Build with the toolchain a majority of build nodes report; say which nodes sit out (ADR 0584, issue 1578)
One build node on another version refused every build of the language. The builder and the checker now follow mesh-sdk's majority agreement; the key's step names a node that sits out ("ace sits out: typescript 26.12.1, agreed 26.13.0"), and a build ask a minority node declined is put again for a node on the agreed version. Check steps compare no toolchain, so clause 2 already holds for them. Pins builder, indexer and checker to mesh-sdk 71be15f.
Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL
This commit is contained in:
@@ -3,7 +3,7 @@ module mesh-delivery-builder
|
||||
go 1.26.0
|
||||
|
||||
require (
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101085317-790d1c82d027
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101142413-71be15f53066
|
||||
github.com/nats-io/nats-server/v2 v2.11.17
|
||||
github.com/nats-io/nats.go v1.54.0
|
||||
)
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101085317-790d1c82d027 h1:a33dsavoaqllh40eVICbhxi6Kero1EGSjt1V+2fa2Hk=
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101085317-790d1c82d027/go.mod h1:ouERNpZizzwZ0HLEtAypsY9FR20FyUgUYRni6J7kVCY=
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101142413-71be15f53066 h1:rpYDAiJ+245xHvmYdta/hE3gSGEoebglr4ujdOoFH/8=
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101142413-71be15f53066/go.mod h1:ouERNpZizzwZ0HLEtAypsY9FR20FyUgUYRni6J7kVCY=
|
||||
github.com/antithesishq/antithesis-sdk-go v0.7.0-default-no-op h1:Z/MZK75wC/NSrkgqeNIa7jexam9uWzhLmFTSCPI/kn0=
|
||||
github.com/antithesishq/antithesis-sdk-go v0.7.0-default-no-op/go.mod h1:FQyySiasQQM8735Ddel3MRojmy4dA1IqCeyJ5jmPMbI=
|
||||
github.com/google/go-tpm v0.9.8 h1:slArAR9Ft+1ybZu0lBwpSmpwhRXaa85hWtMinMyRAWo=
|
||||
|
||||
@@ -317,6 +317,11 @@ func (b *Builder) waitOnce(ctx context.Context, ask delivery.StepAsk, again <-ch
|
||||
defer stop()
|
||||
}
|
||||
step, err := b.Asks.Wait(wait, ask)
|
||||
if d, declined := delivery.DeclinedOf(step); err == nil && declined {
|
||||
// A build node on another toolchain than the agreed one sat out (ADR 0584 clause 2): the ask is put again, for a node
|
||||
// on the agreed one to take. The most askings stays the bound, and the build then fails with these reasons in it.
|
||||
return step, strings.TrimSpace(step.Node + " declined: " + step.Reason + " (lacks " + strings.Join(d.Lacks, ", ") + ")"), errAgain
|
||||
}
|
||||
if err == nil && buildrun.IsLost(step) {
|
||||
// The build seat gave the ask up (issue 734): it can no longer be answered, so it is asked again now, with the
|
||||
// words of why, not at the check bound.
