[
https://issues.apache.org/jira/browse/HADOOP-8163?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13228668#comment-13228668
]
Todd Lipcon commented on HADOOP-8163:
-------------------------------------
Hi Hari,
The creation is not atomic, but the creation of the "info" znode is done after
the creation of the "lock" znode. So, only one NN can try to create an "info"
node at a time (since only one can hold the lock at a time)
The only other possibility is that the NN creates the "lock" znode, then the
session expires, and then creates the "info" znode. But this isn't possible
since, if the session expired, the creation of the "info" node would fail with
an exception.
Do you have a specific error scenario in mind that I'm not covering?
> Improve ActiveStandbyElector to provide hooks for fencing old active
> --------------------------------------------------------------------
>
> Key: HADOOP-8163
> URL: https://issues.apache.org/jira/browse/HADOOP-8163
> Project: Hadoop Common
> Issue Type: Improvement
> Components: ha
> Affects Versions: 0.24.0, 0.23.3
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
> Attachments: hadoop-8163.txt
>
>
> When a new node becomes active in an HA setup, it may sometimes have to take
> fencing actions against the node that was formerly active. This JIRA extends
> the ActiveStandbyElector which adds an extra non-ephemeral node into the ZK
> directory, which acts as a second copy of the active node's information.
> Then, if the active loses its ZK session, the next active to be elected may
> easily locate the unfenced node to take the appropriate actions.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
