Update Azure NetApp management and add NFS protocol version

[wmudge]

Signed-off-by: Webster Mudge [email protected]

[jimright]

Did a simple test of this PR to confirm CDP Azure deployment works correctly.
Steps for test:

• CDP Azure definition included a CML workspace with default root volume size
• Confirmed that Azure NetApp Volume was created with NFSv3 protocol type. Expected from updates.
• Confirmed that {{run__azure_netapp_nfs_mount}} was set and used during the "Construct CDP ML Workspace configurations" task in runtime.
• ML workspace was successfully deployed and Filesystem ID listed in Workspace details matches the NetApp volume mount path.

[jimright]

@wmudge,

Hit an error during the environment teardown. Seems to be related to the changes made in roles/runtime/tasks/initialize_base.yml. See inline comment for details.

[jimright]

Thanks @wmudge - commit 86b519f resolved the teardown error in platform.

I hit another error during infrastructure teardown where the NetApp Capacity Pool could not be removed due to nested resources still being present. When I re-ran teardown this completed successfully so I suspect this is either an eventual consistency issue or a wait is needed on a previous task (e.g. Remove Azure NetApp Volume in cloudera.exe.infrastructure teardown_azure_storage.yml

Snippet of error message is below.

TASK [cloudera.exe.infrastructure : Remove NetApp Capacity Pool] ***************
task path: /runner/project/ansible_dev/collections/ansible_collections/cloudera/exe/roles/infrastructure/tasks/teardown_azure_storage.yml:41
Wednesday 30 March 2022  16:34:03 +0000 (0:02:08.831)       0:36:25.888 *******
<snipped>
The full traceback is:
Traceback (most recent call last):
  File "/tmp/ansible_netapp.azure.azure_rm_netapp_capacity_pool_payload_nfvqr2vg/ansible_netapp.azure.azure_rm_netapp_capacity_pool_payload.zip/ansible_collections/netapp/azure/plugins/modules/azure_rm_netapp_capacity_pool.py", line 204, in delete_azure_netapp_capacity_pool
  File "/usr/local/lib/python3.8/site-packages/azure/mgmt/netapp/operations/_pools_operations.py", line 549, in begin_delete
    raw_result = self._delete_initial(
  File "/usr/local/lib/python3.8/site-packages/azure/mgmt/netapp/operations/_pools_operations.py", line 505, in _delete_initial
    map_error(status_code=response.status_code, response=response, error_map=error_map)
  File "/usr/local/lib/python3.8/site-packages/azure/core/exceptions.py", line 105, in map_error
    raise error
azure.core.exceptions.ResourceExistsError: (CannotDeleteResource) Can not delete resource before nested resources are deleted.
Code: CannotDeleteResource
Message: Can not delete resource before nested resources are deleted.

[wmudge]

I suspect this is an eventual consistency issue and will need to add a guard condition. Still looking into it!

[jimright]

@wmudge - confirmed that both setup and teardown of Azure environment with CML workspace now works with your latest changes. Looks good.