Tell the operator at once when sends wait for the bus's planned step, and say it before the merge (hq issue 336)
mesh/merge-gate pass: builds build-agent, mesh-controller, route-proxy → ace, g14, novox, shanks; no bus step; every machine composes with the change as it…
mesh/repo-check pass: its merge-check.sh passed
mesh/delivery superseded: a newer head of the same pull request

For 28 minutes on 2026-10-08 every send to the control-node was refused for a new bus build that only a
person's bus upgrade moves, and no condition said so: the refusal lived only in each walk's note, and S3
would have called it lateness after half an hour, in words that named neither the bus nor the verb.

- Row S17, bus.<module>.step-waiting: raised on the first watchdog tick after a walk's send is refused for
  the bus, for the operator, naming the machines, what waits, the bus build from and to, since when and
  mesh-controller.bus upgrade. It clears once the bus's machine runs the build the mesh holds.
- S3 leaves out a walk held only by the bus's step.
- A change that builds the bus says in its delivery plan and summary (which mesh/merge-gate carries) that
  merging it needs a person's bus upgrade, and that nothing else reaches its machine until then.
- TestReplay336 fails on the commit before and passes on this one.
This commit is contained in:
jochen
2026-10-09 02:33:57 +02:00
parent 155819f307
commit 1fdc68a8aa
10 changed files with 530 additions and 9 deletions
+25 -4
View File
@@ -97,6 +97,9 @@ type signalFacts struct {
// facts is the snapshot this controller keeps for merge checks (S14).
facts factsFacts
bus busFacts
busErr error
}
// factsFacts is when the newest facts snapshot was taken, when this controller began keeping it, and
@@ -147,6 +150,7 @@ type planFacts struct {
bound time.Duration
waiting string
paused bool
bus bool
}
// waitFacts is one walk waiting for its delivery's word: since its merge opened it.
@@ -320,7 +324,8 @@ func (w *watchdogs) gather(ctx context.Context) *signalFacts {
}
}
f.machines, f.machinesErr = w.gatherMachines(ctx, inv, now)
f.plans, f.waits, f.plansErr = gatherPlans(ctx, inv, now)
f.bus, f.busErr = gatherBus(ctx, inv)
f.plans, f.waits, f.plansErr = gatherPlans(ctx, inv, now, f.bus.heldByTheBus())
f.loop, f.loopErr = w.gatherLoop()
f.mergesPassed, f.merges, f.mergesErr = watchedMerges.last()
if f.mergesErr == nil && !f.mergesPassed.IsZero() && now.Sub(f.mergesPassed) > 3*mergeCatchUpEvery {
@@ -438,8 +443,24 @@ func (w *watchdogs) gatherMachines(ctx context.Context, inv *inventory.Inventory
return out, nil
}
// gatherPlans is every open plan, its tier's bound from what was measured, and what it waits on.
func gatherPlans(ctx context.Context, inv *inventory.Inventory, now time.Time) ([]planFacts, []waitFacts, error) {
// gatherBus is a new bus build waiting for its step and the walks refused for it (S17, novox/hq issue 336).
func gatherBus(ctx context.Context, inv *inventory.Inventory) (busFacts, error) {
b, err := pendingBus(ctx, inv)
if err != nil {
return busFacts{}, fmt.Errorf("what a bus upgrade would do cannot be read: %w", err)
}
plans, err := inv.OpenPlans(ctx)
if err != nil {
return busFacts{}, fmt.Errorf("the open plans cannot be read: %w", err)
}
f := busWaitsOf(plans, b, busRefusedFirst.of)
busRefusedFirst.keepOnly(f.heldByTheBus())
return f, nil
}
// gatherPlans is every open plan, its tier's bound from what was measured, and what it waits on; byTheBus are
// the walks held only by the bus's planned step, which S3 leaves to S17.
func gatherPlans(ctx context.Context, inv *inventory.Inventory, now time.Time, byTheBus map[string]bool) ([]planFacts, []waitFacts, error) {
plans, err := inv.OpenPlans(ctx)
if err != nil {
return nil, nil, fmt.Errorf("the open plans cannot be read: %w", err)
@@ -466,7 +487,7 @@ func gatherPlans(ctx context.Context, inv *inventory.Inventory, now time.Time) (
bound := bounds.of(p.Repository)
out = append(out, planFacts{id: p.ID, repository: p.Repository, commit: p.Commit, tier: p.Tier,
tiers: len(p.Tiers), entered: inTierSince(p), bound: bound,
waiting: planLineWith(p, now, pause, bound), paused: paused})
waiting: planLineWith(p, now, pause, bound), paused: paused, bus: byTheBus[p.ID]})
}
return out, waits, nil
}