[ECS] [request]: Discard Daemons when autoscaling through AWS Batch
- Dominant language
- Shell
- Stars
- 5.4k
- Forks
- 334
- PR merge metrics
- No merged PRs in 30d
Description
### Community Note
* Please vote on this issue by adding a 👍 [reaction](https://blog.github.com/2016-03-10-add-reactions-to-pull-requests-issues-and-comments/) to the original issue to help the community and maintainers prioritize this request
* Please do not leave "+1" or "me too" comments, they generate extra noise for issue followers and do not help prioritize the request
* If you are interested in working on this issue or have submitted a pull request, please leave a comment
**Tell us about your request**
When scheduling Jobs through AWS Batch on ECS EC2 (cluster created by AWS Batch), AWS Batch autoscales nodes based on Job queue, etc.
However when a custom Daemon is deployed on the side (e.g. any Agent - not deployed through AWS Batch), it prevents the autoscaler from removing nodes, which means that the cluster continuously runs at the maximum capacity that was used by AWS Batch.
The request is to discard Daemon in the computation used by the autoscaler, as it does not make sense to keeps only for Daemon.
**Which service(s) is this request for?**
ECS EC2 + AWS Batch
**Tell us about the problem you're trying to solve. What are you trying to do, and why is it hard?**
There's no direct workaround. It could be possible to include Agents through AWS Batch with #1522 but it's far less convenient that having a Daemon.
**Are you currently working around this issue?**
No workarounds at the moment.
**Additional context**
Anything else we should know?
**Attachments**
If you think you might have additional information that you'd like to include via an attachment, please do - we'll take a look. (Remember to remove any personally-identifiable information.)
Contributor guide
Research direction
The issue names no repository files, tests, or entry points. Start by reviewing the AWS Batch on ECS EC2 autoscaling behavior described here and clarify how daemon tasks should be treated; done would require an agreed implementation scope and validation that nodes can scale in when only daemons remain.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- aws
- Domain
- cloud, infrastructure
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100