Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.
And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.
Scheduler doesn't exit when the controlling client exits.
See original GitHub issueWhat happened:
The scheduler should exit after it is idle for a certain amount of time. But if I force kill the client (kill -KILL the process with dask_kubernetes), the containers will continue running. If I set distributed.scheduler.idle-timeout to something low, I can see the containers go to completion status, but then restart again.
What you expected to happen:
In the old default mode (now deploy_mode="local") everything was cleaned up, even if you force kill the process with dask_kubernetes running.
Issue Analytics
- State:
- Created 2 years ago
- Comments:11 (11 by maintainers)
Top Results From Across the Web
Top Related Medium Post
Top Related StackOverflow Question
Troubleshoot Live Code
Top Related Reddit Thread
Top Related Hackernoon Post
Top Related Tweet
Top Related Dev.to Post
Top Related Hashnode Post
Sure, I will find some time for it this week.
Given the lack of activity here and we’ve had no other users reporting these issues I’m going to close this out. Feel free to comment back here if this is still desirable though.