11:51 a.m.
On Wed, Jul 15, 2020 at 10:55 AM Martin Perina <mperina(a)redhat.com> wrote:
On Tue, Jul 14, 2020 at 4:05 PM Michal Skrivanek <michal.skrivanek(a)redhat.com>
wrote:
>
>
>
> On 14 Jul 2020, at 15:33, Yedidyah Bar David <didi(a)redhat.com> wrote:
>
> On Tue, Jul 14, 2020 at 4:21 PM Michal Skrivanek <michal.skrivanek(a)redhat.com>
wrote:
>>
>>
>>
>> > On 14 Jul 2020, at 12:11, Yedidyah Bar David <didi(a)redhat.com> wrote:
>> >
>> > On Tue, Jul 14, 2020 at 11:43 AM Michal Skrivanek
>> > <michal.skrivanek(a)redhat.com> wrote:
>> >>
>> >> Hi all,
>> >> we’re moving to 4.4.z development now and we need to keep a closer eye
on automation results and making sure the build is not broken. For these reasons we’re
considering moving to a similar model as vdsm, having a smaller set of people with merge
rights to make sure the patches get in in the right order and they meet our sanity
standards (OST, bug’s TM)
>> >> Any objections/comments?
>> >
>> > Any reason to not simply branch 4.4? And have the branch maintained by
>> > the stable branches maintainers?
>>
>> just the sheer amount of backports needed (every patch). Doesn’t sound worth the
effort of posting and reviewing(even if just formally) everything twice.
>
>
> If you expect _every_ patch to be backported
We are trying to minimize backports between master and stabilization branches by creating
the stabilization branch as late as possible. This means that both branches use 4.4
database script numbering, which should decrease the need for backports.
It's not clear if you describe what we "want", or what we "do"
(now).
I agree, anyway.
If we would create standard ovirt-engine-4.4 branch, then we would need to introduce 4.5
db scripts numbering on master
Why?
I think you only need to do that once you want to - once you intend to
merge to master functionality that needs a dbscript change, that you
only want for 4.5.
and this would create huge overload when backporting all patches from
master to ovirt-engine-4.4
So current master is meant to be our 4.4.2 release and we have ovirt-engine-4.4.1.z
stabilization branch. Later we will create ovirt-engine-4.4.2.z stabilization branch and
master will become 4.4.3 release and so on.
Ok, but what happens when you actually want to merge 4.5 stuff? Then
you'll be at the same point you are today.
If you do not intend to do this soon, I see no problem in branching
4.4 already. If you are just afraid that someone will accidentally try
to merge some 4.5 dbscript to master, and want to prevent that for
some time, just add some code to check-patch.sh to prevent that.
>
> yes
>
> , just do nothing - let current maintainers do their job, and revert the occasional
bad ones when needed.
>
>
> And how would you prevent breaking the relatively frequent updates? The CI OST is not
working very well (for a long time now) and while we are improving and stabilizing the
infrastrucure it’s not really there yet to consider automated gating.
> Engine is unique in oVirt set of projects, it’s the largest one by far and use
maintainership per team or area (FE, BE, database, API…) and so we have a pretty high
number of people merging patches but far less people keeping up to date with the project’s
planning.
>
>
>
> Otherwise, I think branching is a good approach.
>
>>
>>
>> > --
>> > Didi
>> >
>>
>
>
> --
> Didi
>
>
> _______________________________________________
> Devel mailing list -- devel(a)ovirt.org
> To unsubscribe send an email to devel-leave(a)ovirt.org
> Privacy Statement: https://www.ovirt.org/privacy-policy.html
> oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/
> List Archives:
https://lists.ovirt.org/archives/list/devel@ovirt.org/message/EDI64EBEIJC...
--
Martin Perina
Manager, Software Engineering
Red Hat Czech s.r.o.
--
Didi