You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Roman Khachatryan (Jira)" <ji...@apache.org> on 2020/12/10 15:52:00 UTC
[jira] [Commented] (FLINK-19462) Checkpoint statistics for
unfinished task snapshots
[ https://issues.apache.org/jira/browse/FLINK-19462?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17247326#comment-17247326 ]
Roman Khachatryan commented on FLINK-19462:
-------------------------------------------
I've created a mini-design [doc|https://docs.google.com/document/d/1NVAzo7kATqh9HOomVhEkU-a89DLjbtILIH25g_vkujY/edit?usp=sharing] to clarify the requirements and discuss the design.
Can you take a look [~NicoK]?
cc: [~pnowojski], [~trohrmann] you probably would also like to take a look
> Checkpoint statistics for unfinished task snapshots
> ---------------------------------------------------
>
> Key: FLINK-19462
> URL: https://issues.apache.org/jira/browse/FLINK-19462
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Checkpointing, Runtime / Metrics
> Reporter: Nico Kruber
> Priority: Major
> Labels: usability
>
> If a checkpoint times out, there are currently no stats on the not-yet-finished tasks in the Web UI, so you have to crawl into (debug?) logs.
> It would be nice to have these incomplete stats in there instead so that you know quickly what was going on. I could think of these ways to accomplish this:
> * the checkpoint coordinator could ask the TMs for it after failing the checkpoint or
> * the TMs could send the stats when they notice that the checkpoint is aborted
> Maybe there are more options, but I think, this improvement in general would benefit debugging checkpoints.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)