This simply isn't playing nice :/ On its daily run, it's fine, but on
the weekly, it runs for well over an hour restarting syslogd several
times per second. It hasn't completed yet for this weekly rotate, but:
root@bladehawke /var/log # head syslog; echo ; tail syslog; echo ; \
grep "syslogd 1.3-3: restart" syslog | wc -l
Aug 20 15:02:40 bladehawke syslogd 1.3-3: restart.
Aug 20 15:02:51 bladehawke syslogd 1.3-3: restart.
Aug 20 15:02:51 bladehawke syslogd 1.3-3: restart.
Aug 20 15:02:53 bladehawke syslogd 1.3-3: restart.
Aug 20 15:02:53 bladehawke syslogd 1.3-3: restart.
Aug 20 15:02:56 bladehawke syslogd 1.3-3: restart.
Aug 20 15:02:57 bladehawke syslogd 1.3-3: restart.
Aug 20 15:02:58 bladehawke syslogd 1.3-3: restart.
Aug 20 15:03:06 bladehawke syslogd 1.3-3: restart.
Aug 20 15:03:06 bladehawke syslogd 1.3-3: restart.
Aug 20 15:44:52 bladehawke syslogd 1.3-3: restart.
Aug 20 15:44:52 bladehawke syslogd 1.3-3: restart.
Aug 20 15:44:53 bladehawke syslogd 1.3-3: restart.
Aug 20 15:44:53 bladehawke syslogd 1.3-3: restart.
Aug 20 15:44:53 bladehawke syslogd 1.3-3: restart.
Aug 20 15:44:53 bladehawke syslogd 1.3-3: restart.
Aug 20 15:44:53 bladehawke syslogd 1.3-3: restart.
Aug 20 15:44:53 bladehawke syslogd 1.3-3: restart.
Aug 20 15:44:54 bladehawke syslogd 1.3-3: restart.
Aug 20 15:44:54 bladehawke syslogd 1.3-3: restart.
13478
No matter what kind of rotation is needed, restarting syslogd 13,478
times is a _little_ overboard. If last week's logs are any indication,
I'm about 2/3 of the way through it:
root@bladehawke /var/log # zcat syslog.1.gz |\
grep "syslogd 1.3-3: restart" | grep "Aug 13" | wc -l
21885
--
Anton Graham GPG ID: 0x18F78541
<[EMAIL PROTECTED]> RSA key available upon request
Perhaps no person can be a poet, or even enjoy poetry without a certain
unsoundness of mind.
-- Thomas Macaulay
