Hi Railway team,

I'm having trouble connecting to my Postgres database service. The issue has been intermittent over the past several hours, with a few different symptoms:

Connection refused when connecting via the internal hostname (postgres.railway.internal, port 5432) — psycopg2/Django reports "Connection refused... Is the server running on that host and accepting TCP/IP connections?"

Server closed the connection unexpectedly when connecting via the public proxy (monorail.proxy.rlwy.net, port 50374).

In the Postgres deploy logs, I'm seeing a large volume of could not receive data from client: Connection reset by peer entries with rapidly incrementing client PIDs over a short window, while checkpoint logging otherwise looks normal.

Railway's own built-in DB connection tool (SSH-based) also failed with "Connection terminated unexpectedly."

I redeployed the Postgres service, which changed the internal IPv6 address, but the connection issue persisted afterward — my application still can't reach the database to run migrations.

Could you help me understand:

Whether the Postgres service crashed, was OOM-killed, or hit a resource limit (memory/disk/connection count) around this time

Whether there's an issue on Railway's infrastructure side affecting this database instance

Recommended next steps to restore a stable connection