Upgrading plugins - probably always want to restart the cluster?

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Upgrading plugins - probably always want to restart the cluster?

ppearcy
Last week, I made some tweaks to the hunspell plugin. Specifically, removing all the affix rules that really aren't appropriate for stemming. 

Perhaps this was completely a bone-headed no-no on my part, but I overwrote the existing plugin jar file with the new one, with the expectation that the new plugin would get picked  up when I restarted the cluster during some upcoming maintenance. 

I deleted an index which then triggered every node to crash a the same time. Doh! If you're curious, here is the core dump:
https://gist.github.com/4658564

Did I commit a cardinal sin updating the plugin jar while the cluster was active or should this condition be handled better by elasticsearch? 

Thanks!
Paul

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group, send email to [hidden email].
For more options, visit https://groups.google.com/groups/opt_out.
 
 
Reply | Threaded
Open this post in threaded view
|

Re: Upgrading plugins - probably always want to restart the cluster?

joergprante@gmail.com
It's not Elasticsearch related. It's a JVM bug. Just do not update jars
in the classpath while a JVM is active.  See
http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4425695

Jörg

Am 28.01.13 21:10, schrieb ppearcy:

> Last week, I made some tweaks to the hunspell plugin. Specifically,
> removing all the affix rules that really aren't appropriate for stemming.
>
> Perhaps this was completely a bone-headed no-no on my part, but I
> overwrote the existing plugin jar file with the new one, with the
> expectation that the new plugin would get picked  up when I restarted
> the cluster during some upcoming maintenance.
>
> I deleted an index which then triggered every node to crash a the same
> time. Doh! If you're curious, here is the core dump:
> https://gist.github.com/4658564
>
> Did I commit a cardinal sin updating the plugin jar while the cluster
> was active or should this condition be handled better by elasticsearch?
>
> Thanks!
> Paul
> --
> You received this message because you are subscribed to the Google
> Groups "elasticsearch" group.
> To unsubscribe from this group, send email to
> [hidden email].
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [hidden email].
For more options, visit https://groups.google.com/groups/opt_out.