987530f5-7e51-4b0c-ab55-cc1c28ff2a9f

Hi there, we're actively investigating and working on resolving this

We just pushed an update to the incident as well, you can follow along here https://status.railway.com/cmc6xto9x000lqosffho36pqo

Is there a timeline for resolution?

In addition, I used to have multiple replicas for this service before the migration to metal. Would adding a replica in another region help me here?

We don't have a resolution time yet but we have multiple engineers looking into this right now

If the underlying storage is on us-west metal then adding replicas likely won't help

That's what I figured, thanks for confirming.

But still, can I ask if that was a known issue about removing replicas? Used to have three for my django service, now I just have one in the one region.

Hi,

I also just experienced some issues with database queries on Metal. Specifically, some queries failed to find entries that had been inserted just moments before. I noticed unusually long checkpoint completion times during that period.

Project ID: de51c708-a5bf-4c3e-b101-599f8eb92a4f

Please open your own thread if the issue persists.

!s