABuildInTheScratchNamespaceIsNeverRegistered(t *testing.T) {
if got := inScratch([]link.MadeArtifact{{Name: "server", Reference: "store.internal:5000/app/server@sha256:aa"}}); got != "" {
t.Fatalf("a module's own build reads as a scratch build: %s", got)
}
scratch := "store.internal:5000/scratch/app/server@sha256:aa"
if got := inScratch([]link.MadeArtifact{{Name: "server", Reference: scratch}}); got != scratch {
t.Fatalf("a scratch build was not found: %q", got)
}
if got := inScratch([]link.MadeArtifact{{Name: "look", Reference: "http://store.internal:5000/v2/scratch/app/look/blobs/sha256:bb"}}); got == "" {
t.Fatal("a scratch archive was not found")
}
}
// The walk says per module what it did: "reused from …" or "built: <why>" — and a build seat from before says
// neither, which is said.
func TestAWalksBuildSaysWhetherItReusedOrBuilt(t *testing.T) {
if got := (link.BuildResult{Reused: "reused from the check of novox/shop#7"}).MadeWords(); got != "reused from the check of novox/shop#7" {
t.Fatalf("said %q", got)
}
if got := (link.BuildResult{BuiltWhy: "no check built it"}).MadeWords(); got != "built: no check built it" {
t.Fatalf("said %q", got)
}
if got := (link.BuildResult{}).MadeWords(); !strings.Contains(got, "predates ADR 0350") {
t.Fatalf("said %q", got)
}
}
// A walk's module the check's build stood in for is registered by the digest the check built — the copy in the
// module's namespace — and the walk keeps "reused from …" for it; its first-node gate judges that registered
// build as it judges one the walk built. A scratch reference is refused. Needs the store (MESH_TEST_POSTGRES), and
// is named in testdata/store-tests.
// A walk's leave names the builds its own pull requests' verdicts made at the head each merge merged — never an
// earlier head's — read once per walk, and none when they cannot be read: the builder takes nothing a verdict does
// not name (novox/hq ADR 0350 §2, issue 703).
func TestAWalksLeaveNamesTheBuildsItsPullRequestsVerdictsMade(t *testing.T) {
was := checkedBuildsOf
forgetWalkVerdicts()
t.Cleanup(func() { checkedBuildsOf = was; forgetWalkVerdicts() })
var asked [][]link.PullRef
pull := link.PullRef{Owner: "novox", Repo: "shop", Number: 7}
atHead := link.CheckedBuild{Pull: pull, Check: "build-2", Head: "bbbbbbbbbbbb", Built: link.CheckBuilt{Module: "app", Key: "r1:2"}}
earlier := link.CheckedBuild{Pull: pull, Check: "build-1", Head: "aaaaaaaaaaaa", Built: link.CheckBuilt{Module: "app", Key: "r1:1"}}
checkedBuildsOf = func(_ context.Context, pulls []link.PullRef) ([]link.CheckedBuild, error) {
asked = append(asked, pulls)
return []link.CheckedBuild{atHead, earlier}, nil
}
pulls := []link.PullRef{pull}
heads := map[link.PullRef]string{pull: "bbbbbbbbbbbb0123456789"}
for range 3 {
if got := reuseWithVerdicts(t.Context(), "plan-1", &link.ReuseAsk{Pulls: pulls}, heads); !reflect.DeepEqual(got.Builds,
[]link.CheckedBuild{atHead}) {
t.Fatalf("the leave names %+v, not only the merged head's build", got.Builds)
}
}
if !reflect.DeepEqual(asked, [][]link.PullRef{pulls}) {
t.Fatalf("the verdicts were read %d times for one walk: %v", len(asked), asked)
}
// A merge whose head was not said takes nothing.
if got := reuseWithVerdicts(t.Context(), "plan-1", &link.ReuseAsk{Pulls: pulls}, nil); got.Builds != nil {
t.Fatalf("a merge of no said head names %+v", got.Builds)
}
checkedBuildsOf = func(context.Context, []link.PullRef) ([]link.CheckedBuild, error) {
return nil, errors.New("no bus")
}
if got := reuseWithVerdicts(t.Context(), "plan-2", &link.ReuseAsk{Pulls: pulls}, heads); got.Builds != nil {
t.Fatalf("verdicts that could not be read named %+v", got.Builds)
}
}
// A check is handed only the builds earlier checks of its own pull request at its own head made: another pull
// request's build, or an earlier head's, is never kept (novox/hq issue 703). None when they cannot be read.
func TestACheckIsHandedOnlyItsOwnHeadsBuilds(t *testing.T) {
was := checkedBuildsOf
t.Cleanup(func() { checkedBuildsOf = was })
pull := link.PullRef{Owner: "novox", Repo: "shop", Number: 7}
mine := link.CheckedBuild{Pull: pull, Check: "build-1", Head: "bbbbbbbbbbbb", Built: link.CheckBuilt{Module: "app", Key: "r1:1"}}
earlier := link.CheckedBuild{Pull: pull, Check: "build-0", Head: "aaaaaaaaaaaa", Built: link.CheckBuilt{Module: "app", Key: "r1:1"}}
another := link.CheckedBuild{Pull: link.PullRef{Owner: "novox", Repo: "shop", Number: 8}, Check: "build-9",
Head: "bbbbbbbbbbbb", Built: link.CheckBuilt{Module: "app", Key: "r1:1"}}
var asked [][]link.PullRef
checkedBuildsOf = func(_ context.Context, pulls []link.PullRef) ([]link.CheckedBuild, error) {
asked = append(asked, pulls)
return []link.CheckedBuild{mine, earlier, another}, nil
}
p := link.PullUpdated{Owner: "novox", Repo: "shop", Number: 7, Commit: "bbbbbbbbbbbb0123456789"}
builds := []link.CheckBuild{{Module: "app"}}
if got := namedForTheCheck(t.Context(), p, builds); !reflect.DeepEqual(got, []link.CheckedBuild{mine}) ||
!reflect.DeepEqual(asked, [][]link.PullRef{{pull}}) {
t.Fatalf("the check is handed %+v, read for %v", got, asked)
}
checkedBuildsOf = func(context.Context, []link.PullRef) ([]link.CheckedBuild, error) {
return nil, errors.New("no bus")
}
if got := namedForTheCheck(t.Context(), p, builds); got != nil {
t.Fatalf("verdicts that could not be read named %+v", got)
}
}
func forgetWalkVerdicts() {
walkVerdicts.Lock()
defer walkVerdicts.Unlock()
walkVerdicts.read = map[string]verdictsRead{}
}
// A walk's merged heads are the forge's word for each merge; a merge announced without its head, or a pull request
// merged at two heads in one walk, has none, and its walk takes no check's build (novox/hq issue 703).
func TestAWalksMergedHeadsAreTheForgesWord(t *testing.T) {
merge := func(number int, head string) inventory.PlanMerge {
return inventory.PlanMerge{Repository: "novox/shop", Number: number, Head: head}
}
p := &inventory.Plan{Delivery: &inventory.PlanDelivery{Merges: []inventory.PlanMerge{
merge(7, "aaaaaaaaaaaa1"), merge(8, ""), merge(9, "bbbbbbbbbbbb"), merge(9, "cccccccccccc")}}}
heads := mergedHeadsOf(p)
shop := func(n int) link.PullRef { return link.PullRef{Owner: "novox", Repo: "shop", Number: n} }
if heads[shop(7)] != "aaaaaaaaaaaa1" || heads[shop(8)] != "" || heads[shop(9)] != "" {
t.Fatalf("merged heads %v", heads)
}
}
+203
View File
@@ -0,0 +1,203 @@
package main
import (
"context"
"errors"
"flag"
"fmt"
"os"
"os/exec"
"path/filepath"
"strings"
"time"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/builder"
"github.com/novox/mesh-controller/internal/link"
)
// `check-here` is a pull request's merge check run on the machine at hand **exactly as the build seat
// runs it** (novox/hq issue 286): the same code (builder.Check), the same ask the controller would make of
// the seat — the gate's modules and judge by the planner's own answer over the facts snapshot, the
// repositories beside it at the refs the snapshot says the seat clones them at — in the toolchain image
// the mesh holds, as the user the build seat runs as, against a throwaway store and bus of the versions
// the mesh runs.
//
// **The build seat is the reference.** A merge-check.sh that passed on an agent's machine failed on the
// seat for three reasons that were each the agent's environment, never the change: a newer Go whose gofmt
// lays a file out differently, a sibling checkout at the agent's feature branch where the seat had the
// commit the mesh runs, and a different user. Run here, a check sees what the seat will.
//
// check-here [--tree <checkout>] [--base main] [--registry <artifact store>] [--forge <url of the owner>]
// [--number <n>] [--user uid:gid] [--facts store|<file>] [--keep]
//
// The checkout's HEAD is what is checked, and it must be committed: the seat checks a commit, never a
// working tree.
func checkHereCommand(ctx context.Context, args []string) error {
set := flag.NewFlagSet("check-here", flag.ContinueOnError)
tree := set.String("tree", ".", "the change's checkout; its HEAD is checked")
base := set.String("base", "main", "the branch the change would merge into")
registry := set.String("registry", os.Getenv("MESH_REGISTRY"), "the artifact store holding the facts and the toolchains")
forge := set.String("forge", "", "where the repositories beside it are cloned from, as <forge>/<repository>.git; "+
"the checkout's origin without its own name when not given")
number := set.Int("number", 0, "the pull request's number, when there is one")
// The build seat's service runs as root, and its check containers run as the builder does.
user := set.String("user", "0:0", "the user the check's containers run as: the build seat's")
keep := set.Bool("keep", false, "keep the workspace afterwards")
factsFrom := set.String("facts", "store", "the facts snapshot: `store`, the one the artifact store holds — "+
"what the seat reads — or a file")
if _, err := parseAround(set, args); err != nil {
return err
}
if *registry == "" {
return errors.New("check-here reads the facts and the toolchains from the artifact store: --registry <host:port> or MESH_REGISTRY")
}
dir, err := filepath.Abs(*tree)
if err != nil {
return err
}
git := func(args ...string) (string, error) {
cmd := exec.CommandContext(ctx, "git", args...)
cmd.Dir = dir
out, err := cmd.Output()
return strings.TrimSpace(string(out)), err
}
if dirty, err := git("status", "--porcelain", "--untracked-files=no"); err != nil {
return fmt.Errorf("%s is not a checkout: %w", dir, err)
} else if dirty != "" {
return errors.New("the checkout has changes not committed: the build seat checks a commit, so commit first")
}
head, err := git("rev-parse", "HEAD")
if err != nil {
return err
}
origin, err := git("remote", "get-url", "origin")
if err != nil {
return fmt.Errorf("the checkout has no origin to say which repository it is: %w", err)
}
owner, repo, prefix := ownerRepoOf(origin)
if *forge == "" {
*forge = prefix
}
if _, err := git("fetch", "--quiet", "origin", *base); err != nil {
return fmt.Errorf("cannot fetch %s to say what the change touches: %w", *base, err)
}
// Without rename detection, so a file moved out of a build source is said under its old name as well:
// its going is a change to the build that held it (novox/hq ADR 0267).
changedText, err := git("diff", "--name-only", "--no-renames", "origin/"+*base+"...HEAD")
if err != nil {
return err
}
var paths, removed []string
for _, p := range strings.Split(changedText, "\n") {
if p = strings.TrimSpace(p); p == "" {
continue
}
paths = append(paths, p)
if _, err := os.Stat(filepath.Join(dir, p)); os.IsNotExist(err) {
removed = append(removed, p)
}
}
_ = os.Setenv("MESH_REGISTRY", *registry)
f, err := readFacts(ctx, *factsFrom)
if err != nil {
return fmt.Errorf("the facts snapshot cannot be read: %w", err)
}
entries, read, edges, err := graphOfFacts(f)
if err != nil {
return err
}
p := link.PullUpdated{Owner: owner, Repo: repo, Number: *number, Base: *base, Commit: head, Paths: paths,
Removed: removed, ModuleDirs: moduleDirsIn(dir, paths), ModuleDirsSaid: true}
scope := pullScope(p, entries, read, edges)
_, scriptErr := os.Stat(filepath.Join(dir, builder.CheckScript))
if !scope.gated() && !scope.Mesh {
fmt.Printf("%s: %s, and the repository is not the mesh's: the build seat runs nothing for it\n",
owner+"/"+repo, noModuleTouched)
return nil
}
if !scope.gated() && scriptErr != nil {
fmt.Printf("%s: %s; %s\n", owner+"/"+repo, noModuleTouched, noMergeCheck)
return nil
}
toolchains := map[string]string{}
for language, reference := range f.Versions.Toolchains {
toolchains[language] = catalogue.Rerouted(reference, *registry)
}
if len(toolchains) == 0 {
return errors.New("the facts snapshot names no toolchain: it was taken by a controller from before " +
"issue 286, and the seat's toolchain cannot be known here")
}
beside := map[string]builder.Beside{}
for d, ref := range f.Beside {
from := d
if d == "mesh-controller-main" {
from = "mesh-controller"
}
beside[d] = builder.Beside{Repository: strings.TrimSuffix(*forge, "/") + "/" + from + ".git", Ref: ref}
}
id := fmt.Sprintf("check-here-%d", time.Now().UnixNano())
spec := builder.CheckSpec{ID: id, Repository: dir, Ref: head, Owner: owner, Repo: repo, Number: *number,
Paths: paths, Beside: beside, Modules: scope.Modules, New: scope.New, Manifests: scope.Manifests,
Base: *base, Judge: scope.Judge, Toolchain: toolchains["go"], Toolchains: toolchains, User: *user}
workspace, err := os.MkdirTemp("", "mesh-check-here-")
if err != nil {
return err
}
if !*keep {
defer removeWorkspace(spec.Toolchain, workspace, *user)
}
fmt.Fprintf(os.Stderr, "checking %s/%s at %.8s as the build seat would, against the facts of %s, in %s\n",
owner, repo, head, f.Taken.Format(time.RFC3339), workspace)
v, err := builder.Check(ctx, builder.Command, spec, workspace, *registry, builder.GitCredential{},
func(step, message string) {
if step != "output" {
fmt.Fprintf(os.Stderr, " [%s] %s\n", step, message)
}
})
if err != nil {
return fmt.Errorf("the check could not run — on the seat an error, never a pass: %w", err)
}
fmt.Println(v.Report)
fmt.Println()
fmt.Printf("mesh/merge-gate: %s — %s\n", strings.ToUpper(v.Gate.Verdict), v.Gate.Summary)
if v.Repo != nil {
fmt.Printf("mesh/repo-check: %s — %s\n", strings.ToUpper(v.Repo.Verdict), v.Repo.Summary)
}
for _, l := range []*builder.Layer{v.Gate, v.Repo} {
if l != nil && l.Verdict != "pass" && l.Verdict != "warning" {
return errors.New("the build seat would not pass this change")
}
}
return nil
}
// ownerRepoOf reads owner, repository and the owner's URL from a remote: ssh://git@host:222/novox/mesh-host.git
// → novox, mesh-host, ssh://git@host:222/novox.
func ownerRepoOf(remote string) (string, string, string) {
trimmed := strings.TrimSuffix(strings.TrimSuffix(remote, "/"), ".git")
cut := strings.LastIndexAny(trimmed, "/:")
if cut < 0 {
return "", trimmed, ""
}
repo, prefix := trimmed[cut+1:], trimmed[:cut]
owner := prefix
if at := strings.LastIndexAny(prefix, "/:"); at >= 0 {
owner = prefix[at+1:]
}
return owner, repo, prefix
}
// removeWorkspace removes what the check left, written as the seat's user: by a container of that user
// when it is not this one.
func removeWorkspace(image, workspace, user string) {
if image != "" && user != fmt.Sprintf("%d:%d", os.Getuid(), os.Getgid()) {
// Everything in it — the check's HOME is the workspace, so the toolchain's own files are there too.
_ = exec.Command("docker", "run", "--rm", "--user", user, "--volume", workspace+":/workspace", image,
"sh", "-c", "rm -rf /workspace/* /workspace/.[!.]*").Run()
}
_ = os.RemoveAll(workspace)
}
+44
View File
@@ -0,0 +1,44 @@
package main
import (
"reflect"
"testing"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// 1212/AC1 and AC2: the check request names the pulls bucket of each module claiming the delivery seat, in the order
// of the claimants' names.
func TestACheckNamesThePullsBucketsOfTheSeatsClaimants(t *testing.T) {
for name, c := range map[string]struct {
entries []inventory.Entry
want []string
}{
"only the gate": {[]inventory.Entry{claimingTheSeat("mesh-delivery-gate"), namedOnly("mesh-delivery")}, []string{"mesh-delivery-gate_pulls"}},
"only today's": {[]inventory.Entry{claimingTheSeat("mesh-delivery"), namedOnly("gitea")}, []string{"mesh-delivery_pulls"}},
"both, name order": {[]inventory.Entry{claimingTheSeat("mesh-delivery-gate"), claimingTheSeat("mesh-delivery")}, []string{"mesh-delivery_pulls", "mesh-delivery-gate_pulls"}},
"none": {[]inventory.Entry{namedOnly("gitea")}, nil},
} {
if got := pullsBucketsOf(c.entries); !reflect.DeepEqual(got, c.want) {
t.Errorf("%s: %v, want %v", name, got, c.want)
}
}
}
// 1212/AC1 and AC2: the check a request carries names those buckets (the wiring, not only the helper).
func TestTheCheckRequestCarriesThePullsBuckets(t *testing.T) {
for name, c := range map[string]struct {
entries []inventory.Entry
want []string
}{
"only the gate": {[]inventory.Entry{claimingTheSeat("mesh-delivery-gate")}, []string{"mesh-delivery-gate_pulls"}},
"only today's": {[]inventory.Entry{claimingTheSeat("mesh-delivery")}, []string{"mesh-delivery_pulls"}},
"both, name order": {[]inventory.Entry{claimingTheSeat("mesh-delivery-gate"), claimingTheSeat("mesh-delivery")}, []string{"mesh-delivery_pulls", "mesh-delivery-gate_pulls"}},
} {
got := checkOf(link.PullUpdated{Owner: "novox", Repo: "a", Number: 7}, checkScope{}, c.entries, nil, nil, nil, nil)
if !reflect.DeepEqual(got.Pulls, c.want) || got.Owner != "novox" || got.Number != 7 {
t.Errorf("%s: %+v, want pulls %v", name, got, c.want)
}
}
}
+165
View File
@@ -0,0 +1,165 @@
package main
import (
"context"
"fmt"
"regexp"
"sort"
"strings"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// **A delivery group's member is checked beside the other members' heads** (novox/hq ADR 0382, issue 867).
//
// A mesh-host change and a mesh-controller change that need each other could not both pass their own merge checks:
// each was read beside the other's repository at the commit the mesh runs, which holds neither change. So the
// controller asks mesh-delivery's `groups` once for every pull request's check it asks (§1), and names, in the ask,
// the group's other open pull requests into a base of the same name, in a repository the build seat reads beside the
// check — mesh-lab never — each with its whole head. The build seat reads those repositories at the sibling's merge (§2).
// A pull request for which mesh-delivery answered and named no sibling is asked exactly as before (§7).
// wholeCommit is a commit as the forge says a head: forty hexadecimal characters.
var wholeCommit = regexp.MustCompile(`^[0-9a-f]{40}$`)
// besideRepositories are the repositories (lower case, owner/repo) a check reads beside its own, by the directory each
// is found under — the core's, as besideOf clones them, and mesh-sdk beside the controller; never mesh-lab, which is
// always read at its trunk.
func besideRepositories(entries []inventory.Entry) map[string]string {
read := map[string]string{}
for _, e := range entries {
dir, core := coreModules[e.Manifest.Module]
if !core || e.Provided || e.Source.Repository == "" {
continue
}
read[nameOf(e.Source.Repository)] = dir
if dir == "mesh-controller" && e.Source.Seat != "" {
read[nameOf(siblingOf(e.Source.Repository, "mesh-sdk"))] = "mesh-sdk"
}
}
delete(read, "")
return read
}
// nameOf is a recorded repository as owner/repo in lower case, empty when it names no owner.
func nameOf(repository string) string {
owner, repo := sourceOwner(repository), sourceName(repository)
if owner == "" || repo == "" {
return ""
}
return strings.ToLower(owner + "/" + repo)
}
// siblingsFor is the pull request's siblings, as mesh-delivery's groups say them (ADR 0382 §1): the other members of
// the group its head branch names that are open, unmerged and into a base of its own base's name, in a repository the
// check reads beside it (beside: the directories the ask already names). The second answer is why they could not be
// learned, empty when they could: the group's members carry no head or base — a mesh-delivery from before the rule.
func siblingsFor(p link.PullUpdated, entries []inventory.Entry, beside map[string]link.CheckedOut,
lines []groupsLine) ([]link.CheckSibling, string) {
own := strings.ToLower(p.Owner + "/" + p.Repo)
var group *groupsLine
if p.Head == "" {
// A recheck (mesh-delivery's `recheck` through `delivery check`) carries no head branch name: the group is the one
// whose members hold this very pull request, by repository and number.
learnable := false
for i := range lines {
for _, m := range lines[i].Details {
learnable = true
if p.Number != 0 && m.Number == p.Number && strings.ToLower(m.Repository) == own {
group = &lines[i]
}
}
}
if group == nil {
for _, l := range lines {
if len(l.Details) == 0 && len(l.Members) > 0 && !learnable {
return nil, "mesh-delivery named the groups' members without their pull requests (a mesh-delivery from before novox/hq ADR 0382): " +
"the group of a pull request announced without its head branch cannot be found"
}
}
return nil, "" // a lone pull request
}
} else {
for i := range lines {
if lines[i].ID == p.Head {
group = &lines[i]
}
}
if group == nil {
return nil, ""
}
}
if len(group.Details) == 0 && len(group.Members) > 0 {
return nil, "mesh-delivery named the group's members without their heads and bases (a mesh-delivery from before novox/hq ADR 0382)"
}
read := besideRepositories(entries)
var siblings []link.CheckSibling
for _, m := range group.Details {
repository := strings.ToLower(m.Repository)
if repository == own {
continue
}
if !m.Open {
continue
}
if m.Head == "" || m.Base == "" {
return nil, "mesh-delivery named " + m.ID + " without its head or base (a mesh-delivery from before novox/hq ADR 0382)"
}
dir, readBeside := read[repository]
if m.Base != p.Base || !readBeside || dir == "mesh-lab" {
continue
}
if _, there := beside[dir]; !there {
continue
}
head := strings.ToLower(m.Head)
if !wholeCommit.MatchString(head) {
return nil, fmt.Sprintf("mesh-delivery named %s with a head that is not a whole commit (%q)", m.ID, m.Head)
}
siblings = append(siblings, link.CheckSibling{Repository: m.Repository, Number: m.Number, Head: head, Base: m.Base, Dir: dir})
}
sort.Slice(siblings, func(i, j int) bool { return siblings[i].Dir < siblings[j].Dir })
return siblings, ""
}
// withSiblings asks mesh-delivery's `groups` once and says in the request the pull request's siblings, or why they
// could not be learned (ADR 0382 §1): a check never waits on mesh-delivery past groupAskWithin and never fails for it,
// so a fix to mesh-delivery is still checked without it.
func withSiblings(ctx context.Context, request *link.BuildRequest, p link.PullUpdated, entries []inventory.Entry) {
lines, err := groupsAsked(ctx, deliverySeatHeld(entries))
siblingsFrom(request, p, entries, lines, err)
}
// siblingsFrom is withSiblings from mesh-delivery's answer already had.
func siblingsFrom(request *link.BuildRequest, p link.PullUpdated, entries []inventory.Entry, lines []groupsLine, err error) {
if request.Check == nil {
return
}
if err != nil {
request.Check.SiblingsUnknown = fmt.Sprintf("mesh-delivery did not say the group's pull requests (%v): "+
"the check reads no siblings", err)
return
}
request.Check.Siblings, request.Check.SiblingsUnknown = siblingsFor(p, entries, request.Check.Beside, lines)
}
// composedMember says the pull request is a member of a delivery group whose gate is the composed check's (novox/hq
// ADR 0403 clauses 1 and 7): the group has two members or more and mesh-delivery says so on its line. The group is
// found as siblingsFor finds it, by its head branch name, or, for a recheck that carries none, by the pull request's
// repository and number — not by siblings, which name only what the check reads beside it.
func composedMember(p link.PullUpdated, lines []groupsLine) bool {
own := strings.ToLower(p.Owner + "/" + p.Repo)
for _, l := range lines {
found := p.Head != "" && l.ID == p.Head
for _, m := range l.Details {
if p.Head == "" && p.Number != 0 && m.Number == p.Number && strings.ToLower(m.Repository) == own {
found = true
}
}
if found {
return l.ComposedGate && max(len(l.Details), len(l.Members)) >= 2
}
}
return false
}
+248
View File
@@ -0,0 +1,248 @@
package main
import (
"context"
"encoding/json"
"errors"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// **A delivery group's member is checked beside the other members' heads** (novox/hq ADR 0382 §1, §7, issue 867): the
// controller asks mesh-delivery's `groups` once for every check it asks, and names in the ask the group's other open
// pull requests into a base of the same name, in a repository the build seat reads beside it — mesh-lab never.
const (
headHost = "1111111111111111111111111111111111111111"
headController = "2222222222222222222222222222222222222222"
headSDK = "3333333333333333333333333333333333333333"
headLab = "4444444444444444444444444444444444444444"
headHq = "5555555555555555555555555555555555555555"
)
func theCoreOnTheGitSeat() []inventory.Entry {
onSeat := func(module, repository string) inventory.Entry {
e := fromRepo(module, repository, "")
e.Source.Seat = "git"
return e
}
nats := onSeat("nats", "novox/mesh-catalog")
nats.Source.Path = "modules/nats"
held := catalogue.Manifest{Module: "mesh-delivery", Claims: []catalogue.Claim{{Name: catalogue.DeliverySeat, Scope: catalogue.ScopeMesh}}}
return []inventory.Entry{onSeat("mesh-controller", "novox/mesh-controller"), onSeat("mesh-host", "novox/mesh-host"),
onSeat("node-tools", "novox/mesh-tools"), nats, {Manifest: held, On: []string{"anchor"}}}
}
func besideTheCore() map[string]link.CheckedOut {
return map[string]link.CheckedOut{"mesh-controller": {Repository: "c"}, "mesh-host": {Repository: "h"},
"mesh-tools": {Repository: "t"}, "mesh-catalog": {Repository: "k"}, "mesh-sdk": {Repository: "s"}, "mesh-lab": {Repository: "l"}}
}
func member(repository string, number int, head, base string, open bool) memberLine {
return memberLine{ID: repository + "@" + head[:12], Repository: repository, State: "checked", Number: number, Head: head,
Base: base, Open: open}
}
func group(id string, members ...memberLine) groupsLine {
line := groupsLine{ID: id, Details: members}
for _, m := range members {
line.Members = append(line.Members, m.ID+" ("+m.State+")")
}
return line
}
func TestASiblingIsTheGroupsOtherOpenPullRequestIntoABaseOfTheSameName(t *testing.T) {
entries := theCoreOnTheGitSeat()
pull := link.PullUpdated{Owner: "novox", Repo: "mesh-host", Number: 114, Base: "main", Head: "feat/x", Commit: headHost}
for _, c := range []struct {
what string
lines []groupsLine
want []string
unknown string
}{
{"the other member, open, same base", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "main", true))},
[]string{"mesh-controller=novox/mesh-controller#346@" + headController + " onto main"}, ""},
{"a member into a base of another name is not one", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "release", true))},
nil, ""},
{"a member not open (merged or closed) is not one", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "main", false))},
nil, ""},
{"mesh-lab is never one", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-lab", 5, headLab, "main", true))},
nil, ""},
{"a repository the check reads nothing of is not one", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/hq", 9, headHq, "main", true))},
nil, ""},
{"mesh-sdk, read beside the controller, is one", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-sdk", 7, headSDK, "main", true))},
[]string{"mesh-sdk=novox/mesh-sdk#7@" + headSDK + " onto main"}, ""},
{"another group is none of its business", []groupsLine{group("feat/y",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "main", true))},
nil, ""},
{"a group of its own alone has none", []groupsLine{group("feat/x", member("novox/mesh-host", 114, headHost, "main", true))},
nil, ""},
{"an older mesh-delivery names no heads", []groupsLine{{ID: "feat/x",
Members: []string{"novox/mesh-host@111111111111 (checked)", "novox/mesh-controller@222222222222 (checked)"}}},
nil, "without their heads and bases"},
{"a member without its base", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "", true))},
nil, "without its head or base"},
{"a head that is no commit", []groupsLine{group("feat/x",
member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController[:12]+"zzzzzzzzzzzzzzzzzzzzzzzzzzzz", "main", true))},
nil, "not a whole commit"},
} {
got, unknown := siblingsFor(pull, entries, besideTheCore(), c.lines)
var said []string
for _, s := range got {
said = append(said, s.Dir+"="+s.Repository+"#"+itoa(s.Number)+"@"+s.Head+" onto "+s.Base)
}
if strings.Join(said, ";") != strings.Join(c.want, ";") {
t.Errorf("%s: siblings %v, wanted %v", c.what, said, c.want)
}
if (c.unknown == "") != (unknown == "") || !strings.Contains(unknown, c.unknown) {
t.Errorf("%s: could not learn %q, wanted %q", c.what, unknown, c.unknown)
}
}
// A pull request into a release branch has siblings into the release branch of the same name (ADR 0358 §1).
