Hi,
I've recently set up a syslog box running slack 9. The server is handling massive input from various servers around my network. Our pix, packeteer, etc... are all logging to the syslog. I generate 2gig log files on a daily basis, and it is critical that this syslog bog is reliable. The wierd thing is, syslog appears to randomly stop working, with no mention in the log files or anything. When the syslog stops working, a reboot does not fix it, but a manual "./rc.syslog restart" does fix it. I really have no idea where to go with this because with syslog down there are no log messages to work with. I think the klogd does start though.

The only change I have made to the syslog system is the following two lines in the rc.syslog init script:

Then in the line where the syslogd is called I added:

Forgive me if the paths are incorrect I don't have the box in front of me. I am also unsure of the syslog version, but the slack site says that it runs sysklogd-1.4.1. Any ideas are greatly appreciated!

Mike.

I would start by making sure there is no limit to the size of the log file being generated. When a file reaches the 2GB size it is considered by the OS as a "large file". Some applications have issues with dealing with "large files". I would try to setup a log rotation that rotates the logs and starts writing to a new log once you hit 2GB.

Send some more details to the problem if this doesn't solve it.

BTG

I'll look into size limits and let you know, thanks.

k, syslogd -rd shows a file size limit is being exceeded at 2gb. However I'm running reiserfs 3.6 and I quoth:

this file size max is for version 3.6, and that is what I'm running. I'm wondering if it is a limit of the system, it's a PIII. Any ideas? I'd really rather not go the route of a cron job, I'd prefer having it accept whatever file size we deal out. There is nothing on the system that suggests this limit should exist. ulimit -a shows unlimited in most areas, including file size. Kernel 2.4.27. I'm baffled.

Thanks!