"Cannot use a pool after calling end on the pool" postgres error

Questions
1

I keep getting “Cannot use a pool after calling end on the pool” error using postgres as a chat memory

Every time my query to the llm takes a long time to process I get this error.

Information on your n8n setup

  • n8n version: 1.85.4
  • Database : postgres
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via: Dcoker
  • Operating system: Windows 11
3

Same happens to me all the time using supabase and postgres chat memory.
Does anyone have a workaround?

n8n version : 1.90.2
Database: postgres (supabase)
running n8n via: kubernetes

4

For llm agents that would take long multi-turn actions this will be an issue. I believe this is due to the constant ttl set for the postgres node: n8n/packages/nodes-base/utils/connection-pool-manager.ts at master · n8n-io/n8n · GitHub

Can anyone help make it configurable?

n8n version : 1.94.1
Database: postgres
running n8n via: n8n cloud