release := pull
release.Base = "feat/57"
lines := []groupsLine{group("feat/x", member("novox/mesh-host", 114, headHost, "feat/57", true),
member("novox/mesh-controller", 346, headController, "feat/57", true), member("novox/mesh-tools", 8, headLab, "main", true))}
if got, _ := siblingsFor(release, entries, besideTheCore(), lines); len(got) != 1 || got[0].Dir != "mesh-controller" {
t.Errorf("into a release branch: %+v", got)
}
// A recheck announces the pull request without its head branch: its group is the one holding it (repository, number).
pull.Head = ""
got, unknown := siblingsFor(pull, entries, besideTheCore(), lines)
if len(got) != 1 || got[0].Dir != "mesh-tools" || unknown != "" {
t.Errorf("a recheck of a group member: %+v %q", got, unknown)
}
// A pull request in no group, rechecked, is alone: no sibling, and nothing said.
pull.Number = 999
if got, unknown := siblingsFor(pull, entries, besideTheCore(), lines); got != nil || unknown != "" {
t.Errorf("a lone pull request: %+v %q", got, unknown)
}
if got, unknown := siblingsFor(pull, entries, besideTheCore(), nil); got != nil || unknown != "" {
t.Errorf("no groups at all: %+v %q", got, unknown)
}
// A mesh-delivery that names members without their pull requests cannot find the group of a recheck: it says so.
older := []groupsLine{{ID: "feat/x", Members: []string{"novox/mesh-host@111111111111 (checked)"}}}
pull.Number = 114
if got, unknown := siblingsFor(pull, entries, besideTheCore(), older); got != nil || unknown == "" {
t.Errorf("an older mesh-delivery: %+v %q", got, unknown)
}
}
func itoa(n int) string { b, _ := json.Marshal(n); return string(b) }
// Every check asks mesh-delivery's `groups` once; the ask names the sibling's repository, head and base.
func TestEveryCheckAsksMeshDeliveryOnceAndNamesItsSiblings(t *testing.T) {
entries := theCoreOnTheGitSeat()
asked := 0
was := groupsAsked
t.Cleanup(func() { groupsAsked = was })
groupsAsked = func(_ context.Context, held bool) ([]groupsLine, error) {
asked++
if !held {
t.Error("the seat is held, yet it was asked as not held")
}
return []groupsLine{group("feat/x", member("novox/mesh-host", 114, headHost, "main", true),
member("novox/mesh-controller", 346, headController, "main", true))}, nil
}
request := link.BuildRequest{Check: &link.CheckRequest{Owner: "novox", Repo: "mesh-host", Base: "main", Beside: besideTheCore()}}
withSiblings(t.Context(), &request, link.PullUpdated{Owner: "novox", Repo: "mesh-host", Base: "main", Head: "feat/x"}, entries)
if asked != 1 {
t.Fatalf("mesh-delivery was asked %d times for one check", asked)
}
want := link.CheckSibling{Repository: "novox/mesh-controller", Number: 346, Head: headController, Base: "main", Dir: "mesh-controller"}
if got := request.Check.Siblings; len(got) != 1 || got[0] != want || request.Check.SiblingsUnknown != "" {
t.Fatalf("the ask names siblings %+v, unknown %q", got, request.Check.SiblingsUnknown)
}
body, _ := json.Marshal(request.Check)
if !strings.Contains(string(body), `"siblings":[{"repository":"novox/mesh-controller","number":346,"head":"`+headController+
`","base":"main","dir":"mesh-controller"}]`) {
t.Fatalf("on the wire: %s", body)
}
}
// When mesh-delivery does not answer in time (or is not held, or answered without the fields), the check reads no
// sibling and its request says it could not learn them, so a fix to mesh-delivery is still checked without it.
func TestWhenMeshDeliveryDoesNotAnswerTheCheckReadsNoSiblingAndSaysSo(t *testing.T) {
was := groupsAsked
t.Cleanup(func() { groupsAsked = was })
for _, c := range []struct {
what string
err error
held bool
}{{"it did not answer", context.DeadlineExceeded, true}, {"nothing serves it", errors.New("nothing serves mesh-delivery.groups"), true},
{"no holder on record", errors.New("no holder of mesh-delivery is on record to say it"), false}} {
entries := theCoreOnTheGitSeat()
if !c.held {
entries = entries[:4]
}
groupsAsked = func(_ context.Context, held bool) ([]groupsLine, error) {
if held != c.held {
t.Errorf("%s: held %v", c.what, held)
}
return nil, c.err
}
request := link.BuildRequest{Check: &link.CheckRequest{Owner: "novox", Repo: "mesh-host", Base: "main", Beside: besideTheCore()}}
withSiblings(t.Context(), &request, link.PullUpdated{Owner: "novox", Repo: "mesh-host", Base: "main", Head: "feat/x"}, entries)
if len(request.Check.Siblings) != 0 || !strings.Contains(request.Check.SiblingsUnknown, c.err.Error()) ||
!strings.Contains(request.Check.SiblingsUnknown, "reads no siblings") {
t.Errorf("%s: siblings %+v, unknown %q", c.what, request.Check.Siblings, request.Check.SiblingsUnknown)
}
}
}
// **A pull request for which mesh-delivery answered and named no sibling is asked exactly as before** (ADR 0382 §7):
// byte for byte the request it was without the rule.
func TestAPullRequestWithNoSiblingIsAskedByteForByteAsBefore(t *testing.T) {
entries := theCoreOnTheGitSeat()
was := groupsAsked
t.Cleanup(func() { groupsAsked = was })
for _, answer := range [][]groupsLine{nil, {group("feat/other", member("novox/mesh-controller", 346, headController, "main", true))},
{group("feat/x", member("novox/mesh-host", 114, headHost, "main", true), member("novox/hq", 3, headHq, "main", true))}} {
groupsAsked = func(context.Context, bool) ([]groupsLine, error) { return answer, nil }
build := func() link.BuildRequest {
return link.BuildRequest{ID: "build-1", Repository: "r", Ref: headHost, Check: &link.CheckRequest{Owner: "novox",
Repo: "mesh-host", Number: 114, Base: "main", Beside: besideTheCore(), Modules: []string{"mesh-host"}, Judge: "validator"}}
}
before, _ := json.Marshal(build())
after := build()
withSiblings(t.Context(), &after, link.PullUpdated{Owner: "novox", Repo: "mesh-host", Base: "main", Head: "feat/x"}, entries)
got, _ := json.Marshal(after)
if string(before) != string(got) {
t.Fatalf("the request changed:\nbefore %s\nafter %s", before, got)
}
}
}
// What the build seat's verdict says of its siblings reaches the forge's holder in the controller's `checked`.
func TestTheVerdictCarriesTheSiblingsAndWhereTheJudgeCameFrom(t *testing.T) {
read := []link.CheckSibling{{Repository: "novox/mesh-controller", Number: 346, Head: headController, Base: "main", BaseCommit: headHost}}
c := checkedOf(link.BuildResult{ID: "b", Ref: headHost, Checked: &link.CheckRequest{Owner: "novox", Repo: "mesh-host", Number: 114},
Check: &link.CheckOutcome{Verdict: "pass", Summary: "ok", Siblings: read, JudgeFrom: "sibling novox/mesh-controller#346 merge abc",
Validator: "the change's own validator"}})
if len(c.Siblings) != 1 || c.Siblings[0] != read[0] || c.JudgeFrom == "" || c.Validator == "" {
t.Fatalf("the verdict says %+v", c)
}
body, _ := json.Marshal(c)
for _, want := range []string{`"siblings":[{"repository":"novox/mesh-controller","number":346,"head":"` + headController +
`","base":"main","base_commit":"` + headHost + `"}]`, `"judge_from":"sibling novox/mesh-controller#346 merge abc"`} {
if !strings.Contains(string(body), want) {
t.Fatalf("the verdict lacks %s:\n%s", want, body)
}
}
unknown := checkedOf(link.BuildResult{ID: "b", Check: &link.CheckOutcome{Verdict: "pass", SiblingsUnknown: "mesh-delivery did not answer"}})
if body, _ := json.Marshal(unknown); !strings.Contains(string(body), `"siblings_unknown":"mesh-delivery did not answer"`) ||
strings.Contains(string(body), `"siblings":`) {
t.Fatalf("the verdict that could not learn them says %s", body)
}
}
+517 -22
View File
@@ -2,6 +2,8 @@ package main
import (
"context"
"encoding/json"
"fmt"
"io"
"os"
"path"
@@ -10,18 +12,46 @@ import (
"slices"
"sort"
"strings"
"time"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// What a pull request reaches of the mesh's graph (novox/hq to-be 45 §9, issue 280). The controller no longer checks a
// pull request or says a verdict on it (R4g, ADR 0503 §3, ADR 0501 §5): the checker judges the heads, and mesh-delivery
// asks the operator's Approve on its verdict. What stays here is the mapping of a pull request's changed paths onto the
// mesh's module graph by the planner's own answer (reachOfMerge, touchedBy), which the delivery verbs (`delivery order`,
// the change plan) and the head's declared config (tellPullDeclares) read: **a changed file touches exactly the modules
// whose build reads it** — a module's own directory (the whole repository for one built from its root), or a repository
// its recipe packages — and a directory the change adds a module.json in is a new module.
// A pull request's merge check (novox/hq to-be 45 §9, ADR 0237 as amended 2026-10-06): the forge
// announces every pull request's new head, the controller decides what is checked, asks the build seat to
// check it, and says the verdict as `checked`, which the forge's holder sets as the pull request's
// statuses. **Before merge, never after**: every check the mesh had ran after a merge, on a machine.
//
// **The mesh's module graph decides, not the repository.** The controller holds the graph — every module,
// the repository and directory it is built from — and maps the pull request's changed paths onto it by
// the planner's own answer (reachOfMerge, touchedBy — the one place a changed file is mapped onto modules,
// for the merge handler, the release planner, the merge gate and this check; issue 280): **a changed file
// touches exactly the modules whose build reads it** — a module's own directory (the whole repository for
// one built from its root), or a repository its recipe packages. A file no build reads — a script at the
// root, a README — touches no module. A directory the change adds a module.json in, which the graph does
// not hold yet (said by the head, issue 278), is a new module and is checked too.
//
// - touches a module: the build seat runs **the gate** — `mesh/merge-gate`, the touched manifests, every
// machine composed with the change, the replays — and the repository's own merge-check.sh beside it;
// - touches none, in a repository that is the mesh's (it sources a module on some branch, or shares the
// core's owner): the gate is a pass that says so — a fact, not a missing check — and the repository's
// own merge-check.sh runs as `mesh/repo-check`, a warning when it has none;
// - touches none, anywhere else: the gate is a pass that says so, and nothing more is said.
//
// What is checked is decided here and run there (internal/builder/check.go).
// checkTimeout is how long one check may run on the build seat. Said here so the ask's watchdog (S6)
// and the builder agree on what late means.
const checkTimeout = 45 * time.Minute
// noModuleTouched is the gate's word for a change that touches nothing of the graph.
const noModuleTouched = "the change touches no module of the mesh's graph"
// noMergeCheck is the repository layer's word for a repository of the mesh with no merge-check.sh.
const noMergeCheck = "the repository declares no merge-check.sh: none of its own tests run before it merges"
// coreModules are the modules whose repositories are the mesh's core, by the directory a check finds
// each beside it — and whose owner is the mesh's own.
@@ -45,6 +75,8 @@ type checkScope struct {
// Mesh says the repository is the mesh's: modules are built from it on some branch, its owner is the
// core's, or the change adds a module to it.
Mesh bool
// Judge is who judges the gate (link.JudgeSelf, or the running controller).
Judge string
// From is a module built from the repository, for how the mesh clones it; nil when none is.
From *inventory.Entry
// Reach is the planner's whole answer, which the change plan is made from.
@@ -66,6 +98,10 @@ func pullScope(p link.PullUpdated, entries []inventory.Entry, read map[string][]
Width: len(r.Plan.Modules), Tiers: len(r.Plan.Tiers), Reach: r}
for _, e := range append(append([]inventory.Entry{}, r.Touched...), r.Deleted...) {
s.Manifests = append(s.Manifests, path.Join(strings.Trim(e.Source.Path, "/"), moduleManifestFile))
switch e.Manifest.Module {
case "mesh-controller":
s.Judge = link.JudgeSelf
}
}
for _, d := range r.Added {
s.Manifests = append(s.Manifests, path.Join(d, moduleManifestFile))
@@ -101,37 +137,315 @@ func sourceOwner(repository string) string {
return strings.ToLower(parts[len(parts)-2])
}
// PullUpdated hears a pull request's new head. It asks no check and says no verdict (pullUpdated).
// PullUpdated decides a pull request's merge check, and asks for it when there is something to run.
func (f following) PullUpdated(ctx context.Context, p link.PullUpdated) error {
return f.pullUpdated(ctx, p, os.Stdout)
_, err := f.pullUpdated(ctx, p, os.Stdout)
return err
}
// pullUpdated is PullUpdated saying what it did to w. It asks nothing of anybody and says no verdict (R4g; ADR 0503 §3, §5,
// ADR 0527): the checker judges every repository whose trunk requires `mesh/check`, hq included. All it still does is tell
// mesh-config what the head declares (ADR 0502), whatever judges it.
func (f following) pullUpdated(ctx context.Context, p link.PullUpdated, w io.Writer) error {
// pullUpdated is PullUpdated saying what it did to w. It returns the id of the check it asked of the build seat, or
// "" when it said a verdict itself and asked nothing. A command that answers JSON on standard output (`delivery
// check`) gives it standard error, so its output stays the document alone (novox/hq issue 1311).
func (f following) pullUpdated(ctx context.Context, p link.PullUpdated, w io.Writer) (string, error) {
inv := f.open.inventory
entries, err := inv.Catalogued(ctx)
if err != nil {
return err
return "", err
}
read, err := readForPlanning(ctx, inv)
if err != nil {
return err
return "", err
}
edges, err := inv.Dependencies(ctx)
if err != nil {
return err
return "", err
}
scope := pullScope(p, entries, read, edges)
this := pullAtOf(p)
if this.Truncated {
this.Paths, this.Truncated = truncatedPaths(entries, p.Owner, p.Repo, scope.Modules), false
// **A cycle of edges fails the check, naming every module and edge** (novox/hq ADR 0419 clause 6), a lone pull
// request's as a group's composed check is: no order exists for what it changes.
alone := orderMember{ID: p.Owner + "/" + p.Repo + "@" + twelve(p.Commit), Repository: p.Owner + "/" + p.Repo, Base: p.Base,
Head: p.Commit, Number: p.Number, Paths: p.Paths, PathsTruncated: p.PathsTruncated, Removed: p.Removed, BaseSHA: p.BaseSHA}
if failing, cyclic, err := cycleVerdict(ctx, "", []orderMember{alone},
map[string]orderReach{alone.ID: {Moved: scope.Modules}}, edges, entries); err != nil {
return "", err
} else if cyclic {
failing.Commit, failing.Trunk = p.Commit, twelve(p.BaseSHA)
fmt.Fprintf(w, "%s/%s#%d (%.8s): %s\n", p.Owner, p.Repo, p.Number, p.Commit, failing.Summary)
sayChecked(ctx, w, failing)
return "", nil
}
if scope.gated() || holdsModules(entries, p.Owner, p.Repo) {
tellPullDeclares(ctx, w, []pullAt{this})
// **A member of a group whose gate is the composed check's runs its repository's check alone** (novox/hq ADR 0403
// clause 1): mesh-delivery is asked once, and says so on the group's line. Told nothing, the pull request is
// checked alone, gate included (clause 6).
lines, groupsErr := groupsAsked(ctx, deliverySeatHeld(entries))
composed := groupsErr == nil && composedMember(p, lines)
// The change plan of the commit at hand (ADR 0238): what a merge of it would build and send, posted
// with the verdict whatever the verdict is.
plan := changePlanOf(p.Owner+"/"+p.Repo, p.Base, p.Commit, scope.Reach, entries, func(module string) (inventory.Upgrade, bool) {
u, err := inv.UpgradeOf(ctx, module)
return u, err == nil
})
fmt.Fprint(w, planText(plan))
direct := directVerdict(p, plan)
if composed {
scope.Modules, scope.Dependents, scope.New, scope.Manifests = nil, nil, nil, nil
composedGate(&direct)
}
return nil
switch {
case !scope.gated() && !scope.Mesh:
fmt.Fprintf(w, "%s/%s#%d (%.8s): %s, and the repository is not the mesh's: said, nothing run\n",
p.Owner, p.Repo, p.Number, p.Commit, noModuleTouched)
sayChecked(ctx, w, direct)
return "", nil
case !scope.gated() && p.MergeCheckSaid && !p.MergeCheck:
direct.RepoCheck = &link.CheckLayer{Verdict: "warning", Summary: noMergeCheck}
fmt.Fprintf(w, "%s/%s#%d (%.8s): %s; %s\n", p.Owner, p.Repo, p.Number, p.Commit, noModuleTouched, noMergeCheck)
sayChecked(ctx, w, direct)
return "", nil
}
request, err := checkRequestFor(ctx, f.open, p, scope, entries)
if err != nil {
return "", err
}
request.Check.Plan = &plan
request.Check.Composed = composed
// **The group's other open pull requests are read beside it** (novox/hq ADR 0382 §1).
siblingsFrom(&request, p, entries, lines, groupsErr)
seat := buildSeatHeld(ctx)
ask, err := askOverOn(seat)
if err != nil {
return "", err
}
defer ask.Close()
if err := ask.Ask(ctx, request); err != nil {
return "", err
}
// **A newer request replaces the one still waiting, and a newer key stops the one running** (novox/hq ADR 0406 §3, §4,
// issue 974): the check queue keeps one request a pull request, and a running check watches its key. Nothing is asked
// of the forge here (issue 726).
what := "its own merge-check.sh alone: " + noModuleTouched
if composed {
what = "its own merge-check.sh alone: the gate is its group's composed check's"
}
if scope.gated() {
what = fmt.Sprintf("the gate over %s (a merge would build %d module(s) in %d tier(s))",
strings.Join(append(append([]string{}, scope.Modules...), prefixedAll("new:", scope.New)...), ", "),
scope.Width, scope.Tiers)
}
fmt.Fprintf(w, "%s/%s#%d (%.8s): asked %s to check it before it merges — %s — as %s\n", p.Owner, p.Repo, p.Number,
p.Commit, seat, what, request.ID)
return request.ID, nil
}
// composedWords is what a group member's own verdict says of its gate layer (novox/hq ADR 0403 clause 1).
const composedWords = "judged by its group's composed check"
// gateComposed is a gate layer's verdict when no gate was run for the member alone: never a pass. Mesh-delivery reads it
// as the composed check's to judge, and the forge's holder sets no mesh/merge-gate from it.
const gateComposed = "composed"
// composedGate says a group member's own verdict has no gate of its own: its gate layer, and the verdict that is
// the gate's, read "composed" in place of a pass.
func composedGate(c *link.Checked) {
c.Verdict, c.Summary = gateComposed, composedWords
c.Gate = &link.CheckLayer{Verdict: gateComposed, Summary: composedWords}
}
// directVerdict is the verdict of a change no build judges — it touches no module, and either the repository is not
// the mesh's or the head holds no merge-check.sh. **It names the trunk commit it holds for** (novox/hq ADR 0323 §3):
// the base branch's head the forge announced with the head, so the merge tool can weigh it against the trunk's head
// and does not refuse it for ever (a verdict naming no trunk is never the trunk's head's). Nothing was built to
// judge: the trunk commit is the one the forge said the change was announced against.
func directVerdict(p link.PullUpdated, plan link.ChangePlan) link.Checked {
trunk := p.BaseSHA
if len(trunk) > 12 {
trunk = trunk[:12]
}
if trunk != "" {
plan.Trunk = trunk
}
return link.Checked{Owner: p.Owner, Repo: p.Repo, Number: p.Number, Commit: p.Commit,
ID: link.NewBuildID(time.Now()), Verdict: "pass", Summary: noModuleTouched,
Gate: &link.CheckLayer{Verdict: "pass", Summary: noModuleTouched}, Plan: &plan, Trunk: trunk}
}
// checkRequestFor is the ask for one pull request's head: the repository as the mesh clones it, the head,
// and what is read beside it.
func checkRequestFor(ctx context.Context, open *stores, p link.PullUpdated, scope checkScope,
entries []inventory.Entry) (link.BuildRequest, error) {
shelf := map[string]catalogue.Manifest{}
for _, e := range entries {
shelf[e.Manifest.Module] = e.Manifest
}
world, err := restOfTheMesh(ctx, open, shelf, "")
if err != nil {
return link.BuildRequest{}, err
}
clone := func(s inventory.Source) (string, error) {
if s.Seat == "" {
return s.Repository, nil
}
return clonedFromSeat(world, s.Seat, s.Repository)
}
// As the mesh clones a module built from it; a repository no module is built from, from the forge.
source := inventory.Source{Seat: gitSeat, Repository: p.Owner + "/" + p.Repo}
if scope.From != nil {
source = scope.From.Source
}
repository, err := clone(source)
if err != nil {
return link.BuildRequest{}, err
}
current, err := open.inventory.CurrentBuilds(ctx)
if err != nil {
return link.BuildRequest{}, err
}
beside, err := besideOf(entries, func(module string) string { return current[module].Commit }, clone)
if err != nil {
return link.BuildRequest{}, err
}
builds := checkBuildsOf(p, scope, entries, clone)
return link.BuildRequest{
ID: link.NewBuildID(time.Now()),
Repository: repository,
Ref: p.Commit,
Held: heldBy(ctx),
Seats: seatBases(ctx),
Source: sourceOnSeat(source),
Check: checkOf(p, scope, entries, beside, builds, namedForTheCheck(ctx, p, builds), runningBy(ctx)),
}, nil
}
// checkOf is the check a request carries: the pull request, what the change touches, and the buckets of the pull
// requests' keys of the modules claiming the delivery seat (novox/hq ADR 0437 §3, issue 1212).
func checkOf(p link.PullUpdated, scope checkScope, entries []inventory.Entry, beside map[string]link.CheckedOut,
builds []link.CheckBuild, named []link.CheckedBuild, running map[string]string) *link.CheckRequest {
return &link.CheckRequest{Pulls: pullsBucketsOf(entries), Owner: p.Owner, Repo: p.Repo, Number: p.Number, Base: p.Base,
Trunk: p.BaseSHA, Ready: p.Ready, Paths: p.Paths, Beside: beside, Modules: scope.Modules, Dependents: scope.Dependents,
New: scope.New, Manifests: scope.Manifests, Judge: scope.Judge, Builds: builds, Named: named, Running: running}
}
// besideOf is what a check reads beside its own repository, by the directory it is found under: each core repository
// by the module the mesh builds from it, at besideRefs. running is the commit the mesh runs of a module.
func besideOf(entries []inventory.Entry, running func(module string) string,
clone func(inventory.Source) (string, error)) (map[string]link.CheckedOut, error) {
beside := map[string]link.CheckedOut{}
for _, e := range entries {
dir, core := coreModules[e.Manifest.Module]
if !core || e.Provided || e.Source.Repository == "" {
continue
}
url, err := clone(e.Source)
if err != nil {
return nil, err
}
refs := besideRefs(dir, running(e.Manifest.Module))
beside[dir] = link.CheckedOut{Repository: url, Ref: refs[dir]}
if dir == "mesh-controller" {
// The controller's main stays its main: it judges only a check the running controller predates.
beside["mesh-controller-main"] = link.CheckedOut{Repository: url, Ref: refs["mesh-controller-main"]}
if e.Source.Seat != "" {
for _, sibling := range []string{"mesh-lab", "mesh-sdk"} {
if url, err := clone(inventory.Source{Seat: e.Source.Seat, Repository: siblingOf(e.Source.Repository,
sibling)}); err == nil {
beside[sibling] = link.CheckedOut{Repository: url, Ref: refs[sibling]}
}
}
}
}
}
return beside, nil
}
// trunkOf is the branch a module's source follows: main when it records none, or records a commit.
func trunkOf(s inventory.Source) string {
if t := strings.TrimSpace(followedBranch(s.Ref)); t != "" {
return t
}
return "main"
}
// sourceName is the name of a recorded repository, a path on the git seat or a URL: novox/mesh-host → mesh-host.
func sourceName(repository string) string {
parts := strings.Split(strings.Trim(strings.TrimSuffix(repository, ".git"), "/"), "/")
if len(parts) < 2 {
return ""
}
return parts[len(parts)-1]
}
// namedForTheCheck are the builds earlier checks of the same pull request at the same head named, read once for the
// check (novox/hq issue 703): asked again because only the trunk moved, the check keeps what they built of a key
// still the same. Another pull request's build, or one of an earlier head, is never kept. None when they cannot be
// read, and the check builds, as before.
func namedForTheCheck(ctx context.Context, p link.PullUpdated, builds []link.CheckBuild) []link.CheckedBuild {
if len(builds) == 0 || p.Commit == "" {
return nil
}
pull := link.PullRef{Owner: p.Owner, Repo: p.Repo, Number: p.Number}
read, err := checkedBuildsOf(ctx, []link.PullRef{pull})
if err != nil {
fmt.Printf(" the earlier checks' builds of %s/%s#%d could not be read, so the check builds: %v\n",
p.Owner, p.Repo, p.Number, err)
// empty-on-error: said above; a check handed no named build builds every module, as before issue 703
return nil
}
var named []link.CheckedBuild
for _, b := range read {
if b.Pull == pull && link.SameHead(b.Head, p.Commit) {
named = append(named, b)
}
}
return named
}
// checkBuildsOf are the modules a merge of the change would build, in the walk's order, as the check builds them
// for the walk to take (novox/hq ADR 0350 §1): a module built from the repository checked, at the branch it merges
// into, from the merge the check makes; any other — a module standing on one — from its own repository at the
// branch it follows, as the walk builds it. A module whose build reads the repository checked through a context,
// or that the merge deletes, cannot be built as the walk would: said, and nothing after it is built. A change that
// moves no module builds nothing.
func checkBuildsOf(p link.PullUpdated, scope checkScope, entries []inventory.Entry,
clone func(inventory.Source) (string, error)) []link.CheckBuild {
if len(scope.Reach.Plan.Tiers) == 0 {
return nil
}
m := link.SourceMoved{Owner: p.Owner, Repo: p.Repo, Base: p.Base, CloneURL: p.CloneURL}
byName := map[string]inventory.Entry{}
for _, e := range entries {
byName[e.Manifest.Module] = e
}
skip := map[string]string{}
for _, e := range scope.Reach.Packaging {
skip[e.Manifest.Module] = "its build reads " + p.Owner + "/" + p.Repo + " through a context, which a check " +
"cannot hand the merge"
}
for _, e := range scope.Reach.Deleted {
skip[e.Manifest.Module] = "the merge deletes it"
}
var out []link.CheckBuild
for tier, names := range scope.Reach.Plan.Tiers {
for _, name := range names {
e, known := byName[name]
b := link.CheckBuild{Module: name, Tier: tier, Skip: skip[name]}
switch {
case !known:
b.Skip = "the catalogue does not hold it"
case b.Skip != "":
case sourceIs(e.Source, m):
b.Merged, b.Path = true, e.Source.Path
default:
url, err := clone(e.Source)
if err != nil {
b.Skip = "its repository cannot be cloned: " + err.Error()
break
}
b.Repository, b.Path, b.Ref = url, e.Source.Path, followedBranch(e.Source.Ref)
}
out = append(out, b)
}
}
return out
}
// besideRefs is the ref each repository is cloned at beside a check, by the directory it is found under:
@@ -187,5 +501,186 @@ func sdkPinned() string {
return "main"
}
// siblingOf is another repository of the same owner: novox/mesh-controller → novox/mesh-lab.
func siblingOf(repository, name string) string {
if cut := strings.LastIndex(repository, "/"); cut >= 0 {
return repository[:cut+1] + name
}
return name
}
// sourceOnSeat is a source's seat form, nil for one on no seat.
func sourceOnSeat(s inventory.Source) *link.SourceOnSeat {
if s.Seat == "" {
return nil
}
return &link.SourceOnSeat{Seat: s.Seat, Repository: s.Repository}
}
// checkEvents is where the serving controller says a check's verdict; nil in a command.
var checkEvents link.Bus
// maxCheckReport is how much of a check's report travels in its verdict: enough for the failures and
// the machines, never a log.
const maxCheckReport = 60 << 10
// checked says a merge check's verdict as the controller's `checked`. Nothing is recorded or
// registered: a check builds nothing (issue 240's rule for a dry run, kept for a check).
func checked(ctx context.Context, result link.BuildResult) {
// **A check stopped by its key says nothing** (novox/hq ADR 0406 §4, issue 974, review 2): it stopped because the head or
// the trunk commit moved on, and the newer request will judge. Said as an error it would name the live head — red
// statuses on the forge, a delivery refused — and written to the key it would stand in the newer request's way.
if result.Failed == link.SupersededByKey {
fmt.Printf("%s: stopped, its pull request's key moved on; the newer check will judge\n", result.ID)
return
}
sayChecked(ctx, os.Stdout, checkedOf(result))
}
// checkedOf is what a check's outcome says: each layer, and an error — never a pass — for a check that
// could not run.
func checkedOf(result link.BuildResult) link.Checked {
c := link.Checked{ID: result.ID, On: result.On, Commit: result.Ref}
if result.Checked != nil {
c.Owner, c.Repo, c.Number = result.Checked.Owner, result.Checked.Repo, result.Checked.Number
}
switch {
case result.Check != nil:
c.Verdict, c.Summary, c.Report = result.Check.Verdict, result.Check.Summary, result.Check.Report
c.Gate, c.RepoCheck = result.Check.Gate, result.Check.RepoCheck
// What was judged (novox/hq ADR 0323 §3): the head merged into the trunk commit, and the commits of the
// repositories read beside it.
c.Trunk, c.Beside, c.Took = result.Check.Trunk, result.Check.Beside, result.Check.Took
c.Built = result.Check.Built
c.Validator = result.Check.Validator
c.Siblings, c.SiblingsUnknown, c.JudgeFrom = result.Check.Siblings, result.Check.SiblingsUnknown, result.Check.JudgeFrom
if c.Gate == nil {
// A build seat from before the layers: its verdict is the gate's.
c.Gate = &link.CheckLayer{Verdict: c.Verdict, Summary: c.Summary}
}
case result.Failed != "":
// The check could not run: an error, never read as a pass — on both layers it was asked for.
c.Verdict, c.Summary = "error", "the check could not run: "+firstLine(result.Failed)
default:
c.Verdict, c.Summary = "error", "the build seat answered the check with no verdict"
}
if c.Verdict == "" {
c.Verdict = "error"
}
if result.Check == nil {
c.Gate = &link.CheckLayer{Verdict: c.Verdict, Summary: c.Summary}
c.RepoCheck = &link.CheckLayer{Verdict: c.Verdict, Summary: c.Summary}
}
// **A group member's own check has no gate of its own** (novox/hq ADR 0403 clause 1): what ran was the repository's
// check alone, and its gate layer is "composed", never a pass. A failure or an error stays what it was.
if result.Checked != nil && result.Checked.Composed && result.Check != nil && c.Gate != nil &&
(c.Gate.Verdict == "pass" || c.Gate.Verdict == "warning") {
composedGate(&c)
}
if result.Checked != nil && c.Gate != nil && len(c.Gate.Modules) == 0 {
c.Gate.Modules = append(append([]string{}, result.Checked.Modules...), prefixedAll("new:", result.Checked.New)...)