|
||||
|
||||
@@ -506,10 +506,11 @@ func (b *Builder) stale(m delivery.Module, plan Plan) string {
|
||||
continue // an archive is packed with no toolchain
|
||||
}
|
||||
language := buildrun.LanguageOf[bm]
|
||||
tc, none := delivery.AgreedToolchain(b.Toolchains.Entries(language))
|
||||
if none != "" {
|
||||
return fmt.Sprintf("the %s toolchain is not agreed: %s", language, none)
|
||||
ag := delivery.ToolchainAgreement(b.Toolchains.Entries(language))
|
||||
if ag.Wait != "" {
|
||||
return fmt.Sprintf("the %s toolchain is not agreed: %s", language, ag.Wait)
|
||||
}
|
||||
tc := ag.Toolchain
|
||||
want, ok := m.Toolchains[language]
|
||||
if !ok || want.Version != tc.Version || want.Recipe != tc.Recipe {
|
||||
return fmt.Sprintf("the fingerprint names %s %s (recipe %s) and the nodes agree on %s (recipe %s)", language, want.Version, want.Recipe, tc.Version, tc.Recipe)
|
||||
@@ -543,7 +544,8 @@ func (b *Builder) build(ctx context.Context, module string, m delivery.Module, p
|
||||
repository := delivery.RepositoryName(m.Repository)
|
||||
var lanes []*lane
|
||||
used := map[string]delivery.ToolchainUsed{}
|
||||
var kept []string // build modules whose artifacts stay as the prior done key has them
|
||||
var sitting []string // the nodes that sit out of a language's build, said in the key's step (ADR 0584 clause 2)
|
||||
var kept []string // build modules whose artifacts stay as the prior done key has them
|
||||
for _, bm := range plan.Modules() {
|
||||
if prior != nil && len(only) > 0 && !only[bm] {
|
||||
kept = append(kept, bm)
|
||||
@@ -553,10 +555,12 @@ func (b *Builder) build(ctx context.Context, module string, m delivery.Module, p
|
||||
if bm != ImageBuildModule && plan.NeedsToolchain(bm) {
|
||||
// The ask names the toolchain the fingerprint was computed with; a node building with another refuses it.
|
||||
language := buildrun.LanguageOf[bm]
|
||||
tc, none := delivery.AgreedToolchain(b.Toolchains.Entries(language))
|
||||
if none != "" {
|
||||
return failed("the %s toolchain is not agreed: %s", language, none)
|
||||
ag := delivery.ToolchainAgreement(b.Toolchains.Entries(language))
|
||||
if ag.Wait != "" {
|
||||
return failed("the %s toolchain is not agreed: %s", language, ag.Wait)
|
||||
}
|
||||
tc := ag.Toolchain
|
||||
sitting = append(sitting, SittingOutWords(language, ag)...)
|
||||
if want := m.Toolchains[language]; want.Version != tc.Version || want.Recipe != tc.Recipe {
|
||||
return delivery.BuilderBuild{} // the agreed toolchain moved since try looked: nothing is written
|
||||
}
|
||||
@@ -653,6 +657,9 @@ func (b *Builder) build(ctx context.Context, module string, m delivery.Module, p
|
||||
}
|
||||
}
|
||||
step := "built " + strings.Join(steps, "; ")
|
||||
if len(sitting) > 0 {
|
||||
step += " (" + strings.Join(sitting, "; ") + ")"
|
||||
}
|
||||
if len(used) == 0 {
|
||||
used = nil
|
||||
}
|
||||
@@ -942,3 +949,13 @@ func LogEnded(module string, m delivery.Module, ref, by string, at time.Time, re
|
||||
e.Made = append([]delivery.BuiltArtifact(nil), result.Artifacts...)
|
||||
return e
|
||||
}
|
||||
|
||||
// SittingOutWords says, per node that is not on the agreed toolchain of a language, that it sits out and which version
|
||||
// it has against the agreed one: "ace sits out: typescript 26.12.1, agreed 26.13.0" (novox/hq ADR 0584 clause 2).
|
||||
func SittingOutWords(language string, a delivery.Agreement) []string {
|
||||
var out []string
|
||||
for _, s := range a.SittingOut {
|
||||
out = append(out, fmt.Sprintf("%s sits out: %s %s, agreed %s", s.Node, language, s.Version, a.Toolchain.Version))
|
||||
}
|
||||
return out
|
||||
}
|
||||
|
||||
@@ -223,6 +223,14 @@ var (
|
||||
|
||||
func TestABundleAndAnArchiveAreBuiltOnOneNodeAndDone(t *testing.T) { // 1607/AC7, ADR 0478 clauses 1, 4
|
||||
r := newRig(t, goBundle, archiveA, tsBundle)
|
||||
// ADR 0584: a third node agrees with novox and ace's typescript is the odd one; the build is done and its step says ace sits out.
