CI: exclude bad Phoenix L40S node atl1-1-03-007-31-0 (uncorrectable ECC) - #1750
Merged
Conversation
sbryngelson
enabled auto-merge (squash)
August 23, 2026 02:22
sbryngelson
disabled auto-merge
August 23, 2026 02:27
Contributor
There was a problem hiding this comment.
Pull request overview
This PR updates the Phoenix GPU CI SLURM submission script to avoid scheduling jobs onto a known-bad L40S node that is consistently failing GPU runs with uncorrectable ECC errors, helping keep the GPU CI lane reliable until the hardware is drained/repaired.
Changes:
- Extend the Phoenix GPU
#SBATCH --excludelist to includeatl1-1-03-007-31-0. - Keep the existing exclusion of
atl1-1-03-007-29-0, resulting in a two-node exclusion list for Phoenix GPU jobs.
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #1750 +/- ##
=======================================
Coverage 61.61% 61.61%
=======================================
Files 84 84
Lines 21520 21520
Branches 3180 3180
=======================================
Hits 13259 13259
Misses 6082 6082
Partials 2179 2179 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Adds
atl1-1-03-007-31-0to the Phoenix GPU--excludelist. Net exclude is nowatl1-1-03-007-29-0,atl1-1-03-007-31-0.Why
A second L40S node in the same chassis as the already-excluded
007-29-0(bothatl1-1-03-007-*) is failing every GPU job with the same hardware fault:Confirmed across a CI sweep —
007-31-0is the ECC source in the most recentmasterrun (32601124944, 105 ECC hits) and a PR run, exactly as007-29-0was in the prior batch. Reruns don't help: SLURM has no GPU-health awareness and keeps rescheduling onto the same bad node.Note: the
PMIX ERROR: PMIX_ERR_NO_PERMISSIONS ... dstore_base.clines in these logs are benign (they appear on passing syscheck/pre_process steps too, from the hpcx/2.19-cuda PMIx dstore); they are not the failure cause. The failure is the ECC fault above.Both bad GPUs are being reported to PACE for repair/drain; this exclude is the CI unblock in the meantime.