}
if result.Checked != nil && c.Gate != nil && len(c.Gate.Dependents) == 0 {
c.Gate.Dependents = result.Checked.Dependents
}
if result.Checked != nil {
c.Plan = result.Checked.Plan
if c.Plan != nil && c.Trunk != "" {
// The kept delivery plan records the trunk commit with its other inputs (ADR 0323 §3).
plan := *c.Plan
plan.Trunk = c.Trunk
c.Plan = &plan
}
// A delivery group's composed check (novox/hq ADR 0239): every head it judged, this one first.
if g := result.Checked; g.Group != "" {
c.Group = g.Group
c.Members = append(c.Members, link.CheckedMember{Owner: g.Owner, Repo: g.Repo, Number: g.Number,
Commit: result.Ref, Trunk: c.Trunk})
for _, m := range g.Members {
c.Members = append(c.Members, link.CheckedMember{Owner: m.Owner, Repo: m.Repo, Number: m.Number,
Commit: m.Ref, Trunk: memberTrunk(result.Check, m)})
}
}
}
for _, l := range []*link.CheckLayer{c.Gate, c.RepoCheck} {
if l != nil && l.Verdict == "" {
l.Verdict = "error"
}
}
// **A check run again says so** (novox/hq issue 594). The summaries are the lines the pull request
// shows, clipped by the forge to a few words, so they get only a short marker after the verdict's own
// words; the report gets the whole sentence. Layers copied, so the build's own outcome is left as it came.
if mark := link.RunMarker(result.Delivery); mark != "" {
c.Summary += " " + mark
for _, l := range []**link.CheckLayer{&c.Gate, &c.RepoCheck} {
if *l != nil {
layer := **l
layer.Summary += " " + mark
*l = &layer
}
}
c.Report = link.Restarted(result.Delivery) + "\n\n" + c.Report
}
if len(c.Report) > maxCheckReport {
c.Report = "…" + c.Report[len(c.Report)-maxCheckReport:]
}
return c
}
// memberTrunk is the trunk commit the build seat merged one of a group's heads into (novox/hq ADR 0323 §6), empty
// when it said none.
func memberTrunk(outcome *link.CheckOutcome, m link.GroupMember) string {
if outcome == nil {
return ""
}
for _, j := range outcome.Members {
if strings.EqualFold(j.Owner, m.Owner) && strings.EqualFold(j.Repo, m.Repo) && strings.HasPrefix(m.Ref, j.Commit) {
return j.Trunk
}
}
return ""
}
func prefixedAll(prefix string, items []string) []string {
out := make([]string, 0, len(items))
for _, i := range items {
out = append(out, prefix+i)
}
return out
}
// sayChecked says a merge check's verdict on the bus, where the forge's holder hears it.
func sayChecked(ctx context.Context, w io.Writer, c link.Checked) {
repo := "none"
if c.RepoCheck != nil {
repo = strings.ToUpper(c.RepoCheck.Verdict) + " — " + c.RepoCheck.Summary
}
judged := ""
if c.Trunk != "" {
judged = " merged into trunk " + c.Trunk
}
fmt.Fprintf(w, "%s: %s/%s#%d at %.8s%s checked on %s: gate %s — %s; repository %s\n", c.ID, c.Owner, c.Repo, c.Number,
c.Commit, judged, orSomewhere(c.On), strings.ToUpper(c.Verdict), c.Summary, repo)
if checkEvents == nil {
return
}
body, err := json.Marshal(c)
if err != nil {
return
}
stating, stop := context.WithTimeout(context.WithoutCancel(ctx), 10*time.Second)
defer stop()
if err := checkEvents.PublishSeatEvent(stating, link.MeshControllerSeat, link.KeyChecked, body); err != nil {
fmt.Fprintf(w, "%s: the verdict could not be said, so the pull request is not told it: %v\n", c.ID, err)
}
}
// pullsBucketsOf are the buckets of the pull requests' keys a check names for the build agent: one per module
// claiming the delivery seat, in the order of their names (novox/hq ADR 0437 §3, issue 1212).
func pullsBucketsOf(entries []inventory.Entry) []string {
var out []string
for _, m := range claimantsOf(entries, catalogue.DeliverySeat) {
out = append(out, broker.BucketName(m, "pulls"))
}
return out
}
@@ -0,0 +1,56 @@
//go:build integration
package main
import (
"encoding/json"
"testing"
"time"
"github.com/nats-io/nats.go"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/link"
"github.com/novox/mesh-controller/internal/testbus"
)
func TestACheckStoppedByItsKeyIsAnnouncedAsNoVerdict(t *testing.T) {
js, err := broker.Dial(testbus.URL(t))
if err != nil {
t.Fatal(err)
}
t.Cleanup(js.Close)
if err := broker.AssertMeshStreams(js); err != nil {
t.Fatal(err)
}
legacy, err := js.Conn().JetStream()
if err != nil {
t.Fatal(err)
}
was := checkEvents
checkEvents = link.OverNATS{Conn: js.Conn(), JS: legacy}
t.Cleanup(func() { checkEvents = was })
heard := make(chan *nats.Msg, 4)
sub, err := js.Conn().ChanSubscribe(link.SeatEventSubject(link.MeshControllerSeat, link.KeyChecked), heard)
if err != nil {
t.Fatal(err)
}
defer func() { _ = sub.Unsubscribe() }()
asked := &link.CheckRequest{Owner: "novox", Repo: "mesh-catalog", Number: 3}
checked(t.Context(), link.BuildResult{ID: "stopped", Ref: "abc", Checked: asked, Failed: link.SupersededByKey})
checked(t.Context(), link.BuildResult{ID: "broken", Ref: "abc", Checked: asked, Failed: "the facts snapshot cannot be read"})
select {
case m := <-heard:
var c link.Checked
_ = json.Unmarshal(m.Data, &c)
if c.ID != "broken" || c.Verdict != "error" {
t.Fatalf("the first verdict said is %+v; want only the broken check's error", c)
}
case <-time.After(5 * time.Second):
t.Fatal("nothing was said")
}
select {
case m := <-heard:
t.Fatalf("a second verdict was said: %s", m.Data)
case <-time.After(300 * time.Millisecond):
}
}
+97
View File
@@ -0,0 +1,97 @@
//go:build integration
package main
import (
"context"
"encoding/json"
"fmt"
"testing"
"github.com/nats-io/nats.go"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// Acceptance 5 of milestone 2 (novox/hq ADR 0406 §3, issue 974), live, with the store and a real bus: two merges into one
// repository within a minute, with two open pull requests, leave each pull request with exactly one queued check, by the
// stream's rule. Each merge moves the trunk, and mesh-delivery asks the controller to check both open pull requests again
// against it (`delivery check`, which is the controller's own path for a head announced); nothing moves a request, nothing
// asks the forge, and the queue is read back whole.
func TestTwoMergesWithinAMinuteLeaveEachOpenPullRequestOneQueuedCheck(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
js, _ := aBuildQueue(t)
servingBus.Store(js)
t.Cleanup(func() { servingBus.Store(nil) })
holdTheDeliverySeat(t, open)
was := askDeliveryOwner
t.Cleanup(func() { askDeliveryOwner = was })
askDeliveryOwner = func(_ context.Context, _ *nats.Conn, verb string, _ map[string]any) (json.RawMessage, error) {
if verb != "groups" {
t.Fatalf("asked the owner %s", verb)
}
return json.RawMessage(`[]`), nil
}
// The repository the two pull requests are into, as a module the mesh builds from it.
if err := open.inventory.RegisterModule(ctx, catalogue.Manifest{Module: "mesh-controller", Version: "1"},
inventory.Source{Repository: "novox/mesh-controller", Ref: "main", BuiltFrom: "c0", Head: "c0"}); err != nil {
t.Fatal(err)
}
heads := map[int]string{7: "7777777777777777777777777777777777777777", 8: "8888888888888888888888888888888888888888"}
check := func(number int, trunk string) {
t.Helper()
p := orderMember{ID: fmt.Sprintf("novox/mesh-controller@%.12s", heads[number]), Repository: "novox/mesh-controller",
Base: "main", Head: heads[number], Number: number, Paths: []string{"main.go"}, ModuleDirsSaid: true,
BaseSHA: trunk, Ready: number == 8}.pull()
p.MergeCheck, p.MergeCheckSaid = true, true
if err := (following{open: open, oneShot: true}).PullUpdated(ctx, p); err != nil {
t.Fatal(err)
}
}
// Announced against the trunk as it stands, then asked again after each of two merges.
for _, trunk := range []string{"1111111111111111111111111111111111111111", "2222222222222222222222222222222222222222",
"3333333333333333333333333333333333333333"} {
check(7, trunk)
check(8, trunk)
}
q, err := link.ReadQueue(ctx, js, link.TheBuildMachine)
if err != nil {
t.Fatal(err)
}
seen := map[int]string{}
for _, a := range q.Asks {
c := a.Request.Check
if c == nil {
t.Fatalf("%s is no check", a.ID)
}
if _, twice := seen[c.Number]; twice {
t.Errorf("pull request %d has two queued checks", c.Number)
}
seen[c.Number] = c.Trunk
if a.State != link.AskWaiting {
t.Errorf("%s is %s", a.ID, a.State)
}
if want := "3333333333333333333333333333333333333333"; c.Trunk != want {
t.Errorf("pull request %d is queued against %.8s; want the newest trunk %.8s", c.Number, c.Trunk, want)
}
}
if len(seen) != 2 {
t.Fatalf("queued checks for %v; want one for each of the two open pull requests", seen)
}
// By the streams' own count: the ready one on its stream, the other on its.
for stream, want := range map[string]uint64{broker.ChecksReadyStream: 1, broker.ChecksOtherStream: 1} {
info, err := js.Context().StreamInfo(stream)
if err != nil {
t.Fatal(err)
}
if info.State.Msgs != want {
t.Errorf("%s holds %d request(s); want %d", stream, info.State.Msgs, want)
}
}
}
+143 -2
View File
@@ -1,6 +1,7 @@
package main
import (
"errors"
"strings"
"testing"
@@ -75,8 +76,8 @@ func TestThePullRequestIsMappedOntoTheModuleGraph(t *testing.T) {
t.Errorf("%s: %s %q, wanted %q", c.what, what, got[i], want[i])
}
}
if s.Mesh != c.mesh {
t.Errorf("%s: the mesh's %v, wanted %v", c.what, s.Mesh, c.mesh)
if s.Mesh != c.mesh || s.Judge != c.judge {
t.Errorf("%s: the mesh's %v judged by %q, wanted %v by %q", c.what, s.Mesh, s.Judge, c.mesh, c.judge)
}
if s.gated() != (c.modules != "" || c.new != "") {
t.Errorf("%s: gated %v", c.what, s.gated())
@@ -94,6 +95,28 @@ func TestAPullRequestTouchesWhatPackagesItsRepository(t *testing.T) {
}
}
// Each layer is said; a check that could not run is an error on both, never a pass; a build seat from
// before the layers is read as the gate.
func TestAChecksLayersAreEachSaidAndAnErrorIsNeverAPass(t *testing.T) {
asked := &link.CheckRequest{Owner: "novox", Repo: "mesh-catalog", Number: 3, Modules: []string{"gitea"}}
c := checkedOf(link.BuildResult{ID: "b", Ref: "abc", Checked: asked, Failed: "the facts snapshot cannot be read"})
if c.Verdict != "error" || c.Gate == nil || c.Gate.Verdict != "error" || c.RepoCheck == nil || c.RepoCheck.Verdict != "error" {
t.Fatalf("a check that could not run said %+v", c)
}
if strings.Join(c.Gate.Modules, ",") != "gitea" {
t.Errorf("the gate names %v", c.Gate.Modules)
}
c = checkedOf(link.BuildResult{ID: "b", Ref: "abc", Checked: asked, Check: &link.CheckOutcome{Verdict: "warning", Summary: "wide"}})
if c.Gate == nil || c.Gate.Verdict != "warning" || c.RepoCheck != nil {
t.Fatalf("an outcome without layers said %+v", c)
}
c = checkedOf(link.BuildResult{ID: "b", Ref: "abc", Checked: asked, Check: &link.CheckOutcome{Verdict: "pass",
Gate: &link.CheckLayer{Verdict: "pass"}, RepoCheck: &link.CheckLayer{Verdict: "fail", Summary: "its merge-check.sh failed"}}})
if c.RepoCheck.Verdict != "fail" || c.Gate.Verdict != "pass" {
t.Fatalf("the layers said %+v / %+v", c.Gate, c.RepoCheck)
}
}
// **The check asks the planner, never a mapping of its own** (novox/hq ADR 0238): what a pull request
// reaches is reachOfMerge's answer — the same that plans a merge — so a file at the root touches no
// module in both, and what stands on a touched module is named as its dependents in both.
@@ -148,5 +171,123 @@ func TestAPullRequestReachesWhatAMergeOfItWouldPlan(t *testing.T) {
}
}
// **Only a commit on the trunk is published** (novox/hq ADR 0238): a build of a commit off its repository's
// default branch — a pull request's head, a branch built by hand, a rebuild or replay of one — is recorded
// and never registered, so nothing can send it; a check or a dry run never is either.
func TestABuildOffTheTrunkIsNeverPublished(t *testing.T) {
on := link.BuildResult{ID: "b", Commit: "abc", Trunk: "main", OnTrunk: true}
if err := publishable(on, ""); err != nil {
t.Errorf("a build on the trunk was refused: %v", err)
}
off := link.BuildResult{ID: "b", Commit: "abc", Trunk: "main"}
if err := publishable(off, ""); !errors.Is(err, errOffTheTrunk) || !strings.Contains(err.Error(), "main") {
t.Errorf("a build off the trunk was let through: %v", err)
}
for _, r := range []link.BuildResult{
{ID: "c", Trunk: "main", OnTrunk: true, Check: &link.CheckOutcome{Verdict: "pass"}},
{ID: "d", Trunk: "main", OnTrunk: true, Checked: &link.CheckRequest{}},
{ID: "e", Trunk: "main", OnTrunk: true, DryRun: true},
} {
if publishable(r, "") == nil {
t.Errorf("%s, a check or a dry run, was publishable", r.ID)
}
}
// A module that follows a branch other than the default: that branch is its trunk, and the default
// is not.
follows := link.BuildResult{ID: "g", Commit: "abc", Trunk: "master", Branches: []string{"nox-mesh"}}
if err := publishable(follows, "nox-mesh"); err != nil {
t.Errorf("a commit on the branch a module follows was refused: %v", err)
}
if err := publishable(link.BuildResult{ID: "h", Commit: "abc", Trunk: "master", OnTrunk: true,
Branches: []string{"master"}}, "nox-mesh"); err == nil {
t.Error("a commit on the default but not on the branch the module follows was published")
}
if err := publishable(link.BuildResult{ID: "i", Commit: "abc", Trunk: "main", Branches: []string{"feat/x"}}, ""); err == nil {
t.Error("a feature branch's commit was published")
}
// A build seat older than the rule says nothing: let through and said, so the rule can reach the mesh.
if err := publishable(link.BuildResult{ID: "f", Commit: "abc"}, ""); err != nil {
t.Errorf("a build seat that said nothing was refused: %v", err)
}
}
// **A verdict names the trunk commit it judged** (novox/hq ADR 0323 §3, issue 341): the controller's `checked`
// carries the head and the trunk commit the build seat merged it into, the commits of the repositories read beside
// it, how long it took, and the kept delivery plan records the trunk commit with its other inputs. A group's
// composed check names, per member, the trunk commit that head was merged into (§6).
func TestAVerdictNamesTheTrunkCommitItJudged(t *testing.T) {
plan := &link.ChangePlan{Repository: "novox/mesh-host", Base: "main", Head: "dfb1415aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
Summary: "builds mesh-host"}
asked := &link.CheckRequest{Owner: "novox", Repo: "mesh-host", Number: 66, Base: "main", Plan: plan}
c := checkedOf(link.BuildResult{ID: "b", Ref: "dfb1415aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa", Checked: asked,
Check: &link.CheckOutcome{Verdict: "pass", Took: "4m12s", Head: "dfb1415aaaaa", Trunk: "d28ca95bbbbb",
Gate: &link.CheckLayer{Verdict: "pass"}, RepoCheck: &link.CheckLayer{Verdict: "pass"},
Beside: map[string]string{"mesh-catalog": "0123456789ab", "mesh-lab": "ba9876543210"}}})
if c.Trunk != "d28ca95bbbbb" || c.Commit != "dfb1415aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa" {
t.Fatalf("the verdict names head %s and trunk %q", c.Commit, c.Trunk)
}
if c.Beside["mesh-catalog"] != "0123456789ab" || c.Beside["mesh-lab"] != "ba9876543210" {
t.Fatalf("the verdict names the repositories read beside it as %v", c.Beside)
}
if c.Took != "4m12s" {
t.Fatalf("the verdict says it took %q", c.Took)
}
if c.Plan == nil || c.Plan.Trunk != "d28ca95bbbbb" || c.Plan.Head != plan.Head {
t.Fatalf("the kept plan records %+v", c.Plan)
}
if plan.Trunk != "" {
t.Fatal("the plan the controller asked with was changed: the verdict's is a copy")
}
group := &link.CheckRequest{Owner: "novox", Repo: "mesh-controller", Number: 7, Base: "main", Group: "feat/x",
Members: []link.GroupMember{{Owner: "novox", Repo: "mesh-host", Number: 8, Ref: "cccccccccccc0000", Base: "main"}}}
c = checkedOf(link.BuildResult{ID: "g", Ref: "aaaaaaaaaaaa0000", Checked: group, Check: &link.CheckOutcome{
Verdict: "pass", Head: "aaaaaaaaaaaa", Trunk: "111111111111", Gate: &link.CheckLayer{Verdict: "pass"},
Members: []link.CheckedMember{{Owner: "novox", Repo: "mesh-host", Commit: "cccccccccccc", Trunk: "222222222222"}}}})
if len(c.Members) != 2 || c.Members[0].Trunk != "111111111111" || c.Members[1].Trunk != "222222222222" ||
c.Members[1].Commit != "cccccccccccc0000" {
t.Fatalf("the composed check names its members %+v", c.Members)
}
}
// **The merge made for a check is never registered** (novox/hq ADR 0323 §2): a build of it — a commit on no branch
// of the forge — is refused, naming the trunk. The rule of ADR 0238 §1 stands unchanged; this holds it for the
// merge commit a check makes.
func TestABuildOfACheckMergeIsRefusedNamingTheTrunk(t *testing.T) {
merge := link.BuildResult{ID: "b", Commit: "4b5c6d7e8f90aabbccddeeff0011223344556677", Trunk: "main"}
err := publishable(merge, "")
if !errors.Is(err, errOffTheTrunk) || !strings.Contains(err.Error(), "not on main") {
t.Fatalf("a build of a check's merge was answered %v", err)
}
}
// **A verdict no build judges names its trunk commit too** (novox/hq ADR 0323 §3): a change touching no module of a
// repository with no merge-check.sh (novox/mesh-tools-go) is answered by the controller itself. Naming no trunk, the
// merge tool refused it, asked again, got the same verdict, and refused for ever.
func TestADirectVerdictNamesTheTrunkTheForgeAnnounced(t *testing.T) {
p := link.PullUpdated{Owner: "novox", Repo: "mesh-tools-go", Number: 3, Base: "main", Commit: "1a2b3c4d5e6f7a8b9c0d",
BaseSHA: "7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b"}
c := directVerdict(p, link.ChangePlan{Repository: "novox/mesh-tools-go", Base: "main", Head: p.Commit})
if c.Trunk != "7a8b9c0d1e2f" || c.Plan == nil || c.Plan.Trunk != "7a8b9c0d1e2f" {
t.Fatalf("the direct verdict names trunk %q, its plan %+v", c.Trunk, c.Plan)
}
if c.Verdict != "pass" || c.Commit != p.Commit {
t.Fatalf("the direct verdict is %+v", c)
}
if c := directVerdict(link.PullUpdated{Owner: "novox", Repo: "x", Commit: "abc"}, link.ChangePlan{}); c.Trunk != "" {
t.Fatalf("a forge that said no base head gave the trunk %q", c.Trunk)
}
}
// A head checked again through mesh-delivery (`delivery check --members`, a recheck) reaches the same direct verdict:
// it names the base head the forge last announced, kept on the member, or the merge tool would refuse it for ever.
func TestARecheckedHeadsDirectVerdictNamesItsTrunk(t *testing.T) {
m := orderMember{ID: "novox/mesh-tools-go@1a2b3c4d5e6f", Repository: "novox/mesh-tools-go", Base: "main",
Head: "1a2b3c4d5e6f7a8b9c0d", Number: 3, BaseSHA: "7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b"}
if c := directVerdict(m.pull(), link.ChangePlan{}); c.Trunk != "7a8b9c0d1e2f" {
t.Fatalf("a rechecked head's direct verdict names trunk %q", c.Trunk)
}
}
// A check stopped by its pull request's key moving on says no verdict (issue 974, review 2): an error said for it would name the
// live head, and the newer request is the one that judges. A check that failed for any other reason still says its error.
+7 -16
View File
@@ -4,9 +4,7 @@ import (
"context"
"errors"
"fmt"
"io"
"os"
"strings"
"time"
"github.com/novox/mesh-controller/internal/artifacts"
@@ -217,34 +215,27 @@ func collect(ctx context.Context, inv *inventory.Inventory) {
}
r := sweep(ctx, inv, artifacts.Store{Address: address}, references, kept, afterBuild)
sayCollected(os.Stderr, r)
}
// sayCollected says what one sweep after a build did.
func sayCollected(w io.Writer, r sweepResult) {
if r.HoldersWritten > 0 {
fmt.Fprintf(w, "the artifact store now holds %d more kept archive(s) by a manifest\n", r.HoldersWritten)
fmt.Fprintf(os.Stderr, "the artifact store now holds %d more kept archive(s) by a manifest\n", r.HoldersWritten)
}
if r.Missing > 0 {
fmt.Fprintf(w, "%d archive(s) the mesh keeps are not in the artifact store at all; "+
fmt.Fprintf(os.Stderr, "%d archive(s) the mesh keeps are not in the artifact store at all; "+
"`collection` lists them\n", r.Missing)
}
if len(r.LetGo) > 0 {
// **Each one named** (novox/hq issue 1933): a count could not say which archive a machine's 404 was let go of.
fmt.Fprintf(w, "the artifact store let go of %d artifact(s) the mesh no longer keeps: %s\n", len(r.LetGo),
strings.Join(r.LetGo, ", "))
fmt.Fprintf(os.Stderr, "the artifact store let go of %d artifact(s) the mesh no longer keeps\n", len(r.LetGo))
}
if r.Stopped != "" && !r.Bounded {
fmt.Fprintln(w, r.Stopped)
fmt.Fprintln(os.Stderr, r.Stopped)
}
if r.Held && r.Left > 0 {
fmt.Fprintf(w, "%d more to collect; the next build asks again\n", r.Left)
fmt.Fprintf(os.Stderr, "%d more to collect; the next build asks again\n", r.Left)
}
if r.Skipped > 0 {
fmt.Fprintf(w, "%d artifact(s) the sweep will not address were skipped\n", r.Skipped)
fmt.Fprintf(os.Stderr, "%d artifact(s) the sweep will not address were skipped\n", r.Skipped)
}
if r.Indexes > 0 {
fmt.Fprintf(w, "%d eligible index(es) wait for a collect a person confirms, which takes their platforms too\n", r.Indexes)
fmt.Fprintf(os.Stderr, "%d eligible index(es) wait for a collect a person confirms, which takes their platforms too\n", r.Indexes)
}
}
-6
View File
@@ -233,12 +233,6 @@ func prepareSend(ctx context.Context, open *stores, node string) ([]string, erro
if !slices.Contains(in.BusMinted, "provisioner."+node) {
continue
}
default:
// A bus process's account nobody issued still gets the stand-in, until the composition leaves
// the module off (issue 1753): the warning names the verb (issue 1226).
if line := processAccountWarning(m, node, name, in.BusMinted); line != "" {
sayWarning(ctx, line)
}
}
if _, err := inv.SecretForModule(ctx, node, m.Module, name); err == nil {
did = true
+2 -8
View File
@@ -200,7 +200,6 @@ func composeQuestion(ctx context.Context, open *stores, node string, mode compos
}
sayFound(ctx, d)
out := sendableOf(d)
out.fileReferences = readsFileReferences(in.Input, node)
if !withOrder {
out.Sequence, out.Epoch, out.Generation = 0, 0, 0
}
@@ -239,11 +238,7 @@ func markUnresolvable(err error) error {
// still names anything foreseen is refused, said, and never sent.
func composeSend(ctx context.Context, open *stores, node string) (sendable, error) {
declared, _, err := composeSendValidated(ctx, open, node)
if err != nil {
return declared, err
}
// Never a declaration naming an image the store does not hold (novox/hq issue 1742): that module is left out.
return withoutMissingImages(ctx, open, node, declared), nil
return declared, err
}
// composeSendValidated is composeSend and what the node-engine's validator refuses of the declaration it composed, for the
@@ -286,7 +281,6 @@ func composeSendValidated(ctx context.Context, open *stores, node string) (senda
if len(left) == 0 && len(in.wouldMake) == 0 {
sayFound(ctx, d)
out := sendableOf(d)
out.fileReferences = readsFileReferences(in.Input, node)
// The sender stamps the order it allots (order.stamp); a composition for a send carries none of its own.
out.Sequence, out.Epoch, out.Generation = 0, 0, 0
here, _ := nodeOf(in.Input, node)
@@ -372,7 +366,7 @@ func sayFound(ctx context.Context, d compose.Declared) {
// sendableOf is a composed declaration as the controller's sends know it.
func sendableOf(d compose.Declared) sendable {
out := sendable{Resources: compactKept(d.Resources), Sequence: d.Sequence, Epoch: d.Epoch, Generation: d.Generation,
out := sendable{Resources: d.Resources, Sequence: d.Sequence, Epoch: d.Epoch, Generation: d.Generation,
ReadsModules: d.ReadsModules, moduleOf: d.ModuleOf, buildIDs: d.BuildIDs, modules: d.Modules,
localNetwork: d.LocalNetwork, Received: d.Received, Mesh: d.Mesh, BusUsers: d.BusUsers,
LeftOut: d.LeftOut, leftOutWhy: d.LeftOutWhy, withheld: d.Withheld, unbound: d.Unbound, foreseen: d.Foreseen,
@@ -0,0 +1,68 @@
//go:build integration
package main
import (
"context"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// A member of a group whose gate is the composed check's is asked of the build seat with no module: its repository's
// check alone (novox/hq ADR 0403 clause 1). With no word from mesh-delivery, or one that does not say the gate is
// composed, it is checked alone with its gate (clause 6).
func askedForMember(t *testing.T, answer []groupsLine, answerErr error) link.CheckRequest {
t.Helper()
open := aMesh(t)
ctx := t.Context()
js, _ := aBuildQueue(t)
servingBus.Store(js)
t.Cleanup(func() { servingBus.Store(nil) })
if err := open.inventory.RegisterModule(ctx, catalogue.Manifest{Module: "mesh-controller", Version: "1"},
inventory.Source{Repository: "novox/mesh-controller", Ref: "main", BuiltFrom: "c0", Head: "c0"}); err != nil {
t.Fatal(err)
}
was := groupsAsked
groupsAsked = func(context.Context, bool) ([]groupsLine, error) { return answer, answerErr }
t.Cleanup(func() { groupsAsked = was })
p := link.PullUpdated{Owner: "novox", Repo: "mesh-controller", Number: 346, Base: "main", Head: "feat/x", Commit: headController,
Paths: []string{"main.go"}, ModuleDirsSaid: true, MergeCheck: true, MergeCheckSaid: true}
if err := (following{open: open, oneShot: true}).PullUpdated(ctx, p); err != nil {
t.Fatal(err)
}
q, err := link.ReadQueue(ctx, js, link.TheBuildMachine)
if err != nil || len(q.Asks) != 1 {
t.Fatalf("asks %d: %v", len(q.Asks), err)
}
return *q.Asks[0].Request.Check
}
func TestAComposedMembersCheckIsItsRepositoryChecksAlone(t *testing.T) {
line := group("feat/x", member("novox/mesh-controller", 346, headController, "main", true), member("novox/mesh-catalog", 350, headHost, "main", true))
line.ComposedGate = true
c := askedForMember(t, []groupsLine{line}, nil)
if len(c.Modules) != 0 || len(c.New) != 0 || len(c.Dependents) != 0 || !c.Composed {
t.Fatalf("the member's check is asked as %+v; want no module and composed", c)
}
}
func TestAMemberOfAGroupOfAnOlderMeshDeliveryKeepsItsOwnGate(t *testing.T) {
line := group("feat/x", member("novox/mesh-controller", 346, headController, "main", true), member("novox/mesh-catalog", 350, headHost, "main", true))
c := askedForMember(t, []groupsLine{line}, nil)
if !strings.Contains(strings.Join(c.Modules, ","), "mesh-controller") || c.Composed {
t.Fatalf("an older mesh-delivery's member is asked as %+v; want its own gate", c)
}
}
func TestAMemberWhoseGroupCannotBeLearnedIsCheckedAloneWithItsGate(t *testing.T) {
c := askedForMember(t, nil, context.DeadlineExceeded)
if !strings.Contains(strings.Join(c.Modules, ","), "mesh-controller") || c.Composed ||
!strings.Contains(c.SiblingsUnknown, "reads no siblings") {
t.Fatalf("the member is asked as %+v; want its gate and why its siblings are unknown", c)
}
}
@@ -0,0 +1,136 @@
package main
import (
"context"
"encoding/json"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// A delivery group's gate is the composed check's, and its members' own checks run their repository's check alone
// (novox/hq ADR 0403 clauses 1 and 7). Pure: no store, no bus.
