Solaris

What is the best way to monitor the health of solaris zones? Through the global zone or through the individual zones itself ?

I would say create a shell script which monitors the health of your server and generate mails when the server is down

Chandra

Like your humor

Code:

ping localhost >/dev/null || echo "This server is down" | mail your.address@your.location

---------- Post updated at 01:05 PM ---------- Previous update was at 09:51 AM ----------

Honestly, you must run the ping test from outside!
E.g. the physical system (global zone) can ping the zones. But better you ping from an external monitoring server.
What do you mean by "health"? Just ping-able?

Can you elaborate what do you mean by monitor of solaris health zone ?

If it's only a health check of solaris servers then you can have a script which will monitor the cpu usage, load, monitor IO, memory utilization and check of syslog files for any error and send the mail if any error found or any of the resource reached to its threshold limit and generate a mail to DL/mail id as per requirement and support time.

You can schedule the script after a particular interval with the help of scheduler(Autosys, Tivoli, Control-M, etc.) as per the criticality of the server in the environment.

If case of server down or not pinging job schedule on scheduler for monitoring will get fail and failure alarm/notification will get generated.

Thanks for all the responses.
By health, I did mean that the zone is alive/ping-able/running.

"zoneadm -z zonename list -v" from the global zone will tell you if the zone is alive/running (column STATUS).

You can do ping tests from the global zone but better to do them remotely to rule out cabling or other networking issues.