Managed Postgres cluster offline... why?

Questions / Help
1

Hi there! My team has just started exploring building a Django app deployed to Fly. We noticed today that it’s throwing 500 errors and it appears our Managed Postgres cluster is down - cluster ID d1zj5omwl8ryqkv8.

Exploring the web dashboard, I see no way to examine logs for the cluster or any means of restarting it. The only thing it reports is “Your cluster is experiencing availability issues” with no further details.

Running fly mpg list from the CLI reports that its status is “ready” - but running fly mpg status d1zj5omwl8ryqkv8 reports that its status is “offline”

I tried restoring from a few different backups of varying ages. Each restored cluster also appeared to stay offline for awhile, though some of them eventually came online. One seems to be in yellow status with “Database Availability Degraded”

What am I missing, here? Is there some way to determine what happened to the offline cluster? Thanks!

2

Hi… The Managed Postgres docs mention a “Fly.io Support Portal to log tickets and get help”, which would generally be a much better place to ask questions than here in the community forum.

That falls under “[c]ustomer-facing monitoring and alerting”, which is listed as work in progress, so my guess is that there currently isn’t any option other than the above.

Hope you find a speedy resolution, either way…