Production Postgres stuck QUEUED in sfo — deployment slot wedge
wrumble
PROOP

a month ago

Project: insightful-magic

Project ID: 12191523-8f23-4fa0-aee4-8bf9390032f7

Environment: production

Region: sfo

Service: Postgres (service ID 1d370bec-2712-4116-9634-3c8eca06e525)

Production Postgres stuck QUEUED ("Waiting for deployment slot")

since 13:52 UTC — 15+ minutes. cms-beat also queued.

Sequence:

  • Postgres volume filled, was crash-looping with ENOSPC on pg_wal.

    Volume now reports 490MB / 1000MB.

  • Every new deployment since (manual redeploy + restart) goes

    straight to QUEUED and never starts.

Stuck zombie deployments (visible via GraphQL, hidden from UI):

  • staging/web from 08:49:57 UTC, status=DEPLOYING, stopped=true

  • staging/web from 08:50:11 UTC, status=DEPLOYING, stopped=true,

    queuedReason="Waiting for build slot"

  • staging/web from 13:31:16 UTC, status=DEPLOYING, stopped=true

These show as removed in the UI, railway down succeeds, but they

persist in activeDeployments and appear to hold the build slot.

Public site (cms.myrugby.app) is still serving traffic on the prior

deploy, but all CMS DB-backed writes and Celery beat are down.

Please force-clear the stuck deployments so production Postgres

can come back up. This started just before the platform incidents

were declared today — may be related fallout.

Solved

1 Replies

Status changed to Awaiting Railway Response Railway about 1 month ago


a month ago

Thanks for the precise diagnosis — you're right that the zombie deployments (showing as removed in the UI but still present in activeDeployments) are holding the slot and blocking new deploys from leaving QUEUED. That's consistent with the platform-side fallout from today's incident.

Your data volume is intact and your public site is still serving on the prior deploy. Force-clearing those wedged deployments so the slot frees up is a platform-side action — it can't be done reliably from your end (railway down reports success but they persist). We're escalating it now with your project and service IDs (insightful-magic, Postgres 1d370bec-2712-4116-9634-3c8eca06e525). I don't have a firm ETA yet but will follow up here once the slot is cleared.


Status changed to Awaiting User Response Railway about 1 month ago


Railway
BOT

25 days ago

This thread has been marked as solved automatically due to a lack of recent activity. Please re-open this thread or create a new one if you require further assistance. Thank you!

Status changed to Solved Railway 25 days ago


Welcome!

Sign in to your Railway account to join the conversation.

Loading...