|
||||
r.tools.Put(buildrun.NodeBuildModule, "g14.typescript", tcValue(t, "typescript", "tsc5"), false)
|
||||
r.tools.Put(buildrun.NodeBuildModule, "ace.typescript", tcValue(t, "typescript", "tsc4"), false)
|
||||
defer func() {
|
||||
if st := r.st.vals[r.key]; !strings.Contains(string(st), "ace sits out: typescript tsc4, agreed tsc5") {
|
||||
t.Errorf("the key does not say ace sits out: %s", st)
|
||||
}
|
||||
}()
|
||||
var logged []delivery.BuildLogEntry
|
||||
var lmu sync.Mutex
|
||||
r.b.Node, r.b.Log = "novox", func(e delivery.BuildLogEntry) { lmu.Lock(); logged = append(logged, e); lmu.Unlock() }
|
||||
|
||||
@@ -3,7 +3,7 @@ module mesh-delivery-checker
|
||||
go 1.26.0
|
||||
|
||||
require (
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101022206-8af6e7266915
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101142413-71be15f53066
|
||||
github.com/nats-io/nats-server/v2 v2.11.17
|
||||
github.com/nats-io/nats.go v1.54.0
|
||||
)
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101022206-8af6e7266915 h1:x8k7IeMirMTxUo8BH+xUO+MDnHYgU+636jl1NIw1wvs=
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101022206-8af6e7266915/go.mod h1:ouERNpZizzwZ0HLEtAypsY9FR20FyUgUYRni6J7kVCY=
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101142413-71be15f53066 h1:rpYDAiJ+245xHvmYdta/hE3gSGEoebglr4ujdOoFH/8=
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101142413-71be15f53066/go.mod h1:ouERNpZizzwZ0HLEtAypsY9FR20FyUgUYRni6J7kVCY=
|
||||
github.com/antithesishq/antithesis-sdk-go v0.7.0-default-no-op h1:Z/MZK75wC/NSrkgqeNIa7jexam9uWzhLmFTSCPI/kn0=
|
||||
github.com/antithesishq/antithesis-sdk-go v0.7.0-default-no-op/go.mod h1:FQyySiasQQM8735Ddel3MRojmy4dA1IqCeyJ5jmPMbI=
|
||||
github.com/google/go-tpm v0.9.8 h1:slArAR9Ft+1ybZu0lBwpSmpwhRXaa85hWtMinMyRAWo=
|
||||
|
||||
@@ -3,7 +3,7 @@ module mesh-delivery-indexer
|
||||
go 1.26.0
|
||||
|
||||
require (
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101022206-8af6e7266915
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101142413-71be15f53066
|
||||
github.com/nats-io/nats-server/v2 v2.11.17
|
||||
github.com/nats-io/nats.go v1.54.0
|
||||
)
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101022206-8af6e7266915 h1:x8k7IeMirMTxUo8BH+xUO+MDnHYgU+636jl1NIw1wvs=
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101022206-8af6e7266915/go.mod h1:ouERNpZizzwZ0HLEtAypsY9FR20FyUgUYRni6J7kVCY=
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101142413-71be15f53066 h1:rpYDAiJ+245xHvmYdta/hE3gSGEoebglr4ujdOoFH/8=
|
||||
git.novox.be/novox/mesh-sdk/go v0.1.14-0.20261101142413-71be15f53066/go.mod h1:ouERNpZizzwZ0HLEtAypsY9FR20FyUgUYRni6J7kVCY=
|
||||
github.com/antithesishq/antithesis-sdk-go v0.7.0-default-no-op h1:Z/MZK75wC/NSrkgqeNIa7jexam9uWzhLmFTSCPI/kn0=
|
||||
github.com/antithesishq/antithesis-sdk-go v0.7.0-default-no-op/go.mod h1:FQyySiasQQM8735Ddel3MRojmy4dA1IqCeyJ5jmPMbI=
|
||||
github.com/google/go-tpm v0.9.8 h1:slArAR9Ft+1ybZu0lBwpSmpwhRXaa85hWtMinMyRAWo=
|
||||
|
||||
Reference in New Issue
Block a user