We are noticing that after upgrading Redash to v 10.1.0 the scheduler doesn’t seem to be running for all scheduled queries.
We changed schedules on existing scheduled queries but the scheduler is not picking up the change made via the UI.
Restarted the scheduler pod and things are working but the last updated at date and query results for some queries scheduled to run every 5 minutes does not reflect the run.
Please advise.
How did you deploy Redash? Kubernetes?
Yep, we have deployed it on Kubernetes (AWS).
How many Redis pods do you have? How many postgres pods?
[edit] This sounds like a state mismatch. As though the workers are updating the last_executed_at in one metadata database, and the webserver is reading from a different one.
We’re pointing to existing Redis and Postgres clusters which are not running in K8s pods.
So you are certain you have exactly one postgres and exactly one redis?