Unmanaged Postgres recovery from snapshot - Consul error

Hi Fly Team,
My app upg-tranquil-coral-757 was hosted in GRU and suffered from the recent hardware failure. I managed to create a new volume in IAD from a recent snapshot (vol_4912z9037n0k00or). However, when I try to run a new machine with postgres-flex:15 attached to this volume, it constantly crashes with exit_code=2 (Failed to resolve source machine / Consul error). Could you please help me reset the cluster metadata for upg-tranquil-coral-757 so it can boot as a single-node app in IAD? Thanks!


Hi… It’s usually easiest to create an entire new Postgres app, via fly postgres create --snapshot-id, if you haven’t tried that yet…

https://docs.fly.io/unmanaged-postgres/managing/backup-and-restore#restoring-from-a-snapshot

(Be careful with the --image-ref, since that’s a very common source of problems.)

There are also a few tips for the reattachment phase in this classic forum post:

https://community.fly.io/t/urgency-problems-with-postgres-the-database-is-not-responding/19926/2

(The first line in that sequence, the one with --fork-from, won’t work for you, though, since the volume itself isn’t reachable in your case.)

Hey @limaantonio, as always @mayailurus has some solid advice here, from what I can see you managed to get a single machine healthy shortly after posting, is your postgres back up now?

We strongly suggest running multiple machines and with UPG this shouldn’t be too hard to do though this is completely managed by yourself.