Pain point · Self-hosting & upgrades · Blocks work
Version 0.263.6 crashes about every minute with Litestream and S3
2 source threads · first seen 2024-03 · last seen 2025-06
Summary
On a fresh instance using SQLite with Litestream replicating to S3, NocoDB sends repeated shutdown signals, restarting the container. This is a regression from v0.263.4.
- Affects
- self-hosters on Fly.io
Evidence
Excerpts are copied word for word from the source; follow the link to read it in full.
“NocoDB should not crash and Litestream should properly work with NocoDB as was the case before (v0.263.4 and below)”
“Knex: Timeout acquiring a connection. The pool is probably full”