weird problem w/syslogd

[K.C. Chan]

All, syslogd is not logging. All my messages file is of zero length. When I restarted syslog manually; syslog failed to start, but if I do "./syslog status", it tells me that syslog is running. However, when I run the cmd manually "syslod -m 0" and "echo $?", it returns code "130", I believe it should be a return code of 0. Any one have any idea as to why syslogd is not behaving properly? BTW, this is on RH 7.1 running kernel version 2.4.2-2smp. Thanks.

[Shane Hjorth]

Michael,

Don't know how much help this is going to be but this is the result I get from a Redhat 7.1 system running the same kernel:

[root@mash init.d]# ./syslog status
syslogd (pid 438) is running...
klogd (pid 443) is running...

[root@mash init.d]# syslogd -m 0
syslogd: Already running.
[root@mash init.d]# echo $?
1

So what I can say is this:

run ps -aux | grep syslog, make sure there IS no process running (don't trust the script), if there is manually kill it.

Look at the file /var/run/syslogd.pid and see which process id it thinks syslogd is running under.

If syslogd happened to be running, then kill it, move the messages file across to messages.bak and touch the messages file (giving it the same permissions/owner as the original messages file). Then try restarting syslogd.

Good luck,

Shane.