Commit Graph
22 Commits
Author SHA1 Message Date
jschoubben 6771db22ae Revert L3 (mesh-delivery, merge d827420): its node-engine removed every held module on novox, the bus server among them, at 08:54 CET on 2026-10-29 (issue 774)
mesh/delivery merged: source.main moved to novox/mesh-delivery@c1b4533201c8
mesh/delivery-group group revert/774-l3 merged: every member is merged
mesh/check being carried to 6771db22ae69 on d8274204fbed | from mesh-catalog 1555241393fe | novox/mesh-delivery#109 in group revert/774-l3
L3 lands again with two fixes: a held module is never removed, and the grants cover L3's new subjects.
2026-10-29 09:42:29 +01:00
jschoubben b2351a213d Follow L2 under L3: the builder, checker, indexer and view read the seat mesh-placement's placement, the seat's Leased is the lease handed away, and the view keeps no read of the indexer's placement L2 removed (issue 774, ADR 0567 §8, ADR 0549 §3) 2026-10-28 21:54:47 +01:00
jschoubben a4116a1de7 Builder, checker and indexer count a build node's toolchain only where the catalogue's placement puts its build module, through mesh-sdk's one gate, and read the controller's assignments no more (issue 774, ADR 0549 §3, issue 1940) 2026-10-28 21:54:47 +01:00
jschoubben fc6d5dcbdb mesh-delivery-checker claims its seat and serves try, try-read and try-end by the tools' own code, so the try outlives the controller (ADR 0547 §10, issue 774) 2026-10-28 21:54:47 +01:00
jschoubben 16594feff1 checker_ask_again: refuse while an ask of the check waits in its build seat's stream
Review of fix/lost-check: an ask no node took yet has no step key, so
asking again then would run the check twice (issue 1791's kind). The
checker keeps the asks each running check put in a new state asks, and the
tool refuses unless every one of them has a step key. Asking again under the
same ask id is not done: the stream drops a second put of an id within its
de-duplication window and keeps both past it, so the waiting copy is never
replaced.

Tests: the hand row of the lost-ask table refuses while its ask waits;
counts unchanged.
2026-10-28 20:36:25 +01:00
jschoubben 6f079f13a8 Checker: never lose a step's end, and ask a lost check again (issue 734's kind)
mesh-catalog #553's verdict stayed running after its step mvsgthyjed5786
passed on shanks at 19:09:31Z: the watch of that one step key, opened at
18:46:05Z, heard neither its start nor its end and never closed, so the
wait had no bound running and nothing would ever judge it.

- HeardEnds stands between the asker and each ask's watch: at every step
  end of the same build module (one watch per steps state), at a
  reconnect, the asks waited on read their keys now and are judged; a
  watch that closes is opened again, at most 3 times.
- A step the build module's next process ended (RestartReason, a delivery
  replacing it) counts as a lost ask: asked again under the claim, at most
  LostAskTries, then failed as not heard, judged again when the nodes change.
- checker_ask_again {group, why} ends a running verdict as asked again by
  hand; the checker hears its verdicts, ends the wait and claims it anew.
  Refused while a step of its pull requests runs, or for a verdict that
  is not running.
- checker_status names such checks under lost, and each step's what.

Tests: one table replaces the given-up test (rows for given up, replaced,
deaf and nudged, deaf and asked by hand); counts unchanged (147 unit, 10
integration in the checker).
2026-10-28 20:27:41 +01:00
jschoubben 60c7ffc981 Checker, builder and grants part: hear DEAD_LETTERS again until heard, say when they do not, and compose the same grant (issue 734, review round 1)
mesh-grants moves to the same mesh-sdk (3203358c), so its shadow user list grants the checker and the builder the
DEAD_LETTERS reads the controller's list will, and the take-over is not blocked by a difference. The checker and the
builder hear DEAD_LETTERS through buildrun.DeadLetterHearing: in the background, again at every reconnect and every ask
waited for while deaf, and keep whether they hear it in their new state hearing, which checker_status and builder_stuck
say as dead_letters. checker_try says that a try ended by the not-taken bound leaves its ask in the stream.

