We are frequently asked about our recommend workflow for BRU Server™. While there are many workflows that BRU Server™ can manage depending on your particular needs, this is an example of one used by TOLIS Group for our internal corporate backups.
The purpose of this workflow is to create jobs that properly mitigate a disaster should it happen.
We use an ArGest® LTO-6 TGL2240 internally for our business backups. Each full backup takes 2 tapes. We also have 40TB of RAID 5 disk for Staged D2D in our bruAPP™ Backup Appliance.
In the BRU Server™ environment, we have the destinations in our ArGest® TGL2240 set up as follows:
Week A – Slots 1-2
Week B – Slots 3-4
Week C – Slots 5-6
Week D – Slots 7-8
Monthly Offsite – Slots 13-14
Quarterly Offsite – Slots 15-16
As sub-jobs and destinations, the engineering data all fits on a single LTO-6 tape, so we have
Engineering A – Slot 9
Engineering B – Slot 10
Engineering C – Slot 11
Engineering D – Slot 12
Engineering Offsite – Slot 17
Engineering Quarterly – Slot 18
Slots 19-23 are for emergency jobs – both backup and restore – while slot 24 is the cleaning cartridge
This layout means that the tapes in the left magazine are pretty static while the tapes in the right magazine are the ones being swapped in and out.
Our backup rotations work this way:
Each Friday night, the full job for all client systems run to a Week destination on a 4 week rotation schedule.
The first Saturday of each month, a separate full job for all client systems run on a monthly schedule with each set being taken offsite to one of the employees’ houses where it is rotated back in every 6th month.
The last Sunday of each quarter, a separate full job for all clients is run to the Quarterly Offsite destination and then taken to our bank’s safe deposit box and never brought back in.
Every day at 12:30 and 18:30, an incremental job of all pertinent systems is run to the server’s Disk Stage and that environment is set to a 91 day max stage age – meaning that the system automatically removes any incremental jobs from the disk as soon as they are 92 days old keeping the disk stage automatically groomed.
The engineering schedules are similar, except that the quarterlies are stored by our law firm for software code escrow for customers that have a code escrow agreement with us.
This process keeps 4 weeks of full archives in-house, 6 months of full archives offsite, but rotating back in for overwriting, and a set of legal full jobs for business purposes that are never overwritten. The incremental archives are not needed after a new full job, but we keep them around for 3 months for “oops” coverage.
Again, there are many workflows that we are able to discuss and support, this is simply one example.