On Jun 17, 2008, at 12:38 PM, Sandeep Ghael wrote:
> HI List,
>
> I have began to notice some strangeness when deploying my war files
> to my production environment.
>
> This is my scenario:
> I have a 2 server clustered setup. Resin 3.1.3 pro. Previously, my
> deployments for new versions was simply push out the new war file
> and wait for the
> watchdog to notice the file change. Then resin would expand the
> file and the server would typically restart in 10 seconds or less.
>
> Lately, I have been plagued with extremely long re-start times. And
> I mean reallllly long. On my last deployment, resin took roughly 20
> mintues to start!
> The output shows that resin appears to hang during the "expanding
> war file" stage (or atleast that the message that appears before it
> hangs). Eventually, after a long enough wait, the process gets past
> that step and things expand and deploy properly.
Try looking in the ${resin.root}/admin directory for the logging. The
3.1.3 release was logging too much information, which would cause slow
startup times as the database was validated on startup.
-- Scott
>
>
> In one instance, when this occurred, I decided not to wait for the
> on the fly restart. So i killed the java /resin process and started
> the process from scratch. The restart was still very slow.
>
> I know I'm not giving a lot of info here, but I'm not even sure
> where to begin tracking this down. Anyone seen anything like this,
> or got any advice how to track this down?
>
> regards,
> Sandeep
>
> _______________________________________________
> resin-interest mailing list
> resin-interest@caucho.com
> http://maillist.caucho.com/mailman/listinfo/resin-interest
_______________________________________________
resin-interest mailing list
resin-interest@caucho.com
http://maillist.caucho.com/mailman/listinfo/resin-interest