// A member of a group whose line says composed_gate is a composed member; one whose group says nothing of it (an older
// mesh-delivery), one of a group of a single pull request and a lone pull request are not.
func TestAMemberOfAComposedGroupIsKnownByTheGroupsLine(t *testing.T) {
two := func(composed bool) groupsLine {
l := group("feat/x", member("novox/mesh-host", 114, headHost, "main", true), member("novox/mesh-controller", 346, headController, "main", true))
l.ComposedGate = composed
return l
}
p := link.PullUpdated{Owner: "novox", Repo: "mesh-host", Number: 114, Base: "main", Head: "feat/x", Commit: headHost}
recheck := link.PullUpdated{Owner: "novox", Repo: "mesh-host", Number: 114, Base: "main", Commit: headHost}
for name, c := range map[string]struct {
p link.PullUpdated
lines []groupsLine
want bool
}{
"composed": {p, []groupsLine{two(true)}, true},
"a recheck, by its number": {recheck, []groupsLine{two(true)}, true},
"an older mesh-delivery": {p, []groupsLine{two(false)}, false},
"another group": {p, []groupsLine{group("feat/y", member("novox/mesh-host", 1, headHost, "main", true))}, false},
"no group": {p, nil, false},
} {
if got := composedMember(c.p, c.lines); got != c.want {
t.Errorf("%s: composed %v, want %v", name, got, c.want)
}
}
alone := group("feat/x", member("novox/mesh-host", 114, headHost, "main", true))
alone.ComposedGate = true
if composedMember(p, []groupsLine{alone}) {
t.Error("a group of a single pull request is composed")
}
}
// A composed member's verdict never says its gate passed: the gate layer, and the verdict that is the gate's, read
// "composed"; a failure or an error stays what it was.
func TestAComposedMembersVerdictHasNoPassingGate(t *testing.T) {
asked := &link.CheckRequest{Owner: "novox", Repo: "mesh-host", Number: 114, Composed: true}
pass := link.BuildResult{ID: "b", Ref: headHost, Checked: asked, Check: &link.CheckOutcome{Verdict: "pass", Summary: "no module touched",
Gate: &link.CheckLayer{Verdict: "pass", Summary: "no module touched"}, RepoCheck: &link.CheckLayer{Verdict: "pass", Summary: "ok"}}}
c := checkedOf(pass)
if c.Gate == nil || c.Gate.Verdict != "composed" || c.Verdict != "composed" || c.RepoCheck == nil || c.RepoCheck.Verdict != "pass" {
t.Fatalf("the member's verdict says %+v gate %+v", c, c.Gate)
}
asked.Composed = false
if c := checkedOf(pass); c.Gate.Verdict != "pass" || c.Verdict != "pass" {
t.Fatalf("a member of no composed group has its own gate: %+v", c.Gate)
}
asked.Composed = true
failing := pass
failing.Check = &link.CheckOutcome{Verdict: "fail", Summary: "conflict", Gate: &link.CheckLayer{Verdict: "fail", Summary: "conflict"},
RepoCheck: &link.CheckLayer{Verdict: "fail", Summary: "conflict"}}
if c := checkedOf(failing); c.Gate.Verdict != "fail" || c.Verdict != "fail" {
t.Fatalf("a failure was read as composed: %+v", c.Gate)
}
broken := link.BuildResult{ID: "b", Ref: headHost, Checked: asked, Failed: "the facts snapshot cannot be read"}
if c := checkedOf(broken); c.Verdict != "error" || c.Gate.Verdict != "error" {
t.Fatalf("an error was read as composed: %+v", c)
}
}
// The composed verdict on the bus carries every member, each with its trunk commit, for the forge's holder to set each
// member's mesh/merge-gate from it.
func TestTheComposedVerdictCarriesEveryMember(t *testing.T) {
asked := &link.CheckRequest{Owner: "novox", Repo: "mesh-host", Number: 114, Group: "feat/x",
Members: []link.GroupMember{{Owner: "novox", Repo: "mesh-controller", Number: 346, Ref: headController}}}
c := checkedOf(link.BuildResult{ID: "g", Ref: headHost, Checked: asked, Check: &link.CheckOutcome{Verdict: "pass", Summary: "ok",
Trunk: "aaaaaaaaaaaa", Members: []link.CheckedMember{{Owner: "novox", Repo: "mesh-controller", Commit: headController[:12], Trunk: "bbbbbbbbbbbb"}}}})
body, _ := json.Marshal(c)
for _, want := range []string{`"group":"feat/x"`, `"members":[{"owner":"novox","repo":"mesh-host","number":114,"commit":"` + headHost + `","trunk":"aaaaaaaaaaaa"}`,
`{"owner":"novox","repo":"mesh-controller","number":346,"commit":"` + headController + `","trunk":"bbbbbbbbbbbb"}`} {
if !strings.Contains(string(body), want) {
t.Fatalf("the composed verdict lacks %s:\n%s", want, body)
}
}
}
// Members whose manifests order their modules in a cycle fail the composed check, naming every module and edge, and no
// ask is made of the build seat.
func TestAGroupInACycleFailsItsComposedCheckWithoutABuild(t *testing.T) {
members := []orderMember{
{ID: "novox/x@" + headHost[:12], Repository: "novox/x", Base: "main", Head: headHost, Number: 7, Paths: []string{"module.json"}},
{ID: "novox/y@" + headController[:12], Repository: "novox/y", Base: "main", Head: headController, Number: 9, Paths: []string{"module.json"}},
}
reach := map[string]orderReach{members[0].ID: {Moved: []string{"mx"}}, members[1].ID: {Moved: []string{"my"}}}
was, wasParent := manifestAt, parentOf
t.Cleanup(func() { manifestAt, parentOf = was, wasParent })
parentOf = func(context.Context, string, string, string) (string, error) { return "", nil }
manifestAt = func(_ context.Context, _, repo, _, ref string) (catalogue.Manifest, bool, error) {
if ref == "main" {
return catalogue.Manifest{}, false, nil // what the pull requests merge into holds no such manifest
}
if repo == "x" {
return catalogue.Manifest{Module: "mx", DefinesSeats: []catalogue.SeatDeclaration{{Name: "s", Serves: []catalogue.Verb{{Name: "a"}}}},
Invokes: []string{"seat:s.b"}}, true, nil
}
return catalogue.Manifest{Module: "my", DefinesSeats: []catalogue.SeatDeclaration{{Name: "s", Serves: []catalogue.Verb{{Name: "b"}}}},
Invokes: []string{"seat:s.a"}}, true, nil
}
c, cyclic, err := cycleVerdict(context.Background(), "feat/cycle", members, reach, []inventory.Edge(nil), nil)
if err != nil || !cyclic {
t.Fatalf("no cycle found: %v", err)
}
if c.Verdict != "fail" || c.Group != "feat/cycle" || len(c.Members) != 2 || c.Gate == nil || c.Gate.Verdict != "fail" {
t.Fatalf("the verdict is %+v", c)
}
for _, w := range []string{"mx", "my", "s.a", "s.b"} {
if !strings.Contains(c.Summary, w) {
t.Errorf("the cycle does not name %s: %s", w, c.Summary)
}
}
// Members with no edge between them are no cycle.
manifestsReadMu.Lock()
clear(manifestsRead)
manifestsReadMu.Unlock()
manifestAt = func(context.Context, string, string, string, string) (catalogue.Manifest, bool, error) {
return catalogue.Manifest{}, false, nil
}
if _, cyclic, err := cycleVerdict(context.Background(), "feat/cycle", members, reach, nil, nil); err != nil || cyclic {
t.Fatalf("a cycle where there is none: %v %v", cyclic, err)
}
}
@@ -1,65 +0,0 @@
package main
import (
"io"
"os"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
)
// Issue 1638 (novox/hq ADR 0480 clause 7), taken in with mesh-sdk's container `groups`: the controller sends a container's
// groups only to a node-engine that reports the capability container-groups, and holds the module on any other, naming it.
const n8nWithAGroup = `{"module":"n8n","version":"1",
"config":{"media-gid":{"type":"number","minimum":1,"required":true,"description":"the group that owns the media library"}},
"resources":[{"id":"server","type":"container","name":"n8n",
"image":"n8nio/n8n@sha256:4846eb2f4b874ab04cde7fc1e249d2ddaec66e9aea64439beb2972cfea88e3c0",
"groups":["${config:media-gid}"]}]}`
// printedBy runs reportUnhostable and returns what it printed.
func printedBy(t *testing.T, node string, plan catalogue.Resolution) string {
t.Helper()
stdout := os.Stdout
r, w, err := os.Pipe()
if err != nil {
t.Fatal(err)
}
os.Stdout = w
reportUnhostable(node, plan)
w.Close()
os.Stdout = stdout
out, _ := io.ReadAll(r)
return string(out)
}
// 1638/AC1: a node that reports the capability gets the module; one that does not has it held, the capability named.
func TestAModuleWithGroupsIsHeldOnANodeThatDoesNotReportTheCapability(t *testing.T) {
m, err := catalogue.ParseManifest([]byte(n8nWithAGroup))
if err != nil {
t.Fatal(err)
}
shelf := catalogue.Shelf{"n8n": m}
reads := catalogue.Node{Name: "ace", Capabilities: map[string]bool{"container-runtime": true, "container-groups": true}}
old := catalogue.Node{Name: "g14", Capabilities: map[string]bool{"container-runtime": true}}
got, err := catalogue.Resolve(shelf, []string{"n8n"}, reads, catalogue.World{})
if err != nil || len(got.Unhostable) != 0 || len(got.Modules) != 1 {
t.Fatalf("a node that reads groups: %v, held %+v", err, got.Unhostable)
}
if out := printedBy(t, "ace", got); out != "" {
t.Errorf("a node that reads groups was told: %q", out)
}
got, err = catalogue.Resolve(shelf, []string{"n8n"}, old, catalogue.World{})
if err != nil {
t.Fatal(err)
}
if len(got.Modules) != 0 || len(got.Unhostable) != 1 {
t.Fatalf("n8n was sent to a node that does not report container-groups: hosted %d, held %+v", len(got.Modules), got.Unhostable)
}
if out := printedBy(t, "g14", got); !strings.Contains(out, "n8n") || !strings.Contains(out, "container-groups") {
t.Errorf("the hold does not name the module and the capability: %q", out)
}
}
@@ -1,110 +0,0 @@
//go:build integration
package main
import (
"encoding/json"
"fmt"
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdecl "git.novox.be/novox/mesh-sdk/go/declaration"
"git.novox.be/novox/mesh-sdk/go/overlay"
"git.novox.be/novox/mesh-sdk/go/compose"
)
// 1696/AC2, ADR 0486 §5: a node is sent a file by digest only when its profile reports the capability. A node whose profile lacks
// it, and a node the mesh has no profile of, are sent every content inline; a node that reports it is sent the reference. The test
// goes through composeQuestion and composeSend, the two places readsFileReferences is read (compose_sdk.go).
func TestAFileTravelsByDigestOnlyToANodeThatReportsTheCapability(t *testing.T) {
open := aMesh(t) // anchor and laptop report no file-references
ctx := t.Context()
big := strings.Repeat("0123456789abcdef", 8<<10) // 128 KiB
register(t, open, catalogue.Manifest{Module: "bigfile", Version: "1",
Resources: []map[string]any{{"id": "blob", "type": "file", "path": "/etc/bigfile/blob", "mode": "0644", "content": big}}})
// reader reports the capability; bare has never reported a profile at all.
for i, n := range []struct {
name string
profile map[string]any
}{
{"reader", map[string]any{"capabilities": []any{
map[string]any{"name": "container-runtime", "present": true},
map[string]any{"name": "systemd", "present": true},
map[string]any{"name": sdkdecl.FileReferencesCapability, "present": true}}}},
{"bare", nil},
} {
record, err := open.inventory.AddNode(ctx, n.name)
if err != nil {
t.Fatal(err)
}
if err := open.inventory.SetPlace(ctx, n.name, n.name+".example:51820", "here", false, fmt.Sprintf("10.77.0.%d", i+10)); err != nil {
t.Fatal(err)
}
if n.profile != nil {
if err := open.inventory.RecordProfile(ctx, record.ID, n.profile); err != nil {
t.Fatal(err)
}
}
if err := open.inventory.RecordSealingKey(ctx, record.ID, aPublicKey(t)); err != nil {
t.Fatal(err)
}
if err := open.inventory.RecordOverlayKey(ctx, record.ID, aPublicKey(t)); err != nil {
t.Fatal(err)
}
if _, err := open.inventory.Assign(ctx, n.name, overlay.Name); err != nil {
t.Fatal(err)
}
}
for _, node := range []string{"laptop", "reader", "bare"} {
if _, err := assign(ctx, open, node, "bigfile"); err != nil {
t.Fatalf("assign %s: %v", node, err)
}
}
byDigest := map[string]bool{"laptop": false, "bare": false, "reader": true}
for node, want := range byDigest {
asked, _, err := composeQuestion(ctx, open, node, compose.Reading, false)
if err != nil {
t.Fatalf("%s: %v", node, err)
}
sent, err := composeSend(ctx, open, node)
if err != nil {
t.Fatalf("%s: %v", node, err)
}
for what, s := range map[string]sendable{"question": asked, "send": sent} {
if s.fileReferences != want {
t.Fatalf("%s, %s: fileReferences %v, want %v", node, what, s.fileReferences, want)
}
body, err := s.Body()
if err != nil {
t.Fatal(err)
}
var got struct {
Resources []map[string]any `json:"resources"`
}
if err := json.Unmarshal(body, &got); err != nil {
t.Fatal(err)
}
var found bool
for _, r := range got.Resources {
if r["path"] != "/etc/bigfile/blob" {
continue
}
found = true
inline := r["content"] == big
if want && (inline || r["digest"] == nil || r["size"] == nil) {
t.Fatalf("%s, %s: a node that reads references is sent the content inline", node, what)
}
if !want && (!inline || r["digest"] != nil) {
t.Fatalf("%s, %s: a node that does not report the capability is sent a reference", node, what)
}
}
if !found {
t.Fatalf("%s, %s: the file is not in the declaration", node, what)
}
}
}
}
-253
View File
@@ -1,253 +0,0 @@
package main
import (
"context"
"encoding/json"
"errors"
"fmt"
"sort"
"sync"
"git.novox.be/novox/mesh-sdk/go/compose"
sdkdecl "git.novox.be/novox/mesh-sdk/go/declaration"
"github.com/nats-io/nats.go"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/conditions"
)
// **A declaration's large files travel by their digest to a node that can read one** (novox/hq ADR 0486 §1, §5, issue 1696).
//
// The declaration to a node that reports sdkdecl.FileReferencesCapability carries, in place of the content of every
// file resource of more than sdkdecl.FileReferenceThreshold bytes that may travel so (File.TravelsByDigest: plain content or
// bytes, not sealed, not filled with secrets, not written into another file), its sha256 digest and size. The content is put
// into the bucket of the bus under the hex of the digest before the declaration is sent. A node that does not report the
// capability is sent every content inline, as before: an older node-engine refuses a field it does not know, whole.
// readsFileReferences says the node reports the capability of reading a file given by digest (ADR 0486 §5). A node the mesh has
// no profile of reads none.
func readsFileReferences(in compose.Input, node string) bool {
n, ok := nodeOf(in, node)
return ok && n.Capabilities[sdkdecl.FileReferencesCapability]
}
// referencing is the resources as a node that reads references is sent them, and the content of each file that went by
// digest, by the name of its object (the hex of its digest). The resources it is given are not changed. A file that does not
// travel by digest is passed as it is.
func referencing(resources []map[string]any) ([]map[string]any, map[string][]byte, error) {
out := make([]map[string]any, len(resources))
var objects map[string][]byte
for i, r := range resources {
out[i] = r
if kind, _ := r["type"].(string); kind != string(sdkdecl.TypeFile) {
continue
}
// Cheap first: only a file with a content or bytes of the size is worth reading whole.
content, _ := r["content"].(string)
bytesField, _ := r["bytes"].(string)
if len(content) <= sdkdecl.FileReferenceThreshold && len(bytesField) <= sdkdecl.FileReferenceThreshold {
continue
}
raw, err := json.Marshal(r)
if err != nil {
return nil, nil, fmt.Errorf("resource %v: %w", r["id"], err)
}
var f sdkdecl.File
if err := json.Unmarshal(raw, &f); err != nil {
return nil, nil, fmt.Errorf("resource %v is not a file the controller can read: %w", r["id"], err)
}
if !f.TravelsByDigest() {
continue
}
ref, body, err := f.AsReference()
if err != nil {
return nil, nil, err
}
name, err := sdkdecl.ObjectName(ref.Digest)
if err != nil {
return nil, nil, err
}
with := make(map[string]any, len(r)+2)
for k, v := range r {
with[k] = v
}
delete(with, "content")
delete(with, "bytes")
with["digest"], with["size"] = ref.Digest, ref.Size
out[i] = with
if objects == nil {
objects = map[string][]byte{}
}
objects[name] = body
}
return out, objects, nil
}
// namedObjects are the objects of the bucket a body names: the object name of every file resource given by reference,
// with its size. It reads a declaration as the node does, so it also reads the latest one the bus holds.
func namedObjects(body []byte) (map[string]int64, error) {
var envelope struct {
Resources []struct {
Type string `json:"type"`
Digest string `json:"digest"`
Size int64 `json:"size"`
} `json:"resources"`
}
if err := json.Unmarshal(body, &envelope); err != nil {
return nil, err
}
var out map[string]int64
for _, r := range envelope.Resources {
if r.Type != string(sdkdecl.TypeFile) || r.Digest == "" {
continue
}
name, err := sdkdecl.ObjectName(r.Digest)
if err != nil {
return nil, err
}
if out == nil {
out = map[string]int64{}
}
out[name] = r.Size
}
return out, nil
}
// objectStore is what putting the objects needs of a node's bucket.
type objectStore interface {
// Has says whether an object of that name is there, complete, and whether it carries the sweep's mark of being named by no
// node's latest declaration.
Has(ctx context.Context, name string) (have, marked bool, err error)
// Put writes the object under that name, replacing one of the same name (and so erasing its mark).
Put(ctx context.Context, name string, content []byte) error
}
// putObjects makes sure every object of the declaration is in the bucket before the declaration that names it is sent: a node
// that reads the declaration finds each object. An object already there (the same content, the same name) is not written again,
// **unless the sweep marked it unnamed**: then it is written again, so the mark that would have it removed is gone with the
// old object (issue 1696, review: a mark read just before a send must not take an object the send relies on). Refuses on the
// first it cannot put, naming it, so the declaration is not sent.
func putObjects(ctx context.Context, store objectStore, objects map[string][]byte) error {
names := make([]string, 0, len(objects))
for name := range objects {
names = append(names, name)
}
sort.Strings(names)
for _, name := range names {
have, marked, err := store.Has(ctx, name)
if err != nil {
return fmt.Errorf("whether the object %s is in the bucket: %w", name[:12], err)
}
if have && !marked {
continue
}
if err := store.Put(ctx, name, objects[name]); err != nil {
return fmt.Errorf("putting the object %s (%d bytes) in the bucket: %w", name[:12], len(objects[name]), err)
}
}
return nil
}
// busObjects is a node's bucket on the controller's own connection.
type busObjects struct{ store jetstream.ObjectStore }
// openBucket opens the bucket of a node's files, making it first if it is not there (it is asserted for every node at the
// controller's start; a node adopted since is covered here).
func openBucket(ctx context.Context, conn *nats.Conn, node string) (busObjects, error) {
js, err := jetstream.New(conn)
if err != nil {
return busObjects{}, err
}
bucket, err := broker.DeclarationFilesBucketOf(node)
if err != nil {
return busObjects{}, fmt.Errorf("the object bucket of node %q: %w", node, err)
}
store, err := js.ObjectStore(ctx, bucket)
if errors.Is(err, jetstream.ErrBucketNotFound) {
if err = broker.EnsureDeclarationFiles(ctx, js, node); err == nil {
store, err = js.ObjectStore(ctx, bucket)
}
}
if err != nil {
return busObjects{}, fmt.Errorf("the object bucket %s: %w", bucket, err)
}
return busObjects{store}, nil
}
func (b busObjects) Has(ctx context.Context, name string) (bool, bool, error) {
info, err := b.store.GetInfo(ctx, name)
switch {
case errors.Is(err, jetstream.ErrObjectNotFound):
return false, false, nil
case err != nil:
return false, false, err
}
_, marked := info.Metadata[declarationUnnamed]
return !info.Deleted, marked, nil
}
func (b busObjects) Put(ctx context.Context, name string, content []byte) error {
_, err := b.store.PutBytes(ctx, name, content)
return err
}
// filesLocks hold a node's bucket for one act at a time in this process: a send puts its objects and publishes the declaration that
// names them under the node's lock, and the sweep judges the node's bucket under the same one, so the sweep never removes an
// object between a send's put and its declaration. Another process (a person's push) is covered by the put erasing a mark and by
// the sweep looking at an object again just before it removes it.
var filesLocks sync.Map
func lockFilesOf(node string) func() {
l, _ := filesLocks.LoadOrStore(node, &sync.Mutex{})
mu := l.(*sync.Mutex)
mu.Lock()
return mu.Unlock
}
// kindFilesNotPut is a node whose files that travel by digest could not be put in its bucket, send after send.
const kindFilesNotPut = "declaration-files-not-put"
// filesNotPutAfter is how many sends in a row may fail to put before the condition is raised: the first send to the machine holding
// the bus fails for want of the grant that very send delivers, and says so, and is not yet a fault.
const filesNotPutAfter = 2
var (
filesFailMu sync.Mutex
filesFailed = map[string]int{}
)
func filesNotPutObservation(node, summary string) conditions.Observation {
return conditions.Observation{Scope: conditions.ScopeMachine, ID: node, Kind: kindFilesNotPut, Token: kindFilesNotPut,
Machine: node, Severity: conditions.Warning, Source: "sends", Summary: summary}
}
// notePut records how putting a send's objects went: a failure counts, the second in a row raises the condition, a success
// clears the count and the condition.
func notePut(ctx context.Context, node string, put error) {
filesFailMu.Lock()
if put == nil {
delete(filesFailed, node)
} else {
filesFailed[node]++
}
n := filesFailed[node]
filesFailMu.Unlock()
keeper := conditionsFrom
if keeper == nil {
return
}
key := filesNotPutObservation(node, "").Key()
switch {
case put == nil:
if _, err := keeper.Clear(ctx, key, node+" was sent its files by digest"); err != nil {
fmt.Printf("the condition %s could not be cleared: %v\n", key, err)
}
case n >= filesNotPutAfter:
summary := fmt.Sprintf("%s's files that travel by digest could not be put in its bucket %d sends in a row, so it is sent them "+
"inline: %v", node, n, put)
if _, err := keeper.Observe(ctx, filesNotPutObservation(node, summary)); err != nil {
fmt.Printf("the condition %s could not be raised: %v\n", key, err)
}
}
}
@@ -1,246 +0,0 @@
//go:build integration
package main
import (
"bytes"
"context"
"encoding/json"
"strings"
"testing"
"time"
sdkdecl "git.novox.be/novox/mesh-sdk/go/declaration"
"github.com/nats-io/nats.go"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/link"
"github.com/novox/mesh-controller/internal/testbus"
)
type sigOf struct{}
func (sigOf) Sign(context.Context, []byte) ([]byte, error) { return make([]byte, 64), nil }
// 1696/AC7 (ADR 0486 §1): on a real bus with its default largest message of 1 MiB, a declaration whose files add up to more than
// 3 MiB is sent as references, its objects are in the bucket, each fetched back whole and verified against the digest and size the
// declaration names; the same declaration inline is refused by the controller, naming size and limit. A changed byte in an object
// fails the verification.
func TestADeclarationOfMoreThanThreeMiBIsSentByReferenceOnARealBus(t *testing.T) { // 1696/AC7
s := testbus.Start(t)
conn, err := nats.Connect(s.ClientURL())
if err != nil {
t.Fatal(err)
}
defer conn.Close()
nc, err := conn.JetStream()
if err != nil {
t.Fatal(err)
}
if _, err := nc.AddStream(&nats.StreamConfig{Name: "NODES", Subjects: []string{"mesh.node.*.declare"},
MaxMsgsPerSubject: 1}); err != nil {
t.Fatal(err)
}
ctx, cancel := context.WithTimeout(context.Background(), 60*time.Second)
defer cancel()
api, err := jetstream.New(conn)
if err != nil {
t.Fatal(err)
}
if err := broker.EnsureDeclarationFiles(ctx, api, "novox"); err != nil {
t.Fatal(err)
}
// Run twice: an assertion of an existing bucket is no error and no change.
if err := broker.EnsureDeclarationFiles(ctx, api, "novox"); err != nil {
t.Fatalf("asserting the bucket again: %v", err)
}
var resources []map[string]any
for i := 0; i < 3; i++ {
resources = append(resources, aFileResource("big-"+string(rune('a'+i)), contentOf("big"+string(rune('a'+i)), 1200<<10)))
}
resources = append(resources, aFileResource("small", "tiny\n"))
inline := sendable{Resources: resources}
byDigest := sendable{fileReferences: true, Resources: resources}
bus := link.OverNATS{Conn: conn, JS: nc}
body, err := inline.Body()
if err != nil {
t.Fatal(err)
}
if err := link.Declare(ctx, bus, sigOf{}, "novox", body, 5*time.Second); err == nil ||
!strings.Contains(err.Error(), "1048576") || strings.Contains(err.Error(), "maximum payload exceeded") {
t.Fatalf("the inline declaration of more than 3 MiB: %v", err)
}
bucket, err := openBucket(ctx, conn, "novox")
if err != nil {
t.Fatal(err)
}
objects, err := byDigest.objects()
if err != nil || len(objects) != 3 {
t.Fatalf("objects: %d %v", len(objects), err)
}
if err := putObjects(ctx, bucket, objects); err != nil {
t.Fatal(err)
}
if err := putObjects(ctx, bucket, objects); err != nil {
t.Fatalf("putting them again: %v", err)
}
refBody, err := byDigest.Body()
if err != nil {
t.Fatal(err)
}
if err := link.Declare(ctx, bus, sigOf{}, "novox", refBody, 5*time.Second); err != nil {
t.Fatalf("the declaration by reference: %v", err)
}
// What the stream holds for the node is the reference form, and it names exactly the three objects.
