This mail is targeted to all the yast-storage-ng developers and POs.
I have been working in an approach for LVM-based proposal that offers
quite some cool features, like proposing a VG with several PVs to
optimize the space usage and similar stuff.
But I have a problem with that approach. I'm starting to feel the logic
is too hard to follow.
So I wanted to go back to square one and define a more dumb LVM
proposal. The resulting code will be easier to understand and will reuse
many components with almost no modifications. I just wanted to make sure
we all agree the following restrictions are acceptable.
1) A pre-existing VG will be reused for our purposes only if it's
already big enough, i.e. we never grow existing VGs.
2) If we need to create a new VG, it will contain just one big PV.
Potentially, that means we will need to destroy/resize more partitions,
compared to a more flexible approach with the ability to propose a VG
with several PVs.
2) Extra space freed while making space for the no-lvm partitions (EFI,
PReP, /boot, etc.) will not be turned into PVs to grow our VG (no matter
if it's created or reused).
After all, it's just a proposal.
Is everybody fine with the restrictions?
--
Ancor González Sosa
YaST Team at SUSE Linux GmbH
--
To unsubscribe, e-mail: opensuse-storage+unsubscribe(a)opensuse.org
To contact the owner, e-mail: opensuse-storage+owner(a)opensuse.org
