Darren J Moffat wrote:
> Garrett D'Amore wrote:
>> Darren J Moffat wrote:
>>> Garrett D'Amore wrote:
>>>> Frank Che wrote:
>>>>> The project team would like to change the release binding of this
>>>>> project from micro to patch, for back-porting purpose. If there is
>>>>> any concern on this, please let the project team know.
>>>>>
>>>>> I think this change is qualified for self review. If there is
>>>>> concern, I can upgrade it to a fast track.
>>>>
>>>> I think since you requested micro binding originally, this is
>>>> trivially obvious. Indeed, I think many people these days treat
>>>> micro binding as synonymous with patch binding. (We haven't had a
>>>> "micro" release since Solaris 2.5.1.)
>>>>
>>>> For C-Team's point of view, I think you can just go ahead and
>>>> proceed as if you originally requested patch binding.
>>>
>>> There is a very important difference between micro and patch though
>>> and that is the delivery mechanism. A micro release can be assumed
>>> to be fresh installer or upgraded to but a patch release could need
>>> the feature applied by 'patchadd' as well as being upgraded to in a
>>> fresh bitted release. This case introduces a new package
>>> (SUNWmptsas) so as long as that can be properly dealt with in the
>>> patch release train I have no problem with the change of binding.
>>>
>>
>> C-Team is apparently confused... while the driver is intended for the
>> feature-patch gate (i.e. for delivery as part of a Solaris 10
>> update), they have made a point of requesting the project go back and
>> ask for patch binding.
>
> I don't think the C-Team is confused. Solaris 10 updates are not
> micro releases they are patch binding releases.
Solaris 10 updates have been handled the same way all other updates
since at least Solaris 8 have been handled, as a collection of
patches. So I'm not sure I understand how a micro release would
happen.... unless someone did a Solaris 10.1 or somesuch. The last such
release I recall was Solaris 2..5.1.
Okay, well in this case its obvious enough. In the meantime, I think
case submitters (and their sponsors) should refrain from using micro
release binding -- its not really what they want. They want either
minor or patch binding.
-- Garrett