latest, err := latestDeclarations(ctx, conn, []string{"novox", "ace"})
if err != nil || len(latest) != 1 {
t.Fatalf("the latest declarations: %v %v", len(latest), err)
}
named, err := namedByLatest("novox", latest["novox"])
if err != nil || len(named) != 3 {
t.Fatalf("what the latest declaration names: %v %v", named, err)
}
novoxBucket, err := broker.DeclarationFilesBucketOf("novox")
if err != nil {
t.Fatal(err)
}
store, err := api.ObjectStore(ctx, novoxBucket)
if err != nil {
t.Fatal(err)
}
var envelope struct {
Resources []sdkdecl.File `json:"resources"`
}
if err := json.Unmarshal(refBody, &envelope); err != nil {
t.Fatal(err)
}
for _, f := range envelope.Resources {
if !f.IsReference() {
continue
}
name, _ := sdkdecl.ObjectName(f.Digest)
got, err := store.GetBytes(ctx, name)
if err != nil {
t.Fatalf("the object of %s: %v", f.ID, err)
}
if err := f.Verify(got); err != nil {
t.Errorf("the object does not verify: %v", err)
}
got[len(got)/2] ^= 1
if err := f.Verify(got); err == nil {
t.Errorf("a changed byte in the object of %s passed the verification", f.ID)
}
}
// The sweep against the real bucket: all named, nothing marked; a fourth object named by none is marked, kept a day less a
// minute, and removed at the day.
extra := bytes.Repeat([]byte("x"), 100)
extraName := strings.Repeat("e", 64)
if err := putObjects(ctx, bucket, map[string][]byte{extraName: extra}); err != nil {
t.Fatal(err)
}
b := busSweptBucket{store}
t0 := time.Now()
if r, err := sweepDeclarationFiles(ctx, b, named, t0); err != nil || r.Marked != 1 || r.Removed != 0 || r.Kept != 3 {
t.Fatalf("first sweep: %+v %v", r, err)
}
if r, err := sweepDeclarationFiles(ctx, b, named, t0.Add(24*time.Hour-time.Minute)); err != nil || r.Removed != 0 {
t.Fatalf("a minute short of a day: %+v %v", r, err)
}
if r, err := sweepDeclarationFiles(ctx, b, named, t0.Add(24*time.Hour+time.Second)); err != nil || r.Removed != 1 {
t.Fatalf("a day after: %+v %v", r, err)
}
if infos, err := store.List(ctx); err != nil || len(infos) != 3 {
t.Fatalf("the bucket holds %d objects (%v), want the 3 the declaration names", len(infos), err)
}
}
// 1696/AC8 on a real bucket: the sweep removes an object only if it still carries the mark it judged it by. An object marked
// unnamed, then put again by a send (which erases the mark), is kept when the sweep deletes it by the old mark.
func TestTheSweepsDeleteLeavesAnObjectASendPutAgainAfterItWasMarked(t *testing.T) { // 1696/AC8
s := testbus.Start(t)
conn, err := nats.Connect(s.ClientURL())
if err != nil {
t.Fatal(err)
}
defer conn.Close()
ctx, cancel := context.WithTimeout(context.Background(), 60*time.Second)
defer cancel()
bucket, err := openBucket(ctx, conn, "novox")
if err != nil {
t.Fatal(err)
}
name := strings.Repeat("a", 64)
content := map[string][]byte{name: bytes.Repeat([]byte("y"), 100)}
if err := putObjects(ctx, bucket, content); err != nil {
t.Fatal(err)
}
b := busSweptBucket{bucket.store}
judged := time.Now().Add(-25 * time.Hour).Truncate(time.Second)
if err := b.MarkUnnamed(ctx, name, judged); err != nil {
t.Fatal(err)
}
if _, marked, err := bucket.Has(ctx, name); err != nil || !marked {
t.Fatalf("the object was not marked: %v %v", marked, err)
}
// The sweep has read the mark; a send now puts the object again, which erases it.
if err := putObjects(ctx, bucket, content); err != nil {
t.Fatal(err)
}
if have, marked, err := bucket.Has(ctx, name); err != nil || !have || marked {
t.Fatalf("a send did not put the marked object again: have %v marked %v %v", have, marked, err)
}
gone, err := b.Delete(ctx, name, judged)
if err != nil || gone {
t.Fatalf("the sweep deleted by an old mark: gone %v, %v", gone, err)
}
if have, _, err := bucket.Has(ctx, name); err != nil || !have {
t.Fatalf("the object a send relies on is gone: %v %v", have, err)
}
// And the mark still on it is honoured: marked again and deleted by that mark, it goes.
if err := b.MarkUnnamed(ctx, name, judged); err != nil {
t.Fatal(err)
}
if gone, err := b.Delete(ctx, name, judged); err != nil || !gone {
t.Fatalf("an object still carrying the mark it was judged by is kept: gone %v, %v", gone, err)
}
}
// 1696/AC9 on a real bus: the sweep of a node's bucket takes the node's lock first, so it waits for a send that has put its objects
// and not yet published the declaration that names them.
func TestTheSweepOfANodesBucketHoldsTheNodesLock(t *testing.T) { // 1696/AC9
s := testbus.Start(t)
conn, err := nats.Connect(s.ClientURL())
if err != nil {
t.Fatal(err)
}
defer conn.Close()
ctx, cancel := context.WithTimeout(context.Background(), 60*time.Second)
defer cancel()
const node = "sweep-lock-node"
if _, err := openBucket(ctx, conn, node); err != nil { // the bucket exists, or the sweep would return before the lock
t.Fatal(err)
}
release := lockFilesOf(node) // a send is between its put and its publication
done := make(chan struct{})
go func() {
_, _ = sweepNode(ctx, conn, node)
close(done)
}()
select {
case <-done:
t.Fatal("the sweep judged the node's bucket while a send held it")
case <-time.After(300 * time.Millisecond):
}
release()
select {
case <-done:
case <-time.After(20 * time.Second):
t.Fatal("the sweep did not go on once the send let go")
}
}
@@ -1,534 +0,0 @@
package main
import (
"context"
"encoding/base64"
"encoding/hex"
"encoding/json"
"errors"
"fmt"
"strings"
"testing"
"time"
sdkdecl "git.novox.be/novox/mesh-sdk/go/declaration"
"github.com/novox/mesh-controller/internal/conditions"
"github.com/novox/mesh-controller/internal/link"
)
// contentOf is a deterministic content of n bytes that does not compress to nothing and holds no character JSON escapes.
func contentOf(seed string, n int) string {
var b strings.Builder
for i := 0; b.Len() < n; i++ {
fmt.Fprintf(&b, "%s-%07d-%x;", seed, i, i*2654435761%1000003)
}
return b.String()[:n]
}
func aFileResource(id, content string) map[string]any {
return map[string]any{"id": id, "type": "file", "path": "/etc/" + id, "mode": "0644", "content": content}
}
// controlNodeShape is a declaration shaped like the control node's on 2026-10-21 (issue 1689), the sizes measured then: a bus users
// file of 252,678 bytes, the vault's export of 138,475 bytes (compact), the tool runtime 21 KB, the packet filter 19 KB, the launcher
// 14 KB, and about 600 small resources for the rest (about 286 KB). `more` scales the count of small ones: 1 is the shape, 1.4 forty
// percent more of them.
func controlNodeShape(more float64) []map[string]any {
res := []map[string]any{
aFileResource("nats.bus-users", contentOf("users", 252678)),
aFileResource("mesh-vault.kept", contentOf("kept", 138475)),
aFileResource("node-tools.runtime", contentOf("runtime", 21<<10)),
aFileResource("nftables.filtering", contentOf("filter", 19<<10)),
aFileResource("mesh-host.launcher", contentOf("launcher", 14<<10)),
}
for i := 0; i < int(600*more); i++ {
res = append(res, aFileResource(fmt.Sprintf("module-%03d.config", i), contentOf(fmt.Sprintf("small%d", i), 380)))
}
return res
}
func wireSizeOf(t *testing.T, s sendable) int {
t.Helper()
body, err := s.Body()
if err != nil {
t.Fatal(err)
}
// The signature the controller's key makes is 64 bytes.
msg, err := link.Envelope(body, make([]byte, 64))
if err != nil {
t.Fatal(err)
}
return len(msg)
}
// 1696/AC1: for a node that reports the capability, the two big files of a control-node-shaped declaration carry their digest and size
// and no content, and the declaration is far under the bus's 1 MiB.
func TestAFileOver64KiBTravelsByDigestToANodeThatReadsOne(t *testing.T) { // 1696/AC1
users, kept := contentOf("users", 252678), contentOf("kept", 138475)
s := sendable{fileReferences: true, Resources: []map[string]any{
aFileResource("nats.bus-users", users), aFileResource("mesh-vault.kept", kept),
aFileResource("small", "tiny\n"),
}}
body, err := s.Body()
if err != nil {
t.Fatal(err)
}
var got struct {
Resources []map[string]any `json:"resources"`
}
if err := json.Unmarshal(body, &got); err != nil {
t.Fatal(err)
}
for i, want := range []string{users, kept} {
r := got.Resources[i]
digest := sdkdecl.DigestOf([]byte(want))
if _, has := r["content"]; has || r["digest"] != digest || r["size"] != float64(len(want)) {
t.Errorf("%v goes with content %v, digest %v, size %v; want no content, %s and %d", r["id"], r["content"] != nil,
r["digest"], r["size"], digest, len(want))
}
if r["path"] == nil || r["mode"] != "0644" {
t.Errorf("%v lost its path or mode: %v", r["id"], r)
}
}
if got.Resources[2]["content"] != "tiny\n" || got.Resources[2]["digest"] != nil {
t.Errorf("a small file is not inline: %v", got.Resources[2])
}
if strings.Contains(string(body), users[:200]) || strings.Contains(string(body), kept[:200]) {
t.Error("the content of a file that went by digest is in the body")
}
if n := wireSizeOf(t, s); n > 100<<10 {
t.Errorf("the declaration is %d bytes on the wire, want under 100 KB", n)
}
// The objects to put are the contents, each under the hex of its digest, and a reference verifies against its object.
objects, err := s.objects()
if err != nil || len(objects) != 2 {
t.Fatalf("objects: %v %v", objects, err)
}
for _, want := range []string{users, kept} {
name, _ := sdkdecl.ObjectName(sdkdecl.DigestOf([]byte(want)))
if string(objects[name]) != want {
t.Errorf("the object %s does not hold the content", name[:12])
}
}
named, err := namedObjects(body)
if err != nil || len(named) != 2 {
t.Fatalf("the objects the body names: %v %v", named, err)
}
for name := range objects {
if _, ok := named[name]; !ok {
t.Errorf("the body does not name the object %s it was sent with", name[:12])
}
if _, err := hex.DecodeString(name); err != nil || len(name) != 64 {
t.Errorf("the object name %q is not 64 hex characters", name)
}
}
}
// 1696/AC2: a node that does not report the capability is sent every content inline, as before, and has no object put.
func TestANodeWithoutTheCapabilityIsSentEveryContentInline(t *testing.T) { // 1696/AC2
users := contentOf("users", 252678)
s := sendable{Resources: []map[string]any{aFileResource("nats.bus-users", users)}}
body, err := s.Body()
if err != nil {
t.Fatal(err)
}
var got struct {
Resources []map[string]any `json:"resources"`
}
if err := json.Unmarshal(body, &got); err != nil {
t.Fatal(err)
}
if got.Resources[0]["content"] != users || got.Resources[0]["digest"] != nil || got.Resources[0]["size"] != nil {
t.Fatalf("a node that reads no references is not sent the content inline: %v", got.Resources[0]["digest"])
}
if objects, err := s.objects(); err != nil || len(objects) != 0 {
t.Fatalf("objects for a node that reads none: %v %v", len(objects), err)
}
}
// 1696/AC3: only a file that may travel by digest does: one of exactly 64 KiB, a sealed one, one filled with secrets, one written
// into another file, and a directory stay inline, whatever the node reads.
func TestOnlyAFileThatMayTravelByDigestDoes(t *testing.T) { // 1696/AC3
exact := contentOf("exact", 64<<10)
big := contentOf("big", 64<<10+1)
sealed := aFileResource("sealed", big)
sealed["sealed"], sealed["content"] = "box", ""
sealed["content"] = big // a sealed file's content is none; the body must pass it as it is, whatever it holds
secret := aFileResource("secret", big)
secret["secrets"] = map[string]any{"token": "x"}
into := aFileResource("into", big)
into["into"] = "/etc/other"
s := sendable{fileReferences: true, Resources: []map[string]any{
aFileResource("exact", exact), sealed, secret, into,
{"id": "dir", "type": "directory", "path": "/etc/d", "content": big},
aFileResource("big", big),
}}
body, err := s.Body()
if err != nil {
t.Fatal(err)
}
var got struct {
Resources []map[string]any `json:"resources"`
}
if err := json.Unmarshal(body, &got); err != nil {
t.Fatal(err)
}
for i, id := range []string{"exact", "sealed", "secret", "into", "dir"} {
if got.Resources[i]["id"] != id || got.Resources[i]["digest"] != nil || got.Resources[i]["content"] == nil {
t.Errorf("%s went by digest or lost its content: %v", id, got.Resources[i]["digest"])
}
}
if big := got.Resources[5]; big["id"] != "big" || big["digest"] == nil || big["content"] != nil {
t.Errorf("a plain file of 64 KiB and a byte did not go by digest: %v", big["digest"])
}
}
// A file given as bytes goes by digest on its decoded length, with the decoded bytes in the object.
func TestAFileOfBytesTravelsByDigestOnItsDecodedLength(t *testing.T) {
raw := []byte(contentOf("png", 70<<10))
r := map[string]any{"id": "icon", "type": "file", "path": "/usr/share/icon.png", "bytes": stdBase64(raw)}
s := sendable{fileReferences: true, Resources: []map[string]any{r}}
objects, err := s.objects()
if err != nil || len(objects) != 1 {
t.Fatalf("%v %v", len(objects), err)
}
name, _ := sdkdecl.ObjectName(sdkdecl.DigestOf(raw))
if string(objects[name]) != string(raw) {
t.Fatal("the object is not the decoded bytes")
}
if r["bytes"] == nil {
t.Fatal("the composed resource was changed in place")
}
}
type memObjects struct {
have map[string][]byte
marked map[string]bool
puts []string
fail map[string]error
}
func (m *memObjects) Has(_ context.Context, name string) (bool, bool, error) {
_, ok := m.have[name]
return ok, m.marked[name], nil
}
func (m *memObjects) Put(_ context.Context, name string, c []byte) error {
if err := m.fail[name]; err != nil {
return err
}
m.puts = append(m.puts, name)
m.have[name] = c
delete(m.marked, name)
return nil
}
// 1696/AC4: an object already in the bucket is not put again, one absent is, and a put that fails stops the send, naming the object.
func TestObjectsAreInTheBucketBeforeTheDeclarationIsSent(t *testing.T) { // 1696/AC4
objects := map[string][]byte{strings.Repeat("a", 64): []byte("one"), strings.Repeat("b", 64): []byte("two")}
b := &memObjects{have: map[string][]byte{strings.Repeat("a", 64): []byte("one")}}
if err := putObjects(t.Context(), b, objects); err != nil || len(b.puts) != 1 || b.puts[0] != strings.Repeat("b", 64) {
t.Fatalf("put %v (%v), want the one absent", b.puts, err)
}
// An object the sweep marked unnamed is put again, which erases the mark: a send relies on it.
b = &memObjects{have: map[string][]byte{strings.Repeat("a", 64): []byte("one"), strings.Repeat("b", 64): []byte("two")},
marked: map[string]bool{strings.Repeat("a", 64): true}}
if err := putObjects(t.Context(), b, objects); err != nil || len(b.puts) != 1 || b.puts[0] != strings.Repeat("a", 64) || b.marked[strings.Repeat("a", 64)] {
t.Fatalf("a marked object: put %v (%v) marked %v, want it put again and the mark gone", b.puts, err, b.marked)
}
b = &memObjects{have: map[string][]byte{}, fail: map[string]error{strings.Repeat("a", 64): errors.New("permissions violation")}}
err := putObjects(t.Context(), b, objects)
if err == nil || !strings.Contains(err.Error(), "aaaaaaaaaaaa") || !strings.Contains(err.Error(), "permissions violation") {
t.Fatalf("a failed put: %v", err)
}
}
// fallbackDelivery refuses the first declaration it is sent with the bucket's failure, as a bus that has not yet been given the
// controller's grant does, and takes the second.
type fallbackDelivery struct {
bodies [][]byte
first error
}
func (f *fallbackDelivery) grant(context.Context, []readyNode) error { return nil }
func (f *fallbackDelivery) declare(_ context.Context, _ readyNode, body []byte) (string, error) {
f.bodies = append(f.bodies, body)
if len(f.bodies) == 1 && f.first != nil {
return "", fmt.Errorf("x: %w: %w", errFilesNotPut, f.first)
}
return "digest", nil
}
// 1696/AC5: when the bucket cannot be written yet (the grant travels in the declaration of the machine holding the bus), that send goes
// with every content inline and is made; a send that fails for another reason is not retried inline.
func TestASendWhoseFilesCannotBePutGoesInlineOnce(t *testing.T) { // 1696/AC5
users := contentOf("users", 100<<10)
s := readyNode{node: "novox", declared: sendable{fileReferences: true, Resources: []map[string]any{aFileResource("nats.bus-users", users)}}}
f := &fallbackDelivery{first: errors.New("timeout")}
digests, err := deliver(t.Context(), f, "novox", []readyNode{s})
if err != nil || digests["novox"] != "digest" || len(f.bodies) != 2 {
t.Fatalf("%v %v %d", digests, err, len(f.bodies))
}
if strings.Contains(string(f.bodies[0]), users[:100]) || !strings.Contains(string(f.bodies[1]), users[:100]) {
t.Error("the first body should name the file by digest and the second carry it inline")
}
other := &fallbackDelivery{}
other.first = nil
failing := &failingDelivery{err: errors.New("the bus is down")}
if _, err := deliver(t.Context(), failing, "novox", []readyNode{s}); err == nil || failing.calls != 1 {
t.Fatalf("a failure of another kind: %v after %d send(s)", err, failing.calls)
}
}
type failingDelivery struct {
err error
calls int
}
func (f *failingDelivery) grant(context.Context, []readyNode) error { return nil }
func (f *failingDelivery) declare(context.Context, readyNode, []byte) (string, error) {
f.calls++
return "", f.err
}
// A bucket for the sweep.
type sweepBucketFake struct {
objects map[string]time.Time // zero: named at the last look
deleted []string
}
func (b *sweepBucketFake) Objects(context.Context) ([]bucketObject, error) {
var out []bucketObject
for n, at := range b.objects {
out = append(out, bucketObject{Name: n, UnnamedSince: at})
}
return out, nil
}
func (b *sweepBucketFake) MarkUnnamed(_ context.Context, n string, at time.Time) error {
b.objects[n] = at
return nil
}
func (b *sweepBucketFake) ClearUnnamed(_ context.Context, n string) error {
b.objects[n] = time.Time{}
return nil
}
func (b *sweepBucketFake) Delete(_ context.Context, n string, markedAt time.Time) (bool, error) {
if at, ok := b.objects[n]; !ok || !at.Equal(markedAt) {
return false, nil // put again since it was judged: its mark is gone
}
delete(b.objects, n)
b.deleted = append(b.deleted, n)
return true, nil
}
// 1696/AC6: an object a latest declaration still names is not removed, however long; one unnamed is marked at the first look and removed
// only when the mark is 24 hours old; one named again loses its mark.
func TestAnObjectIsRemovedADayAfterNoLatestDeclarationNamesIt(t *testing.T) { // 1696/AC6
named, unnamed, back := strings.Repeat("1", 64), strings.Repeat("2", 64), strings.Repeat("3", 64)
t0 := time.Date(2026, 10, 21, 12, 0, 0, 0, time.UTC)
b := &sweepBucketFake{objects: map[string]time.Time{named: t0.Add(-100 * time.Hour), unnamed: {}, back: t0.Add(-5 * time.Hour)}}
named3 := map[string]bool{named: true, back: true}
r, err := sweepDeclarationFiles(t.Context(), b, named3, t0)
if err != nil || r.Marked != 1 || r.Cleared != 2 || r.Removed != 0 || len(b.deleted) != 0 {
t.Fatalf("first look: %+v %v %v", r, err, b.deleted)
}
if !b.objects[named].IsZero() || !b.objects[back].IsZero() || !b.objects[unnamed].Equal(t0) {
t.Fatalf("marks after the first look: %v", b.objects)
}
// Still unnamed 23 hours 59 minutes later: kept. A day after the mark: removed. The named ones stay for ever.
if r, err = sweepDeclarationFiles(t.Context(), b, map[string]bool{named: true, back: true}, t0.Add(24*time.Hour-time.Minute)); err != nil ||
r.Removed != 0 || len(b.deleted) != 0 {
t.Fatalf("a minute short of a day: %+v %v %v", r, err, b.deleted)
}
if r, err = sweepDeclarationFiles(t.Context(), b, map[string]bool{named: true, back: true}, t0.Add(24*time.Hour)); err != nil ||
r.Removed != 1 || len(b.deleted) != 1 || b.deleted[0] != unnamed {
t.Fatalf("a day after: %+v %v %v", r, err, b.deleted)
}
if _, ok := b.objects[named]; !ok {
t.Fatal("an object a latest declaration names was removed")
}
// An object named again before the day is up loses its mark and is not removed.
b = &sweepBucketFake{objects: map[string]time.Time{unnamed: t0}}
if _, err := sweepDeclarationFiles(t.Context(), b, map[string]bool{unnamed: true}, t0.Add(23*time.Hour)); err != nil || !b.objects[unnamed].IsZero() {
t.Fatalf("named again: %v %v", b.objects, err)
}
}
// What a node's latest declaration names is read from the signed envelope as stored on the bus; an unreadable one is an error and
// never "names none", which would mark an object a node needs.
func TestTheObjectsALatestDeclarationNamesAreReadFromTheBus(t *testing.T) {
s := sendable{fileReferences: true, Resources: []map[string]any{aFileResource("a", contentOf("a", 70<<10))}}
body, _ := s.Body()
msg, _ := link.Envelope(body, []byte("sig"))
named, err := namedByLatest("novox", msg)
name, _ := sdkdecl.ObjectName(sdkdecl.DigestOf([]byte(contentOf("a", 70<<10))))
if err != nil || len(named) != 1 || !named[name] {
t.Fatalf("%v %v", named, err)
}
if none, err := namedByLatest("ace", []byte(`{"declaration":"e30=","signature":""}`)); err != nil || len(none) != 0 {
t.Fatalf("a declaration naming none: %v %v", none, err)
}
if _, err := namedByLatest("novox", []byte("not json")); err == nil {
t.Fatal("an unreadable latest declaration was read as naming none")
}
}
// 1698/AC2 (ADR 0486 §9): a declaration shaped like the control node's, for a node that reads references, stays under half of the
// bus's largest message on the wire; the inline form of the same does not fit at all; and the same shape with 40 percent more resources
// fails the test's own limit, so a change that eats the room fails in a pull request and not on the live mesh.
func TestTheControlNodesDeclarationInTheReferenceFormStaysUnderHalfTheBusLimit(t *testing.T) { // 1698/AC2
const limit = 1048576
shape := sendable{fileReferences: true, Resources: controlNodeShape(1)}
n := wireSizeOf(t, shape)
t.Logf("the control node's shape in the reference form: %d bytes on the wire, %.1f%% of %d", n, 100*float64(n)/limit, limit)
if n > limit/2 {
t.Errorf("the reference form is %d bytes on the wire, over half (%d) of the bus's largest message", n, limit/2)
}
if inline := wireSizeOf(t, sendable{Resources: controlNodeShape(1)}); inline < limit*3/4 {
t.Errorf("the inline form is %d bytes, which is not the shape that was over the limit", inline)
}
grown := wireSizeOf(t, sendable{fileReferences: true, Resources: controlNodeShape(1.4)})
if grown <= limit/2 {
t.Errorf("the shape with 40%% more resources is %d bytes, which the test would let pass", grown)
}
}
func stdBase64(b []byte) string { return base64.StdEncoding.EncodeToString(b) }
// A sweep that read an object's mark before a send put it again must not remove it: the object, now without the mark, is left.
func TestASweepDoesNotRemoveAnObjectASendPutAgainAfterItWasJudged(t *testing.T) { // 1696/AC8
n := strings.Repeat("4", 64)
t0 := time.Date(2026, 10, 21, 12, 0, 0, 0, time.UTC)
b := &sweepBucketFake{objects: map[string]time.Time{n: t0}}
judged := bucketObject{Name: n, UnnamedSince: t0}
b.objects[n] = time.Time{} // the send put it again: no mark
if gone, err := b.Delete(t.Context(), judged.Name, judged.UnnamedSince); err != nil || gone || len(b.deleted) != 0 {
t.Fatalf("an object put again was removed: %v %v %v", gone, err, b.deleted)
}
}
// One node's bucket is held by one act at a time in this process: the sweep waits for a send that has put its objects and not yet
// published the declaration that names them (1696/AC9).
func TestTheSweepOfANodesBucketWaitsForASendThatHasPutItsObjects(t *testing.T) { // 1696/AC9
unlock := lockFilesOf("lock-test-node")
got := make(chan struct{})
go func() {
defer lockFilesOf("lock-test-node")()
close(got)
}()
select {
case <-got:
t.Fatal("the sweep took the node's bucket while a send held it")
case <-time.After(50 * time.Millisecond):
}
other := lockFilesOf("lock-test-other-node") // another node's bucket is not held
other()
unlock()
select {
case <-got:
case <-time.After(2 * time.Second):
t.Fatal("the sweep did not take the node's bucket once the send was done")
}
}
// A put that fails send after send raises a condition; the first failure does not (the grant it waits for travels in that very
// declaration), and a put that works clears it (1696/AC10).
func TestAPutThatFailsSendAfterSendRaisesAConditionAndClearsWhenItWorks(t *testing.T) { // 1696/AC10
store := conditions.NewInMemory()
k := conditions.NewKeeper(t.Context(), conditions.Options{Store: store, History: store,
Teller: &conditions.Told{}, Now: func() time.Time { return time.Date(2026, 10, 21, 14, 0, 0, 0, time.UTC) }})
t.Cleanup(func() { k.Close(t.Context()) })
was := conditionsFrom
conditionsFrom = k
t.Cleanup(func() { conditionsFrom = was })
key := filesNotPutObservation("cond-node", "").Key()
open := func() bool {
_, found, err := conditions.ReadOne(t.Context(), store, key)
if err != nil {
t.Fatal(err)
}
return found
}
boom := errors.New("permissions violation")
notePut(t.Context(), "cond-node", boom)
if open() {
t.Fatal("the first failure raised the condition")
}
notePut(t.Context(), "cond-node", boom)
if !open() {
t.Fatal("the second failure in a row did not raise the condition")
}
notePut(t.Context(), "cond-node", nil)
if open() {
t.Fatal("a put that worked did not clear the condition")
}
notePut(t.Context(), "cond-node", boom)
if open() {
t.Fatal("the count did not start again after a put that worked")
}
}
// aBusThatSays is a bus that keeps the declarations published to it and, at each, asks whether the node's bucket is held.
type aBusThatSays struct {
link.Bus
published chan string
heldAtPublish func() bool
}
func (b *aBusThatSays) PublishDeclaration(_ context.Context, node string, _ []byte) error {
if b.heldAtPublish != nil && !b.heldAtPublish() {
return errors.New("published while the node's bucket was not held")
}
b.published <- node
return errors.New("stop here: nothing is recorded")
}
type aSignerOfZeros struct{}
func (aSignerOfZeros) Sign(context.Context, []byte) ([]byte, error) { return make([]byte, 64), nil }
// whileHeld says whether the node's lock is held by someone right now, without keeping it.
func whileHeld(node string) bool {
got := make(chan struct{})
go func() {
defer lockFilesOf(node)()
close(got)
}()
select {
case <-got:
return false
case <-time.After(100 * time.Millisecond):
// held: the goroutine stays waiting and takes and releases the lock once it is let go.
return true
}
}
// 1696/AC9: a send holds the node's lock from the put to the publication, so the sweep never removes an object between them. The
// test goes through overTheBus.declare: while another holds the lock the declaration is not published, and at the publication the
// lock is held.
func TestASendHoldsTheNodesLockFromThePutToThePublication(t *testing.T) { // 1696/AC9
const node = "lock-send-node"
bus := &aBusThatSays{published: make(chan string, 1)}
bus.heldAtPublish = func() bool { return whileHeld(node) }
send := overTheBus{server: link.SendingOn(bus), signer: aSignerOfZeros{}}
release := lockFilesOf(node) // a sweep is judging this node's bucket
done := make(chan error, 1)
go func() {
_, err := send.declare(t.Context(), readyNode{node: node}, []byte(`{}`))
done <- err
}()
select {
case n := <-bus.published:
t.Fatalf("%s was sent its declaration while the sweep held the node's bucket", n)
case <-time.After(150 * time.Millisecond):
}
release()
select {
case <-bus.published:
case <-time.After(5 * time.Second):
t.Fatal("the send did not go on once the sweep let go")
}
if err := <-done; err == nil || !strings.Contains(err.Error(), "nothing is recorded") {
t.Fatalf("the send did not reach the publication (a lock held at it): %v", err)
}
}
-96
View File
@@ -1,96 +0,0 @@
package main
import (
"bytes"
"context"
"encoding/json"
"fmt"
"strings"
"github.com/novox/mesh-controller/internal/conditions"
"github.com/novox/mesh-controller/internal/link"
)
// **A declaration is one bus message, and the bus's largest message is a cliff** (novox/hq issue 1689). On 2026-10-21
// the control node's declaration passed it by 2 KB and nothing reached that node until two assignments were undone.
