Closed tardfree closed 2 years ago
I'll need to talk with the rest of the team, but I'm not sure we'll have a fix for this. It look like key 'job-schedule-cluster' was added to the zapi job-schedule-cron-get-iter sometime after 9.1
But @tardfree if you really need this to work, what you can do in your local copy is in this file, remove 'job-schedule-cluster' from the 'key_fields'. https://github.com/ansible-collections/netapp.ontap/blob/main/plugins/modules/na_ontap_info.py#L1053 that should fix the error above for you 'job-schedule-cluster'.
So were going to fix this for the next version as it an easy fix. We have an internal story DEVOPS-4804 to track this
Thank you that's great. For my purposes I'm just excluding it with gather_subset "!job_schedule_cron_info" until I can upgrade all my remaining 9.1 systems (apart from any FAS2240 stuck on 9.1 of course).
@tardfree the fix for this has been merged in. It will be part of 21.18.0 release, which is expected on April 6th.
I'll leave this story open until then.
fixed in 21.18.0
Summary
Very basic playbook, calling na_ontap_info with no gather_subsets fails on job schedules with ontap 9.1. Documentation for this module says supported for ontap 9.1 onwards. Using gather_subset to exclude job_schedule_cron_info does work, so the issue is minor and isolated.
I totally understand that this is not likely worth the effort to fix due to 9.1 being way out of support - perhaps just document it.
This problem is likely caused by cron jobs moving from outside into a cluster vserver. Example cli outputs to highlight this: Ontap 9.1
Ontap 9.7 (has the cluster column in the table)
Component Name
na_ontap_info
Ansible Version
ONTAP Collection Version
ONTAP Version
Playbook
Steps to Reproduce
Playbook has hardcoded host/username/password as it's very basic.
Expected Results
The module should run without exception and register the results in the named variable. Excluding a subset shouldn't be required to work.
Running against a cluster running 9.5 or 9.7 works without issue. I don't have a 9.3 cluster to test with to see if that works.
Actual Results