[
https://issues.apache.org/jira/browse/HDFS-17244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18028898#comment-18028898
]
ASF GitHub Bot commented on HDFS-17244:
---------------------------------------
github-actions[bot] closed pull request #6237: [HDFS-17244] Avoid affecting the
overall logic of checkpoint when I/O…
URL: https://github.com/apache/hadoop/pull/6237
> Avoid affecting the overall logic of checkpoint when I/O exceptions occur
> -------------------------------------------------------------------------
>
> Key: HDFS-17244
> URL: https://issues.apache.org/jira/browse/HDFS-17244
> Project: Hadoop HDFS
> Issue Type: New Feature
> Affects Versions: 3.2.1
> Reporter: liying
> Assignee: liying
> Priority: Major
> Labels: pull-request-available
> Attachments: image-2023-10-30-15-27-47-567.png,
> image-2023-10-30-15-28-14-552.png
>
>
> Sometimes it will checkpoint repeatedly on our cluster with multiple
> SNNS（hdfs verison ：3.2.1）. We check the codes， and find that it will send
> fsimage file to all other namenode. When it sends the fimage file to the
> other snns， it will fail because of other snns will refuse to receive the
> file。 It will affect the whole login of checkpointing。For example， we check
> the result of transfering fsimage to snn by feature‘s geting method， and it
> will throw the IOException and don't update the lastUploadTime and the
> lastCheckpointTime，then it will checkpoint repeatedly.
> !image-2023-10-30-15-27-47-567.png!
> !image-2023-10-30-15-28-14-552.png!
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
