Hi Jesse, Jesse Klaasse schrieb: > After two months with a system running smoothly, we are currently > experiencing the same problems again (albeit less often). > > A little update: > - The system's memory has been upgraded to 16 GB, Tomcat's memory settings > are Xms4096m and Xmx10240m now.
When using big heaps, you need to take extra effort to get your GC settings right. Do you have GC-Logs? What are the JVM options you use to start Tomcat? > - We are using commons-dbcp 1.2.2 instead of Tomcat's internal pooling > mechanism, so with org.apache.commons.dbcp.BasicDataSourceFactory as data > source factory > - Java version is Sun JDK 1.5.0.15 now. We are still at IIS 6.0, JK 1.2.26, > Tomcat 5.5.20 and APR 1.1.14. > - We have no automatic Tomcat reboots > - Currently this 503 hang happens about once or twice a week > > Attached you will find a part of the isapi_redirector.log of today (the > whole morning 503 errors occurred). > I also tried to create a thread dump using StackTrace while Tomcat was > hanging, but it didn't really work. The other JVM's were able to produce a > thread dump, and since all JVM's are part of the same group of services, I > have included this thread dump. The attached thread dumps do not belong to Tomcat. They are from the Gauss VIP Content Server. > Any idea, anyone? First of all I assume you are talking about the event starting at 07:09? The initial problems are Timeouts (Winsock 10060). So again we need to find out, why things took to long on the backend. Thread Dumps would be a good start ... As we don't have them: - Check without APR - Check whether your GC takes to long > http://www.nabble.com/file/p20930602/20081210_isapi_redirect.zip > 20081210_isapi_redirect.zip > http://www.nabble.com/file/p20930602/20081210_threaddump.txt > 20081210_threaddump.txt Regards, Rainer --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
