You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@forrest.apache.org by "Tim Williams (JIRA)" <ji...@apache.org> on 2008/07/07 13:09:31 UTC
[jira] Commented: (FOR-1082) Concurrency testing of the locationmap
throws errors
[ https://issues.apache.org/jira/browse/FOR-1082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12611041#action_12611041 ]
Tim Williams commented on FOR-1082:
-----------------------------------
Happened across this article that appears both interesting and relevant. As I mentioned on list, maybe the ConcurrentHashMap will be the answer to not synchronize the whole getAttribute method.
http://www.theserverside.com/tt/articles/article.tss?l=ScalingYourJavaEEApplications
> Concurrency testing of the locationmap throws errors
> ----------------------------------------------------
>
> Key: FOR-1082
> URL: https://issues.apache.org/jira/browse/FOR-1082
> Project: Forrest
> Issue Type: Bug
> Components: Locationmap
> Reporter: Thorsten Scherler
>
> http://marc.info/?t=120180448900001&r=1&w=2
> The problem that I encounter is that the locationmap is throwing an
> error as soon as there are concurrent user.
> Try
> <map:match pattern="test">
> <map:generate src="lm://transform.html.document"/>
> <map:serialize />
> </map:match>
> Doing some testing in my case I use jmeter with 250 threads and 100 loop
> where the ramp-up time is 20.
> This provokes that the input module is requested on the startup with
> more then one request. It seems that the requests are issued as fast
> that the module seems to fail to finish to load the first time the
> locationmap.
> I changed
> public synchronized Object getAttribute(...)
> which seems to fix the problem.
> However we may want to look into a possibility to prevent using the synchronized in the whole method.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.