rgw flags set to storeName even for multisite deployment

[olivierbouffet]

Is this a bug report or feature request?

• Bug Report

Deviation from expected behavior:
Hello,

When creating a multisite objectstore the rgw_zone and rgw_zonegroup flags are set to objectstore name instead of zone and zonegroup.

How to reproduce it (minimal and precise):
create realm :

apiVersion: ceph.rook.io/v1
kind: CephObjectRealm
metadata:
  name: realm-1
  namespace: rook-ceph

create a zonegroup:

apiVersion: ceph.rook.io/v1
kind: CephObjectRealm
metadata:
  name: realm-1
  namespace: rook-ceph
root@dockerdev-263:/home/code# cat 05_zg-1.yaml
apiVersion: ceph.rook.io/v1
kind: CephObjectZoneGroup
metadata:
  name: zonegroup-1
  namespace: rook-ceph
spec:
  realm: realm-1

create a zone :

apiVersion: ceph.rook.io/v1
kind: CephObjectRealm
metadata:
  name: realm-1
  namespace: rook-ceph
root@dockerdev-263:/home/code# cat 05_zg-1.yaml
apiVersion: ceph.rook.io/v1
kind: CephObjectZoneGroup
metadata:
  name: zonegroup-1
  namespace: rook-ceph
spec:
  realm: realm-1
root@dockerdev-263:/home/code# cat 06_oz-1.yaml
apiVersion: ceph.rook.io/v1
kind: CephObjectZone
metadata:
  name: zone-1
  namespace: rook-ceph
spec:
  zoneGroup: zonegroup-1
  metadataPool:
    replicated:
      size: 1
      requireSafeReplicaSize: false
    preservePoolsOnDelete: false
  dataPool:
    replicated:
      size: 1
      requireSafeReplicaSize: false
    preservePoolsOnDelete: false

create a multisite object store :

apiVersion: ceph.rook.io/v1
kind: CephObjectStore
metadata:
  name: object-multi-1
  namespace: rook-ceph # namespace:cluster
spec:
  gateway:
    port: 80
    # securePort: 443
    instances: 1
  zone:
    name: zone-1
  # service endpoint healthcheck
  healthCheck:
    bucket:
      disabled: false
      interval: 60s
    # Configure the pod liveness probe for the rgw daemon
    livenessProbe:
      disabled: false

and then get the ceph config dump :

[root@rook-ceph-tools-5b54fb98c-k5qk9 /]# ceph config dump | grep rgw
    client.rgw.object.multi.1.a        advanced  rgw_enable_usage_log                   true
    client.rgw.object.multi.1.a        advanced  rgw_log_nonexistent_bucket             true
    client.rgw.object.multi.1.a        advanced  rgw_log_object_name_utc                true
    client.rgw.object.multi.1.a        advanced  rgw_zone                               object-multi-1                            *
    client.rgw.object.multi.1.a        advanced  rgw_zonegroup                          object-multi-1                            *

Environment:

• OS (e.g. from /etc/os-release): debian 10 buster
• Kernel (e.g. uname -a):
• Cloud provider or hardware configuration:
• Rook version (use rook version inside of a Rook Pod): v1.7.8
• Storage backend version (e.g. for ceph do ceph -v): ceph version 16.2.6
• Kubernetes version (use kubectl version):
• Kubernetes cluster type (e.g. Tectonic, GKE, OpenShift):
• Storage backend status (e.g. for Ceph use ceph health in the Rook Ceph toolbox):

[travisn]

@olivierbouffet Before this issue is closed, please open a PR against master. As commented here, looks like this was only fixed in the 1.7 branch.

[olivierbouffet]

hi @travisn , apparently my PR has been also merge to master.
Sorry for all the confusion , about my PR.

[travisn]

hi @travisn , apparently my PR has been also merge to master. Sorry for all the confusion , about my PR.

@olivierbouffet In which PR was it merged to master? I'm not seeing it, thanks.

[olivierbouffet]

Hi @travisn , I think in this one : #9261