httpd-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Webmaster" <webmas...@filarescue.org>
Subject [users@httpd] uh oh...
Date Fri, 16 Jul 2004 13:15:39 GMT
*This message was transferred with a trial version of CommuniGate(tm) Pro*

The server normally has a load average of almost 0 (way more server than needed for the little
bit of traffic it gets).

But this morning I woke up to the following:


last pid: 99942;  load averages: 20.07, 19.90, 19.52                                     
                            up 2+21:00:47  07:29:18
78 processes:  21 running, 56 sleeping, 1 zombie
CPU states: 83.3% user,  0.0% nice, 16.7% system,  0.0% interrupt,  0.0% idle
Mem: 228M Active, 601M Inact, 125M Wired, 39M Cache, 112M Buf, 8780K Free
Swap: 2048M Total, 68K Used, 2048M Free

  PID USERNAME   PRI NICE   SIZE    RES STATE    TIME   WCPU    CPU COMMAND
13445 www        127    0 20964K 10936K RUN     40:11  4.49%  4.49% httpd
13984 www        128    0 20900K 10928K RUN     15:42  4.39%  4.39% httpd
13444 www        127    0 20964K 10944K RUN     40:54  4.25%  4.25% httpd
13773 www        127    0 21104K 11148K RUN     25:46  4.25%  4.25% httpd
13443 www        127    0 20968K 11016K RUN     24:14  4.25%  4.25% httpd
13985 www        127    0 20900K 10928K RUN     15:41  4.25%  4.25% httpd
13837 www        127    0 21100K 11256K RUN      0:29  4.25%  4.25% httpd
13457 www        127    0 20948K 10992K RUN     39:31  4.20%  4.20% httpd
13816 www        127    0 20916K 10972K RUN     15:50  4.20%  4.20% httpd
13983 www        127    0 20900K 10928K RUN     15:42  4.15%  4.15% httpd
13772 www        127    0 20908K 10960K RUN     15:33  4.15%  4.15% httpd
13988 www        128    0 20908K 10956K RUN     13:32  4.15%  4.15% httpd
13987 www        127    0 20900K 10940K RUN     15:33  4.10%  4.10% httpd
13820 www        126    0 20916K 10968K RUN     15:54  4.05%  4.05% httpd
13986 www        127    0 20900K 10928K RUN     15:38  4.05%  4.05% httpd
13774 www        127    0 20896K 10932K RUN     26:20  3.96%  3.96% httpd
13441 www        126    0 21124K 11168K RUN     24:55  3.91%  3.91% httpd
13449 www        126    0 21140K 11164K RUN     30:34  3.86%  3.86% httpd


This was in my system log:

Jul 15 23:51:02 server kernel: pid 26512 (httpd), uid 0: exited on signal 10 (core dumped)
Jul 15 23:53:57 server kernel: pid 13402 (httpd), uid 80: exited on signal 11
Jul 15 23:54:03 server kernel: pid 13403 (httpd), uid 80: exited on signal 11
Jul 15 23:54:10 server kernel: pid 13404 (httpd), uid 80: exited on signal 11
Jul 15 23:54:22 server kernel: pid 13405 (httpd), uid 80: exited on signal 11

(I now have /usr/local/httpd.core, over 3 megs)


This was in the httpd-error.log:

[Thu Jul 15 23:55:54 2004] [info] (57)Socket is not connected: core_output_filter: writing
data to the network
[Thu Jul 15 23:56:07 2004] [info] (57)Socket is not connected: core_output_filter: writing
data to the network
[Thu Jul 15 23:59:51 2004] [info] (32)Broken pipe: core_output_filter: writing data to the
network


It had been running wonderfully, but I was getting strange error messages in my error log
because of mod_security, so I commented it out of the httpd.conf file and restarted. Now I
get the above (worse) problem.

So... I uncommented the mod_security, figuring I'd rather have the mutex error entries in
the log file than a runaway httpd, but when I tried to restart, I get this:

[Fri Jul 16 07:55:22 2004] [notice] SIGHUP received.  Attempting to restart
[Fri Jul 16 07:55:23 2004] [notice] seg fault or similar nasty error detected in the parent
process

A second attempt at starting worked, though - with no changes... just waited a couple of minutes
and tried again. I'm still waiting for the server load to skyrocket. Right now it's normal.

It's a FreeBSD 5.2.1-RELEASE box with the Apache 2.0.50

As for the core dump, I used (for the first time) gdb to look at the core dump. When I first
started gdb, it gave me:

Core was generated by `httpd'.
Program terminated with signal 4, Illegal instruction.
#0  0x28f5f712 in ?? ()

Not sure what else to do with gdb, so I'll stop working with that for now.

*banging head*

Chris



---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
   "   from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org


Mime
View raw message