Oh, does this sound familiar. I used to run into PermGen exhaustion all the time. The only thing to do was to kill -KILL Tomcat, raise MaxPermSize, restart, and watch it with e.g. LambdaProbe to see if we were getting into trouble again. I eventually settled on the same number you did: MaxPermSize=128M.
The state that it can get out of with patience sounds to me like you are short of heap, as you apparently suspect. My guess is that it's spending much time in the garbage collector but is gleaning enough memory to make a very little progress before it has to GC again. You should find it helpful to sample the VM's memory management stat.s repeatedly and see how close to your limits you are running. If I thought that GC was being triggered by memory starvation more than a few times a day, I'd raise the heap maximum until the periodic GC is able to maintain sufficient free memory for all but the most demanding usage spikes. More than that probably just makes GC cycles more noticeable without any benefit. -- Mark H. Wood, Lead System Programmer [email protected] Balance your desire for bells and whistles with the reality that only a little more than 2 percent of world population has broadband. -- Ledford and Tyler, _Google Analytics 2.0_
pgpxgGwj7TVIv.pgp
Description: PGP signature
------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev
_______________________________________________ DSpace-tech mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dspace-tech
