CASSANDRA-21523 jenkins shouldn't retry on fail-fast exceptions - #4939
Closed
michaelsembwever wants to merge 2 commits into
Closed
CASSANDRA-21523 jenkins shouldn't retry on fail-fast exceptions#4939michaelsembwever wants to merge 2 commits into
michaelsembwever wants to merge 2 commits into
Conversation
michaelsembwever
changed the title
netudima
reviewed
Jul 15, 2026
netudima
approved these changes
Jul 16, 2026
michaelsembwever
force-pushed
the
mck/jenkins-fail-fast-dont-retry/5.0
branch
2 times, most recently
from
July 26, 2026 22:11
e0a811c to
d65252c
Compare
To jenkins k8s add ephemeral-storage requests and limits to agents, and log peak disk usage Agents were evicted at the node level, and the node's message named the cause: neither container declared an ephemeral-storage request, which leaves the pod BestEffort for storage, invisible to the scheduler and first out when the node fills. /var/lib/docker was an unbounded emptyDir besides. Kubelet charges the pod for every emptyDir, the workspace included, together with the containers' writable layers and logs, all against the containers' limits summed. Each template now budgets 80Gi of the ~89Gi that a 100GiB node allocates, as ci-cassandra's agent nodes are and pre-ci's now are, and docker's own volume carries a 60Gi sizeLimit. Exceeding either evicts the pod naming what it was, which the node's own threshold does not. That volume moves into the pod template's raw `yaml:` under a name of its own, the only way it can keep a sizeLimit: the plugin's EmptyDirVolume has no such field, and a volume declared under `volumes:` is generated as `volume-N`, whose generated copy wins any merge with a raw entry of that name. A deployed agent's pod spec showed the sizeLimit being dropped exactly so. Also extract inline sh blocks to functions for readability
michaelsembwever
force-pushed
the
mck/jenkins-fail-fast-dont-retry/5.0
branch
from
July 27, 2026 22:10
d65252c to
2196038
Compare
run-ci must not silently drop a site's helm values customisations Long-lived instances like pre-ci.cassandra.apache.org carry values absent from jenkins-deployment.yaml, and any `helm upgrade -f jenkins-deployment.yaml` drops them. Keep them in a separate file and pass `--values-override`; helm merges it over the repo's. Before upgrading, compare the deployed values against what is about to be applied and require confirmation for anything only the site holds. This catches added keys and dropped list items; it cannot tell a customisation from a key the repo has since removed, hence a prompt, and it is blind to a key present in both with a locally edited value, an agent.podTemplates entry for instance. Aborts when not on a tty. Also keep the jenkins-home claim on uninstall, so --only-tear-down no longer destroys build history; jenkins-test.sh validates .jenkins/ without a cluster, including the yaml embedded in the pod templates and JCasC scripts that the chart never parses; a workflow runs it and the run-ci tests on changes to either. The requests timeouts and unused wait_for_jenkins_http argument are drive-bys to get run-ci's own lint command clean, so the workflow can enforce it.
michaelsembwever
force-pushed
the
mck/jenkins-fail-fast-dont-retry/5.0
branch
from
July 28, 2026 13:04
2196038 to
b6180d8
Compare
netudima
approved these changes
Jul 28, 2026
Member
Author
|
committed 209ef2b |
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
2 participants
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.
https://issues.apache.org/jira/browse/CASSANDRA-21523
To jenkins k8s add ephemeral-storage limits to agents and log peak disk usage
Agents were being evicted at the node level due to missing ephemeral-storage requests/limits and no sizeLimit on the /var/lib/docker emptyDir.
Also extract inline sh blocks to functions for readability