Sorry, I've been out of town for a few days. >-----Original Message----- >From: Gerhard Froehlich [mailto:[EMAIL PROTECTED]] >Sent: Friday, February 22, 2002 8:37 AM >To: [EMAIL PROTECTED]; [EMAIL PROTECTED] >Subject: RE: Uninterruptable Sleep > > >Colin, > >>I am using a source distribution I downloaded from from the >apache dist site >>at the end of January. > >Must be 2.01, or?
Yes, it's 2.01 > >>Some further information that might be of interest: >> - My testing environment did NOT exhibited this behaviour >even after I >>bombarded it for a full hour using jmeter running 30+ threads. > >Ahh jmeter on the same machine? I made very negative >experience with that >tool. For stressing my applications, I use the Microsoft Webapplication >Stress tool: ><http://www.microsoft.com/technet/treeview/default.asp?url=/Tec hNet/itsolutions/intranet/downloads/w >ebstres.asp> Thanks, I'll try it out. > >It's from MS <spit>, but it's (ahem) good. > >> - The production server went into this state within 2 >minutes of running a >>scaled back version (6 threads) of the same jmeter test. > >Try it without JMeter again... > >>Production is RedHat 6.1 (2.2.12-20) jdk1.3.1_01 >>Test is SuSe 6.4 (2.2.14) jdk1.3.0 >>both are tomcat 3.2.3 >> >>My config is: >> Tomcat: >> -Xmx 256M -Xms 128M >> StoreJanitor: >> <parameter name="freememory" value="64000000"/> >> <parameter name="heapsize" value="256000000"/> >> >>After reading the links you've provided, I see that I may have to >>re-consider these settings. > >Indeed: >How many RAM has your machine? This is another strange thing. There is 512M, but top only shows 64M...I think that I should update top >- Don't set the -Xms parameter, I read that somewhere in a performace >article. I have to update the Cocoon docs. >- Don't set the -Xmx == RAM you have. Your machine needs some >memory, too. >- Set the heapsize always lower as the -Xmx paramter, >otherwise your JVM >can run out of Memory. >- Don't set free memory that high, I think 10megs should be >enough. Consider, >this formel: > if ((JVM totalMemory > heapsize AND JVM freeMemory() < >freememory) == true) >then the cleanup thread kicks in. Wrong settings can cause to >many thread >iteration. > OK, I'll use this config. >>Are you aware of any kernel or JDK version issues that may >cause problems? > >Nope, unfortunately it's always the code ;-). > >>Thanks for the information! > >No problem, you're welcome. > >BTW: can you try the latest CVS, just for comparing? > I will try it out. Thanks again! > ~Gerhard > >---------------------------- >I just found the last bug... >---------------------------- > > >--------------------------------------------------------------------- >Please check that your question has not already been answered in the >FAQ before posting. <http://xml.apache.org/cocoon/faqs.html> > >To unsubscribe, e-mail: <[EMAIL PROTECTED]> >For additional commands, e-mail: <[EMAIL PROTECTED]> --------------------------------------------------------------------- Please check that your question has not already been answered in the FAQ before posting. <http://xml.apache.org/cocoon/faqs.html> To unsubscribe, e-mail: <[EMAIL PROTECTED]> For additional commands, e-mail: <[EMAIL PROTECTED]>
