[release/2.2] Clean up disk space in node e2e workflow#13548
Conversation
What are the changes compared to that original commit? If minimal, it'd be good to set the author to @dims like the original one, or at least add him as co-author with the Edit: Very minimal, can you set @dims as author and yourself as co-author? |
Add disk cleanup step before checking out repositories in node e2e workflow. GitHub Actions ubuntu-24.04 runners come preloaded with large toolchains that exhaust disk space during compilation of Kubernetes test binaries and containerd binaries. Cleaning up unused toolchains avoids kubelet DiskPressure condition and subsequent pod eviction failures. This change isolates and backports the workflow disk cleanup step originally introduced to main in commit 9d5ee65. Co-authored-by: Chris Henzie <[email protected]> Assisted-by: Antigravity Signed-off-by: Chris Henzie <[email protected]>
147eea0 to
1500e58
Compare
Done, thanks for catching this |
|
/cherrypick release/2.1 |
|
@samuelkarp: new pull request created: #13552 DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
|
@samuelkarp: new pull request created: #13553 DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
|
@samuelkarp: new pull request created: #13554 DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
Add disk cleanup step before checking out repositories in node e2e workflow. GitHub Actions ubuntu-24.04 runners come preloaded with large toolchains that exhaust disk space during compilation of Kubernetes test binaries and containerd binaries. Cleaning up unused toolchains avoids kubelet DiskPressure condition and subsequent pod eviction failures.
This change isolates and backports the workflow disk cleanup step originally introduced to main in commit
9d5ee65.
Assisted-by: Antigravity