Uploaded image for project: 'OpenShift Pipelines'
  1. OpenShift Pipelines
  2. SRVKP-3339

Add anti-affinity rules to controller's replicas

XMLWordPrintable

    • Icon: Story Story
    • Resolution: Done
    • Icon: Major Major
    • Pipelines 1.20.0
    • None
    • Operator
    • None
    • False
    • Hide

      None

      Show
      None
    • False
    • Hide
      As part of improving high availability (HA) for the Tekton controllers, this update adds a preferredDuringSchedulingIgnoredDuringExecution pod anti-affinity rule. This ensures that multiple replicas of the controller are preferably scheduled on different nodes, reducing the risk of service disruption in case of a node failure.
      Show
      As part of improving high availability (HA) for the Tekton controllers, this update adds a preferredDuringSchedulingIgnoredDuringExecution pod anti-affinity rule. This ensures that multiple replicas of the controller are preferably scheduled on different nodes, reducing the risk of service disruption in case of a node failure.
    • Enhancement
    • Done
    • 5
    • Pipelines Sprint pioneers 32

      As part of the HA setup, we probably want to add anti-affinity rules to our deployment to make sure replicas do not end up in the same node.

      See https://github.com/knative/serving/pull/8944 as an example

              jkhelil abdeljawed khelil
              vdemeest Vincent Demeester
              Sai Raju Manthina Sai Raju Manthina
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

                Created:
                Updated:
                Resolved: