CGI Loop?

Hi all, I was changing a couple of group memberships and when I got to
one of my groups to remove a member it just hung and when I looked at
the server with top, there was an httpd process running at 90% processor
power. So I restarted apache and everything went back to normal. When I
tried again it did the same thing, steadily in increased until it hit
about 95%. Everything else seems to be working fine. I was thinking
memory leak or something, but my boss said it sounds more to him like a
cgi loop. Can anyone help me out with this one? Thanks for your help!

Joe

Joseph Hause Continental Visinet Broadband
jhause@visi.net http://www.visi.net
Network Operations 757-873-4500/888-323-4500
"Far better it is to dare
mighty things than to take
rank with those poor timid
spirits who know neither
victory nor defeat"
Theodore Roosevelt (1899)

Hi all, I was changing a couple of group memberships and when I got to
one of my groups to remove a member it just hung and when I looked at
the server with top, there was an httpd process running at 90% processor
power. So I restarted apache and everything went back to normal. When I
tried again it did the same thing, steadily in increased until it hit
about 95%. Everything else seems to be working fine. I was thinking
memory leak or something, but my boss said it sounds more to him like a
cgi loop. Can anyone help me out with this one? Thanks for your help!

What’s in the logs? What’s the database doing? What versions of things
are you running?

Mysql 4.0.21, apache 1.33, mod_perl 1.29, rt3.2.2

Top output:

527 www 123 0 96412K 92044K RUN 1:28 89.94% 89.94% httpd
465 mysql 20 0 57264K 27176K kserel 0:16 6.20% 6.20% mysqld

Apache error log:

[Thu Dec 23 15:30:21 2004] [notice] caught SIGTERM, shutting down
[Thu Dec 23 15:30:42 2004] [notice] Apache/1.3.33 (Unix) mod_perl/1.29
configured – resuming normal operations
[Thu Dec 23 15:30:42 2004] [notice] Accept mutex: flock (Default: flock)
Out of memory during request for 992 bytes, total sbrk() is 513054720 bytes!
Callback called exit.
[Thu Dec 23 15:50:19 2004] [error] [client 206.246.194.253] File does
not exist: /opt/rt3/share/html/favicon.ico
[Thu Dec 23 16:04:28 2004] [warn] child process 6199 did not exit,
sending another SIGHUP
[Thu Dec 23 16:04:28 2004] [warn] child process 6200 did not exit,
sending another SIGHUP
[Thu Dec 23 16:04:28 2004] [warn] child process 6201 did not exit,
sending another SIGHUP
[Thu Dec 23 16:04:28 2004] [warn] child process 6288 did not exit,
sending another SIGHUP
[Thu Dec 23 16:04:28 2004] [warn] child process 6203 did not exit,
sending another SIGHUP
[Thu Dec 23 16:04:28 2004] [warn] child process 6204 did not exit,
sending another SIGHUP
[Thu Dec 23 16:04:28 2004] [warn] child process 6205 did not exit,
sending another SIGHUP
[Thu Dec 23 16:04:28 2004] [warn] child process 6206 did not exit,
sending another SIGHUP
[Thu Dec 23 16:04:28 2004] [warn] child process 6289 did not exit,
sending another SIGHUP
[Thu Dec 23 16:04:28 2004] [warn] child process 6290 did not exit,
sending another SIGHUP
[Thu Dec 23 16:04:29 2004] [warn] child process 6199 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:04:29 2004] [warn] child process 6200 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:04:29 2004] [warn] child process 6201 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:04:29 2004] [warn] child process 6288 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:04:29 2004] [warn] child process 6203 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:04:29 2004] [warn] child process 6204 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:04:29 2004] [warn] child process 6205 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:04:29 2004] [warn] child process 6206 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:04:29 2004] [warn] child process 6289 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:04:29 2004] [warn] child process 6290 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:04:33 2004] [error] child process 6199 still did not
exit, sending a SIGKILL
[Thu Dec 23 16:04:33 2004] [error] child process 6200 still did not
exit, sending a SIGKILL
[Thu Dec 23 16:04:33 2004] [error] child process 6201 still did not
exit, sending a SIGKILL
[Thu Dec 23 16:04:33 2004] [error] child process 6288 still did not
exit, sending a SIGKILL
[Thu Dec 23 16:04:33 2004] [error] child process 6203 still did not
exit, sending a SIGKILL
[Thu Dec 23 16:04:33 2004] [error] child process 6204 still did not
exit, sending a SIGKILL
[Thu Dec 23 16:04:33 2004] [error] child process 6205 still did not
exit, sending a SIGKILL
[Thu Dec 23 16:04:33 2004] [error] child process 6206 still did not
exit, sending a SIGKILL
[Thu Dec 23 16:04:33 2004] [error] child process 6289 still did not
exit, sending a SIGKILL
[Thu Dec 23 16:04:33 2004] [error] child process 6290 still did not
exit, sending a SIGKILL
[Thu Dec 23 16:04:33 2004] [notice] SIGHUP received. Attempting to restart
[Thu Dec 23 16:04:42 2004] [notice] Apache/1.3.33 (Unix) mod_perl/1.29
configured – resuming normal operations
[Thu Dec 23 16:04:42 2004] [notice] Accept mutex: flock (Default: flock)
[Thu Dec 23 16:06:58 2004] [notice] caught SIGTERM, shutting down
[Thu Dec 23 16:13:46 2004] [notice] Apache/1.3.33 (Unix) mod_perl/1.29
configured – resuming normal operations
[Thu Dec 23 16:13:46 2004] [notice] Accept mutex: flock (Default: flock)
[Thu Dec 23 16:14:52 2004] [warn] child process 501 still did not exit,
sending a SIGTERM
[Thu Dec 23 16:14:56 2004] [error] child process 501 still did not exit,
sending a SIGKILL
[Thu Dec 23 16:14:56 2004] [notice] caught SIGTERM, shutting down
[Thu Dec 23 16:15:11 2004] [notice] Apache/1.3.33 (Unix) mod_perl/1.29
configured – resuming normal operations
[Thu Dec 23 16:15:11 2004] [notice] Accept mutex: flock (Default: flock)

