Bug 2060697

More to do here before this is able to go through QE, need to update vendor directories in MAPA and MAO

Waiting for available nightly build to verify bug.

Hi Joel, 
I tried to verify this bug on 4.11.0-0.nightly-2022-03-08-191358, but there are two problems about it. Please help to take a look. Thanks!

1. "partitionNumber" will be removed, I need to:

Disable CVO: oc scale deployment -n openshift-cluster-version cluster-version-operator --replicas 0
Disable MAO: oc scale deployment -n openshift-machine-api machine-api-operator --replicas 0
Delete the mutatingwebhookconfiguration: oc delete mutatingwebhookconfiguration machine-api

then "partitionNumber" will not be removed, but other fields, for example 
            group:
              name: partitionpg
don't need to do this.

2. "partitionNumber" still cannot work, I create 3 machineset:
first time

          placement:
            availabilityZone: us-east-2b
            region: us-east-2
            group:
              name: partitionpg
              partitionNumber: 3
machine created successfully, check on AWS console "Partition number" shows 1. But it should be 3.

second time

          placement:
            availabilityZone: us-east-2b
            region: us-east-2
            group:
              name: partitionpg
              partitionNumber: 3
machine created successfully, check on AWS console "Partition number" shows 2. But it should be 3.

third time 
          placement:
            availabilityZone: us-east-2b
            region: us-east-2
            group:
              name: partitionpg
              partitionNumber: 8
machine created successfully, check on AWS console "Partition number" shows 3. But the machine should create failed, as the Partition of partitionpg is 7.

I just tested this with cluster bot and the partitionNumber field is working for me with the latest nightly.

The partition number in your example is not on the right level, it needs to be on the same level as group, not within the group :) 

The other validations you've mentioned haven't been implemented yet.

Hi Joel,
I'm very sorry for my mistake. Yes, I just changed the partitionNumber to be on the same level as group, and it worked. The other validations are not within the scope of this bug, so move this to Verified.

liuhuali@Lius-MacBook-Pro huali-test % oc get clusterversion
NAME      VERSION                              AVAILABLE   PROGRESSING   SINCE   STATUS
version   4.11.0-0.nightly-2022-03-08-191358   True        False         9h      Cluster version is 4.11.0-0.nightly-2022-03-08-191358

Steps:
1.access to the AWS console to manually create a partition Placement Group with 5 Number of partitions

2. create awsplacementgroup
liuhuali@Lius-MacBook-Pro huali-test % oc get awsplacementgroup partitionpg2 -o yaml
apiVersion: machine.openshift.io/v1
kind: AWSPlacementGroup
metadata:
  creationTimestamp: "2022-03-10T05:30:46Z"
  generation: 1
  name: partitionpg2
  namespace: openshift-machine-api
  resourceVersion: "113597"
  uid: d98ea863-7493-4750-bea2-d56a4ef11492
spec:
  credentialsSecret:
    name: aws-cloud-credentials
  managementSpec:
    managed:
      groupType: Partition
      partition:
        count: 5
    managementState: Unmanaged

3.reference the awsplacementgroup in the MachineSet.

          placement:
            availabilityZone: us-east-2b
            region: us-east-2
            group:
              name: partitionpg2
            partitionNumber: 3

liuhuali@Lius-MacBook-Pro huali-test % oc get machine
NAME                                                 PHASE     TYPE         REGION      ZONE         AGE
huliu-aws411-k2t9g-a2-9ngvp                          Running   m6i.large    us-east-2   us-east-2b   11m

4.check on AWS console "Partition number" shows 3.

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (Important: OpenShift Container Platform 4.11.0 bug fix and security update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHSA-2022:5069