disaster recoveryWell, I suppose we can’t forget to actually document how to recover the servers themselves. The best thing that has ever happened for planning DR is virtualization. It gives a unified recovery method for greatly differing VMs on disparate operating systems with different hardware configurations.

The key item that saves the most time is — rather than documenting 30 servers and having each one have a full recovery method — grouping servers together. For example, document that server X, Y, Z are all virtual, and are recovered by following X procedure. That way, as you create servers and change things around, you can simply add them to the list of how they are recovered.

Don’t forget to document the different recovery methods! For example, if you have a stretch cluster across a campus, with HA configured properly, nothing needs to be done for a single server room/building failure. But, if you have a whole campus failure, your recovery method at the DR site is different. Also remember that while all virtual servers might be in the stretch HA at the main site, they may not all be replicated to the DR site, so some may need to come from backups instead.

Since we’ve now consolidated our documentation down to a few recovery methods rather than one for each server, take the time to make this extremely detailed. If the building goes up in flames while you’re out of town on vacation, you want someone else to be able to recover things without needing your assistance. Make sure to document pre-requisite servers as well for everything. A properly configured data flowchart for a system will take care of documenting interdependencies. Additionally, passwords can be huge in a DR scenario; remember that the password repository you keep on the file server is now down. You may need that Directory Services Restore Mode password that you set six years ago and never changed to even be able to get to that file server up and running again.

Check out last week’s post on how to construct a proper disaster recovery plan, or skip around the rest of the series:

Are you in need of assistance in creating and maintaining a solid backup and disaster recovery plan? Email us or give us a call at 502-240-0404 to get started.