Description of problem:


ceph-ansible-4.0.11-1.el7cp
We have a RGW multisite setup with Dashboard enabled on both clusters.
When trying to upgrade the multisite secondary rolling_update.yaml fails trying to create RGW system user.
This is because the expected behaviour has changed from RGW stand point, and we need to add "--yes-i-really-mean-it" to any user creation on multiste secondary.

This creates a problem since this is not required on primary cluster, hence creating an RFE as we need to figure this out.

Ansible code :
radosgw-admin user create --uid={{ dashboard_rgw_api_user_id }} --system


Failure step:

TASK [ceph-dashboard : create radosgw system user] *******************************************************************************************************************************************************************************************
Thursday 23 January 2020  08:09:41 +0000 (0:00:00.905)       1:27:43.977 ******
FAILED - RETRYING: create radosgw system user (3 retries left).
FAILED - RETRYING: create radosgw system user (2 retries left).
FAILED - RETRYING: create radosgw system user (1 retries left).
fatal: [extensa010.ceph.redhat.com -> extensa010.ceph.redhat.com]: FAILED! => changed=true 
  attempts: 3
  cmd:
  - timeout
  - --foreground
  - -s
  - KILL
  - '20'
  - radosgw-admin
  - --cluster
  - ceph
  - user
  - create
  - --uid=ceph-dashboard
  - --display-name=Ceph dashboard
  - --system
  delta: '0:00:00.156309'
  end: '2020-01-23 08:09:58.168867'
  msg: non-zero return code
  rc: 22
  start: '2020-01-23 08:09:58.012558'
  stderr: |-
    Please run the command on master zone. Performing this operation on non-master zone leads to inconsistent metadata between zones
    Are you sure you want to go ahead? (requires --yes-i-really-mean-it)
  stderr_lines: <omitted>
  stdout: ''
  stdout_lines: <omitted>