Inngest says its database connection poolers collapsed twice on Oct 10, at 14:00 and 15:30 UTC, stalling run scheduling and queue processing for most accounts. Both poolers lost all of their connections to the database and could not rebuild them. The mechanism is the part worth studying: a timeout did not just abandon a query. It cost the pool a connection, every time, until there was no pool left.
The 38-second reel: Inngest, the database pool that destroyed its own connections.
What happened
Inngest’s services reach the primary database through a pair of connection poolers, PgBouncer, run in AWS. On Oct 10, both poolers lost all of their connections to the database at 14:00, and again at 15:30, and could not rebuild them. Run scheduling and queue processing stalled for most accounts in the 14:00 to 14:48 and 15:30 to 16:25 windows. Some runs were left Running and can be cancelled and replayed. A replay starts from the beginning, so completed steps run again.
The mechanism
When a lookup’s deadline expired while its query was still waiting in PgBouncer for a database connection, the driver cancelled the query and closed its client connection. PgBouncer had nothing running to cancel, so it later ran the abandoned query on a database connection and then closed that connection entirely, instead of returning it to the pool.
Inngest calls this a known interaction between the pgx driver and PgBouncer. We did not test either one, and this is Inngest’s own account, not ours.
The loop that ate the pool
Each pooler opens new database connections one at a time. Connections were being destroyed faster than they could be replaced, so the pools collapsed to nearly zero usable connections. Thousands of clients queued behind them, and their deadlines expired too, which kept the cycle going.
Two ingredients fed the loop. Both collapses started at the top and bottom of the hour, when cron-scheduled work roughly triples the query rate, and a newly rolled-out set of lookups was failing in large numbers at the same moments. Restarting the poolers did not reliably recover them. Raising the pool size did not help, because opening connections, not the pool limit, was the bottleneck. Inngest says it reproduced the behavior in a test environment.
The fix
A lookup’s deadline now ends the caller’s wait without cancelling the underlying query, which removes the trigger for the October 10 collapse on the lookups involved. The remaining database calls on the execution path and their timeouts are under review, and Inngest is consolidating to a single data center, in progress. The fix covers the lookups involved. It is not a claim that the fix is complete.
Our reading, not Inngest’s
A deadline that abandons queued work can still cost the database. That is the question to carry into your own stack: what does your pool do when every client times out at once?
What we did not verify
This is a single source, Inngest’s own incident report from Oct 15, 2026, and every claim above is attributed to it. We did not run PgBouncer or the driver, so we cannot confirm the mechanism independently. The report also covers a separate Oct 9 network outage, which is not covered here. The voiceover in the reel is AI-generated, and the script was written with an AI model made by Anthropic. Watch the reel on Threads.
This post started as an Instagram post →
- https://www.instagram.com/reel/De0BISzEzDH/
- https://www.threads.com/@dataeng.ai/post/De0BS0Yk_BD
- https://www.youtube.com/shorts/-hzimVYe4Fs
- https://www.inngest.com/blog/2026-10-15-incident-report-for-october-9-and-10-2026
Numbers above trace to these sources. If one moved, tell us and we fix it.


Talk it through
Argue with us on Instagram.