a month ago
Update: Solved.
Root cause was an old Redis deployment from before recent Railway platform/networking changes. Redis showed Online, but the private endpoint was stale/unreachable from app services.
Fix:
- Redeployed the Redis service, not just restarted it.
- Redeployed Celery Beat, Celery Worker, and backend afterward.
- Private Redis TCP now works from backend/Beat.
- Backend routes are fast again.
- Celery Beat broker/database health is OK.
Leaving this here in case it helps someone else: if a database service is very old and private networking starts timing out while DNS still resolves, try redeploying the database service from the Deployments page.
1 Replies
a month ago
Your Redis service's last deployment was on 2026-03-06 UTC, which is significantly older than your other services. Platform networking changes since then can cause a running database to become unreachable over the private network even though it appears online. The fix is to redeploy (not just restart) the Redis service - you can do this from the Redis service's deployments page by clicking the three-dot menu on the latest deployment and selecting "Redeploy". After the Redis redeploy completes, redeploy your Celery Beat and Celery Worker services as well so they establish fresh connections.
Status changed to Awaiting User Response Railway • about 1 month ago
Status changed to Solved grilm14 • about 1 month ago