Solaris

This came to our attention when one of our scripts didn't run over the weekend prompting us to check our messages logging in /var/adm/messages. To our surprise, there was no system error/activity logging to the file except dates. Simply our /var/adm/message* were all filled with just your typical date outputs. We have other zones on the same machine that do NOT have this issue, so it's isolated to just this zone.

Any suggestions aside from logger which, we did and it posted to the messages file ?

Initially I would get hold of the /etc/syslog.conf file from a working local zone and compare that to the file on the non-logging zone and see what the difference is.

Did you check syslog.conf on the problem zone - if there is any doubt about the contents do a diff against a syslog.conf file with working logging.

.... assuming you are not using rsyslog and svcadm enable turned on messaging.

Edit: Dennis beat me to it.

Thanks, using diff helped me figure out what was missing.

I compared both global and local zone against the zone which is not logging. The difference was not much if not, anything warranted for a change. This is actually baffling as this appears to be somewhat of a common problem on solaris systems after a search on the net. I'll still figure out what puzzle piece is missing but I've been on this for about 2 weeks and even Oracle support doesn't know what's going on. They've been just providing me with scripts.

What does the

Code:

svcs

command show? You can have a service in different status modes.

Status: enabled for syslogd

Also checked the log provided within svcs and nothing out of the ordinary