Happened to me too, very weird behaviour
I'll go ask about it

Team's working on a fix

Sorry for the trouble

Angelo from the team explained what happened here if you're curious

Hey @alex73630 -

We just got the fix in, we are going to be doing a few things.

Railway sparingly moves workloads around as a function of our ability to handle capacity. Box gets full, we move 'em. Box needs an upgrade? Move 'em.

This time, we did a type change on how we handle multi-region deployments because we need to keep track of regions across cloud providers GCP <-> Railway Metal

However, because we didn't account for that, we sent them to a new, un-working region: Region 0
Moving forward, we will:

• Put a Railway banner whenever the platform takes an action on their behalf. Avoiding confusion if we do a restart.

• Not send them to Region0

In your case, your existing workload is fine an operable yes?

New deploys should have gone out just fine.

Hey Angelo,
Thanks for the response.

Our current deployment is working as expected, only the newer deployments have failed to even build.
Luckily they didn't contain any new backend code and so it wasn't a huge issue.

I'll try to do trigger a deployment to see if it's fixed.

Can confirm, I was able to deploy a new build 👍
Thanks for the help !

Thanks for using Railway

(and reporting)

ty @Fragly for escalating it to us

My level of trust in Railway is the same as my current Metal usage, 99% 😄
Keep up the great work!

!s