myxce/.wolf/bugs/bug-1907.md

id	status	status_updated	status_note	component	severity	timestamp	last_seen	occurrences	files	tags	related	fix_commits
bug-1907	fixed	2026-10-02		controller	wrong-result	2026-10-01T15:49:05Z	2026-10-01T15:49:05Z	1	src/controller/src/job.c src/harness/test.c	harness job-control not-found job-array	bug-1896	47551fb2

bug-1907: harness setpend and jobrun cannot report not-found: the controller acks status 0 first

Description

Observed: Found by code reading, no failing run captured. Both controller handlers reply before looking the job up:

src/controller/job.c:2246 handle_jobdispatch: init_reply_by_req(); reply.status = 0; xce_reply(...)   /* then the work */
src/controller/job.c:3869 do_set_preason:     init_reply_by_req(); reply.status = 0; xce_reply(...)   /* then the work */

So harness jobrun -j <id> and setpend -j <id> succeed for a job (or array element) that does not exist.

Expected: like the other single-job commands, jobrun/setpend on an unknown job ID report "not found" (job-array-one-based-index Requirements 4.7 and 5.13). Impact: wrong result: a test script that targets a wrong element (e.g. $jobid[0] on a new array) passes silently instead of failing. Found by: review (job-array-one-based-index task 4.3 implementation notes and feature validation, 2026-10-01)

Environment

• OS: no /etc/os-release; Docker container; NixOS kernel; kernel Linux 6.18.33 x86_64
• libc: glibc 2.42-47 (interpreter /nix/store/wb6rhpznjfczwlwx23zmdrrw74bayxw4-glibc-2.42-47/lib/ld-linux-x86-64.so.2)
• Toolchain: gcc (GCC) 15.2.0, meson 1.9.1
• Build: asan (src/build.jai)
• Code: scheduler-framework @ efff6b65-dirty
• Deployment: any (controller request handling)

Reproduction

1. Start a controller (xce-deploy-ha) and run xce_test against it.
2. jobrun -j 0-0-1-5 (an ID that was never submitted) and setpend -j 0-0-1-5 -p 1: both report success (before 47551fb2).
3. Compare jobkill -j 0-0-1-5, which reports not found. (jobrun is a client-context command; setpend is in the scheduler context.)

• Frequency: deterministic (1/1 live, 2026-10-01)
• Evidence: /home/claude/tmp/bug-1907-red.out

Root cause

handle_jobdispatch (src/controller/job.c:2246) and do_set_preason (job.c:3869) send their reply with status 0 before touching the job; a missing job is only logged afterwards (set_preason PERROR, dispatch_a_job_). The early acknowledgement is deliberate: the scheduler sends these requests in batches and its publish/retry/redelivery logic (scheduler.c, test_scheduler_decision_publish_retry.c, test_scheduler_decision_redelivery.c) depends on it. The harness commands jobrun and setpend used the same requests and so reported success for any job ID.

Fix

Harness only; the controller/scheduler protocol is unchanged. New job_exists_or_report() in src/harness/test.c looks the job up with the same interruptible xcecli_jobinfo call jobinfo uses and judges the result with cmdutil_job_lookup_report(); cmd_jobrun and cmd_setpend call it before sending their request and return -ENOENT with ": job not found" (plus the <array-id>[k] hint for a bare identity).

Verification

• Test: live harness script against a 3-node controller deploy (no unit seam: the commands need a controller). RED before the fix: jobrun -j 0-0-1-5 printed nothing and setpend -j 0-0-1-5 -p 1 printed ok, script exit 0 (/home/claude/tmp/bug-1907-red.out). GREEN after: both print job 0-0-1-5 not found + hint and exit 1; jobrun -j $jobid[3] on a missing element reports it not found; setpend/jobrun on the real $jobid still succeed (exit 0)
• Suites: harness meson suite 7/7 OK (asan, src/build.jai)
• Live: the GREEN run above (xce-deploy-ha, controllers only, 2026-10-01)

References

• Related: none