// So the size is measured at every send: a warning condition from 75% of the limit (ADR 0486 §8, issue 1698: it was 90% while every file travelled inline), urgent over it, cleared when
// the declaration is back under 75%. The limit is the connection's own, not a constant.
// kindDeclarationNearLimit is a declaration at or near the bus's largest message.
const kindDeclarationNearLimit = "declaration-near-limit"
// declarationRoomSource is the condition's source: the sends.
const declarationRoomSource = "sends"
func declarationRoomObservation(node string, size, limit int64) conditions.Observation {
o := conditions.Observation{Scope: conditions.ScopeMachine, ID: node, Kind: kindDeclarationNearLimit,
Token: kindDeclarationNearLimit, Machine: node, Severity: conditions.Warning, Source: declarationRoomSource}
if size > limit {
o.Severity = conditions.Urgent
}
o.Summary = fmt.Sprintf("%s's declaration is %d bytes on the wire, %d%% of the bus's largest message of %d bytes",
node, size, size*100/limit, limit)
if size > limit {
o.Summary += ": over it, so it cannot be sent"
}
return o
}
func init() {
link.DeclarationMeasured = func(ctx context.Context, node string, size, limit int64) {
settleDeclarationRoom(ctx, conditionsFrom, node, size, limit)
}
}
// settleDeclarationRoom raises or clears the condition of a declaration's size. A nil keeper (a tool, not the
// serving controller) says nothing.
func settleDeclarationRoom(ctx context.Context, keeper *conditions.Keeper, node string, size, limit int64) {
if keeper == nil || limit <= 0 {
return
}
obs := declarationRoomObservation(node, size, limit)
if float64(size) < link.NearTheLimit*float64(limit) {
if _, err := keeper.Clear(ctx, obs.Key(), fmt.Sprintf("%s's declaration is %d bytes, under the limit's margin", node, size)); err != nil {
fmt.Printf("the condition %s could not be cleared: %v\n", obs.Key(), err)
}
return
}
fmt.Printf("NEAR THE LIMIT: %s\n", obs.Summary)
if _, err := keeper.Observe(ctx, obs); err != nil {
fmt.Printf("the condition %s could not be raised: %v\n", obs.Key(), err)
}
}
// compactKept writes the vault's export compact. The composition (the SDK's keptFile) writes it indented, and its 334
// sealed secrets were 164,763 bytes of it against 138,474 compact: a fifth of the declaration's room, in whitespace,
// base64-encoded again on the wire. Every reader parses JSON, so the spacing is nobody's. A resource that is not
// the export, or whose content is not JSON, is left as it is. The resources are copied, never changed in place.
func compactKept(resources []map[string]any) []map[string]any {
out := make([]map[string]any, len(resources))
for i, r := range resources {
out[i] = r
id, _ := r["id"].(string)
path, _ := r["path"].(string)
content, ok := r["content"].(string)
if !ok || (id != "kept" && !strings.HasSuffix(id, ".kept")) || !strings.HasSuffix(path, "/export.json") {
continue
}
var dst bytes.Buffer
if json.Compact(&dst, []byte(content)) != nil {
continue
}
if strings.HasSuffix(content, "\n") {
dst.WriteByte('\n')
}
if dst.String() == content {
continue
}
c := make(map[string]any, len(r))
for k, v := range r {
c[k] = v
}
c["content"] = dst.String()
out[i] = c
}
return out
}
@@ -1,71 +0,0 @@
package main
import (
"strings"
"testing"
"time"
"github.com/novox/mesh-controller/internal/conditions"
)
// The vault's export is sent compact (issue 1689): same document, no indentation, one rewrite.
func TestTheVaultsExportIsSentCompactAndNothingElseIsTouched(t *testing.T) {
indented := "{\n \"export\": 1,\n \"kept\": [\n {\n \"name\": \"a\"\n }\n ]\n}\n"
other := "{\n \"x\": 1\n}\n"
in := []map[string]any{
{"id": "mesh-vault.kept", "type": "file", "path": "/srv/vault/export.json", "mode": "0600", "content": indented},
{"id": "mesh-vault.config", "type": "file", "path": "/srv/vault/config.json", "content": other},
{"id": "other.kept", "type": "file", "path": "/srv/other/notes.json", "content": other},
}
out := compactKept(in)
want := "{\"export\":1,\"kept\":[{\"name\":\"a\"}]}\n"
if out[0]["content"] != want || out[0]["mode"] != "0600" || out[0]["path"] != "/srv/vault/export.json" {
t.Fatalf("the export is %q (%v), expected %q with its mode and path kept", out[0]["content"], out[0], want)
}
if out[1]["content"] != other || out[2]["content"] != other {
t.Fatalf("a file that is not the export was rewritten: %v %v", out[1], out[2])
}
if in[0]["content"] != indented {
t.Fatalf("the composed resource was changed in place")
}
if again := compactKept(out); again[0]["content"] != want {
t.Fatalf("compacting twice changed it: %q", again[0]["content"])
}
}
// From 75% of the bus's largest message (ADR 0486 §8, issue 1698) the declaration is a warning; over it, urgent; back under 75%, cleared.
func TestADeclarationNearTheBusLimitRaisesAConditionAndClearsWhenSmaller(t *testing.T) {
store := conditions.NewInMemory()
k := conditions.NewKeeper(t.Context(), conditions.Options{Store: store, History: store,
Teller: &conditions.Told{}, Now: func() time.Time { return time.Date(2026, 10, 21, 14, 0, 0, 0, time.UTC) }})
t.Cleanup(func() { k.Close(t.Context()) })
const limit = 1000
key := declarationRoomObservation("novox", 0, limit).Key()
read := func() (conditions.Condition, bool) {
t.Helper()
c, found, err := conditions.ReadOne(t.Context(), store, key)
if err != nil {
t.Fatal(err)
}
return c, found
}
settleDeclarationRoom(t.Context(), k, "novox", 749, limit)
if _, found := read(); found {
t.Fatal("a declaration at 74.9% raised a condition")
}
settleDeclarationRoom(t.Context(), k, "novox", 750, limit)
c, found := read()
if !found || c.Severity != conditions.Warning || !strings.Contains(c.Summary, "750 bytes") ||
!strings.Contains(c.Summary, "1000") || c.Subject.Machine != "novox" {
t.Fatalf("at 75%% the condition is %v %+v", found, c)
}
settleDeclarationRoom(t.Context(), k, "novox", 1001, limit)
if c, _ := read(); c.Severity != conditions.Urgent {
t.Fatalf("over the limit the condition is %q, expected urgent", c.Severity)
}
settleDeclarationRoom(t.Context(), k, "novox", 700, limit)
if _, found := read(); found {
t.Fatal("the condition stayed after the declaration fell under 75%")
}
}
-273
View File
@@ -1,273 +0,0 @@
package main
import (
"context"
"encoding/json"
"errors"
"fmt"
"sort"
"strings"
"time"
"github.com/nats-io/nats.go"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// **An object is removed a day after no node's latest declaration names it** (novox/hq ADR 0486 §3, issue 1696), so a node that is
// asleep and receives its latest declaration later still finds every object that declaration names. What each node's latest
// declaration names is read from the bus itself, the last message of that node's subject of the stream NODES: nothing else
// is kept of it. When an object is first found named by none, the time is written in its own metadata (declarationUnnamed);
// found named again, that is erased; found unnamed for a day, it is removed. Each node has a bucket of its own, judged against that
// node's latest declaration alone. Objects are immutable and regenerable from the
// controller's state, so a removal loses nothing.
// declarationUnnamed is the key of an object's metadata holding when it was first found named by no node's latest declaration.
const declarationUnnamed = "unnamed-since"
// sweepEvery is how often the bucket is judged: the removal is exact to within this.
const sweepEvery = time.Hour
// bucketObject is one object of the bucket as the sweep sees it.
type bucketObject struct {
Name string
// UnnamedSince is when it was first found named by none, zero when it was named at the last look.
UnnamedSince time.Time
}
// sweptBucket is what the sweep needs of the bucket.
type sweptBucket interface {
Objects(ctx context.Context) ([]bucketObject, error)
MarkUnnamed(ctx context.Context, name string, since time.Time) error
ClearUnnamed(ctx context.Context, name string) error
// Delete removes the object if it still carries the mark it was judged by, and says whether it did: a send that put it again
// since has erased the mark, and the object is left.
Delete(ctx context.Context, name string, markedAt time.Time) (bool, error)
}
// filesSweep says what a sweep did.
type filesSweep struct{ Kept, Marked, Cleared, Removed int }
// sweepDeclarationFiles judges every object against the names of the latest declarations: named ones are kept (and a
// mark of being unnamed is erased), unnamed ones are marked at the first look and removed once the mark is a day old.
func sweepDeclarationFiles(ctx context.Context, bucket sweptBucket, named map[string]bool, now time.Time) (filesSweep, error) {
var r filesSweep
objects, err := bucket.Objects(ctx)
if err != nil {
return r, err
}
sort.Slice(objects, func(i, j int) bool { return objects[i].Name < objects[j].Name })
for _, o := range objects {
switch {
case named[o.Name]:
r.Kept++
if !o.UnnamedSince.IsZero() {
if err := bucket.ClearUnnamed(ctx, o.Name); err != nil {
return r, fmt.Errorf("the object %s: %w", o.Name[:12], err)
}
r.Cleared++
}
case o.UnnamedSince.IsZero():
if err := bucket.MarkUnnamed(ctx, o.Name, now); err != nil {
return r, fmt.Errorf("the object %s: %w", o.Name[:12], err)
}
r.Marked++
case now.Sub(o.UnnamedSince) >= broker.DeclarationFilesKeptFor:
gone, err := bucket.Delete(ctx, o.Name, o.UnnamedSince)
if err != nil {
return r, fmt.Errorf("the object %s: %w", o.Name[:12], err)
}
if gone {
r.Removed++
} else {
r.Kept++
}
default:
r.Kept++
}
}
return r, nil
}
// namedByLatest is the name of every object this latest declaration of a node names. A body that cannot be read is an error, never
// "names none": an object the node still needs would be marked for removal.
func namedByLatest(node string, signed []byte) (map[string]bool, error) {
var s link.Signed
if err := json.Unmarshal(signed, &s); err != nil {
return nil, fmt.Errorf("%s's latest declaration is not readable: %w", node, err)
}
names, err := namedObjects(s.Declaration)
if err != nil {
return nil, fmt.Errorf("%s's latest declaration is not readable: %w", node, err)
}
named := make(map[string]bool, len(names))
for name := range names {
named[name] = true
}
return named, nil
}
// latestDeclarations are the last signed declarations of the nodes, by node, as the stream NODES holds them. A node with none has
// no entry.
func latestDeclarations(ctx context.Context, conn *nats.Conn, nodes []string) (map[string][]byte, error) {
js, err := conn.JetStream(nats.Context(ctx))
if err != nil {
return nil, err
}
out := map[string][]byte{}
for _, node := range nodes {
msg, err := js.GetLastMsg("NODES", link.DeclareSubject(node), nats.Context(ctx))
if errors.Is(err, nats.ErrMsgNotFound) {
continue
}
if err != nil {
return nil, fmt.Errorf("the latest declaration of %s: %w", node, err)
}
out[node] = msg.Data
}
return out, nil
}
// busSweptBucket is the bucket on the controller's connection.
type busSweptBucket struct{ store jetstream.ObjectStore }
func (b busSweptBucket) Objects(ctx context.Context) ([]bucketObject, error) {
infos, err := b.store.List(ctx)
if errors.Is(err, jetstream.ErrNoObjectsFound) {
return nil, nil
}
if err != nil {
return nil, err
}
out := make([]bucketObject, 0, len(infos))
for _, info := range infos {
o := bucketObject{Name: info.Name}
if at, err := time.Parse(time.RFC3339, info.Metadata[declarationUnnamed]); err == nil {
o.UnnamedSince = at
}
out = append(out, o)
}
return out, nil
}
func (b busSweptBucket) meta(ctx context.Context, name string, change func(map[string]string)) error {
info, err := b.store.GetInfo(ctx, name)
if err != nil {
return err
}
meta := info.ObjectMeta
kept := map[string]string{}
for k, v := range meta.Metadata {
kept[k] = v
}
change(kept)
meta.Metadata = kept
return b.store.UpdateMeta(ctx, name, meta)
}
func (b busSweptBucket) MarkUnnamed(ctx context.Context, name string, since time.Time) error {
return b.meta(ctx, name, func(m map[string]string) { m[declarationUnnamed] = since.UTC().Format(time.RFC3339) })
}
func (b busSweptBucket) ClearUnnamed(ctx context.Context, name string) error {
return b.meta(ctx, name, func(m map[string]string) { delete(m, declarationUnnamed) })
}
func (b busSweptBucket) Delete(ctx context.Context, name string, markedAt time.Time) (bool, error) {
info, err := b.store.GetInfo(ctx, name)
if errors.Is(err, jetstream.ErrObjectNotFound) {
return false, nil
}
if err != nil {
return false, err
}
if at, perr := time.Parse(time.RFC3339, info.Metadata[declarationUnnamed]); perr != nil || !at.Equal(markedAt) {
return false, nil
}
return true, b.store.Delete(ctx, name)
}
// sweepingDeclarationFiles judges the bucket every sweepEvery for as long as this controller serves and holds the lease.
func sweepingDeclarationFiles(ctx context.Context, inv *inventory.Inventory, conn *nats.Conn) {
tick := time.NewTicker(sweepEvery)
defer tick.Stop()
failing := ""
for {
select {
case <-ctx.Done():
return
case <-tick.C:
}
if epoch, err := theLease.epoch(ctx); err != nil || epoch == 0 {
continue
}
why := sweepOnce(ctx, inv, conn)
if why != failing {
if why != "" {
fmt.Printf("the bucket of declaration files cannot be swept: %s\n", why)
} else {
fmt.Printf("the bucket of declaration files is swept again\n")
}
failing = why
}
}
}
// sweepOnce judges the bucket of every node against that node's latest declaration, one bucket at a time under the node's lock, with
// the declaration read again under it: a send that put an object and published its declaration is seen, and one that has not yet
// is waiting for the lock.
func sweepOnce(ctx context.Context, inv *inventory.Inventory, conn *nats.Conn) string {
nodes, err := inv.Nodes(ctx)
if err != nil {
return fmt.Sprintf("the nodes cannot be read: %v", err)
}
var problems []string
var total filesSweep
for _, n := range nodes {
r, err := sweepNode(ctx, conn, n.Name)
total.Kept, total.Marked, total.Cleared, total.Removed = total.Kept+r.Kept, total.Marked+r.Marked, total.Cleared+r.Cleared, total.Removed+r.Removed
if err != nil {
problems = append(problems, fmt.Sprintf("%s: %v", n.Name, err))
}
}
if total.Removed+total.Marked+total.Cleared > 0 {
fmt.Printf("the buckets of declaration files: %d object(s) kept, %d found unnamed, %d named again, %d removed\n",
total.Kept, total.Marked, total.Cleared, total.Removed)
}
return strings.Join(problems, "; ")
}
func sweepNode(ctx context.Context, conn *nats.Conn, node string) (filesSweep, error) {
working, cancel := context.WithTimeout(ctx, 2*time.Minute)
defer cancel()
js, err := jetstream.New(conn)
if err != nil {
return filesSweep{}, err
}
bucket, err := broker.DeclarationFilesBucketOf(node)
if err != nil {
return filesSweep{}, fmt.Errorf("the bucket of node %q: %w", node, err)
}
store, err := js.ObjectStore(working, bucket)
if errors.Is(err, jetstream.ErrBucketNotFound) {
return filesSweep{}, nil // a node never sent a reference has no bucket yet
}
if err != nil {
return filesSweep{}, fmt.Errorf("the bucket cannot be read: %w", err)
}
defer lockFilesOf(node)()
latest, err := latestDeclarations(working, conn, []string{node})
if err != nil {
return filesSweep{}, err
}
named := map[string]bool{}
if body, ok := latest[node]; ok {
if named, err = namedByLatest(node, body); err != nil {
return filesSweep{}, err
}
}
return sweepDeclarationFiles(working, busSweptBucket{store}, named, time.Now())
}
@@ -1,89 +0,0 @@
//go:build integration
package main
import (
"encoding/json"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
)
// 1840/AC5 (ADR 0507 §5): the source of the declared-packages mirror composes each node and names, for a package two modules
// declare and one declared absent, the modules that declare it; the same records give the same bytes on a second pass, so
// the mirror writes nothing; a node with no package has a key that says so.
func TestTheDeclaredPackagesMirrorNamesWhatEachNodesModulesDeclare(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
register(t, open, catalogue.Manifest{Module: "viewer", Version: "1", Resources: []map[string]any{
{"id": "pkg-curl", "type": "package", "package": "curl"},
{"id": "pkg-nano", "type": "package", "package": "nano", "absent": true}}})
register(t, open, catalogue.Manifest{Module: "fetcher", Version: "1", Resources: []map[string]any{
{"id": "pkg-curl", "type": "package", "package": "curl"}}})
for _, module := range []string{"viewer", "fetcher"} {
if _, err := assign(ctx, open, "anchor", module); err != nil {
t.Fatalf("assign anchor %s: %v", module, err)
}
}
source := declaredPackagesMirror(open)
first, err := source(ctx)
if err != nil {
t.Fatal(err)
}
var got inventory.DeclaredPackagesMirrored
if err := json.Unmarshal(first["anchor"], &got); err != nil {
t.Fatalf("anchor has no value: %v", err)
}
// Every node also runs the mesh's own modules, which declare packages of their own: curl and nano are looked
// up by name, and no package names viewer or fetcher but those they declare.
byName := map[string]inventory.PackageDeclared{}
for _, p := range got.Packages {
byName[p.Package] = p
}
same := func(a, b []string) bool {
if len(a) != len(b) {
return false
}
for i := range a {
if a[i] != b[i] {
return false
}
}
return true
}
for _, p := range []inventory.PackageDeclared{{Package: "curl", Modules: []string{"fetcher", "viewer"}},
{Package: "nano", Modules: []string{"viewer"}, Absent: true}} {
if g, ok := byName[p.Package]; !ok || g.Absent != p.Absent || !same(g.Modules, p.Modules) {
t.Errorf("anchor declares %s as %+v, want %+v (all: %+v)", p.Package, g, p, got.Packages)
}
}
for _, p := range got.Packages {
if p.Package == "curl" || p.Package == "nano" {
continue
}
for _, m := range p.Modules {
if m == "viewer" || m == "fetcher" {
t.Errorf("%s is said to be declared by %s, which does not declare it", p.Package, m)
}
}
}
var laptop inventory.DeclaredPackagesMirrored
if err := json.Unmarshal(first["laptop"], &laptop); err != nil || laptop.Node != "laptop" || laptop.Packages == nil {
t.Errorf("a node assigned neither module has no value of its own: %s (%v)", first["laptop"], err)
}
for _, p := range laptop.Packages {
if p.Package == "curl" || p.Package == "nano" {
t.Errorf("laptop is assigned neither module and declares %+v", p)
}
}
second, err := source(ctx)
if err != nil {
t.Fatal(err)
}
for node, value := range first {
if string(second[node]) != string(value) {
t.Errorf("the value of %s changed between two passes over the same records:\n%s\n%s", node, value, second[node])
}
}
}
+200 -9
View File
@@ -815,21 +815,20 @@ func deliveryPlanDocument(m orderMember, entries []inventory.Entry, read map[str
// delivery walks [-n 50] [--plan <id>]
func deliveryCommand(ctx context.Context, args []string) error {
if len(args) == 0 {
return errors.New("delivery order|go|stop|walks")
return errors.New("delivery plan|order|check|go|stop|walks")
}
sub, rest := args[0], args[1:]
if sub == "plan" {
// Removed (ADR 0450 §8; issue 1318): refused by name before any flag is read, naming its replacement.
return removedByName("delivery-plan")
}
if sub == "check" {
// Removed (ADR 0503 §3, ADR 0527; issue 1318): the checker judges every head, so the controller asks nothing.
return removedByName("delivery-check")
}
set := flag.NewFlagSet("delivery "+sub, flag.ContinueOnError)
repository := set.String("repository", "", "owner/repository")
base := set.String("base", "main", "the branch it merges into")
head := set.String("head", "", "the commit at hand")
paths := set.String("paths", "", "the files it changes, comma-separated")
moduleDirs := set.String("module-dirs", "", "the directories holding a module.json at the head, comma-separated")
removed := set.String("removed", "", "the files it deletes, comma-separated")
membersJSON := set.String("members", "", "a group's members, as JSON")
membersFile := set.String("members-file", "", "a file of a group's members, as JSON: the same as --members, for a list "+
"too long for an argument (novox/hq issue 1272)")
group := set.String("group", "", "a delivery group's id")
ready := set.Bool("ready", false, "the head's pull request is marked ready: its check is queued before every other check "+
"(novox/hq ADR 0406 §3)")
by := set.String("by", catalogue.DeliverySeat, "who says it")
@@ -873,6 +872,19 @@ func deliveryCommand(ctx context.Context, args []string) error {
return enc.Encode(v)
}
switch sub {
case "plan":
if *repository == "" || *head == "" && *paths == "" {
return errors.New("delivery plan --repository owner/repo --head <commit> --paths a,b")
}
m := changeMember(*repository, *base, *head, splitList(*paths), splitList(*removed), moduleDirsOf(*moduleDirs))
entries, read, edges, err := theGraph(ctx, inv)
if err != nil {
return err
}
return answer(deliveryPlanDocument(m, entries, read, edges, func(module string) (inventory.Upgrade, bool) {
u, err := inv.UpgradeOf(ctx, module)
return u, err == nil
}))
case "order":
if len(members) == 0 {
return errors.New("delivery order --members <json>: a group has members")
@@ -886,6 +898,27 @@ func deliveryCommand(ctx context.Context, args []string) error {
return fmt.Errorf("what the members say of their manifests cannot be read: %w", err)
}
return answer(orderOf(members, reachOfMembers(members, entries, read, edges), edges, changes))
case "check":
if *group == "" && len(members) == 1 {
// One head, checked again as the forge's announcement would have it (a recheck): the controller's
// own path for a pull request, run because the delivery's owner asked.
return recheckOneHead(members[0].ID, answer, func() (string, error) {
var asked string
err := sayingOnTheBus(ctx, inv, func() (err error) {
asked, err = (following{open: open, oneShot: true}).pullUpdated(ctx, members[0].pull(), os.Stderr)
return err
})
return asked, err
})
}
if *group == "" || len(members) < 2 {
return errors.New("delivery check --group <id> --members <json> (two heads or more), or --members <one head>")
}
id, err := askGroupCheck(ctx, open, *group, members)
if err != nil {
return err
}
return answer(map[string]any{"asked": id, "group": *group})
case "go":
if len(positionals) != 1 {
return errors.New("delivery go <plan> [--by <who>] [--why <text>]")
@@ -1060,6 +1093,133 @@ func theGraph(ctx context.Context, inv *inventory.Inventory) ([]inventory.Entry,
return entries, read, edges, nil
}
// cycleVerdict is the failing verdict of a group whose modules are ordered in a cycle, and whether there is one: the
// group's composed check, said without a build. The manifests' change is read from the forge; an answer it cannot give
// is an error, never read as no cycle.
func cycleVerdict(ctx context.Context, group string, members []orderMember, reach map[string]orderReach,
edges []inventory.Edge, entries []inventory.Entry) (link.Checked, bool, error) {
changes, err := changesOf(ctx, changeSourcesOf(members, entries, reach))
if err != nil {
return link.Checked{}, false, fmt.Errorf("what the group's members say of their manifests cannot be read: %w", err)
}
var moved []string
for _, m := range members {
moved = appendNew(moved, reach[m.ID].Moved...)
}
said := orderEdges(moved, changes, edges).CycleSaid()
if said == "" {
return link.Checked{}, false, nil
}
first := members[0].pull()
c := link.Checked{ID: link.NewBuildID(time.Now()), Owner: first.Owner, Repo: first.Repo, Number: first.Number,
Commit: first.Commit, Verdict: "fail", Summary: said, Report: said, Group: group,
Gate: &link.CheckLayer{Verdict: "fail", Summary: said}}
for _, m := range members {
if group == "" {
break // a lone pull request's verdict names no group and no member
}
p := m.pull()
c.Members = append(c.Members, link.CheckedMember{Owner: p.Owner, Repo: p.Repo, Number: p.Number, Commit: p.Commit,
Trunk: twelve(p.BaseSHA)})
}
return c, true, nil
}
// twelve is a commit as twelve characters.
func twelve(c string) string {
if len(c) > 12 {
return c[:12]
}
return c
}
// groupPrimary is the head a group's composed check is run from: the one moving the controller, which then
// judges the group by itself; else the one moving the node-engine, whose validator judges; else the first.
func groupPrimary(members []orderMember, reach map[string]orderReach) int {
for _, want := range []string{"mesh-controller", "mesh-host"} {
for i, m := range members {
if slices.Contains(reach[m.ID].Moved, want) {
return i
}
}
}
return 0
}
// askGroupCheck asks the build seat for a group's composed check: every head laid over the mesh in turn,
// judged as one future state (novox/hq ADR 0239). The verdict comes back as `checked` with the group's id,
// for mesh-delivery; the forge's holder sets no status from it.
func askGroupCheck(ctx context.Context, open *stores, group string, members []orderMember) (string, error) {
entries, read, edges, err := theGraph(ctx, open.inventory)
if err != nil {
return "", err
}
reach := reachOfMembers(members, entries, read, edges)
// **A cycle of edges fails the group's composed check, naming every module and edge** (novox/hq ADR 0419 clause 6),
// before the build seat is asked: no composition of the heads has an order to be sent in.
if failing, cyclic, err := cycleVerdict(ctx, group, members, reach, edges, entries); err != nil {
return "", err
} else if cyclic {
sayChecked(ctx, os.Stderr, failing)
return failing.ID, nil
}
primary := groupPrimary(members, reach)
p := members[primary].pull()
scope := pullScope(p, entries, read, edges)
// The gate runs over what any member moves; a judge from the primary alone.
for _, m := range members {
r := reach[m.ID]
scope.Modules = appendNew(scope.Modules, r.Moved...)
scope.Dependents = appendNew(scope.Dependents, r.Dependents...)
