hadoop-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chris Douglas <cdoug...@apache.org>
Subject Re: removing old releases
Date Wed, 27 Apr 2011 21:01:01 GMT
No need to wait a week, if you have time to clear it out. Thanks for
taking care of this, Ian. -C

On Wed, Apr 27, 2011 at 5:19 AM, Ian Holsman <ian@holsman.net> wrote:
> I've had a look at hadoop's dist area (http://www.apache.org/dist/hadoop/ )
> and was wondering which of these people think we should do here.
> If no one objects within a week, I will remove everything except the hadoop-0.20.2,
> hadoop-21.0, and stable directories there.
> Thanks
> Ian
> Begin forwarded message:
>> From: Mark Thomas <m>
>> Date: April 27, 2011 8:07:53 AM EDT
>> Subject: Re: [ACTION REQUIRED] www.apache.org/dist/<tlp> housekeeping
>> PMC members,
>> Six weeks ago, the infrastructure team sent you the e-mail below. To the
>> projects that were already following the release guidelines and to those
>> projects that have since cleaned up their /dist area - thank you. The
>> infrastructure team really does appreciate you doing this.
>> Regrettably, a large number of PMCs have chosen to ignore the message
>> below. To those PMCs, you have one month (until 31 May 2011) to clean up
>> your dist area or the infrastructure team will simply remove all files
>> that are more than twelve months old.
>> Mark
>> on behalf of the ASF Infrastructure Team
>> On 10/03/2011 08:16, Mark Thomas wrote:
>>> PMC members,
>>> As the ASF grows in size, so does the total size of the distribution
>>> artefacts we ask our mirror community to support for us. The larger this
>>> total size, the greater the strain on both ASF infrastructure and on the
>>> mirroring system.
>>> As per the release guidelines [1], only current releases should be
>>> available at http://www.apache.org/dist/. Monitoring of
>>> http://www.apache.org/dist/ [2] shows that some projects are not
>>> removing old releases. This is placing an unnecessary strain on both ASF
>>> infrastructure and on our mirror volunteers.
>>> Thanks to those PMCs that have been removing old releases from their
>>> distribution directory. The infrastructure appreciates you keeping on
>>> top of this.
>>> PMCs that have not been removing old releases are required to review
>>> their current distribution directory and remove any old releases.
>>> - PMCs using svnpubsub should remove old releases via svn.
>>> - PMCs not using svnpubsub should remove old releases directly from
>>> /www/www.apache.org/dist/<tlp> on people.apache.org at. Note that any
>>> deletions may take up to 24 hours to replicate to http://www.apache.org/dist
>>> In both cases it may take longer for changes to replicate to mirrors.
>>> Old releases removed from http://www.apache.org/dist/ are not lost.
>>> Release are automatically copied to http://archive.apache.org/dist/ and
>>> are never deleted.
>>> This inevitably raises the question what is a current release and what
>>> is an old release. To some extent, this varies from project to project
>>> but typically it amounts to the following:
>>> a) latest release of the current branch
>>> b) latest stable release of the current branch
>>> c) latest stable release of previous branches
>>> It is hard to give concrete examples that apply to all projects since
>>> each project is free to use its own release numbering scheme. However, a
>>> project that includes versions 2.1.0, 2.1.1 and 2.1.2 in its release
>>> directory almost certainly has some cleaning up to do. A project that
>>> includes 1.0.6, 1.1.5 and 2.0.7 probably doesn't.
>>> If you have any questions about how to manage your distribution
>>> directory please contact the infrastructure team.
>>> Thanks in advance,
>>> Mark
>>> on behalf of the ASF Infrastructure Team
>>> [1] http://www.apache.org/dev/release.html
>>> [2] http://people.apache.org/~henkp/tlps/
> --
> Ian Holsman
> Ian@Holsman.net
> PH: +1-703 879-3128 AOLIM: ianholsman Skype:iholsman
> Never explain - your friends do not need it and your enemies will not believe you anyway.
Elbert Hubbard

View raw message