Our daily checklist is actually a checklist on a web page. There’s a checkbox for each machine on each client with a status next to the checkbox.

If the status is “OK”, it means all the checks have passed. If the status is “WARNING” or “ERROR”, it means one or more of the checks has problems.

Every weekday we “do the dailies”, and we don’t check the checkbox until we have investigated the problem, or at least made a ticket to investigate the problem.

This particularly helps for the “OK” checks — we don’t have to look at anything, and when 2/3 or more of your machines have a status of “OK”, it saves tons of time.

And when the remaining 1/3 or fewer come out as not OK, it’s easy to see what’s not OK. Maybe the backup failed, maybe the available free tablespace is starting to get low (we page when it’s really low, but our daily checks catch it when it is getting there, so we have plenty of time to fix it if we need to — adding tablespaces in MySQL requires a restart).