On 05/02/2015 09:11, 토로치 wrote: > Hello. > > > I am a PLM system developers. > > We PLM system operates by JAVA. > > We are using the Tomcat version 7.0.55. > > The system Down occurs when we are using a PLM system. > > The hs-err_pid5684.mdmp file over 20GB is created in the Tomcat root folder. > > When the contents of the hs_err_pid5512.log file shown that a JVM bug.
If you think you have hit a JVM bug, why are you posting here? <snip/> > Current thread (0x00000000384c9800): JavaThread "http-apr-9500-exec-226" > daemon [_thread_in_native, id=18668, > stack(0x000000005a630000,0x000000005a730000)] That tells me you are using the APR/native connector. It is possible that that connector is casing the problem but the stack trace isn't what I'd expect to see in that case. I recommend switching to the BIO HTTP connector and seeing if the crash still occurs. <snip/> > JAVA_HOME=C:\Java\x64\jdk1.7.0_03 That is a very old JVM. You should also try updating that to the latest 7.0.x release and see if the bug reoccurs. On a related topic, make sure you are using the latest native component for the APR/native connector (1.1.32 as I type this). An upgrade to the latest Tomcat 7.0.x release wouldn't hurt either. In short: - get everything up to date - test HTTP BIO vs HTTP APR/native Mark --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
