Disaster Recovery Planning

When it comes to DR, failure to plan is planning to fail. You simply must be thinking about what to do in the event of a disaster. You must have a plan. And at each step of the plan you need to ask yourself a very simple question:

“What will we do if…?”

Here is an example:

“What will we do if the server melts?”

“We have SAN replication and would just need to failover and boot up the DR server.”

“What will we do if the SAN replication had stopped working?”

“We would recover from tape to restore the server to last night.”

“What will we do if the tapes are no good?”

“We would recover from the last known good tapes.”

“What will we do if the DR server does not boot in a timely manner, after locating those tapes?”

“We would build a new server.”

“What will we do if there is no other server available to be built?”

“We would go buy another server.”

“What will we do if we cannot buy a compatible server to the one we lost?”

“We could take the test server and redeploy it as production for the time being.”

Now, compare that series of questions to the following ones (which seems to be a more familiar and likely conversation that takes place):

“What will we do if the server suffers a hard drive failure?”

“We have the disks in a RAID5 array.”

“What will we do if a second disk fails before the array is rebuilt?”

“The chances of two disks failing is minimal.”

“So…what will we do if two disks fail?”

“There is little chance of that happening.”

“Um…OK…so, what will we do if that happens?”

“Don’t worry about it, it won’t happen.”

Sorry, but that doesn’t leave me warm and fuzzy inside, and you should not feel comfortable either. When it comes to DR, you can never ask enough questions.

Leave a Reply