Tests no longer race the wall clock; the checker invokes compose-candidate again

Conversation 1 Commits 2 Files Changed 18

+176 -125

2 Commits

Author	SHA1	Message	Date
jschoubben	34a607a1f5	Give the checker back its compose-candidate invoke, now that the new indexer runs mesh/delivery merged: source.main moved to novox/mesh-delivery@a62276bcafad Details mesh/check being carried to 34a607a1f5cf on a62276bcafad \| from 15d28ddf5bd3 The invoke was left out (mesh-delivery #142) only while the indexer built before the verb ran; the new indexer and the node-engine's compose-candidate run on every machine since 22:57, so a check composes on the machines again instead of passing with a warning (ADR 0593 §7, §9). Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL	2026-11-02 23:04:44 +01:00
jschoubben	f73b0bf6f6	Tests wait on events or the clock they are given, never on milliseconds a loaded machine outlasts The check of 2026-11-02 22:35 on a loaded shanks failed two buildrun tests on wall-clock bounds of 60 ms, the second failure of this kind tonight (the first: mesh-sdk's busretry tests, f02f142). Quick fix: give the two buildrun asker tests bounds of seconds. Thorough fix (chosen, because the kind recurs across every module of the repository): every test that raced a wall clock now waits on an event, steps a clock it is given, or has a bound only a failing test reaches. - buildrun asker tests: the steps' own times stand for the clock (bound an hour, starts and phases written in the past), no goroutine paces phases with sleeps. - builder: the decline bound reads Builder.Now; the decline test steps it; the stream-wait test waits for five check reads instead of 100 ms. - checker: the lost-ask try bounds an untaken ask at 200 ms only where that is what ends it; the program says when a merge turn changes, and the integration test waits on that line instead of 300 ms. - teller: the bed makes the teller's own ends bucket (a 10 s stall at start the tests raced by 0.5 s), waits for its warrant subscription instead of 500 ms pauses, and the package-level now is restored with t.Cleanup. - watchman: the recount, run in its own goroutine, is waited for instead of 10 ms. - polls of 2-5 s become 10 s (unit) and 20 s (integration). Under a throttled cgroup (half a core shared with two busy loops) beside the node-engine's suite, the buildrun tests failed 5 of 20 runs before and passed 30 of 30 after; no test added. Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL	2026-11-02 23:04:44 +01:00

Author

SHA1

Message

Date

jschoubben

34a607a1f5

Give the checker back its compose-candidate invoke, now that the new indexer runs

mesh/delivery merged: source.main moved to novox/mesh-delivery@a62276bcafad

Details

mesh/check being carried to 34a607a1f5cf on a62276bcafad | from 15d28ddf5bd3

The invoke was left out (mesh-delivery #142) only while the indexer built before the verb ran; the new indexer and the
node-engine's compose-candidate run on every machine since 22:57, so a check composes on the machines again instead of
passing with a warning (ADR 0593 §7, §9).

Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL

2026-11-02 23:04:44 +01:00

jschoubben

f73b0bf6f6

Tests wait on events or the clock they are given, never on milliseconds a loaded machine outlasts

The check of 2026-11-02 22:35 on a loaded shanks failed two buildrun tests on wall-clock bounds of 60 ms, the
second failure of this kind tonight (the first: mesh-sdk's busretry tests, f02f142).

Quick fix: give the two buildrun asker tests bounds of seconds.
Thorough fix (chosen, because the kind recurs across every module of the repository): every test that raced a
wall clock now waits on an event, steps a clock it is given, or has a bound only a failing test reaches.
- buildrun asker tests: the steps' own times stand for the clock (bound an hour, starts and phases written in the
past), no goroutine paces phases with sleeps.
- builder: the decline bound reads Builder.Now; the decline test steps it; the stream-wait test waits for five
check reads instead of 100 ms.
- checker: the lost-ask try bounds an untaken ask at 200 ms only where that is what ends it; the program says when a
merge turn changes, and the integration test waits on that line instead of 300 ms.
- teller: the bed makes the teller's own ends bucket (a 10 s stall at start the tests raced by 0.5 s), waits for its
warrant subscription instead of 500 ms pauses, and the package-level now is restored with t.Cleanup.
- watchman: the recount, run in its own goroutine, is waited for instead of 10 ms.
- polls of 2-5 s become 10 s (unit) and 20 s (integration).

Under a throttled cgroup (half a core shared with two busy loops) beside the node-engine's suite, the buildrun
tests failed 5 of 20 runs before and passed 30 of 30 after; no test added.

Claude-Session: https://claude.ai/code/session_01NcbgfJ9Gum47zkFKWRaovL

2026-11-02 23:04:44 +01:00

Tests no longer race the wall clock; the checker invokes compose-candidate again #143