[jira] [Comment Edited] (FLINK-9945) RocksDB state backend Checkpointing Failed

["Stefan Richter (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLINK-9945?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16558160#comment-16558160 ] 

Stefan Richter edited comment on FLINK-9945 at 7/26/18 10:47 AM:
-----------------------------------------------------------------

Ok, there is no exception but it clearly says that the checkpoint expired before you can see the exception. An expired checkpoint gets cancelled and you only see they exception during the cancellation. 

Almost sure that this is not a bug, so the question is why the checkpoint expires. Did you change the expiration time or maybe your checkpoints are just very big and need a longer timeout?

Edit, seems like your timeout is 10 minutes, but there are late messages at around 11 minutes, so maybe try to set the timeout to 20 or 30 minutes.


was (Author: srichter):
Ok, there is no exception but it clearly says that the checkpoint expired before you can see the exception. An expired checkpoint gets cancelled and you only see they exception during the cancellation. 

Almost sure that this is not a bug, so the question is why the checkpoint expires. Did you change the expiration time or maybe your checkpoints are just very big and need a longer timeout?

> RocksDB state backend Checkpointing Failed
> ------------------------------------------
>
>                 Key: FLINK-9945
>                 URL: https://issues.apache.org/jira/browse/FLINK-9945
>             Project: Flink
>          Issue Type: Bug
>    Affects Versions: 1.4.2
>            Reporter: xymaqingxiang
>            Priority: Major
>         Attachments: image-2018-07-25-16-57-45-617.png, image-2018-07-26-18-31-21-429.png
>
>
> Checkpoint failed.
> The log is:
> !image-2018-07-25-16-57-45-617.png!



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)