aws-samples / aws-samples/custom-scheduler-eks
Support custom plugin configuration
- Dominant language
- Go Template
- Stars
- 35
- Forks
- 13
- Avg merge
- 15h 7m
- Merged PRs (30d)
- 1
Description
We've been trying this helm chart with some CI workloads but see this warning, which causes our CI pods to fail.
> WARNING: Event retrieved from the cluster: Node didn't have enough resource: cpu, requested: 3000, used: 45030, capacity: 47710
We'd like to configure the NodeResourcesFit plugin to use the [RequestedToCapacityRatio strategy for bin packing](https://kubernetes.io/docs/concepts/scheduling-eviction/resource-bin-packing/#enabling-bin-packing-using-requestedtocapacityratio). However, since the plugin configuration is hard-coded, it's not possible to do this.
Contributor guide
Research direction
Locate the Helm chart's hard-coded NodeResourcesFit plugin configuration and inspect how scheduler settings are rendered. Check the chart's available validation or deployment checks; done means the RequestedToCapacityRatio strategy can be configured for CI workloads without the reported scheduling failure.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- helm, kubernetes
- Domain
- infrastructure
- Issue type
- Feature
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100