Every day, I'm assuming when the server goes on crunching the logs, my RaQ3 becomes unresponsive. This takes 2 hours! I have 256mb and I don't use any scripting beyond SSI, I don't use webalizer, I have all the updates from the cobalt site.

When I run "top", I see "swatch" and "parseReport.pl" taking up most of the CPU and Memory (over 95% between them). I even erased my entire logs at one point to see if it helps (it didn't).

While I do get about 30-50gb traffic a month, this was the case last year as well and I wasn't getting this problem...

Any ideas how I can resolve this?

Search for web.stats, ftp.stats and mail.stats. Delete these files and the .cache ones and this should resolve things. Seems like it's trying to turn over big log files.

I wiped all ".stats" files over 10mb and all ".cache*" files over 10mb, this perl script is still hanging my server...

Any ideas?

I remember having a similar problem on my RAQ3 a year ago. I killed the swatch pid, Rebooted and thn ran top to see if swatch ran again. If it did i killed it and that fixed things.

kill -9 1234

replace 1234 with the PID number

Ok the problem is most likely anlog I had many servers doing this and so I disbaled anlog as I had my own webstats anway

cp /usr/local/sbin/analog /usr/local/sbin/analog.old

rm /usr/local/sbin/analog

pico /usr/local/sbin/analog then add this to that file

!/bin/sh
exit 0

then chmod 755 /usr/local/sbin/analog


now its disbaled and should help a lot