#Cluster recovery using .7z backups - field feedback needed

1 messages · Page 1 of 1 (latest)

flat thicket
#

Hi all, I am today curious to gather feedback regarding this activity, especially on the steps after a real disaster... Thanks in advance !

digital spear
#

after a real disaster it's probably quicker and easier to setup a new cluster and restore the data from your backup (SnapMirror/Tape)

#

but if you're really curious, here is the procedure

#

the main benefit for a cluster DB recovery is if you still have your disks with aggregates/data intact, as the recovery will make sure the correct volumes are attached to the correct SVMs. If you don't have the data anymore you need to restore from backup anyways

deft sundial
#

This right here is what makes snapmirror awesome. And for longer term customers that have gone through a refresh cycle or three, you end up doing this natural cascade of older systems becoming secondary mirror/vault targets that you can pretty quickly recover from (subjective, depends on bandwidth/amount-of-data) using reversed snapmirrors or vault recoveries from snaps within those mirrored volumes.