Skip to content

Fix RJob failure propagation and submission logging - #2587

Open
JifeiShan wants to merge 1 commit into
open-compass:mainfrom
JifeiShan:codex/propagate-rjob-failures
Open

Fix RJob failure propagation and submission logging#2587
JifeiShan wants to merge 1 commit into
open-compass:mainfrom
JifeiShan:codex/propagate-rjob-failures

Conversation

@JifeiShan

Copy link
Copy Markdown
Contributor

Summary

  • propagate non-zero runner task results so the OpenCompass process fails instead of only logging errors
  • emit an explicit RJob submitted event only after rjob submit succeeds
  • avoid logging an empty command error for successful submissions

This prevents planned RJob names whose submission failed from being interpreted as successfully completed jobs by external lifecycle integrations.

Tests

  • python3 tests/runners/test_base.py
  • python3 -m compileall -q opencompass/runners/base.py opencompass/runners/rjob.py tests/runners/test_base.py
  • git diff --check

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants