intended internal disk update fails

[phillxnet]

The user consequence here is that a refresh of the disk page is required after some disk/pool operations to ensure the information displayed is correct:

Exception while updating disk state: 'PoolDetailView' object has no attribute '_update_disk_state'

the code in question stopped working as intended as from the following commit:

89d018b

which was part of the reviewed pr:

#1436

however that commit / pr does simplify our objects and addressed the stated issues and may help to divide concerns and removed a large inheritance for a single method call: so I suspect the fix is not to simply add back this inheritance.

But a viable mechanism to enact a disk status update is required to take the place of that call.

I have added a TODO: in the following commit to highlight this current bug:
b025e21

So the requirement is:

"How do we best call / trigger DiskMixin._update_disk_state() from within pool.py given the new lack of inheritance therein."

[phillxnet]

@MFlyer I take it the two commits associated with this issue are in error?
EDIT:
OK, just worked it out: you have "23" rather than the intended 32

[MFlyer]

@phillxnet yup!

[phillxnet]

@MFlyer No worries. I was just surprised to see them

[phillxnet]

Thanks to @FroggyFlox for re-highlighting this issue's potential relevance in our pool mount state update slackness, ie 2 Web-UI updates required to display Pool status changes. I.e. bottom up pool info informed by disk state (pool wise) that is not first updated. Need further research as we move towards simplifying our pool update strategy such as is approached in currently pending pr:

"pool resize disk removal unknown internal error and no UI counterpart. Fixes #1722" #2010
which contains a number of simplifications on disk scanning and consequent pool info propagation mechanisms.

[phillxnet]

Now we have the previously referenced pr merged:

"pool resize disk removal unknown internal error and no UI counterpart. Fixes #1722" #2010

We can again look to addressing this usability bug. It may be we can make progress via auspicious use of:

Model.refresh_from_db

(introduced in Django version 1.8). Taking care that we don't undo the performance enhancements that we have gained from #1722