Open ricetj opened 5 years ago
Comment by ShawnArnwine Sunday Jul 24, 2016 at 18:37 GMT
Mostly covered in our ATO-related SOPs and IRP/ISCP...but needs to be turned into checklists and properly assigned.
Comment by dkhenry Monday Jan 23, 2017 at 22:56 GMT
We should also test out the recovery
(09/5/2017)
(11/29/2017) Pet instances (CSR's) basically just need to be snapshotted so we can restore them if needed.
(01/17/2018) blocked by DRP tabletop, will result in more specific tix to support this effort
(05/16/2018) Still relevant?
I lean toward yes, but will leave this up to Wyatt.
I'm envisioning a brief breakdown of the types of state and systems we manage, and how we generally provide backups/restoration for those systems. Critical persistent state in RDS and whichever EBS volumes we rely on, and how we restore infrastructure with terraform/ansible. It would be useful to summarize what we'll restore from scratch vs what we need to back up.
@wyattwalter Ported this ticket over. This looks like more of an ATO issue, but may come into the ops scope. Wanted to get your take on this. Would this be a shared effort with Ops and Product Support?
@ricetj Does this ticket need VSP work?
@andreahewitt-odd This is another ticket that was ported over from Vets.gov repo. https://github.com/department-of-veterans-affairs/vets.gov-team/issues/1542 This will most likely be an ATO ticket in the future but may touch a few teams. I think we need to wait till we have ATO up and going before taking this on.
@ricetj to put this on top of the list of things to do.
@ricetj Is this still active / pending / to-be-discovered?
This is something that needs to be in the Q yes. Something the team needs to at the very least has an eye on for the future.
We should create new initiatives to address this based on the state of current infrastructure. Leaving this until we can create new issues for that work.
If the whole system comes crashing down, how do we:
The goal of the issue is slightly vague, but we can refine as we look into this.