[
https://issues.apache.org/jira/browse/HDFS-1969?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13039218#comment-13039218
]
Todd Lipcon commented on HDFS-1969:
-----------------------------------
Hi Konstantin. I spent some time this morning trying to do this, but it would
actually require a bit of refactoring to avoid introducing lots of duplicated
code. The {{setFields}} method has lots of side effects beyond just setting up
some properties. I think this refactoring is a good idea, but I'd prefer to
separate it from this blocker so we can make progress towards getting the
release out the door.
Please note that this patch isn't the first case where we have a check against
layoutVersion while writing the VERSION file -- we already have such a check
for federation-related fields.
> Running rollback on new-version namenode destroys namespace
> -----------------------------------------------------------
>
> Key: HDFS-1969
> URL: https://issues.apache.org/jira/browse/HDFS-1969
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: name-node
> Affects Versions: 0.22.0
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
> Priority: Blocker
> Fix For: 0.22.0
>
> Attachments: hdfs-1969.txt, hdfs-1969.txt
>
>
> The following sequence leaves the namespace in an inconsistent/broken state:
> - format NN using 0.20 (or any prior release, probably)
> - run hdfs namenode -upgrade on 0.22. ^C the NN once it comes up.
> - run hdfs namenode -rollback on 0.22 (this should fail but doesn't!)
> This leaves the name directory in a state such that the version file claims
> it's an 0.20 namespace, but the fsimage is in 0.22 format. It then crashes
> when trying to start up.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
