I scripted our backup jobs to send an email to me if there’s a problem. This works well enough, usually.
However this week the backups started failing because of a dodgy disk, but it seemed I wasn’t getting the alerts. It wasn’t until one of the sysadmins mentioned the problem with the disk that I found out there was a problem. I made a mental note to check out why I hadn’t been alerted but, so far, hadn’t got any further than that.
As it turns out the alerts weren’t failing, but were being created with no ‘Sent’ date, and for some reason I must have chosen to display the ‘Sent’ time rather than the ‘Received’ time in Outlook, and to sort by it.
So, the email alerts were coming in to the bottom of my Inbox where in the normal course of events I’d never see them. Doh!
I need to implement some sort of positive check on the success of the backups – in previous jobs I’ve set up a directory listing of the directory with the backup files in and incorporated that in a daily check. I’ll probably do the same again.