OLM-2690: Subscription affinity

[perdasilva]

This PR brings down changes to the Subscription API that allows users to override the pod.spec.affinity of the operator deployment to enable deployment into multi-arch clusters

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: perdasilva

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [perdasilva]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[perdasilva]

/hold for qe approval

[oceanc80]

/lgtm

[perdasilva]

/retest

[michaelryanpeter]

/label docs-approved

[perdasilva]

/retest

[anik120]

/retest

[perdasilva]

/retest

[jianzhangbjz]

/assign @bandrade

[perdasilva]

@bandrade any ETA on this one?

[bandrade]

@perdasilva @jianzhangbjz I'm working on it. I should finish tomorrow. Sorry for the late response

[bandrade]

oc exec catalog-operator-5b668f4d9d-f2krm  -n openshift-operator-lifecycle-manager -- olm --version
OLM version: 0.19.0
git commit: c5a0ad3b230f04b5ec47525854fcf306f6250d5c

Testing scenario 1 - CR with NodeAffinity defined, POD should be scheduled in a specific node

apiVersion: monitoring.coreos.com/v1
kind: Prometheus
metadata:
  name: example
  namespace: test
spec:
  evaluationInterval: 30s
  serviceMonitorSelector: {}
  alerting:
    alertmanagers:
      - namespace: monitoring
        name: alertmanager-main
        port: web
  affinity:
    nodeAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
        nodeSelectorTerms:
          - matchExpressions:
              - values:
                  - ci-ln-9mfwdi2-72292-rw7jx-worker-c-cqwmw
                key: kubernetes.io/hostname
                operator: In
  probeSelector: {}
  podMonitorSelector: {}
  scrapeInterval: 30s
  ruleSelector: {}
  replicas: 2
  serviceAccountName: prometheus-k8s

  oc get pods -n test -o wide
  NAME                                   READY   STATUS    RESTARTS   AGE   IP            NODE                                       NOMINATED NODE   READINESS GATES
  prometheus-example-0                   2/2     Running   0          36s   10.131.0.21   ci-ln-9mfwdi2-72292-rw7jx-worker-c-cqwmw   <none>           <none>
  prometheus-example-1                   2/2     Running   0          36s   10.131.0.22   ci-ln-9mfwdi2-72292-rw7jx-worker-c-cqwmw   <none>           <none>
  prometheus-operator-787d8d5fb6-wf7zv   1/1     Running   0          13m   10.131.0.19   ci-ln-9mfwdi2-72292-rw7jx-worker-c-cqwmw   <none>           <none>

Testing scenario 2 - CR with anti-affinity defined, POD should not be scheduled in a node with a specific label defined.

  oc label node ci-ln-9mfwdi2-72292-rw7jx-worker-c-cqwmw app=dev                                                                                                                                                                        
  node/ci-ln-9mfwdi2-72292-rw7jx-worker-c-cqwmw labeled

  apiVersion: monitoring.coreos.com/v1
  kind: Prometheus
  metadata:
    name: example
    namespace: test
  spec:
    evaluationInterval: 30s
    serviceMonitorSelector: {}
    alerting:
      alertmanagers:
        - namespace: monitoring
          name: alertmanager-main
          port: web
    affinity:
      podAntiAffinity:
        requiredDuringSchedulingIgnoredDuringExecution:
          - labelSelector:
              matchExpressions:
                - values:
                    - dev
                  key: app
                  operator: In
            topologyKey: kubernetes.io/hostname
    probeSelector: {}
    podMonitorSelector: {}
    scrapeInterval: 30s
    ruleSelector: {}
    replicas: 2
    serviceAccountName: prometheus-k8s

  oc get pods -n test -o wide
   NAME                                   READY   STATUS    RESTARTS   AGE   IP            NODE                                       NOMINATED NODE   READINESS GATES
   prometheus-example-0                   2/2     Running   0          30s   10.131.0.29   ci-ln-9mfwdi2-72292-rw7jx-worker-c-cqwmw   <none>           <none>
   prometheus-example-1                   2/2     Running   0          30s   10.128.2.14   ci-ln-9mfwdi2-72292-rw7jx-worker-b-6zgrh   <none>           <none>
   prometheus-operator-787d8d5fb6-ft8z4   1/1     Running   0          78s   10.129.2.14   ci-ln-9mfwdi2-72292-rw7jx-worker-a-vfvnp   <none>           <none>

/lgtm
/unhold

[bandrade]

/unhold

[openshift-ci-robot]

/retest-required

Remaining retests: 0 against base HEAD fceaf8a and 2 for PR HEAD e40c77d in total

[openshift-ci]

@perdasilva: all tests passed!

Full PR test history. Your PR dashboard.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here.