}
request, err := checkRequestFor(ctx, open, p, scope, entries)
if err != nil {
return "", err
}
request.Check.Group = group
// A group's check is queued as ready when any member's pull request is marked ready.
request.Check.Ready = slices.ContainsFunc(members, func(m orderMember) bool { return m.Ready })
world, err := restOfTheMesh(ctx, open, shelfOf(entries), "")
if err != nil {
return "", err
}
for i, m := range members {
if i == primary {
continue
}
mp := m.pull()
// As the mesh clones a module built from it; a repository no module is built from, from the forge.
source := inventory.Source{Seat: gitSeat, Repository: mp.Owner + "/" + mp.Repo}
for _, e := range entries {
if !e.Provided && sameRepository(e.Source.Repository, link.SourceMoved{Owner: mp.Owner, Repo: mp.Repo}) {
source = e.Source
break
}
}
url := source.Repository
if source.Seat != "" {
url, err = clonedFromSeat(world, source.Seat, source.Repository)
}
if err != nil {
return "", fmt.Errorf("%s cannot be cloned for the group's check: %w", m.ID, err)
}
request.Check.Members = append(request.Check.Members, link.GroupMember{Owner: mp.Owner, Repo: mp.Repo,
Number: mp.Number, Repository: url, Ref: mp.Commit, Paths: mp.Paths, Base: mp.Base})
}
seat := buildSeatHeld(ctx)
ask, err := askOverOn(seat)
if err != nil {
return "", err
}
defer ask.Close()
if err := ask.Ask(ctx, request); err != nil {
return "", err
}
fmt.Fprintf(os.Stderr, "group %s: asked %s to check %d head(s) composed together, as %s\n", group, seat,
len(members), request.ID)
return request.ID, nil
}
// appendNew appends what is not there yet.
func appendNew(to []string, items ...string) []string {
for _, i := range items {
@@ -1070,6 +1230,15 @@ func appendNew(to []string, items ...string) []string {
return to
}
// shelfOf is the catalogue's manifests by name.
func shelfOf(entries []inventory.Entry) map[string]catalogue.Manifest {
shelf := map[string]catalogue.Manifest{}
for _, e := range entries {
shelf[e.Manifest.Module] = e.Manifest
}
return shelf
}
// sayPlanMoved says a walk kept in a new state as the controller's `plan-moved`, the plan whole: what the
// delivery it walks reads its steps from. Never in the way of the walk: said beside it, and a save that could
// not be said is logged — the delivery's owner, which also asks for the walks it follows, finds it by
@@ -1211,3 +1380,25 @@ func overNATS(conn *nats.Conn) (link.Bus, error) {
}
return link.OverNATS{Conn: conn, JS: js}, nil
}
// recheckAnswer is what `delivery check` answers for one head: the id of the check asked of the build seat (empty when
// the controller said the verdict itself and asked nothing), which mesh-delivery reads, and the head rechecked.
func recheckAnswer(asked, member string) map[string]any {
return map[string]any{"asked": asked, "rechecked": member}
}
// recheckOneHead runs a one-head `delivery check` and answers it as data. Standard output is the answer's JSON alone,
// or the seat verb reads no answer from it (novox/hq issue 1311): while the check runs, whatever it prints to standard
// output (the controller's own path prints its prose there, as do the stores it reads) goes to standard error.
func recheckOneHead(member string, answer func(any) error, ask func() (string, error)) error {
real := os.Stdout
os.Stdout = os.Stderr
asked, err := func() (string, error) {
defer func() { os.Stdout = real }()
return ask()
}()
if err != nil {
return err
}
return answer(recheckAnswer(asked, member))
}
@@ -0,0 +1,85 @@
package main
import (
"bytes"
"context"
"encoding/json"
"fmt"
"os"
"strings"
"testing"
"github.com/novox/mesh-controller/internal/link"
)
// `delivery check` with one head answers the id of the check it asked, as data (novox/hq issue 1311): the prose
// the check says goes to standard error, so standard output is the document alone and answerOf reads it.
func TestDeliveryCheckOfOneHeadAnswersTheRequestID(t *testing.T) {
argv := []string{"delivery", "check", "--members", "[]"}
doc, err := json.Marshal(recheckAnswer("check-42", "novox/x@abc"))
if err != nil {
t.Fatal(err)
}
// Seen failing first: the prose the command used to print on standard output before the document.
polluted := append([]byte("novox/x#7 (abc): asked the build seat as check-42\n"), doc...)
if got := answerOf(argv, polluted, "", true); got.Answer != nil {
t.Fatalf("prose before the document should leave no answer; got %v", got.Answer)
}
got := answerOf(argv, doc, "novox/x#7 (abc): asked the build seat as check-42\n", true)
m, ok := got.Answer.(map[string]any)
if !ok || m["asked"] != "check-42" || m["rechecked"] != "novox/x@abc" {
t.Fatalf("the answer is %v; want asked check-42 and rechecked novox/x@abc", got.Answer)
}
if !strings.Contains(got.Output, "asked the build seat") {
t.Fatalf("the prose is still said in the text: %q", got.Output)
}
}
// The text of the other callers does not change: sayChecked writes the same line to the writer it is given, and
// the controller's own path still gives PullUpdated standard output.
func TestSayCheckedWritesTheSameLineToItsWriter(t *testing.T) {
var w bytes.Buffer
sayChecked(context.Background(), &w, link.Checked{ID: "c1", Owner: "novox", Repo: "x", Number: 7, Commit: "0123456789abcdef",
Verdict: "pass", Summary: "fine"})
want := "c1: novox/x#7 at 01234567 checked on a machine that did not say which: gate PASS — fine; repository none\n"
if w.String() != want {
t.Fatalf("sayChecked said %q; want %q", w.String(), want)
}
}
// Whatever the check prints to standard output while it runs (checks.go namedForTheCheck, the inventory's pointer
// repair) must not reach the answer's standard output: it is exactly one JSON document (novox/hq issue 1311).
func TestOneHeadRecheckLeavesStandardOutputTheAnswerAlone(t *testing.T) {
r, w, err := os.Pipe()
if err != nil {
t.Fatal(err)
}
e, ew, err := os.Pipe()
if err != nil {
t.Fatal(err)
}
realOut, realErr := os.Stdout, os.Stderr
os.Stdout, os.Stderr = w, ew
runErr := recheckOneHead("novox/x@abc", func(v any) error { return json.NewEncoder(os.Stdout).Encode(v) },
func() (string, error) {
fmt.Println(" the earlier checks' builds could not be read, so the check builds")
fmt.Fprintln(os.Stdout, "repaired a pointer")
return "check-42", nil
})
os.Stdout, os.Stderr = realOut, realErr
w.Close()
ew.Close()
if runErr != nil {
t.Fatal(runErr)
}
var out, errOut bytes.Buffer
_, _ = out.ReadFrom(r)
_, _ = errOut.ReadFrom(e)
var doc map[string]any
if err := json.Unmarshal(out.Bytes(), &doc); err != nil || doc["asked"] != "check-42" || doc["rechecked"] != "novox/x@abc" {
t.Fatalf("standard output %q is not the answer alone (%v)", out.String(), err)
}
if !strings.Contains(errOut.String(), "repaired a pointer") {
t.Fatalf("the prose should be on standard error: %q", errOut.String())
}
}
+1 -10
View File
@@ -41,7 +41,7 @@ type stalledLine struct {
// waiting for the owner's word and what holds it (novox/hq issues 362, 529).
Walk string `json:"walk,omitempty"`
// Waiting are the merge checks a line of the state `unanswered` waits on, as mesh-delivery says each:
// "mesh/check pending since <UTC time>", "mesh/check never set" (novox/hq issue 438).
// "mesh/merge-gate pending since <UTC time>", "mesh/repo-check never set" (novox/hq issue 438).
Waiting []string `json:"waiting,omitempty"`
// Checks are the same merge checks as data, which the controller words in the operator's own time (novox/hq
// issue 443): a time inside a finished sentence cannot be said again in another zone. A mesh-delivery from before
@@ -58,17 +58,8 @@ type stalledLine struct {
// Severity is "urgent" for a group held or half merged and a refused member (novox/hq issues 1275, 1278); empty
// or anything else is a warning, as before.
Severity string `json:"severity,omitempty"`
// Needs is "new-head" for a delivery whose cure is its author's new head (novox/hq ADR 0505 §5), Author that author
// as the operator reads it ("the author of feat/x", with the session that pushed it when its pull request names one)
// and Branch its branch. A mesh-delivery from before says none, and the words are as before.
Needs string `json:"needs,omitempty"`
Author string `json:"author,omitempty"`
Branch string `json:"branch,omitempty"`
}
// needsNewHead is what a stalled line needs when its cure is its author's new head (ADR 0505 §5).
const needsNewHead = "new-head"
// severity is the condition's severity a stalled line asks: urgent when it says so, a warning otherwise.
func (l stalledLine) severity() conditions.Severity {
if l.Severity == string(conditions.Urgent) {
@@ -181,3 +181,32 @@ func TestATruncatedChangeIsReadAsChangingTheManifestsOfWhatItMoves(t *testing.T)
t.Fatalf("the member's change is read as %+v", src)
}
}
// A lone pull request's cycle fails its check, naming every module and edge, with no group and no members.
func TestALonePullRequestInACycleFailsWithNoGroup(t *testing.T) {
was, wasParent := manifestAt, parentOf
t.Cleanup(func() { manifestAt, parentOf = was, wasParent })
manifestsReadMu.Lock()
clear(manifestsRead)
manifestsReadMu.Unlock()
const head = "3333333333333333333333333333333333333333"
parentOf = func(context.Context, string, string, string) (string, error) { return "", nil }
manifestAt = func(_ context.Context, _, _, file, ref string) (catalogue.Manifest, bool, error) {
if ref == "main" {
return catalogue.Manifest{}, false, nil
}
if file == "modules/x/module.json" {
return catalogue.Manifest{Module: "mx", DefinesSeats: []catalogue.SeatDeclaration{{Name: "s", Serves: []catalogue.Verb{{Name: "a"}}}},
Invokes: []string{"seat:s.b"}}, true, nil
}
return catalogue.Manifest{Module: "my", DefinesSeats: []catalogue.SeatDeclaration{{Name: "s", Serves: []catalogue.Verb{{Name: "b"}}}},
Invokes: []string{"seat:s.a"}}, true, nil
}
m := orderMember{ID: "novox/r@" + head[:12], Repository: "novox/r", Base: "main", Head: head, Number: 5,
Paths: []string{"modules/x/module.json", "modules/y/module.json"}}
c, cyclic, err := cycleVerdict(context.Background(), "", []orderMember{m}, map[string]orderReach{m.ID: {Moved: []string{"mx", "my"}}}, nil, nil)
if err != nil || !cyclic || c.Verdict != "fail" || c.Group != "" || len(c.Members) != 0 || !strings.Contains(c.Summary, "s.a") ||
!strings.Contains(c.Summary, "my") {
t.Fatalf("the verdict is %+v (cyclic %v, %v)", c, cyclic, err)
}
}
@@ -59,23 +59,8 @@ func TestAMergeWaitsForItsDeliverysWordAndThePersonsWordWorksWithoutIt(t *testin
}
}
// heardOf has this mesh's own view of the plans take in the plan as the bucket holds it. A verb of the
// controller (deliveryCommand, plansCommand) opens an inventory of its own and writes the plan there; this
// mesh's inventory learns of the write from the bucket's watch, a moment later, and a loaded machine makes the
// moment long enough for advanceHeld to read the plan as it was, still waiting for its word, and ask nothing
// (issue 1399). Reading the plan from the bucket puts it into the view at once, with no wait to bound.
heardOf := func(id string) {
t.Helper()
if _, err := inv.PlanByID(ctx, id); err != nil {
t.Fatal(err)
}
}
// No holder on record: the merge starts its walk, as before.
// A cut takes its first tier from the builder's keys and a step of the walk records the take (the walk asks no
// build), so each walk that starts is stepped once before its takes are counted.
p := merge("c1aaaaaaaa", "modules/app/index.ts")
advanceHeld(ctx, open)
if p.Waiting() || len(*asked) != 1 {
t.Fatalf("with no holder on record the walk waited (%v) or asked %v", p.Waiting(), *asked)
}
@@ -105,7 +90,6 @@ func TestAMergeWaitsForItsDeliverysWordAndThePersonsWordWorksWithoutIt(t *testin
if err := deliveryCommand(ctx, []string{"go", p.ID}); err == nil {
t.Fatal("a walk was let go twice")
}
heardOf(p.ID)
advanceHeld(ctx, open)
if len(*asked) != 2 {
t.Fatalf("the word did not start the walk: %v", *asked)
@@ -119,7 +103,6 @@ func TestAMergeWaitsForItsDeliverysWordAndThePersonsWordWorksWithoutIt(t *testin
// ADR 0276): the merge's batch is cut once it ended.
finish(p.ID)
p = merge("c3cccccccc", "modules/mesh-delivery/main.go")
advanceHeld(ctx, open)
if p.Waiting() || len(*asked) != 3 {
t.Fatalf("mesh-delivery's own walk waited for mesh-delivery: %v %v", p.Waiting(), *asked)
}
@@ -136,7 +119,6 @@ func TestAMergeWaitsForItsDeliverysWordAndThePersonsWordWorksWithoutIt(t *testin
if err := plansCommand(ctx, []string{"go", p.ID, "--why", "mesh-delivery is down"}); err != nil {
t.Fatal(err)
}
heardOf(p.ID)
advanceHeld(ctx, open)
got, _ = inv.PlanByID(ctx, p.ID)
if got.Waiting() || !strings.HasPrefix(got.Delivery.By, "a person") || len(*asked) < 4 {
@@ -161,9 +143,7 @@ func TestAMergeWaitsForItsDeliverysWordAndThePersonsWordWorksWithoutIt(t *testin
t.Fatal(err)
}
asks := len(*asked)
p = merge("c6ffffffff", "modules/app/index.ts")
advanceHeld(ctx, open)
if p.Waiting() || len(*asked) != asks+1 {
if p = merge("c6ffffffff", "modules/app/index.ts"); p.Waiting() || len(*asked) != asks+1 {
t.Fatalf("with the holder gone the walk waited: %v", p.Waiting())
}
}
+4
View File
@@ -136,6 +136,10 @@ func TestTheDeliveryVerbsComposeTheirCommands(t *testing.T) {
{"delivery-walks", map[string]any{}, []string{"delivery", "walks"}},
{"delivery-walks", map[string]any{"plan": "plan-1"}, []string{"delivery", "walks", "--plan", "plan-1"}},
{"delivery-order", map[string]any{"members": "[]"}, []string{"delivery", "order", "--members", "[]"}},
{"delivery-check", map[string]any{"group": "feat/x", "members": "[]"},
[]string{"delivery", "check", "--group", "feat/x", "--members", "[]"}},
{"delivery-plan", map[string]any{"repository": "novox/a", "paths": "x", "head": "c0"},
[]string{"delivery", "plan", "--repository", "novox/a", "--paths", "x", "--head", "c0"}},
{"plans", map[string]any{"go": "plan-1", "why": "down"}, []string{"plans", "go", "plan-1", "--why", "down"}},
} {
got, err := argvFor(c.verb, c.args)
+10
View File
@@ -35,6 +35,16 @@ func TestAMergeOfABaseTakesWhatStandsOnItAlong(t *testing.T) {
t.Fatalf("%s was rebuilt and stands on nothing that moved (%v)", n, names)
}
}
// The dependents come in base order when the merge orders them: the runtime, then shop, then
// the plugin that stands on shop.
ordered := orderByBases(append([]inventory.Entry{entry("mesh-tools")}, got...), against)
pos := map[string]int{}
for i, e := range ordered {
pos[e.Manifest.Module] = i
}
if !(pos["mesh-tools"] < pos["shop"] && pos["shop"] < pos["shop-plugin"]) {
t.Fatalf("not in base order: %v", ordered)
}
// Nothing moved: nothing follows.
if more := dependentsOf(nil, entries, against); len(more) != 0 {
t.Fatalf("with nothing moved, %d module(s) were rebuilt", len(more))
+1 -1
View File
@@ -669,7 +669,7 @@ func localDoctor(ctx context.Context) (*doctor, func(), error) {
if err != nil {
return nil, nil, err
}
js, err := aBus()
js, err := dialTheBus()
if err != nil {
open.Close()
return nil, nil, err
@@ -0,0 +1,20 @@
//go:build integration
package main
import (
"encoding/json"
"testing"
"github.com/novox/mesh-controller/internal/link"
)
func TestADryRunsOutcomeIsTakenInByNothing(t *testing.T) {
err := builds{}.Built(t.Context(), link.BuildResult{
ID: "build-1", Repository: "ssh://forge/app.git", Ref: "unreviewed", Module: "app", DryRun: true,
Manifest: json.RawMessage(`{"module":"app","version":"1"}`),
})
if err != nil {
t.Fatalf("a dry run's outcome was not simply set aside: %v", err)
}
}
+29
View File
@@ -0,0 +1,29 @@
package main
import (
"encoding/json"
"testing"
"github.com/novox/mesh-controller/internal/link"
)
// A dry run's outcome is looked at, never taken in (novox/hq issue 240). The daemon here holds no
// store at all, so anything that tried to record or register would fail rather than pass quietly.
// The mark survives the wire both ways: asked as a dry run, answered as one.
func TestTheDryRunMarkTravelsWithTheBuild(t *testing.T) {
raw, _ := json.Marshal(link.BuildRequest{ID: "build-1", Repository: "r", DryRun: true})
var asked link.BuildRequest
if err := json.Unmarshal(raw, &asked); err != nil || !asked.DryRun {
t.Fatalf("the request lost its dry-run mark: %s", raw)
}
raw, _ = json.Marshal(link.BuildResult{ID: "build-1", DryRun: true})
var answered link.BuildResult
if err := json.Unmarshal(raw, &answered); err != nil || !answered.DryRun {
t.Fatalf("the outcome lost its dry-run mark: %s", raw)
}
raw, _ = json.Marshal(link.BuildResult{ID: "build-2"})
if string(raw) != `{"id":"build-2","repository":"","on":""}` {
t.Fatalf("an ordinary outcome carries a dry-run mark: %s", raw)
}
}
+20
View File
@@ -12,6 +12,7 @@ import (
"time"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// What the core's bounds are set from (novox/hq to-be 45 Phase 0).
@@ -23,6 +24,25 @@ import (
// controller as it hears each — a send's first report, a machine's next word, a plan leaving a tier, a
// build's outcome — and summarised here per machine, repository or module.
// recordBuildDuration measures one build from its ask to its outcome heard.
func recordBuildDuration(ctx context.Context, inv *inventory.Inventory, result link.BuildResult, asked time.Time) {
if asked.IsZero() || result.ID == "" {
return
}
subject := result.Module
if subject == "" {
subject = result.Repository
}
detail := "built"
if result.Failed != "" {
detail = "failed: " + firstLine(result.Failed)
}
if err := inv.RecordDuration(ctx, inventory.Duration{Kind: inventory.DurationBuild, Subject: subject,
Node: result.On, Ref: result.ID, Started: asked, Took: time.Since(asked), Detail: detail}); err != nil {
fmt.Fprintf(os.Stderr, "%s: how long it took could not be recorded: %v\n", result.ID, err)
}
}
// durationSummary is one subject's measurements of one kind.
type durationSummary struct {
Kind string `json:"kind"`
+12
View File
@@ -19,6 +19,7 @@ import (
snapshot "git.novox.be/novox/mesh-sdk/go/facts"
"github.com/novox/mesh-controller/internal/artifacts"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/builder"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
@@ -240,6 +241,17 @@ func gatherFacts(ctx context.Context, open *stores, busVersion string) (snapshot
scrub.Site(o.Site)
}
// What a merge check runs in and reads beside it, as the build seat would be asked for it.
if held, err := inv.Held(ctx); err == nil {
for language, reference := range builder.ToolchainsOf(held) {
if f.Versions.Toolchains == nil {
f.Versions.Toolchains = map[string]string{}
}
f.Versions.Toolchains[language] = catalogue.Recorded(reference)
}
} else {
return snapshot.Facts{}, err
}
for _, e := range entries {
if dir, core := coreModules[e.Manifest.Module]; core && !e.Provided && e.Source.Repository != "" {
for d, ref := range besideRefs(dir, current[e.Manifest.Module].Commit) {
@@ -0,0 +1,84 @@
//go:build integration
package main
import (
"context"
"errors"
"testing"
"time"
"git.novox.be/novox/mesh-sdk/go/asks"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/identity"
"github.com/novox/mesh-controller/internal/link"
)
// aForgeAndABuildSeat gives a test what a real ask of a build needs: the forge holding the git seat, assigned to anchor, so
// the clone URL composes; the identity store; the build seat's queue on a real bus; and the environment of the serving controller, which is
// never the controller's terminal (a verb's, in effect: startedAtTheTerminal is false).
func aForgeAndABuildSeat(t *testing.T, open *stores) *broker.JetStream {
t.Helper()
js, _ := aBuildQueue(t)
// The ask signs what it sends with the controller's identity, as the serving controller does.
identity.ForTest(t)
forge, err := catalogue.ParseManifest([]byte(`{"module":"gitea","version":"1",
"claims":[{"name":"git","scope":"mesh"}],"provides":[{"name":"git","scope":"mesh"}],
"serves":{"git":{"scheme":"http","port":3000}}}`))
if err != nil {
t.Fatal(err)
}
register(t, open, forge)
if _, err := assign(t.Context(), open, "anchor", "gitea"); err != nil {
t.Fatal(err)
}
t.Setenv(servedVar, "1")
return js
}
// 1146/AC9, through the real ask (askABuild, buildOneAsked, verbMayAsk) and a real bus: the first build of a module a
// merge adds on the operator's Approve is queued on the build seat; the same ask without that warrant, and the warrant's
// ask for a path the merge does not hold, are refused by the guard of a build through a verb.
func TestTheApprovedFirstBuildPassesTheGuardOfAVerbAndIsQueued(t *testing.T) {
const head, commit = "a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1", "0177ea7a00000000"
open := aMesh(t)
ctx := t.Context()
js := aForgeAndABuildSeat(t, open)
a := aDeliveryMergeOf(t, head, map[string]string{"novox/mesh-delivery@a1a1a1a1a1a1": head})
wasKept, wasAsk := keptDeliveryMerge, routerAskOf
keptDeliveryMerge = func(context.Context, string, string, int, string) (deliveryMerge, bool, error) {
return a.merge, true, nil
}
routerAskOf = func(context.Context, catalogueReader, string, string) (*asks.Ask, *asks.Warrant, error) {
return &a.ask, &a.warrant, nil
}
t.Cleanup(func() { keptDeliveryMerge, routerAskOf = wasKept, wasAsk })
// Without the warrant the guard refuses, whatever the repository holds.
_, err := askABuild(ctx, buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "", "main")
var policy *heldAtTheTerminal
if !errors.As(err, &policy) {
t.Fatalf("1146/AC9: a first build asked without the warrant was not refused: %v", err)
}
// And a path the merge does not hold is refused on the warrant.
warranted := withFirstBuildWarrant(ctx, "novox/mesh-delivery", []string{""}, "the Approve")
if _, err := askABuild(warranted, buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "other", "main"); !errors.As(err, &policy) {
t.Fatalf("1146/AC9: the warrant covered a path the merge does not hold: %v", err)
}
m := link.SourceMoved{Owner: "novox", Repo: "mesh-delivery", Base: "main", HeadSHA: head, Commit: commit, Number: 7,
MergedAt: time.Now().Add(-time.Hour).UTC().Format(time.RFC3339Nano), Paths: []string{"module.json", "main.go"},
ModuleDirs: []string{"."}, ModuleDirsSaid: true}
if err := (following{open: open}).SourceMoved(ctx, m); err != nil {
t.Fatal(err)
}
queue, err := link.ReadQueue(ctx, js, link.TheBuildMachine)
if err != nil {
t.Fatal(err)
}
if len(queue.Asks) != 1 {
t.Fatalf("1146/AC9: the build queue holds %+v, want the one first build", queue)
}
}
@@ -0,0 +1,29 @@
package main
import "testing"
// 1146/AC9: the warrant the controller checked covers exactly the repository on the git seat and the module paths the
// merge holds; nothing else passes verbMayAsk on it.
func TestTheFirstBuildWarrantCoversOnlyItsRepositoryAndItsModulePaths(t *testing.T) {
ctx := withFirstBuildWarrant(t.Context(), "novox/mesh-delivery", []string{"", "view"}, "the Approve")
w := firstBuildWarrantIn(ctx)
for _, c := range []struct {
source buildSource
path string
covers bool
}{
{buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "", true},
{buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "/view/", true},
{buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "other", false},
{buildSource{Repository: "agent/mesh-delivery", Seat: "git"}, "", false},
{buildSource{Repository: "novox/mesh-delivery"}, "", false},
{buildSource{Repository: "novox/mesh-delivery", Seat: "elsewhere"}, "", false},
} {
if got := w.covers(c.source, c.path); got != c.covers {
t.Errorf("1146/AC9: %v at %q covered = %v, want %v", c.source, c.path, got, c.covers)
}
}
if firstBuildWarrantIn(t.Context()).covers(buildSource{Repository: "novox/mesh-delivery", Seat: "git"}, "") {
t.Error("1146/AC9: a context with no warrant covers an ask")
}
}
+21 -77
View File
@@ -5,7 +5,6 @@ import (
"encoding/json"
"errors"
"fmt"
"regexp"
"strings"
"time"
@@ -47,7 +46,7 @@ func (f following) firstBuilds(ctx context.Context, m link.SourceMoved, entries
case len(added) == 0:
return false, nil
}
if err := askFirstBuilds(ctx, f.open, m, now, added, why); err != nil {
if err := askFirstBuilds(ctx, inv, m, now, added, why); err != nil {
return true, notNow(err)
}
return true, nil
@@ -68,35 +67,27 @@ func keepFirstMerge(ctx context.Context, inv *inventory.Inventory, m link.Source
return err
}
// askFirstBuilds keeps the merge as answered and registers its modules from the builder's done keys (novox/hq ADR 0501
// §3, §6): nothing asks their first builds. The merge kept is the registration's warrant (approvedMove reads it by the
// commit the indexer's trunk value names); a module whose key is not done yet is registered when the key's write wakes
// the take-in (new_modules.go).
func askFirstBuilds(ctx context.Context, open *stores, m link.SourceMoved, now time.Time, added []string,
// askFirstBuilds keeps the merge, asks the builds, and marks the merge as no longer waiting for them.
func askFirstBuilds(ctx context.Context, inv *inventory.Inventory, m link.SourceMoved, now time.Time, added []string,
why string) error {
inv := open.inventory
repository := m.Owner + "/" + m.Repo
if err := keepFirstMerge(ctx, inv, m, now, false); err != nil {
return err
}
fmt.Printf("%s merged into %s (%.8s) on the operator's Approve (%s): the mesh builds nothing from it yet, so %s "+
"are registered from the builder's done keys, asking no build\n", repository, m.Base, m.Commit, why,
strings.Join(added, ", "))
// Answered: it no longer waits, so the word's arrival again does nothing twice.
if err := inv.AnswerMerge(ctx, repository, m.Commit, inventory.WaitingRecord, false); err != nil {
return err
fmt.Printf("%s merged into %s (%.8s) on the operator's Approve (%s): the mesh builds nothing from it yet, so the "+
"first build of %s is asked\n", repository, m.Base, m.Commit, why, strings.Join(added, ", "))
// The ask carries the warrant just checked, for this repository and these module paths alone: the guard of a build
// through a verb (verbMayAsk) lets exactly that through.
paths := make([]string, 0, len(added))
for _, dir := range added {
if dir == "." {
dir = ""
}
paths = append(paths, dir)
}
modules, err := newModulesOf(ctx, repository)
switch {
case err != nil:
// Said, not failed: the merge is answered, and the look at every module when a controller starts, or the next
// write of the builder or the indexer for one of them, registers them.
fmt.Printf(" the indexer's modules of %s cannot be read now (%v): they are registered when the builder's next "+
"write of one wakes the take-in\n", repository, err)
case len(modules) > 0:
inBackground(func() { takeNewModules(context.WithoutCancel(ctx), open, modules...) })
}
return nil
askBuilds(withFirstBuildWarrant(ctx, repository, paths, why), inv, m, buildSource{Repository: repository, Seat: gitSeat}, added)
// Answered: it no longer waits, so the word's arrival again asks nothing twice.
return inv.AnswerMerge(ctx, repository, m.Commit, inventory.WaitingRecord, false)
}
// resumeFirstBuilds runs firstBuilds for the first merges waiting for the word mesh-delivery just announced for a pull
@@ -111,7 +102,7 @@ func (f following) resumeFirstBuilds(ctx context.Context, d deliveryMerge) error
}
for _, w := range waiting {
k := announcedOf(w)
if !k.FirstBuild || len(k.Held) > 0 || !strings.EqualFold(k.Owner+"/"+k.Repo, d.Owner+"/"+d.Repo) || k.Number != d.Number ||
if !k.FirstBuild || !strings.EqualFold(k.Owner+"/"+k.Repo, d.Owner+"/"+d.Repo) || k.Number != d.Number ||
k.HeadSHA != d.HeadSHA {
continue
}
@@ -134,7 +125,7 @@ func (f following) resumeFirstBuilds(ctx context.Context, d deliveryMerge) error
case err != nil:
return notNow(err)
case len(added) > 0:
if err := askFirstBuilds(ctx, f.open, k.SourceMoved, time.Now().UTC(), added, why); err != nil {
if err := askFirstBuilds(ctx, inv, k.SourceMoved, time.Now().UTC(), added, why); err != nil {
return notNow(err)
}
}
@@ -181,29 +172,6 @@ func firstBuildsOf(ctx context.Context, inv catalogueReader, m link.SourceMoved,
return nil, "", err
}
// heldFirstBuild says a first build was refused because its manifest uses a word the running build-agent does not read,
// and its merge is kept waiting for a build-agent that reads it (novox/hq ADR 0482 §5, issue 1173). Read from a merge kept
// before R4e, never written since: nothing asks a first build (ADR 0501 §6), and a controller that starts answers such a
// merge and registers its modules from their keys (answerHeldMerges).
type heldFirstBuild struct {
Module string `json:"module"`
Path string `json:"path"`
Word string `json:"word"`
At time.Time `json:"at"`
}
// unknownWord is how the strict reading of a manifest says a word is not one it knows: encoding/json's own words, which
// the build-agent's answer carries whole. The build's answer has no typed field for it (Failed is the builder's words).
var unknownWord = regexp.MustCompile(`json: unknown field "([^"]+)"`)
// refusedWord is the word a build's failure says the reader did not know; empty when it says none.
func refusedWord(failed string) string {
if m := unknownWord.FindStringSubmatch(failed); m != nil {
return m[1]
}
return ""
}
// probeFirstBuildsID is the probe that reads the first merges waiting for mesh-delivery's word.
const probeFirstBuildsID = "D18"
@@ -235,10 +203,9 @@ func firstBuildObservations(waiting []inventory.BatchedMerge, now time.Time) []c
Headline: fmt.Sprintf("The first build of %s is not asked: no Approve was announced", repository),
Explanation: fmt.Sprintf("The merge %.8s into %s adds modules the mesh builds nothing from yet. It registers them "+
"only when mesh-delivery announces that it merged the pull request on your Approve of that head, and no such "+
"announcement has come in %s. The merge is kept, and the modules are registered from the builder's keys the moment "+
"the announcement is kept.",
"announcement has come in %s. The merge is kept, and the builds are asked the moment the announcement is kept.",
k.Commit, repository, for_),
Resolved: "mesh-delivery announces the merge, or the operator registers the modules at the controller's terminal",
Resolved: "mesh-delivery announces the merge, or a hand `build` registers the modules",
})
}
return out
@@ -250,28 +217,5 @@ func probeFirstBuilds(ctx context.Context, d *doctor) ([]conditions.Observation,
if err != nil {
return nil, err
}
now := time.Now().UTC()
return append(firstBuildObservations(waiting, now), newModuleObservations(now)...), nil
}
// answerHeldMerges answers every merge a controller of before held for a build-agent that reads its new modules'
// manifests (issue 1173): nothing asks their first builds any more, and the look at every module when this controller
// starts registers them from their keys (ADR 0501 §6, §7).
func answerHeldMerges(ctx context.Context, inv *inventory.Inventory) {
waiting, err := inv.WaitingMerges(ctx)
if err != nil {
fmt.Printf("the merges held for a newer build-agent could not be read: %v\n", err)
return
}
for _, w := range waiting {
if k := announcedOf(w); len(k.Held) > 0 {
if err := inv.AnswerMerge(ctx, w.Repository, w.Commit, inventory.WaitingRecord, false); err != nil {
fmt.Printf("the merge %.8s of %s held for a newer build-agent could not be answered: %v\n", w.Commit,
w.Repository, err)
continue
}
fmt.Printf("the merge %.8s of %s was held for a newer build-agent; its modules are registered from the "+
"builder's keys instead\n", w.Commit, w.Repository)
}
}
return firstBuildObservations(waiting, time.Now().UTC()), nil
}
@@ -5,48 +5,71 @@ package main
import (
"context"
"encoding/json"
"slices"
"sync"
"testing"
"time"
"github.com/nats-io/nats.go/jetstream"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
"github.com/novox/mesh-controller/internal/testbus"
)
// 1146/AC7 on keys (ADR 0501 §3, §6; issue 1851, A5), whole against a real bus and the store, in both orders of the two
// events: mesh-delivery #7 merged on the operator's Approve into a repository the mesh builds nothing from. The forge's
// merge, heard as the controller hears it (SourceMoved), and mesh-delivery's word, published on the subject the
// controller consumes with the router's record of the ask in its bucket, come in either order; each way the module the
// merge commit holds is registered from the builder's done key of it, and nothing asks a build. Without the Approve in
// the record it is not registered. Store-backed: skips without one.
func TestAMergeOnTheApproveIntoANewRepositoryRegistersItsModuleFromItsKey(t *testing.T) {
// 1146/AC7, whole against a real bus and the store, in both orders of the two events (novox/hq ADR 0432, issue 1146):
// mesh-delivery #7 merged on the operator's Approve into a repository the mesh builds nothing from. The forge's merge,
// heard as the controller hears it (SourceMoved), and mesh-delivery's word, published on the subject the controller
// consumes with the router's record of the ask in its bucket, come in either order; each way the first build of the
// module the merge commit holds is asked exactly once, its outcome registers the module, and it is in the catalogue.
