Hey there! We've found the following might help you get unblocked faster:

• 🧵 Unable to Create or See Tables in Railway Postgres from Deployed Backend

• 🧵 Connection with Postgres

• 🧵 Postgres database unreachable

If you find the answer from one of these, please let us know by solving the thread!

For the itm-backend, you're making requests to the VECTOR_DATABASE_URL right?

If so, you're round tripping ALL the way back to GCP's edge, then to our metal edge

Could you replace that public URL (viaduct.rlwy) with a private URL (postgres.railway.internal:5432)? It'll also save you on bandwidth

hey jake its actually the DATABASE_URL (we don't really use the vector db for critical operations, just as a vector store for some of our LLM stuff). The DATABASE_URL is pointed to ${{Postgres.DATABASE_PRIVATE_URL}}.

here are the connection pool settings (I use Prisma)

https://gist.github.com/saarimz/0a5b27023ffc300fbf2262438114108f

I've been trying to optimize the connection pool to get the errors down but to no avail

Gotchya. I've re-enabled the GCP region. Please let me know if that works for you!

Hey Saarim,

Relaying here as Rahul escalated your ask internally, you still have time with him on the board but wondering if you have gone ahead and migrated back to cloud?