Make CI lab names unique per build - #75
Merged
Merged
Conversation
envsubst does not expand the ${BUILD_NUMBER:-0} default form, and the
sudo environment does not carry BUILD_NUMBER, so every Jenkins lab ran
as jenkins-<lane>-0. With /var/lib/libvirt/images on shared NFS across
the lab runners, a lane scheduled on a different host than the
previous build collided with the prior build's same-named, still
running lab: qemu image locking refused the disks ("Is another process
using the image?") and the lane failed at compute boot. Use the plain
${BUILD_NUMBER} form, which envsubst renders to the real build number;
these configs are only consumed through the pipeline's envsubst step.
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Tag build 6: the four alma lanes passed, and both Ubuntu lanes failed at compute boot with
Is another process using the image?— qemu image locking on the shared NFS/var/lib/libvirt/images. Root cause:envsubstcannot expand the${BUILD_NUMBER:-0}default form (and sudo dropsBUILD_NUMBER), so every build's labs were namedjenkins-<lane>-0; build 5's successful labs were left running, and build 6's Ubuntu lanes landed on the opposite hosts, colliding with build 5's same-named disks on shared storage. Plain${BUILD_NUMBER}renders the real number at the pipeline's envsubst step (these configs are only consumed through that path). Paired with a ci-cd-platform change makingciIntegrationLabdestroy labs after the run.