// Without the Approve in the record nothing is asked. Store-backed: skips without one.
func TestAMergeOnTheApproveIntoANewRepositoryAsksTheFirstBuildAndItsOutcomeRegistersTheModule(t *testing.T) {
const head, commit = "a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1a1", "0177ea7a00000000"
for _, c := range []struct {
what string
wordFirst bool
approved bool // the router's record holds the operator's Approve of the merged head
wantRegister bool
what string
wordFirst bool
approved bool // the router's record holds the operator's Approve of the merged head
wantRegiter bool
}{
{what: "the forge's merge before mesh-delivery's word", approved: true, wantRegister: true},
{what: "mesh-delivery's word before the forge's merge", wordFirst: true, approved: true, wantRegister: true},
{what: "the forge's merge before mesh-delivery's word", approved: true, wantRegiter: true},
{what: "mesh-delivery's word before the forge's merge", wordFirst: true, approved: true, wantRegiter: true},
{what: "a word whose record holds no Approve", wordFirst: true},
} {
t.Run(c.what, func(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
inv := open.inventory
noBuildAsked(t)
b := aBuilderOnTheBus(t)
js := b.js
// The real ask, guard and queue included: only its answers are kept, so the build's outcome can be matched.
aForgeAndABuildSeat(t, open)
// Written by the controller's merge path on its own goroutine, read by the test: under one lock.
var mu sync.Mutex
var asks [][3]string
var ids []string
asked := func() [][3]string { mu.Lock(); defer mu.Unlock(); return slices.Clone(asks) }
realAsk := askABuild
askABuild = func(c context.Context, source buildSource, path, ref string) (string, error) {
id, err := realAsk(c, source, path, ref)
if err == nil {
mu.Lock()
asks = append(asks, [3]string{source.Repository, path, ref})
ids = append(ids, id)
mu.Unlock()
}
return id, err
}
t.Cleanup(func() { askABuild = realAsk })
if err := inv.RegisterModule(ctx, catalogue.Manifest{Module: "operator-channel", Version: "1",
DefinesSeats: []catalogue.SeatDeclaration{{Name: broker.AsksSeat, Records: []string{"asks"}}}},
inventory.Source{Repository: "novox/mesh-catalog", Seat: "git", Path: "modules/operator-channel", BuiltFrom: "c0ffee"}); err != nil {
t.Fatal(err)
}
js, err := broker.Dial(testbus.URL(t))
if err != nil {
t.Fatal(err)
}
t.Cleanup(js.Close)
if err := broker.AssertMeshStreams(js); err != nil {
t.Fatal(err)
}
@@ -64,14 +87,10 @@ func TestAMergeOnTheApproveIntoANewRepositoryRegistersItsModuleFromItsKey(t *tes
a := aDeliveryMergeOf(t, head, map[string]string{"novox/mesh-delivery@a1a1a1a1a1a1": head})
if c.approved {
record, _ := json.Marshal(map[string]any{"state": "closed", "ask": a.ask, "warrant": a.warrant})
if _, err := records.Put(ctx, sdkdelivery.TellerModule+"."+a.merge.AskID, record); err != nil {
if _, err := records.Put(ctx, theDelivery+"."+a.merge.AskID, record); err != nil {
t.Fatal(err)
}
}
// The indexer's value and the builder's done key of the module the merge commit holds.
fp := fpOf('9')
manifest := b.indexed(t, "mesh-delivery-view", "1", "novox/mesh-delivery", ".", commit, fp)
b.built(t, "mesh-delivery-view", "1", fp, manifest, sdkdelivery.BuildDone)
following := following{open: open}
server := link.ConnectNats(js, nil, noListener{})
@@ -85,6 +104,8 @@ func TestAMergeOnTheApproveIntoANewRepositoryRegistersItsModuleFromItsKey(t *tes
done := make(chan struct{})
go func() { defer close(done); _ = server.Serve(serving) }()
t.Cleanup(func() { stop(); <-done })
servingBus.Store(js)
t.Cleanup(func() { servingBus.Store(nil) })
word := func() {
t.Helper()
@@ -112,27 +133,37 @@ func TestAMergeOnTheApproveIntoANewRepositoryRegistersItsModuleFromItsKey(t *tes
forge()
} else {
forge()
time.Sleep(300 * time.Millisecond)
if _, known := inventoryHolds(t, inv, "mesh-delivery-view"); known {
t.Fatal("1146/AC7: registered before mesh-delivery's word was kept")
if len(asked()) != 0 {
t.Fatalf("1146/AC7: the merge asked %v before mesh-delivery's word was kept", asked())
}
word()
}
if !c.wantRegister {
time.Sleep(time.Second)
if _, known := inventoryHolds(t, inv, "mesh-delivery-view"); known {
t.Fatal("1146/AC7: a merge whose record holds no Approve registered its module")
if !c.wantRegiter {
if len(asked()) != 0 {
t.Fatalf("1146/AC7: a merge whose record holds no Approve asked %v", asked())
}
return
}
becomes(t, 10*time.Second, "the module registered", func() bool {
_, known := inventoryHolds(t, inv, "mesh-delivery-view")
return known
})
want := [3]string{"novox/mesh-delivery", "", "main"}
becomes(t, 10*time.Second, "the first build asked", func() bool { return len(asked()) >= 1 })
if got := asked(); len(got) != 1 || got[0] != want {
t.Fatalf("1146/AC7: the merge asked %v, want %v once", got, want)
}
mu.Lock()
first := ids[0]
mu.Unlock()
build := onTrunk(first, "novox/mesh-delivery", "git", "", map[string]any{"module": "mesh-delivery-view", "version": "1"})
build.Commit = commit
if _, _, err := takeIn(ctx, inv, build); err != nil {
t.Fatalf("1146/AC7: the outcome of the first build was refused: %v", err)
}
src, err := inv.SourceOf(ctx, "mesh-delivery-view")
if err != nil || src.Repository != "novox/mesh-delivery" {
t.Fatalf("1146/AC7: the module is built from %q (%v)", src.Repository, err)
}
if _, known := inventoryHolds(t, inv, "mesh-delivery-view"); !known {
t.Fatal("1146/AC7: the module is not in the catalogue")
}
})
}
}
+2 -2
View File
@@ -64,7 +64,7 @@ func registerFrom(t *testing.T, forms *sourceForms) (int64, error) {
mesh := &aMeshRegisteredFromTheURL{anEmptyMesh{asked: true,
request: inventory.BuildRequest{ID: "build-1", Repository: aRepoOnTheForge, For: "build"}}, aRepoOnTheForge}
return mayRegisterIn(t.Context(), mesh, forms, "mesh-delivery-view", inventory.Source{Repository: aRepoOnTheForge},
registrationAsk{Build: "build-1"}, "", "main")
"build-1", "", "main")
}
// 1277/AC2: an address unread twice, then read, is waited out and the forge is asked.
@@ -112,7 +112,7 @@ func TestAKnownAddressAndAnotherHostIsStillNotOnTheForge(t *testing.T) {
const elsewhere = "http://elsewhere.example/x/y.git"
mesh := &aMeshRegisteredFromTheURL{anEmptyMesh{asked: true, request: inventory.BuildRequest{ID: "build-1", For: "build"}}, elsewhere}
_, err := mayRegisterIn(t.Context(), mesh, sourceFormsRead(context.Background(), time.Now, read), "mesh-delivery-view",
inventory.Source{Repository: elsewhere}, registrationAsk{Build: "build-1"}, "", "main")
inventory.Source{Repository: elsewhere}, "build-1", "", "main")
if !errors.Is(err, errNotItsSource) || !strings.Contains(err.Error(), "is not on the mesh's forge") {
t.Fatalf("1277/AC4: the refusal changed: %v", err)
}
+1 -1
View File
@@ -68,7 +68,7 @@ func TestTheForgeNotAnsweringInTimeIsNotAskedAgain(t *testing.T) {
// An ask that hangs past the bound is cut off at the bound, not at its own wait, and the failure still says it
// waited: the receive loop is held no longer than the bound (novox/hq issue 486).
func TestAnAskHangingPastTheBoundIsCutOffAtIt(t *testing.T) {
w := waitOut{bound: 100 * time.Millisecond, first: 10 * time.Millisecond, most: 50 * time.Millisecond,
w := waitOut{bound: 200 * time.Millisecond, first: 10 * time.Millisecond, most: 50 * time.Millisecond,
now: time.Now, sleep: sleepOrDone}
asked := 0
began := time.Now()
+2 -6
View File
@@ -444,12 +444,8 @@ func TestAMergeThatDeletesAModulePlansNothingToBuildForIt(t *testing.T) {
if err := inv.SavePlan(ctx, &plan); err != nil {
t.Fatal(err)
}
if err := inv.RecordBuild(ctx, inventory.Build{ID: "build-k", Module: "kept", Repository: "novox/mesh-catalog",
At: time.Now().UTC(), Asked: asked0, Failed: "http://forge/novox/mesh-catalog.git has no module.json at modules/kept, so " +
"there is nothing saying what it is: open …/module.json: no such file or directory"}); err != nil {
t.Fatal(err)
}
advancePlans(ctx, open)
planBuilt(ctx, open, "kept", "", "http://forge/novox/mesh-catalog.git has no module.json at modules/kept, so "+
"there is nothing saying what it is: open …/module.json: no such file or directory", asked0, "build-k")
p, _ := inv.PlanByID(ctx, "plan-deleted")
if p.State == inventory.PlanFailed || p.Modules["kept"].State != planDeleted {
t.Fatalf("a module deleted at its source failed the plan: %s %+v", p.State, p.Modules["kept"])
-168
View File
@@ -1,168 +0,0 @@
package main
import (
"context"
"errors"
"fmt"
"sort"
"strings"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/inventory"
)
// The grants issued (novox/hq ADR 0528 clause 2, ADR 0429 §6 rule 6, issue 1394): what the assigner's sixth rule reads,
// that a module's bus credential for a fingerprint exists. A module's permissions on the bus are what the bus's user list
// composes from its registered manifest (inv.BusRecords), and a module's manifest is registered at a fingerprint when the
// controller takes it in from the builder's key; so once the list is sent to the machine that runs the bus, the
// credential of every module it carried is issued at the fingerprint of the build it is registered at. The controller
// writes `<module>.<fingerprint hex>` then, naming the accounts it composed for the module, never removed. A module one
// of whose accounts has no password yet (the list leaves it out) has no grant until a later send carries it.
// grantsFrom is the grant of every module whose accounts all have a password in hashes, at the fingerprint current names
// for it; a module whose current build names no fingerprint of the new form has none. principals are the bus's users
// as broker.Users composes them.
func grantsFrom(principals []broker.Principal, hashes map[string]string, current map[string]string) map[string]sdkdelivery.ControllerGrant {
accounts, missing := map[string][]string{}, map[string]bool{}
for _, p := range principals {
if p.Module == "" || (p.Kind != broker.KindModule && p.Kind != broker.KindProcess) {
continue
}
if hashes[p.Username()] == "" {
missing[p.Module] = true
continue
}
accounts[p.Module] = append(accounts[p.Module], p.Username())
}
out := map[string]sdkdelivery.ControllerGrant{}
for module, fp := range current {
if missing[module] || !strings.HasPrefix(fp, sdkdelivery.SourcePrefix2) {
continue
}
key, err := sdkdelivery.ControllerGrantKey(module, fp)
if err != nil {
continue
}
names := accounts[module]
sort.Strings(names)
out[key] = sdkdelivery.ControllerGrant{Accounts: names}
}
return out
}
// currentFingerprints is, by module, the fingerprint of the build each registered module is registered at: its
// manifest's commit (built_from), that commit's newest successful build, and the fingerprint that build names.
func currentFingerprints(ctx context.Context, inv *inventory.Inventory) (map[string]string, error) {
builds, err := inv.CurrentBuilds(ctx)
if err != nil {
return nil, err
}
out := map[string]string{}
for module, b := range builds {
if b.Commit == "" {
continue
}
id, err := inv.BuildOf(ctx, module, b.Commit)
if err != nil || id == "" {
continue
}
build, found, err := inv.BuildByID(ctx, id)
if err != nil || !found {
continue
}
out[module] = build.SourceFingerprint
}
return out, nil
}
// grantsKV is what writing the grants needs of their bucket.
type grantsKV interface {
Get(ctx context.Context, key string) (jetstream.KeyValueEntry, error)
Create(ctx context.Context, key string, value []byte, opts ...jetstream.KVCreateOpt) (uint64, error)
Update(ctx context.Context, key string, value []byte, revision uint64) (uint64, error)
}
// keepGrants writes every grant the list just sent issued that the bucket does not hold as it is: a create for a new one,
// a compare-and-set for one whose accounts changed. It answers how many it wrote.
func keepGrants(ctx context.Context, kv grantsKV, grants map[string]sdkdelivery.ControllerGrant) (int, error) {
keys := make([]string, 0, len(grants))
for k := range grants {
keys = append(keys, k)
}
sort.Strings(keys)
wrote := 0
var errs []error
for _, key := range keys {
value, err := sdkdelivery.EncodeControllerGrant(grants[key])
if err != nil {
errs = append(errs, fmt.Errorf("%s: %w", key, err))
continue
}
e, err := kv.Get(ctx, key)
switch {
case errors.Is(err, jetstream.ErrKeyNotFound):
_, err = kv.Create(ctx, key, value)
case err != nil:
default:
if before, derr := sdkdelivery.DecodeControllerGrant(e.Value()); derr == nil && sdkdelivery.SameControllerGrant(before, withVersion(grants[key])) {
continue
}
_, err = kv.Update(ctx, key, value, e.Revision())
}
if err != nil {
errs = append(errs, fmt.Errorf("%s: %w", key, err))
continue
}
wrote++
}
return wrote, errors.Join(errs...)
}
func withVersion(g sdkdelivery.ControllerGrant) sdkdelivery.ControllerGrant {
g.V = sdkdelivery.Version
return g
}
// grantsIssued writes the grants the bus's user list just sent to its machine issued (ADR 0528 clause 2). Said, never
// fatal: a grant not written now is written at the next send of the list, and the assigner's wants wait for it.
func grantsIssued(ctx context.Context, inv *inventory.Inventory, js jetstream.JetStream, indent string) {
kv, err := js.KeyValue(ctx, broker.GrantsBucket)
if err != nil {
fmt.Printf("%sthe grants issued are not written: the bucket %s: %v\n", indent, broker.GrantsBucket, err)
return
}
records, err := inv.BusRecords(ctx)
if err != nil {
fmt.Printf("%sthe grants issued are not written: %v\n", indent, err)
return
}
principals, err := broker.Users(records)
if err != nil {
fmt.Printf("%sthe grants issued are not written: %v\n", indent, err)
return
}
kept, err := inv.BusUsers(ctx)
if err != nil {
fmt.Printf("%sthe grants issued are not written: %v\n", indent, err)
return
}
hashes := make(map[string]string, len(kept))
for name, u := range kept {
hashes[name] = u.PasswordHash
}
current, err := currentFingerprints(ctx, inv)
if err != nil {
fmt.Printf("%sthe grants issued are not written: %v\n", indent, err)
return
}
wrote, err := keepGrants(ctx, kv, grantsFrom(principals, hashes, current))
if wrote > 0 {
fmt.Printf("%s%d grant(s) issued written for the assigner\n", indent, wrote)
}
if err != nil {
fmt.Printf("%ssome grants issued are not written: %v\n", indent, err)
}
}
@@ -1,91 +0,0 @@
//go:build integration
package main
import (
"context"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/nats-io/nats.go"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
"github.com/novox/mesh-controller/internal/testbus"
)
// 1394/AC1 of C9 (ADR 0528 clause 2): when the controller has issued mesh-x's credential for fingerprint F (mesh-x
// registered at a build of F, its account minted, the bus's machine sent its declaration with the user list, through
// overTheBus.declare as a push sends it), then grants.mesh-x.F exists on the bus,
// and a second pass writes it again as nothing new.
func TestTheGrantOfAnIssuedCredentialIsOnTheBus(t *testing.T) {
open := aMesh(t)
ctx := t.Context()
inv := open.inventory
fp := grantFP('d')
m := catalogue.Manifest{Module: "mesh-x", Version: "1", OwnSecrets: catalogue.OwnSecrets{"broker": {Path: "/var/lib/mesh-x/broker"}}}
if err := inv.RecordBuild(ctx, inventory.Build{ID: "build-x1", Repository: "novox/mesh-x", Module: "mesh-x", Commit: "c0ffee1",
SourceFingerprint: fp}); err != nil {
t.Fatal(err)
}
if err := inv.RegisterModule(ctx, m, inventory.Source{Repository: "novox/mesh-x", BuiltFrom: "c0ffee1", Head: "c0ffee1"}); err != nil {
t.Fatal(err)
}
if _, err := assign(ctx, open, "anchor", "mesh-x"); err != nil {
t.Fatal(err)
}
user := broker.Principal{Kind: broker.KindModule, Node: "anchor", Module: "mesh-x"}.Username()
if _, err := inv.MintBusPassword(ctx, inventory.BusUser{Username: user, Kind: inventory.BusModule, Node: "anchor", Module: "mesh-x"}); err != nil {
t.Fatal(err)
}
conn, err := nats.Connect(testbus.URL(t))
if err != nil {
t.Fatal(err)
}
t.Cleanup(conn.Close)
js, err := jetstream.New(conn)
if err != nil {
t.Fatal(err)
}
if err := broker.OnConn(conn).EnsureControllerBuckets(); err != nil {
t.Fatal(err)
}
// The live path: a send of the bus's machine's declaration carrying the user list (overTheBus.declare), which records
// the list sent and then writes the grants it issued.
legacy, err := conn.JetStream()
if err != nil {
t.Fatal(err)
}
_ = js.DeleteStream(ctx, "GRANTS_DECLARED")
if _, err := js.CreateStream(ctx, jetstream.StreamConfig{Name: "GRANTS_DECLARED", Subjects: []string{link.DeclareSubject("anchor")}}); err != nil {
t.Fatal(err)
}
t.Cleanup(func() { _ = js.DeleteStream(context.Background(), "GRANTS_DECLARED") })
send := overTheBus{open: open, server: link.SendingOn(link.OverNATS{Conn: conn, JS: legacy}), signer: aSignerOfZeros{}}
ready := readyNode{node: "anchor", declared: sendable{BusUsers: "the user list the bus's machine is sent"}}
if _, err := send.declare(ctx, ready, []byte(`{}`)); err != nil {
t.Fatalf("the send of the bus's machine: %v", err)
}
kv, err := js.KeyValue(ctx, broker.GrantsBucket)
if err != nil {
t.Fatal(err)
}
key, _ := sdkdelivery.ControllerGrantKey("mesh-x", fp)
e, err := kv.Get(ctx, key)
if err != nil {
t.Fatalf("no grant %s after the list was issued: %v", key, err)
}
g, err := sdkdelivery.DecodeControllerGrant(e.Value())
if err != nil || len(g.Accounts) != 1 || g.Accounts[0] != user {
t.Fatalf("the grant reads %+v, %v; want the account %s", g, err, user)
}
if _, err := send.declare(ctx, ready, []byte(`{}`)); err != nil {
t.Fatalf("the second send: %v", err)
}
if again, _ := kv.Get(ctx, key); again.Revision() != e.Revision() {
t.Errorf("a second pass wrote the grant again (revision %d, was %d)", again.Revision(), e.Revision())
}
}
+34
View File
@@ -9,7 +9,9 @@ import (
"strings"
"testing"
"git.novox.be/novox/mesh-sdk/go/catalogue"
"github.com/novox/mesh-controller/internal/inventory"
"github.com/novox/mesh-controller/internal/link"
)
// The grants reach the bus before the gate (novox/hq issue 490).
@@ -174,3 +176,35 @@ func TestTheGrantsStepKeepsOneMachinesBuilds(t *testing.T) {
t.Error("the grants step keeps the builds of a machine it does not send")
}
}
// The delivery plan says the step before the merge.
func TestAChangePlanSaysTheGrantsStep(t *testing.T) {
const catalogue_ = "http://forge.internal:20000/novox/mesh-catalog.git"
entry := func(name string, on ...string) inventory.Entry {
e := fromRepo(name, catalogue_, "modules/"+name)
e.Source.BuiltFrom = "old"
e.On = on
return e
}
nats := entry("nats", "novox")
nats.Manifest.BusUsers = "/etc/nats/users.conf"
nats.Manifest.Claims = []catalogue.Claim{{Name: catalogue.BrokerSeat, Scope: catalogue.ScopeMesh}}
restic := entry("restic", "ace", "novox", "shanks")
only := entry("bus-tools", "novox")
entries := []inventory.Entry{nats, restic, only}
rollsOut := func(string) (inventory.Upgrade, bool) { return inventory.Upgrade{RollOut: true}, true }
plan := func(paths ...string) link.ChangePlan {
m := link.SourceMoved{Owner: "novox", Repo: "mesh-catalog", Base: "main", Commit: "head", Paths: paths}
return changePlanOf("novox/mesh-catalog", "main", "head", reachOfMerge(m, entries, nil, nil), entries, rollsOut)
}
text := planText(plan("modules/restic/module.json"))
if !strings.Contains(text, "grants step: if this changes what the bus's user list says (a new tool, subject or "+
"user), novox is sent that list alone, every build there kept, before restic is judged on its first machine") {
t.Errorf("the change plan does not say the grants step:\n%s", text)
}
// A module only on the bus's machine goes there with the list in its own send: no step of its own.
if text := planText(plan("modules/bus-tools/module.json")); strings.Contains(text, "grants step") {
t.Errorf("a module on the bus's machine alone reads:\n%s", text)
}
}
-106
View File
@@ -1,106 +0,0 @@
package main
import (
"context"
"strings"
"testing"
"time"
sdkdelivery "git.novox.be/novox/mesh-sdk/go/delivery"
"github.com/nats-io/nats.go/jetstream"
"github.com/novox/mesh-controller/internal/broker"
)
func grantFP(c byte) string { return sdkdelivery.SourcePrefix2 + strings.Repeat(string(c), 64) }
// 1394/AC (unit half): the list sent issues the credential of every module whose accounts all have a password, at the
// fingerprint its build names; a module with an account the list leaves out, and one registered at a fingerprint not of
// the new form, get no grant; a module with no account of its own is granted with none.
func TestTheGrantsAListIssues(t *testing.T) {
x := broker.Principal{Kind: broker.KindModule, Node: "novox", Module: "mesh-x"}
xp := broker.Principal{Kind: broker.KindProcess, Node: "novox", Module: "mesh-x", Process: "run"}
y := broker.Principal{Kind: broker.KindModule, Node: "ace", Module: "mesh-y"}
node := broker.Principal{Kind: broker.KindNode, Node: "novox"}
hashes := map[string]string{x.Username(): "h1", xp.Username(): "h2", node.Username(): "h3"}
got := grantsFrom([]broker.Principal{x, xp, y, node}, hashes,
map[string]string{"mesh-x": grantFP('a'), "mesh-y": grantFP('b'), "web": grantFP('c'), "old": "src1:abc"})
xKey, _ := sdkdelivery.ControllerGrantKey("mesh-x", grantFP('a'))
webKey, _ := sdkdelivery.ControllerGrantKey("web", grantFP('c'))
if len(got) != 2 {
t.Fatalf("grants = %v, want mesh-x's and web's", got)
}
if g := got[xKey]; len(g.Accounts) != 2 || g.Accounts[0] != x.Username() && g.Accounts[1] != x.Username() {
t.Errorf("mesh-x's grant = %+v, want both its accounts", g)
}
if g, ok := got[webKey]; !ok || len(g.Accounts) != 0 {
t.Errorf("web, with no account of its own, is granted %+v, %v", g, ok)
}
}
// fakeGrants is the grants bucket in memory.
type fakeGrants struct {
values map[string][]byte
creates int
updates int
}
type fakeEntry struct {
key string
value []byte
rev uint64
}
func (e fakeEntry) Bucket() string { return broker.GrantsBucket }
func (e fakeEntry) Key() string { return e.key }
func (e fakeEntry) Value() []byte { return e.value }
func (e fakeEntry) Revision() uint64 { return e.rev }
func (e fakeEntry) Created() time.Time { return time.Time{} }
func (e fakeEntry) Delta() uint64 { return 0 }
func (e fakeEntry) Operation() jetstream.KeyValueOp { return jetstream.KeyValuePut }
func (f *fakeGrants) Get(_ context.Context, key string) (jetstream.KeyValueEntry, error) {
v, ok := f.values[key]
if !ok {
return nil, jetstream.ErrKeyNotFound
}
return fakeEntry{key: key, value: v, rev: 1}, nil
}
func (f *fakeGrants) Create(_ context.Context, key string, value []byte, _ ...jetstream.KVCreateOpt) (uint64, error) {
if _, ok := f.values[key]; ok {
return 0, jetstream.ErrKeyExists
}
f.values[key] = value
f.creates++
return 1, nil
}
func (f *fakeGrants) Update(_ context.Context, key string, value []byte, _ uint64) (uint64, error) {
f.values[key] = value
f.updates++
return 2, nil
}
// 1394/AC (unit half): a grant is created once; a second send of the same list writes nothing; a reissue for other
// accounts is a new revision of the same key.
func TestAGrantIsWrittenOnceAndMovedOnAReissue(t *testing.T) {
kv := &fakeGrants{values: map[string][]byte{}}
key, _ := sdkdelivery.ControllerGrantKey("mesh-x", grantFP('a'))
grants := map[string]sdkdelivery.ControllerGrant{key: {Accounts: []string{"novox.mesh-x"}}}
for range 2 {
if _, err := keepGrants(t.Context(), kv, grants); err != nil {
t.Fatal(err)
}
}
if kv.creates != 1 || kv.updates != 0 {
t.Errorf("creates %d, updates %d, want one create", kv.creates, kv.updates)
}
grants[key] = sdkdelivery.ControllerGrant{Accounts: []string{"ace.mesh-x"}}
if wrote, err := keepGrants(t.Context(), kv, grants); err != nil || wrote != 1 || kv.updates != 1 {
t.Errorf("a reissue wrote %d (%v), updates %d", wrote, err, kv.updates)
}
if g, err := sdkdelivery.DecodeControllerGrant(kv.values[key]); err != nil || g.Accounts[0] != "ace.mesh-x" {
t.Errorf("the grant reads %+v, %v", g, err)
}
}
-3
View File
@@ -64,9 +64,6 @@ var handActVerbs = []handActVerb{
{Verb: "plans stop", Decision: "the operator's answer to an ask is their decision, not a repair (ADR 0259)",
DecidedFor: []string{conditions.CauseOperatorAnswer}},
{Verb: "plans close"},
// A module registered at references by hand, because the builder had no key for it (ADR 0526 clause 3): a repair
// the builder should have made, so it counts.
{Verb: "register"},
// A walk started by a person instead of its delivery's owner (novox/hq ADR 0239): the owner down, or
// not trusted with it — either is a repair the owner should have made. Unless the operator chose it on
// a warrant (ADR 0259).

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