[Pkg-sympa-devel] Let's stay with 6.0 for Squeesh (6.1 is unstable)

[Jonas Smedegaard]

On Mon, Oct 11, 2010 at 11:03:44AM +0200, Stefan Hornburg (Racke) wrote:
>On 10/11/2010 10:56 AM, Jonas Smedegaard wrote:
>>On Mon, Oct 11, 2010 at 10:02:21AM +0200, Stefan Hornburg (Racke) 
>>wrote:
>>>On 10/10/2010 04:06 PM, Jonas Smedegaard wrote:
>>>>Hi,
>>>>
>>>>I just realized that the 6.1 branch is unstable upstream, so 
>>>>probably not something we want to maintain a beta release of for 
>>>>several years in Debian stable.
>>>>
>>>>Luckily the packaging currently in testing is 6.0 - I suggest we 
>>>>keep it at that.
>>>>
>>>>More detailed, I suggest we...
>>>>
>>>>a) Release my almost ready updated 6.1b.7 packaging for experimental
>>>>b) Request ftpmasters to drop current 6.1b.4 packaging from unstable
>>>>c) Release newest stable 6.0 release for unstable
>>>>d) Request unblocking of that 6.0 packaging
>>>>
>>>>
>>>>Any objections to this?
>>>>
>>>>Alternatively we can continue 6.1 in unstable, but then we lack a
>>>>staging area for newer 6.0 packaging, which I am sure the release
>>>>team won't like (and I don't like it myself!).
>>>>
>>>
>>>We don't need to drop 6.1 from unstable. The RC bug in Squeeze Sympa 
>>>can be fixed via an upload to proposed-updates.
>>
>>True, but then we lack a staging area for newer 6.0 packaging, which I 
>>am sure the release team won't like (and I don't like it myself!).
>>
>>...as I wrote right above :-P
>>
>>What I mean by "staging area" is a repository where our packaging is 
>>exposed to others that might spot flaws in our packaging. Unstable 
>>acts as a 10-day staging area for testing, with a lot of Debian 
>>developers as users. proposed-updates is less of a staging area in the 
>>regards of spotting errors, as it has much fewer users - relying 
>>mostly on the release team to (double-)check for flaws.
>>
>
>proposed-updates is just for package updates which can't go through 
>testing.

True.  That does not contradict any of what I wrote.

Do you have an opinion on my concerns?


>In our case it was bad luck that the freeze and the 6.1 upload happened 
>at almost the same time.

The freeze had been in the make since spring or so, so should not have 
come as a surprise to us.  Also, we should never release unstable code 
to "unstable": The term refers to stability not of code itself but its 
packaging - proper place to release unstable code is "experimental"!

I am not complaining here - I even suspect that I did that switch to 6.1 
myself :-(


>I'm really sure that 6.1 will be rock solid before the Wheezy freeze 
>:-).

:-)


>>But sure - we can try leave it as-is first. Will you get in touch with 
>>the release team about using proposed-updates?
>>
>
>Yes.

Great!

As you might have noticed, I already branched off the 6.1 packaging. I 
will now similarly setup a testing branch, unless anyone objects (fast! 
I might do it within an hour from now...!).

Any opinions on whether we should then package most recent upstream 6.0 
release which contains some bugfixes, or stay with code currently in 
testing and only fix RC bugs?  Perhaps you will ask the release team 
thir opinion on this, Stefan?


  - Jonas

-- 
  * Jonas Smedegaard - idealist & Internet-arkitekt
  * Tlf.: +45 40843136  Website: http://dr.jones.dk/

  [x] quote me freely  [ ] ask before reusing  [ ] keep private
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.alioth.debian.org/pipermail/pkg-sympa-devel/attachments/20101011/2bc108ca/attachment-0001.pgp>