Comment # 14 on bug 1078495 from Ancor Gonzalez Sosa

(In reply to Ancor Gonzalez Sosa from comment #13)
> (In reply to Ancor Gonzalez Sosa from comment #12)
> >
> > So the bug, if any, is not in the form. It was in control.xml or, at most,
> > in the proposal code (if it's supposed to work in a different way, which I
> > don't see).

So the options to fix skelcd are:

 - Set disable_order to nil for swap. That way, the initial proposal will never
try to disable swap, although the user can still change it.
 - Set disable_order in root to some value lower than swap. That way, snapshots
will be deactivated before killing swap.
 - Both.
 - None. Of course. :-)

I will leave that decision to Ludwig.

> According to
> https://github.com/yast/yast-storage-ng/blob/master/doc/old_and_new_proposal.
> md#the-volumes-subsection there is indeed a bug in the proposal. Since
> disable_order is not specified for root, it should have never disabled the
> snapshots during the initial attempt. So the sequence should have been:
> 
> 
>   Trying proposal with initial settings
>   Trying proposal after disabling '/home'
>   Trying proposal after disabling 'adjust_by_ram' for 'swap'
>   Trying proposal after disabling 'swap'
>   NO PROPOSAL POSSIBLE

This PR fixes the proposal so the behavior matches the documentation. That's in
fact even worse regarding this bug report, but now is more correct.

https://github.com/yast/yast-storage-ng/pull/597