-
Notifications
You must be signed in to change notification settings - Fork 13.8k
[FLINK-38487][build] bump vmImage version to ubuntu-24.04 #27095
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
@gaborgsomogyi The CI is still failing - were you expecting this change to fix the "[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x16596ae]" ? |
666a722
to
0530e1e
Compare
Seems like the new one is doing something so probably not going to blow up, but let's see... |
There is a test failure but it's unrelated:
@1996fanrui any idea, because we've just touched the unaligned checkpoint part? |
I analyzed the log from https://dev.azure.com/apache-flink/apache-flink/_build/results?buildId=70122&view=results It seems the TM is crashed since NPE, resulting in the resource is not sufficient, so it is not caused by UC. I think it should be a problem with the test itself. Code : flink/flink-tests/src/test/java/org/apache/flink/test/checkpointing/UnalignedCheckpointTestBase.java Line 1050 in 47478ed
|
0530e1e
to
b49c6cd
Compare
Should we add a null check there?🤔 Until the azure is not fixed we're not able to test it... |
Never mind, now it's passing. Hope that the whole pipeline will be green now... |
Now it's passing🎉 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the fix. LGTM.
What is the purpose of the change
Now there are e2e failures on azure: https://dev.azure.com/apache-flink/98463496-1af2-4620-8eab-a2ecc1a2e6fe/_build/results?buildId=70100
I've had a deeper look and found the following: it started breaking because the Microsoft-hosted Ubuntu image the pipeline runs on changed overnight (Oct 3 → Oct 4). That update brings a newer toolchain (Ubuntu 22.04.5 + Docker 28.0.4, and a FIPS-hardened Go/kubectl stack). The kube step then crashes (kubectl exits with status 2) and your bash step dies with “STDIO streams did not close…”, so the Flink Kubernetes e2e never even really starts. It’s an agent/image regression, not a Flink change.
Brief change log
Bump vmImage version to ubuntu-24.04 to match the compatibility + use latest kubectl.
Verifying this change
Azure run.
Does this pull request potentially affect one of the following parts:
@Public(Evolving)
: noDocumentation