Register now or log in to join your professional community.
As a DBA Backup and Restore (BnR) tests are regular in our day to day operations. On-Site backups can be divided into two broad categories: 1) Software (Apps and DB) 2) Machine level - I.T shops nowadays are highly virtualized where it is common for a Systems Admin to schedule regular backups of VMs
How to validate the backup? Will it get restored and recovered in a disaster?
1.a) Most database vendors ship tools to validate there DB backups
1.b) During a clone request of DB and Application from PROD to TEST environment. It is recommended to record the details of TEST environment before and after the clone activity. This will serve the purpose of BnR activity. It can be submitted as proof during compliance audits.
1.c) To ensure your off-site backups on tape comply with your Organizations BnR policy and restore can happen with the defined RTO, a quarterly restore excersise from tape is recommended.
2) Restore of PROD VM backup makes less sense at DB & application level because restoring different parts of the system to different points in time can lead to inconsistent data. DB and Applications restore and recovery work in there own fashion.
Yes, VM restore is helpful in going back to good known/working configuration of O.S.
As an IT Manager Facing disaster is something you should always think about! So from my point of view Virtualization test environment is the key / keep Mimic servers for the most critical production server and do the restore / disaster recovery test in clear schedule