CI: Prevent zombie MAD RTE perftest server#11630
Open
Alexey-Rivkin wants to merge 1 commit into
Open
Conversation
3d49ad0 to
e942c04
Compare
d949803 to
bd1126d
Compare
A failed run leaves behind a zombie server container: docker_stop_srv cats the log before docker stop, and when the log is missing the cat aborts the script (set -e) before the stop runs. Being a detached container, the zombie survives the build and keeps the MAD agent registration on the HCA, so every following build's server fails with EINVAL and the client hangs until the job timeout - the stage breaks for all PRs until the zombie is removed by hand. Stop the container first and sweep zombie ucx_perftest_* containers before starting a new server, so one bad run cannot wedge the stage.
bd1126d to
0f13015
Compare
tvegas1
approved these changes
Jul 8, 2026
dpressle
approved these changes
Jul 9, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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?
Stop the MAD perftest server container reliably, and clear any zombie
ucx_perftest_*container before starting a new one.Why?
On a failed run,
docker_stop_srvrunscaton a missing log, which aborts underset -ebeforedocker stop. The detached server survives as a zombie and holds the MAD registration on the HCA, so every later build's server dies withmad_register_server_via ... Invalid argumentand the client hangs until timeout. Broke all PRs until zombies were cleared by hand (e.g. build 127843).How?
docker_stop_srv:docker stopbeforecat.docker_run_srv:docker rm -fsweep ofucx_perftest_*to heal zombies from a canceled or crashed prior run.