[
https://issues.apache.org/jira/browse/SOLR-5040?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13708120#comment-13708120
]
Mark Triggs commented on SOLR-5040:
-----------------------------------
Thanks Mark.
Chiming in one data point, I started looking at the SnapShooter code because I
was wondering how to start a snapshot from a script and then (programatically)
wait for it to complete.
In the end the only way I could see to do this was to poll the replication
handler with 'command=details' to wait for its state to change. But this
wouldn't have been necessary if the replication handler waited the request
until the snapshot completed.
> SnapShooter doesn't create a lock as it runs
> --------------------------------------------
>
> Key: SOLR-5040
> URL: https://issues.apache.org/jira/browse/SOLR-5040
> Project: Solr
> Issue Type: Bug
> Components: replication (java)
> Reporter: Mark Triggs
> Priority: Trivial
> Attachments: snapshooter-locking.diff
>
>
> Hi there,
> While messing around with the replication handler recently, I noticed that
> the snapshooter didn't seem to be writing a lock file. I had a look at the
> SnapShooter.java code, and to my untrained eye it seemed like it was creating
> a Lock object but never actually taking a lock.
> I modified my local installation to use lock.obtain() instead of
> lock.isLocked() and verified that I'm now seeing lock files. I've attached a
> very small patch just in case this is a genuine bug.
> Cheers,
> Mark
--
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
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
