Sometimes my RAQ stops at the time of logrotation

I have to manually restart httpd then to get is back up again.
The problem is that apache is trying to restart but a few httpd processes are probably still running. So it can't restart.

Is there an easy way to let the restart wait till the processes end, or that it kills the processes still running. The last one is preferrable since when i logged into the box after ten minutes downtime i still had to kill two httpd processes before i could restart.

Below you'll find a part of the log file. Maybe this makes things more clear.

[Tue Nov 6 04:03:05 2001] [notice] SIGUSR1 received. Doing graceful restart
[Tue Nov 6 04:18:25 2001] [notice] Apache/1.3.20 (Unix) mod_throttle/3.1.2 mod_gzip/1.3.19.1a PHP/4.0.6 mod_perl/1.26 PHP/4.0.6 mod_auth_pam/1.0a mod_ssl/2.8.4 OpenSSL/0.9.6b configured -- resuming normal operations


Thanks

I had the same problem before. Maybe its because your server is pretty loaded and your logs are huge. When they are being rotated your server is way overloaded and crashes. Try gzipping your logs and keep them somewhere safe. Then you have a fresh log file.

I only have one (test)site on it running now. Since i couldn't be near the box all the time it was to unreliable to have (commercial) sites on, so i moved them to another server.

So the box may get a few hits a day on that test domain, and still it crashes, not every day, only sometimes