lei.chen/myxce
1
0
Fork 1
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

docs(buglog): bug-1923 invariant audit (scheduler dims 1/4) clean

Browse source 
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This commit is contained in:
Rongsong Shen 2026-10-08 17:32:35 +00:00
parent dc624bfca5
commit 7f1c7953a1
Signed by untrusted user: rshen
GPG key ID: 80533536B0EB651C
2 changed files with 3 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
.wolf/bugs/bug-1923.md
View file

@ -80,6 +80,8 @@ Two separate defects on the SnapshotEntry save/load path, both pre-existing (ori
- Suites: scheduler meson suite Ok 100 / Fail 0, asan (`src/xce.build` in the fix worktree); `scheduler repl trace baseline test` OK with the regenerated `set_output.replay`.
- Live: `xce_scheduler --sim /home/claude/tmp/bug-1923/fixC/contention.replay --interactive` on the fixed asan build: `info resource license` shows `value=2.00 used=0.00`, then `value=2.00 used=2.00` after `step`, and jobs 102 and 103 DISPATCH (before the fix: `value=(none)`, and both PEND). A variant-A replay regenerated with the fixed writer also loads `value=2.00`. No HA soak was needed: the change is limited to the scheduler's snapshot save/load and does not touch the controller or WAL.
UPDATE (2026-10-08): Invariant audit (xce-audit-invariants scheduler, dimensions 1 and 4) run 2026-10-08 at dc624bfc: CLEAN. A temporary round-trip probe (24 resources, sparse shared slots 3..23, every value type, three workers with sparse/single/no per-worker values) went live core, sched_record_start, sched_sim_driver_start, then an in-memory save/load, and matched every slot by name, type and value; it failed when either half of the fix was reverted (save .caps -> r07 dropped; load empty-host -> r03 dropped). The probe source is kept at /home/claude/tmp/bug-1923/test_snapshot_resval_probe.c. The real-binary --sim-verify of a license-contention replay from a mid-run snapshot verifies rounds 3000-10000; the pre-fix fixture had extra mismatches in 6000-10000. Dimension 2 is covered by test_core_isolation and dimension 3 is untouched; the HA soak is controller-only and n/a.
## References
- Related: bug-1568 (same `snapshot_wire.c` worker-resval loop; that bug added the missing-resource diagnostic)
- Docs: .kiro/steering/persistence.md (Snapshot Save/Load Field-Completeness)