Skip to content

Custom Node Scheduling

Brennen Murray requested to merge custom-scheduling into main

What does this MR do and why?

This MR will allow consumers of this Helm chart to specify various vars for customizing scheduling of the analytics stack workloads.

The following components can be set for each component of the deployment or global in .Values.global: tolerations, affinity, nodeSelector, topologySpreadConstraints.

The main reasoning behind these additions is to allow consumers of the helm chart to ensure that the analytics stack gets deployed to the correct nodes or instance types they desire. This can often alleviate noisy neighbor situations, autoscaling issues, and organizational components.

MR acceptance checklist

  • [ X ] The correct type labels have been applied to this MR.
  • [ X ] This MR has been made as small as possible, to improve review efficiency and code quality.
  • [ X ] This MR has been self-reviewed per the code review guidelines.
  • [ X ] The changes have undergone manual testing and are functioning as intended.
  • This MR has updated the Chart.yaml version number following SemVer versioning practices.
  • This MR documents any breaking changes in the MR description, and the upgrade path has been documented in the first commit as well as in MR description.

How to set up and validate

  1. Modify the global values or individual chart values for the previously mentioned YAML vars
  2. Run helm template . -f values.yaml in the root dir.

Numbered steps to set up and validate the change are strongly suggested.

How to deploy upon merging

Numbered steps to explain how this change needs to be deployed. For instance, if there are any changes that should be made outside of the code changes themselves.

Merge request reports