[
https://issues.apache.org/jira/browse/HDFS-955?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12832757#action_12832757
]
Todd Lipcon commented on HDFS-955:
----------------------------------
bq. because the problem stated here has a simple solution, and because it is
important to fix it
I'm still waiting for that simple solution. I don't believe it does have one.
Your solution of reordering the loops shrinks the time window of the problem
occurring, but it still occurs. Note that the "state 5" problem in my comment
above occurs regardless of the number of storage directories.
I'll upload a patch momentarily which includes another unit test that shows the
issue.
> FSImage.saveFSImage can lose edits
> ----------------------------------
>
> Key: HDFS-955
> URL: https://issues.apache.org/jira/browse/HDFS-955
> Project: Hadoop HDFS
> Issue Type: Bug
> Affects Versions: 0.20.1, 0.21.0, 0.22.0
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
> Priority: Blocker
> Attachments: hdfs-955-unittest.txt, PurgeEditsBeforeImageSave.patch
>
>
> This is a continuation of a discussion from HDFS-909. The FSImage.saveFSImage
> function (implementing dfsadmin -saveNamespace) can corrupt the NN storage
> such that all current edits are lost.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
