GetTaskQueueUserData API Always Takes 4 Minutes on Temporal v2.28.0 (PostgreSQL)

Hi Team,

We’re observing that the internal API call:

/temporal.server.api.matchingservice.v1.MatchingService/GetTaskQueueUserData

consistently takes exactly 4 minutes (240 seconds) in our Temporal setup:

System Info:

  • Temporal version: v2.28.0
  • Database: PostgreSQL
  • Observability: Prometheus + Grafana
  • Issue: GetTaskQueueUserData latency is always exactly 240s, never lower, even under low load

Questions:

  1. Is it normal for GetTaskQueueUserData to hit a 240s response time under any condition?
  2. Could a DB lock or cold cache explain this behavior?
  3. Any known fixes or improvements in later versions beyond v2.28.0?

We would appreciate guidance on how to debug this and prevent the latency. Please let us know if logs, metrics, or DB schema details would help.

Thanks!