Clean up cancelled GPU image prewarm pods - #964
Merged
Merged
Conversation
bradhilton
force-pushed
the
joshua/prewarm-cancel-cleanup-20260925
branch
from
September 25, 2026 07:03
adfdb69 to
141774f
Compare
bradhilton
marked this pull request as ready for review
September 26, 2026 16:00
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Cancelling the GPU image workflow during node prewarming can leave ownerless one-shot pull pods behind. They have fixed node-derived names but no run identity, so workflow cleanup cannot safely distinguish them from another run.
This change labels tag-check and node-prewarm one-shots with the GitHub run ID/attempt, then deletes only pods matching both that run identity and the prewarm name. The build script performs the cleanup from its existing exit trap, and the workflow repeats it from the existing
always()cleanup step. Persistent prewarm DaemonSets and peer-run pods do not match the selector.Validation:
bash -n scripts/build-gpu-image.shuv run --no-project --python /home/brad/github/openpipe/caladan/.venv/bin/python --with pytest python -m pytest -q -o addopts='' tests/unit/test_gpu_image_build_context.py(2 passed)git diff --checkSource only. This does not cancel or modify the currently running image build, persistent prewarm DaemonSets, shared caches, or GPU workloads.