Hi:
Our web application is running on tomcat 5.5.23 and apache 2.x (on
Linux 2.6.18-194.3.1.el5). After 12 hours the application would just halt. The
application log and the catalina.out does not indicate any problems (i.e. no
exception thrown). JVM is java 1.5.20 (Sun)
I noticed that each time this occur, apache would spawn many child processes. I
did an strace and here's what I found
# strace of parent httpd process
select(0, NULL, NULL, NULL, {1, 0}) = 0 (Timeout)
wait4(-1, 0x7ffff3636e8c, WNOHANG|WSTOPPED, NULL) = 0
select(0, NULL, NULL, NULL, {1, 0}) = 0 (Timeout)
wait4(-1, 0x7ffff3636e8c, WNOHANG|WSTOPPED, NULL) = 0
select(0, NULL, NULL, NULL, {1, 0}) = 0 (Timeout)
wait4(-1, 0x7ffff3636e8c, WNOHANG|WSTOPPED, NULL) = 0
select(0, NULL, NULL, NULL, {1, 0}) = 0 (Timeout)
# strace of the child httpd process
Process 26138 attached - interrupt to quit
read(21, <unfinished ...>
Process 26138 detached
#strace of tomcat
Process 3566 attached - interrupt to quit
accept(29,
Our application use SiteMinder as authentication mechanism, and SiteMinder
would spew logs (on httpd error_log) get trying to get semaphore and SiteMinder
agent running.
Any ideas on how should I proceed to troubleshoot? Does anyone encounter this
problem before?
Thanks in advance.
Regards,
Ludwig_________________________________________________
Philippine Linux Users' Group (PLUG) Mailing List
http://lists.linux.org.ph/mailman/listinfo/plug
Searchable Archives: http://archives.free.net.ph