Not done as asked: a module health tool. The catalogue refuses a tool check on a module with no runtime of its own and no
other check beside it (ADR 0227 rule 8, ADR 0366 §1), so the deaf state is said in the status tools and the log.
2026-10-26 14:44:53 +01:00
jschoubben 6b85220fa4 Checker: an ask the build seat gives up on ends its try as error and asks a check again; a try no node takes ends at the step bound; checker_try_end ends one by hand (issue 734)
A check ask handed out five times during a bus reload was given up into DEAD_LETTERS and no build node wrote a step for it,
so try-1793017340003993788 stayed running for ever, held one of the four try places, and checker_try of its head kept
answering its id. The checker now hears the copies DEAD_LETTERS keeps of the build seats' given-up asks (an ordered
consumer from its start; the grant is mesh-sdk/mesh-controller DeadLetterReads, branch fix/734-lost-ask) and ends the
step it waits for: a try ends as error saying why, a check is asked again under its claim, up to three times, then fails
as not heard. A try whose ask no node takes within the step bound ends the same way (a check's backlog stays no fault,
ADR 0525 §9). checker_try_end ends a try by hand, freeing its place and stopping its judging.
2026-10-26 14:44:40 +01:00
jschoubben c66fe11e16 checker: claim no check of a repository an approved merge holds a turn on, and claim it when the turn goes (ADR 0548 §3; issue 568 D1) 2026-10-26 11:29:25 +01:00
jschoubben 991589fdd2 A try of a branch accepts a mesh-sdk pin at the head of mesh-sdk's branch of the same name, so a milestone's repositories can be tried before their pull requests (issue 1921)
mesh/delivery delivered
mesh/check checking 991589fdd20a on 4c1d36c9cd7f
2026-10-24 16:41:39 +02:00
jschoubben 3553cbaf08 Count only the toolchain reports of nodes a build module is assigned to, so a node it left does not make the builders disagree (issue 1940)
mesh/delivery delivered
mesh-build-node was unassigned from ace on 2026-10-23 and its ace.typescript key stayed in
mesh-build-node_toolchains at v26.12.0: nothing deletes it, so the indexer, the builder and the
checker took it for a builder and every Node fingerprint said 'builders disagree'. The three now
read mesh-controller_assignments and pass a toolchain report on only while its build module has an
open assignment on its node; builders that run different versions still disagree.
2026-10-24 12:48:41 +02:00
jschoubben 527f87f94c Merge branch 'feat/r4g-asks' into feat/r4g
# Conflicts:
#	modules/mesh-delivery-checker/cmd/mesh-delivery-checker/judge.go
#	modules/mesh-delivery-checker/manifest_test.go
#	modules/mesh-delivery-checker/module.json
2026-10-23 16:20:17 +02:00
jschoubben d9d1e57e3a Delete the checker's hold, its own capacity and places, its node moves and its read of the controller's plans-building, so a check waits only in the build seats' streams (issue 1887 A5b, closes 1685; ADR 0525 §4, §7) 2026-10-23 16:09:34 +02:00
jschoubben 5091cab652 Read the assigner's composition into the checker's verdict in place of 'not yet run', failing a check whose composition failed, naming its machine (issue 1297, subtask B3d; ADR 0528 clause 5) 2026-10-23 15:27:59 +02:00
jschoubben 7abe230701 Judge a pushed head as a try on the checker, through checker_try and checker_try_read (ADR 0501 §5)
mesh-controller's try asked build-agent, which leaves the checking at R4e. A try is now a key in the checker's state
tries: the checker judges the head merged onto its trunk as it judges a pull request's head, through the same
judge, places and holds, writes its end in the key and owes the forge nothing (no status, no preview, no verdict).
One try per head, at most four at once; hq and a repository whose trunk the gate does not name are refused with
"no level of <repo> is checked by the checker".
2026-10-22 18:03:28 +02:00
jschoubben d83156fb21 Add checker_previews and checker_status read tools so the live acceptance of issues 1680 and 1685 can be read (C1, C2) 2026-10-22 17:25:11 +02:00
jschoubben b519c8dde2 Keep nothing for a workspace given siblings, place the siblings where the tests state them, decide the hold from the listed plan keys (review D1-D3)
mesh/repo-check head b519c8dde2da merged into trunk a5aa69a4e815 beside mesh-catalog#423@12ff1cfb +1 more — pass: its merge-check.sh passed
mesh/merge-gate head b519c8dde2da merged into trunk a5aa69a4e815 — pass: group feat/the-checker composed: every machine composes with the change as it did …
mesh/delivery delivered
mesh/delivery-group group feat/the-checker merged: every member is merged
2026-10-21 21:52:41 +02:00
jschoubben a20af30d54 A group's Go member naming a mesh-sdk commit that is neither on the trunk nor the group's head fails the check naming the member and the commit; a pin only a head carries names issue 1148 (issue 1624) 2026-10-21 21:34:35 +02:00
jschoubben 4cdba63b9b Read tools for the checker's verdicts, a pull request's report and the kept verdicts (issue 1686) 2026-10-21 21:32:08 +02:00
jschoubben 96ba6bd3dc Say the verdict in a line of at most 140 characters and owe the status mesh/check in the checker's bucket, which the forge bridge performs once per revision; the checker's stream checks wakes the readers of its buckets (issues 1682, 1683) 2026-10-21 21:20:42 +02:00
jschoubben 4242fb5b0b The indexer gives a candidate's touched module the fingerprint its merge preview gives it, and wakes on the checker's preview.written without a restart; pin mesh-sdk to the R4c-5a head (issues 1445, 1681) 2026-10-21 21:17:09 +02:00
jschoubben 27072a01b5 Add the checker module: it watches the candidates, claims a verdict once, owes the status mesh/check and ends a check past its bound (issue 1678) 2026-10-21 21:07:14 +02:00