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 KiB
| id | status | status_updated | status_note | component | severity | timestamp | last_seen | occurrences | files | tags | related | legacy | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| bug-1621 | fixed | 2026-09-24 | lib | leak | 2026-08-22T08:12:00Z | 2026-09-19T09:00:00Z | 1 |
|
|
|
true |
bug-1621: LeakSanitizer (run with ASAN_OPTIONS=detect_leaks=1, overriding this test suite's own…
Description
LeakSanitizer (run with ASAN_OPTIONS=detect_leaks=1, overriding this test suite's own detect_leaks=0 convention): 'Direct leak of 8 byte(s) ... #1 make_host_account ../lib/xcealloc.c:119 #2 make_job_account ../lib/xcealloc.c:188' -- an xce_host_account_t's host field (strdup'd in make_host_account(), src/lib/xcealloc.c:119) is never freed.
Environment
Not recorded (migrated from buglog.json).
Reproduction
Not recorded (migrated from buglog.json).
Root cause
free_host_account() (src/lib/misc.c) frees acct->accounts[] (each via free_res_account()) and then FREEUP(acct) itself, but never FREEUP(acct->host) -- the strdup'd host string make_host_account() (src/lib/xcealloc.c:119, haccount->host = strdup(halloc->host);) allocates. Pre-existing, NOT introduced by scheduler-sim-decision-parity task 2.3: confirmed by tracing the leak's allocation stack, which bottoms out in make_job_account()/make_host_account() (xcealloc.c) -- the exact same tree-building code the LIVE dispatch path already uses (scheduler.c's sched_commit_decision() -> make_job_account()), so this leak equally exists on the live, production dispatch path today, just previously undetected because src/scheduler/testsuite/meson.build runs this whole suite with ASAN_OPTIONS=detect_leaks=0 (fixtures are deliberately not torn down in this test family, per test file header conventions -- masks this along with every other intentional fixture leak).
Fix
FIXED (2026-09-08, follow-up session). Applied exactly the fix this entry already recommended: added FREEUP(acct->host); to free_host_account() (src/lib/misc.c), immediately before FREEUP(acct) -- mirrors free_res_account()'s existing FREEUP(acct->resname) pattern in the same file. Discovered as the remaining residual leak while fixing bug-1723's test-cleanup gaps: after adding the missing free_JobAlloc_ptr()/free_HostAlloc_list() calls those two tests needed, xcelib_test still leaked 12 bytes/2 allocations (exactly 2x "host0"/"host1", 6 bytes each) at make_host_account (xcealloc.c:119) via make_job_account -- this exact function/line. Since free_host_account() is called from free_job_account(), which real production dispatch code uses (scheduler.c, core_events.c), this fix has genuine production value, not just test-cleanliness. Verified: xcelib_test now 1/1 Ok with zero LeakSanitizer reports; full xce-build.sh all asan --incremental: 94/94 Ok, zero regressions.
Verification
Not recorded (migrated from buglog.json).
References
- Related: bug-1723