Hi Ancor, On 18.5.2016 18:03, Ancor Gonzalez Sosa wrote:
I have been planning to send this mail for quite some time. Since we were lately focused on implementing features for SP2 I had put it on hold. In this sprint we have some storage-ng related PBIs again so...
Well, the focus is not exactly where I would like it to be (due to some left-overs from the last sprint), but at least it is getting on track.
This is kind of a suggestion for our product owner about sorting team-S' PBIs.
Following the agile approach of trying to have something shippable as soon as possible (and always shippable), I would like to see an installable Tumbleweed image using libstorage-ng the sooner the better. Something that can turn our own instance of openQA green for, at least, one scenario.
The three main "users" of libstorage (from the top of my head) are:
- (a) The partitioning proposal - (b) Yast-Bootloader - (c) The expert partitioner
For (a) we already have working code for very simple scenarios (ms-dos style partitions with contiguous space).
(c) is not absolutely needed to perform an installation[1].
Well, even (b) may not be considered necessary for the very first step. Would not really produce a system you would like to use, but I would consider partitioning the system according to the proposal and mounting it correctly as the first step.
Yast-Bootloader has been recently refactored to the ground, killing a lot of old code and replacing SCR with the more object-oriented CFA. It should be in a sane-enough state to be touched.
Agreed here.
So, as much as I would like to keep working in improving the proposal and supporting LVM (two things that are indeed quite interesting from a developer's point of view), I think we should postpone improving what we already have and focus on having a first version of the stuff that is missing (i.e. stuff that is preventing us from having that installable ISO). (b) comes to mind, but sure I'm forgetting something.
Looking above the waterline of the sprint backlog, I see the cards more as finishing what was started before than improving stuff (but that may be a different point of view). On the other hand, your suggestion is IMO a good main focus point for the next sprint (or the one following, given the length of the next sprint and the not-yet-obvious bug situation for SP2).
Cheers.
[1] Moreover, we should probably use the opportunity to have the UI of the expert partitioner redesigned (or at least reviewed), but that's another story.
I gave Ken a heads-up that there will be some work for him in this area some weeks ago :-) Jiri -- Regards, Jiri Srain Project Manager --------------------------------------------------------------------- SUSE LINUX, s.r.o. e-mail: jsrain@suse.com Lihovarska 1060/12 tel: +420 284 084 659 190 00 Praha 9 fax: +420 284 084 001 Czech Republic http://www.suse.com -- To unsubscribe, e-mail: opensuse-storage+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-storage+owner@opensuse.org