Self-hosted Query in Queue for event

Issue Summary

Some of my queries are only saying “Query in queue”
any way to see or clear the queue

A summary of the issue and the browser/OS environment in which it occurs.

Technical details:

  • Redash Version: Version: 8.0.0+b32245 (a16f551e)
  • Browser/OS: client chrome macos / server ubuntu
  • How did you install Redash: docker
1 Like

Is it random? Do they have anything in common? If you wait do they eventually execute?

Query in queue means that an execution task has been placed in Redis but has not been picked up by a worker to be executed. This can happen if you have lots of queries running simultaneously without enough workers to handle the load.

You can see the status of your queues from with Redash by viewing the system status.

Also, I strongly encourage you to upgrade to V10.1 which we just release. The v8 codebase you run today is more than three years old and we don’t support it any longer.

I just upgraded to 10 Version: 10.1.0 (2589bef1). Still an issue.

you mentioned may not be enough workers. is there a query I can run to determine how many are in the queue?

1 Like

It seems to be every query I start

There’s not a query per se. You can check the system status page to show all the queues, how many jobs are waiting, how many workers there are etc. Just click your user avatar and then System Information.

It seems to have 88 queries queued up but no workers

How would I check if the worker is doing?

Did you start all your services in docker? Looks like you have no workers which means nothing will execute.

I rebooted and the workers are now showing but they are not doing anything.

What does your docker-compose file look like? Have you tried executing a query since you bounced everything?

it runs now why are there 403 queries on this page?

What does this mean? Does it mean that a query can execute successfully? If this is the case, do you see your workers behaving as expected in the System Information panel?

It will be easiest to help you if you can provide information about the nature of the errors. 403’s are authentication errors. Where do you see them? In the network inspector? Which URLs return 403’s?

count of queries was 403 now it is 44344 queued jobs

:joy: unlucky number since 403 is also an HTTP response code.

It looks like none of your workers are monitoring the periodic queue, which can fill up. Although I would expect the jobs to timeout eventually.

1 Like

how many queues are there? the setup instructions didn’t have that in it

There are six queues. If you follow the setup instructions they should all be covered.

"periodic", "emails", "default", "scheduled_queries", "queries", "schemas"

[edit] The docs around setup still need to be revised for V10, though, so your confusion is surely justified.

That did it the number of queued jobs is going down.
and the scheduled worker is getting to work

1 Like