Rongsong Shen
d26ff8870d
The buglog was one 27K-line JSON file whose fields covered only symptom,
root cause and fix, as free text: 2% of entries named the code version a
bug was found on, 12% the build profile, 36% a way to reproduce, and 58% of
the file was OpenWolf hook auto-detections ("Missing guard clause",
"Function not marked async" on .md files).
- .claude/skills/buglog/bug-record.md: the record standard. Eight
sections (Summary as the title, Description, Environment, Reproduction;
Root cause, Fix, Verification, References once fixed), each with the
content it must contain -- e.g. Environment needs the OS, the libc the
failing binary loads (from its ELF interpreter: glibc 2.42 in the nix
shell vs 2.40 on the host), toolchain, build profile and commit.
- .wolf/bugs/<id>.md: 500 records migrated from buglog.json; every field
verified to round-trip. They carry legacy: true and an explicit "not
recorded" where the old log had no such field; component derived from
file paths, severity from keywords (else unknown), fix_commits from
hashes in status_note/fix that git resolves (91 of 463 fixed).
- .wolf/buglog-archive.json: the 710 hook auto-detections, verbatim.
- .wolf/buglog.json is now an index generated from the records, keeping
the fields OpenWolf's hooks and `openwolf bug search` read.
- scripts/buglog.py rewritten: new (template + Environment collected from
the machine and --binary), check, status (refuses a status the record
doesn't support; fixed needs Root cause, Fix, Verification and
--commit hashes git resolves), note, recur, set, show, list, triage,
env, index. 29 unittest cases in scripts/tests/test_buglog.py.
- scripts/git-hooks/pre-commit runs `check --staged`: incomplete staged
records, a legacy record hand-changed to fixed without the fix-time
sections, a new record claiming legacy: true, or a stale index refuse
the commit.
- scripts/gitea-sync.py reads the records: issue body = the record's
sections plus linked fix commits (origin remote, GITEA_COMMIT_REPO to
override), and a body hash so an issue's text is refreshed when the
record changes.
- bug-1866 rewritten to the full standard as the worked example.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
38 lines
2.5 KiB
Markdown
38 lines
2.5 KiB
Markdown
---
|
|
id: "bug-1577"
|
|
status: "fixed"
|
|
status_updated: "2026-09-24"
|
|
status_note: ""
|
|
component: "unknown"
|
|
severity: "crash"
|
|
timestamp: "2026-08-17T00:00:00Z"
|
|
last_seen: "2026-09-19T09:00:00Z"
|
|
occurrences: 1
|
|
files: [".claude/skills/xce-test-resources/run-resource-verify.sh"]
|
|
tags: ["resource-value-verification", "xce-deploy-ha", "worker-daemon", "bug-1572", "task-6.1"]
|
|
related: ["bug-1572"]
|
|
legacy: true
|
|
---
|
|
|
|
# bug-1577: No explicit error -- discovered via live testing task 6.1 (resource-value-verification)…
|
|
|
|
## Description
|
|
No explicit error -- discovered via live testing task 6.1 (resource-value-verification): xce-deploy-ha's deploy-ha.sh Step 3 opt-in-by-presence worker branch (predates this feature) generates start-worker.sh whenever a worker binary (xcecd_virtual) happens to exist alongside the controller build, and generated start-all.sh always launches it. run-resource-verify.sh's own deployment (task 5.1+6.1) unconditionally hit this since a full `/xce-build all` leaves xcecd_virtual built. Result: a REAL worker daemon ('worker-1') ran for the whole duration of every xce-test-resources run, silently violating Requirement 2.1's design intent ('no real worker daemon process required') and periodically re-reporting its own resources (~every 15s) -- a live risk of triggering the still-open bug-1572 (scheduler double-free/crash on a resource's 2nd periodic report) if a run happened to last that long.
|
|
|
|
## Environment
|
|
_Not recorded (migrated from buglog.json)._
|
|
|
|
## Reproduction
|
|
_Not recorded (migrated from buglog.json)._
|
|
|
|
## Root cause
|
|
deploy-ha.sh's worker-detection/start-all.sh generation is out of this task's edit boundary (xce-deploy-ha files are done/approved for this feature) and has no flag to skip the worker step -- it only conditionally starts one when a binary is present, with no opt-out.
|
|
|
|
## Fix
|
|
start_all() now exports XCE_WORKER_BIN=/bin/true before invoking the generated start-all.sh. The generated start-worker.sh already reads ${XCE_WORKER_BIN:-<real path>} at RUN time from its own inherited environment, so pointing it at a real no-op binary makes the worker step exit immediately with no persistent process, without touching deploy-ha.sh itself. Verified via a /proc/*/exe-based watcher (not pgrep -f text matching, which self-matches -- see bug-1576) across a full default-resource-set run: zero xcecd_virtual/xcecd/xcecd_worker processes at any point post-fix, versus one running the entire time pre-fix.
|
|
|
|
## Verification
|
|
_Not recorded (migrated from buglog.json)._
|
|
|
|
## References
|
|
- Related: bug-1572
|