[
https://issues.apache.org/jira/browse/FLINK-7167?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16197144#comment-16197144
]
Aljoscha Krettek commented on FLINK-7167:
-----------------------------------------
Thanks for pointing this out, [~phoenixjiangnan]. There is actually a (very
outdated) PR that also tries to tackle this issue:
https://github.com/apache/flink/pull/3522. I think the solution we will go for
in the end is to get rid of this configuration parameter, this should make
things easier.
> Job can't set checkpoint meta directory to cover the system setting
> state.checkpoints.dir
> ------------------------------------------------------------------------------------------
>
> Key: FLINK-7167
> URL: https://issues.apache.org/jira/browse/FLINK-7167
> Project: Flink
> Issue Type: Improvement
> Components: State Backends, Checkpointing
> Affects Versions: 1.3.1
> Reporter: yuqi
> Assignee: Bowen Li
>
> If we want to recover a failed job use checkpoint, till now, as all job
> checkpoint meta data are in the same directory and do not have specific
> identification, we have to traverse all file in the directory to find the
> data of this job. this is rather troublesome. so seting this configuration
> in the job level is preferable.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
