license install error

[tepid drum]

I am installing licenses with legacy 28-char license codes. With play like:

    - name: Add licenses
      na_ontap_license:
        state: present
        license_codes: "{{ licenses }}"
        <<: *login
      tags:
        - license

After run the play, got these errors:

TASK [Add licenses] *********************************************************************************************************************************************************************************************
fatal: [localhost]: FAILED! => {"changed": false, "msg": "Error: mismatch in license package names: 'fabricpool'.  Expected: dict_keys(['installed_licenses', 'base', 'nfs', 'cifs', 'iscsi', 'fcp', 'snaprestore', 'snapmirror', 'flexclone', 'snapvault', 'snaplock', 'snapmanagersuite', 'snapprotectapps', 'v_storageattach', 'insight_balance', 'ocshift', 'tpm', 've', 'dp_optimized', 'fabricpool', 'snapmirror_sync']), found: dict_keys(['installed_licenses', 'base', 'nfs', 'cifs', 'iscsi', 'fcp', 'snaprestore', 'snapmirror', 'flexclone', 'snapvault', 'snaplock', 'snapmanagersuite', 'snapprotectapps', 'v_storageattach', 'insight_balance', 'ocshift', 'tpm', 've', 'dp_optimized'])."}

Howevever, looking at filer side with system license show -owner {node}, it seems all licenses were installed successfully.

What's causing the complaint? My filer os is 9.10.1P12. netapp ansible collection 22.5.0.

Could it be a bug in this version of netapp.ontap collection? it's passing dictionary not expected by the filer.

[versed ledge]

I’ll need to check this when tomorrow when I’m in. The issue appear to be it expecting after the install for that list of license to be there. But after the e install it gets the second list which is missing fabricpool.
I’ll need to check where in the code we do this check

[tepid drum]

I think maybe it's a static list of keys. 'fabricpool' does not require license, and is not in the "{{ licenses }}" list.

[versed ledge]

There might be a bug, but may need some addtional logs here to see what going on.

We don't have a static list for this
At the start of the module we create a list of licenses on your machine
https://github.com/ansible-collections/netapp.ontap/blob/main/plugins/modules/na_ontap_license.py#L654

Later on after adds and removed are complete we do a comapre
https://github.com/ansible-collections/netapp.ontap/blob/main/plugins/modules/na_ontap_license.py#L426

Expecting the original list and the new list to be the same.

Would it be possible to rerun this with out api debuging enabled
https://github.com/ansible-collections/netapp.ontap/wiki/Debugging

It would produce a file called /tmp/ontap_apis.log that you can email to [e-mail] so we can see which REST api where called and what path it took through the code.

[tepid drum]

ok, it's not always producible. tried another time and not seeing same error.

PLAY [update license for filer] *****************************************************************

TASK [Add licenses] *****************************************************************************
changed: [localhost] => {"changed": true, "updated_licenses": ["nfs", "cifs", "iscsi", "fcp", "snaprestore", "snapmirror", "flexclone"]}

PLAY RECAP **************************************************************************************
localhost                  : ok=1    changed=1    unreachable=0    failed=0    skipped=0    rescued=0    ignored=0   

licenses seem to be installed fine.

in this case, it gives me updated_licenses with 7 features, but there is 13 features installed. Not sure that means anything to you?

[tepid drum]

@versed ledge How do I change the trace_api log filename and/or location? Right now it's saved to /tmp/ontap_api.log, when multiple users are running different playbooks with debug, it can have permissions issues & mixing up logs.

[versed ledge]

currently you can't, it a hardcode value (https://github.com/ansible-collections/netapp.ontap/blob/main/plugins/module_utils/netapp.py#L110). We expected user only to run it once for debuging and not have it on for more than that.

I can go back to the team and see make it a setable variable if it something you would expect to be on for more than debuging.

[tepid drum]

It'd be nice, since this issue my users encounter is only intermittent, had to turn on trace for a while. So different users can tramp on each other.

[tepid drum]

hi @versed ledge I've encountered this issue again, and have sent the trace log to ng-ansibleteam.

[versed ledge]

Thanks (Netapp has a global 4 day holiday this week). So we may not be able to get back to you until Tuesday. But we did get the trace are are looking in to it

[tepid drum]

@versed ledge Any findings? Thanks.