We're experiencing recurring unexpected PostgreSQL connection terminations on our service, causing cascading failures across all queue workers.

Environment:

- Database: Railway-managed PostgreSQL

- Connection library: pg-pool (node-postgres)

What's happening:

At 2026-02-19 17:14:11 UTC, all active database connections were terminated simultaneously, resulting in ECONNRESET errors across every queue worker:

- billing, connection-messages, webhooks, webhookWorker, secret-agent, linkedin-invites, emails, follow-up-actions

Error stack trace:

```

Database query error: Connection terminated unexpectedly

at /app/node_modules/pg-pool/index.js:45:11

at process.processTicksAndRejections (node:internal/process/task_queues:105:5)

at async Object.query (/app/src/config/database.js:65:17)

at async processJobs (/app/src/services/bulkCollectionProcessor.js:78:20)

Key observations:

- All connections dropped at the exact same second, suggesting a server-side disconnection (not a client timeout)

- This is not an isolated incident — it has been recurring

- Our application uses connection pooling via pg-pool

Questions:

1. Was there any PostgreSQL maintenance, failover, or resource limit enforcement at that time?

2. Are there any known issues with the PostgreSQL instance or underlying infrastructure?

3. Is there a connection limit or idle timeout configured on the Railway side that could be causing this?

We'd appreciate any insight into what's triggering these disconnections. Happy to provide additional logs or details.

Thanks!