You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "Pedro Mázala (Jira)" <ji...@apache.org> on 2024/05/21 15:46:00 UTC

[jira] [Commented] (FLINK-35388) Flink 1.18.1 get stuck after creating highly available job result storage directory

    [ https://issues.apache.org/jira/browse/FLINK-35388?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17848259#comment-17848259 ] 

Pedro Mázala commented on FLINK-35388:
--------------------------------------

I found this https://stackoverflow.com/questions/78329856/flink-checkpoints-stalling-and-timing-out-with-latency-related-errors

> Flink 1.18.1 get stuck after creating highly available job result storage directory
> -----------------------------------------------------------------------------------
>
>                 Key: FLINK-35388
>                 URL: https://issues.apache.org/jira/browse/FLINK-35388
>             Project: Flink
>          Issue Type: Bug
>          Components: Runtime / Checkpointing
>    Affects Versions: 1.18.1
>            Reporter: Pedro Mázala
>            Priority: Major
>         Attachments: output2.json
>
>
> After upgrading to Flink 1.18.1 (from 1.17.2) some pipelines get stuck after creating highly available job result storage directory.
> I use GCS for storage and I did not find anything unusual in the logs. 
> The pod on k8s stays unhealthy until a manual restart (kill JM pod)



--
This message was sent by Atlassian Jira
(v8.20.10#820010)