does puppetdb 4.x handle brief postgresql backend outage

[MarkB]

Does anyone know the best way to handle an upgrade of postgresql that is being used as the backend of PuppetDB?  I've got some PuppetDB 4.2 instances talking to Postgresql 9.4.7 and want to do a minor upgrade to 9.4.15.  I've already tested the procedure using a cloned Postgresql instance, and it only takes 10 minutes (this is on AWS using RDS by the way).

My thought is that PuppetDB will automatically recover after Postgresql comes back, but I'm not sure, as I can't find any documentation anywhere about this scenario.

I could also simply shutdown the PuppetDB while the Postgresql upgrade is happening, but not sure if that would be necessary.  I seem to remember reading somewhere that PuppetDB will queue up any reports/facts and such that come in while the backend is down, but I've not been able to dig up any docs that say that, so I'm not sure if I'm remembering correctly or not.

Anyone have any experience or thoughts on this scenario?

Thanks,

   --Mark

[Peter Bauer]

I have an older version, however PuppetDB reconnects on the first Agent run so the first one fails but subsequent runs will succeed.