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>
3.3 KiB
| id | status | status_updated | status_note | component | severity | timestamp | last_seen | occurrences | files | tags | related | legacy | gitea_issue | gitea_synced_status | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| bug-1674 | fixed | 2026-09-24 | apiserver | unknown | 2026-08-24T13:40:00Z | 2026-09-19T09:00:00Z | 1 |
|
|
true | 41 | fixed |
bug-1674: Gitea issue #41 ("job output timestamp is incorrect"): reporter's transcript shows `xcecli jobs…
Description
Gitea issue #41 ("job output timestamp is incorrect"): reporter's transcript shows xcecli jobs list -l <jobid> succeeding but omitting submit/start/done timestamps. Reproduced live on branch scheduler-framework, but the actual failure is more severe than reported: xcecli jobs list -l <jobid> (and bare xcecli jobs list -l) fail OUTRIGHT with query request failed: [invalid parameters: unsupported query field: pending_reason] -- no output is produced at all, not even a partial listing missing timestamps. Likely a regression introduced after the original issue was filed (this branch is under active scheduler-framework development).
Environment
Not recorded (migrated from buglog.json).
Reproduction
Not recorded (migrated from buglog.json).
Root cause
Both the -l/fields-filtered job-list code paths (jobs.go:553 and the near-duplicate at jobs.go:901) validate the requested field list via utils.GetSupportedFields(models.JobSubmitRequest{}) -- the job SUBMISSION request struct -- instead of models.JobInfo{}, the job query/response struct (the correct pattern, confirmed already used correctly for hosts at hosts.go:105/307 with models.HostInfo{}, and for at least one other job code path at jobs.go:1131 with models.JobInfo{}). Read-only/response-only fields like pending_reason, start_time, end_time, pids, pgids don't exist on JobSubmitRequest, so the first such field hit in Go's randomized map-iteration order triggers a hard unsupported query field rejection, making -l (and any custom -o field list touching these fields) fail unconditionally.
Fix
FIXED (2026-08-24). src/frontend/apiserver/controllers/jobs.go:553,901 -- both field-validation call sites now use utils.GetSupportedFields(models.JobInfo{}) instead of models.JobSubmitRequest{}, matching the already-correct pattern at jobs.go:1131. Caught during live verification (not source review alone): core_limit/mem_limit exist on JobSubmitRequest but not JobInfo/JobAdditionalFields -- switching wholesale would have newly broken those two previously-valid -o/-l fields. Fixed by adding CoreLimit/MemLimit (CustomLimit) to models.JobAdditionalFields (models/jobs.go), the existing merge-in bucket already used for pids/pgids/interactive_job_url. Added TestJobsListFieldValidation_UsesJobInfoNotSubmitRequest (jobs_test.go) covering all 24 fields in xcecli's real JobFieldsMap vocabulary. Full apiserver/controllers suite passes (CGO_ENABLED=0, this package's pre-existing required mode). Live-verified: jobs list -l and -o "pending_reason start_time core_limit mem_limit pid pgid" all now succeed -- every one previously errored with "unsupported query field".
Verification
Not recorded (migrated from buglog.json).
References
- Gitea: #41