PDA

View Full Version : Outage 3/15/2008


andy
Mar 16th, 2008, 10:37 am
What happened yesterday, why have we been offline?

Yesterday we put another milestone in terms of our backup strategy into production. We now have an offsite database server, shadowing the production data up to the second. Everytime something really fundamental like this is put into production we have to do something that is known in DB terms as 'Cold Backup' What that means is basically in order to get a guaranteed consistent status of the DB we have to shut down the DB, then take a snap shot and after that it can be restarted.

Well, in essence that is what we did. In order to get our offsite DB backup server online and synced up with the main server we had to do a cold backup. Then after transfering the backup to our offsite server we have now a database (and everything else) stored offsite so that even in the event of a natural desaster our data is not lost.