Please Fix
The log file for database ‘databasename’ is full. Back up the transaction log for the database to free up some log space.
At the end of the email were two words: Please fix.
Okay, we should be able to do that. Of course, we have other issues now. See, the log drive for this box was about 70Gb. He did not just fill up his log, he filled up the disk. Do you know what happens to other databases that cannot use the disk to write their logs? Yeah, I know you do. Too bad he did not. Of course, this junior DBA figured that with the database in simple recovery mode that the log would not grow, and assumed we should be able to fix things in about two minutes.
Why the hell not call us at the start? As a result, the server was out of service for a few hours as we tried to get things back in order. All the while he complains that if we had done things right to begin with, none of this would have happened. My thinking? Yeah, if we had never hired you to begin with, then none of this would have happened.
Anyway, this snowballs into more problems. Now people want to know why we are not monitoring the available free space on our disk drives. Huh? We are! We knew the drive had filled up! “Well, why didn’t you do anything?” Do what? Go to the box and start killing spids? I am not the one that kicked this thing off, you did. How about having some understanding of the basic technology you are being asked to work with? Once we saw the issue we starting working to fix it. Any idea how long it takes to undo a transaction that filled up a 70Gb drive? It takes more than a few minutes.
The OZ
Aw Comeon, surely you could have just magically kicked off a monitor process for things such as run away log files and solved his problem before he knew he had one, while your at it, could you please have that script monitor all spid’s for this on all your servers by oh say 5pm this evening. Never mind that if someone didnt approve the access that allowed this to happen in the first place this would have never happened.