a month ago
Production Postgres service Postgres in project era-workflow (service id e33be437-9219-4fd5-acf2-ea4d2a5c11ce, US West) became unreachable ~23:48 UTC while still showing Online — 0 vCPU / 0 B memory, console attach fails, clients time out connecting. Restart returned "Problem processing request" twice; a Redeploy started 00:03 UTC has been stuck ~30 minutes at volume migration with no container logs. Please force-kill the old container / release the volume so the redeploy can finish. "A fresh-image redeploy (Redeploy source image) was also tried at ~00:41 UTC.
1 Replies
Status changed to Awaiting Railway Response Railway • about 1 month ago
a month ago
Update: recovered. The command palette "Redeploy source image" succeeded (~9 min) after the menu Redeploy failed at volume migration (~35 min, no logs). The app service also needed a manual restart afterward to clear its dead connection pool. Leaving this for your visibility since multiple US West Postgres services hit the same state tonight; feel free to close.
Status changed to Solved wintershroom • about 1 month ago