Jesse Vincent wrote:

Hi all, I was changing a couple of group memberships and when I got to
one of my groups to remove a member it just hung and when I looked at
the server with top, there was an httpd process running at 90% processor
power. So I restarted apache and everything went back to normal. When I
tried again it did the same thing, steadily in increased until it hit
about 95%. Everything else seems to be working fine. I was thinking
memory leak or something, but my boss said it sounds more to him like a
cgi loop. Can anyone help me out with this one? Thanks for your help!

What’s in the logs? What’s the database doing? What versions of things
are you running?


Joe


Joseph Hause Continental Visinet Broadband
jhause@visi.net http://www.visi.net
Network Operations 757-873-4500/888-323-4500

#############################################################
"Far better it is to dare
mighty things than to take
rank with those poor timid
spirits who know neither
victory nor defeat"
Theodore Roosevelt (1899)


http://lists.bestpractical.com/cgi-bin/mailman/listinfo/rt-users

Be sure to check out the RT wiki at http://wiki.bestpractical.com

Joe

Joseph Hause Continental Visinet Broadband
jhause@visi.net http://www.visi.net
Network Operations 757-873-4500/888-323-4500
"Far better it is to dare
mighty things than to take
rank with those poor timid
spirits who know neither
victory nor defeat"
Theodore Roosevelt (1899)

That is similar to my problem.

I have it blowing up like that on an endless recursion in the
SearchBuilder. The difference in my case is that it ends up eating all
memory. It is a dual opteron and it takes it about a minute or so to do
that. If your machine is slower and you have CPU ulimits set you are
likely to run into them first.

Same version of RT. Rest is newer (debian sarge).

A.

Joe Hause wrote: