couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Noah Slater <nsla...@apache.org>
Subject Re: [PROPOSAL] New release procedure
Date Tue, 12 Mar 2013 14:50:34 GMT
Lazy consensus established. We are moving forward with our new release
procedure. Yay!


On 8 March 2013 01:42, Noah Slater <nslater@apache.org> wrote:

> Hello devs,
>
> I would like to propose a new release procedure:
>
> http://wiki.apache.org/couchdb/Release_Procedure
>
> This proposal will pass on lazy consensus, so speak up if you want to
> object.
>
> How this differs from the old release procedure:
>
> 1) Code level tests (CLI, etap, etc) have been removed. The test procedure
> has been updated to include an additional "make distcheck" step. This will
> allow release candidates to go up very quickly, and puts the onus on you
> guys to test things thoroughly.
>
> 2) The hashing and signing step has been split out from the build. This
> allows the build to take place on a remote host that is not trusted with
> your GPG key. I have also added an explicit step for checking
> the authenticity of the tarball that is produced.
>
> 3) As much as possible has been automated. The scripts are a bit rough and
> ready, but they will do, provided you follow the instructions precisely.
> Once again, the hope is that this will make releasing less burdensome for
> the release manager.
>
> Please take the time to review the document, and to review the scripts.
>
> The scripts are in the couchdb-admin.git repository:
>
> http://wiki.apache.org/couchdb/SourceCode
>
> Please take the time to review these also.
>
> The release procedure, and indeed the scripts, are not finished. They
> never will be. I will be updating them over the next day or two, and during
> the 1.3.0 release. But they are finished enough that I am comfortable
> making this proposal to the list.
>
> Also, sorry for the noise. Testing against Subversion is... Tricky. :)
>
> Thanks,
>
> --
> NS
>



-- 
NS

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message