[
https://issues.apache.org/jira/browse/BOOKKEEPER-472?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13534285#comment-13534285
]
Ivan Kelly commented on BOOKKEEPER-472:
---------------------------------------
[~umamaheswararao] I think this would be better done at the script level, to
avoid having autorecovery in the same jvm process. I think isolating like this
is better for stability. How I imagined this, is that you have an variable in
conf/bkenv.sh or in conf/bk_server.conf, which bin/bookkeeper-daemon.sh uses to
decide whether to start a autorecovery alongside bookie or not.
> Provide an option to start Autorecovery along with Bookie Servers
> -----------------------------------------------------------------
>
> Key: BOOKKEEPER-472
> URL: https://issues.apache.org/jira/browse/BOOKKEEPER-472
> Project: Bookkeeper
> Issue Type: Sub-task
> Components: bookkeeper-auto-recovery
> Reporter: Uma Maheswara Rao G
> Assignee: Uma Maheswara Rao G
> Attachments: BOOKKEEPER-472.patch
>
>
> We can also have an option to start the Autorecovery along with Bookie
> servers.
> If some users are not having too much load on the servers, they can even
> start them along the Bookie servers. If they feel, Auditor would disturb
> Bookie performance, they can anyway start as separate process.
> In another case, deployment overhead will reduce a bit as Monitoring process
> need not monitor one more process in their lifcycles etc.
> Thoughts?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
