Hey Rodrigo o/
Thanks for the adjusted debdiffs, I'll take a look. However, for now:
Comment #32:
> About the test steps, I mentioned in item (1) that numerous contributors
> and driver maintainers have validated the fix internally and confirmed
> that it worked without issues. Unfortunately that's the best we have.
Yes, I was aware about item 1 in [Test Plan] when I wrote comment #28,
but I didn't think that was the end of options thus suggested an idea:
Comment #28:
> For `Test Steps`, I can appreciate the limitations.
> However, we'd need at least a minimal way to verify the code change,
> even if that is synthetically or with a mocked up call
> (AFAICT/IIUIC the code change does look correct and safe).
"""
I also appreciate the existing validation by contributors/drive maintainers
is 'the best we have', but is that really 'the best we _could_ have'?
Apparently, we _could_ do something.
Is it _not possible_ to do at least _a mocked up test_ for this code path?
That seems to be standard practice in test suites of openstack components.
From what I can look at the patch, one could mock `send_request` to return
normally or trigger the expected exception (with/without the right message),
and check that the caller returns or reraises as expected in the 3 cases.
Right? If that is not possible, please clarify the reason, as I understand
this SRU is needed, so we'd need to figure tests out so it can be released.
It's difficult to SRU a change that can't be _minimally_ tested in _any way_,
or even regression tested in indirect ways (to at least check it does not
break other things -- that is a valid test too, when direct tests are hard).
I'll continue to review and hopefully upload the changes, and it might even
be accepted into -proposed depending on the reviewer/discussion, but without
some way to test this, I'd imagine it would be gated at release to -updates.
Thanks again for all your work on this.
** Changed in: manila (Ubuntu Noble)
Status: New => Incomplete
** Changed in: manila (Ubuntu Jammy)
Status: New => Incomplete
** Changed in: cloud-archive/bobcat
Status: New => Incomplete
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2069125
Title:
[SRU] Manila driver error with ONTAP SVM-scoped user
To manage notifications about this bug go to:
https://bugs.launchpad.net/cloud-archive/+bug/2069125/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
