Uploaded image for project: 'Network Observability'
  1. Network Observability
  2. NETOBSERV-1116

Console plugin / FLP (kafka) podAffinity

XMLWordPrintable

    • Icon: Story Story
    • Resolution: Duplicate
    • Icon: Undefined Undefined
    • None
    • None
    • Console Plugin
    • 3
    • False
    • None
    • False
    • OCPSTRAT-1207 - Improve Network Observability Operator performance with latest eBPF enhancements (bpfman, Tcx hook latest kernel & RHEL9.4)
    • TODO (release note required)

      Following SRE work, some alerts has been raised on console plugin side:

      deployment_validation_operator_minimum_three_replicas Validation error for app-sre-observability: Indicates when a deployment uses less than three replicas 
      deployment_validation_operator_no_anti_affinity Validation error for app-sre-observability: Indicates when deployments with multiple replicas fail to specify inter-pod anti-affinity, to ensure that the orchestrator attempts to schedule replicas on different nodes.

      it would be usefull to be able to specify pod affinity / anti affinity especially using more than one replica

      https://redhat-internal.slack.com/archives/C03PUMXPZGQ/p1687406803548959 

      The default replica value can stay 1 for now as we don't expose in developer view but will need to be updated in the future.

       

      This also apply to FLPs when using KAFKA mode

              Unassigned Unassigned
              jpinsonn@redhat.com Julien Pinsonneau
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

                Created:
                Updated:
                Resolved: