Successful recovery plans and implementation

[Jeremy Evans]

Hello, 

I am attempting to use barman for postgres recovery.  I am using streaming backup (postgres) and can successfully restore to a new standalone node.

The issue is what to do now.  I use pgautofailover for my cluster environment, and if I restore to a single node in a cluster, then bring the node online, pgautofailover complains about the system identifier being incorrect (because this is a new cluster so it is different).  

Is there a correct process for what I am attempting to do in this situation?  This procedure would need to be done in the case of catastrophic cluster failure.

My next step would be to implement always on archiving for true PIT recovery (different scenario than the above).  However, there seems to be limited information on recovery procedures in that area as well.  Would someone be able to point me to more information on that?

Thanks in advance.