Best Practice for Redundant VMs
Posted on 2013-02-03
Proper greeting of the day,
I'm running my network of 15 VMs using vSphere 5 across 3 hosts on 1 NetApp 2050 filer, in 1 NFS store. I have a FAS 270 that is also configured and has the same size as the NFS store on the 2050.
I do not have access to any 3rd party backup software, and in a month I hand over the keys to another administrator. My backup solution right now is snapshots on both the filer and vSphere, but I have no redundancy if the 2050 fails. My question is, what's the best way to ensure that my core services can continue to function/can be recovered quickly in the case of a catastrophic equipment failure? We've considered SnapMirror, but I'm worried about the performance of the live database as well as the time it would take to recover from a failure. I could build redundant VMs on the 270, but my hosts can't handle the load. Is there a way, within vSphere, to have the VMs exist on 2 data stores at once? Any help would be appreciated.