Just as a note the project releases page for common/hdfs/mapreduce (e.g. http://hadoop.apache.org/mapreduce/releases.html ) do not have an entry for the 0.20.1 release, but going to the download link shows 0.20.1 as the stable release for the .20 branch.
I suspect that is a cause of the confusion.
From: Edward Capriolo [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, September 15, 2009 9:44 AM
To: [EMAIL PROTECTED]
Subject: Re: Hadoop 0.20.1 (a big secret?)
On Tue, Sep 15, 2009 at 11:35 AM, Todd Lipcon <[EMAIL PROTECTED]> wrote:
> Hi Stephen,
> There was an email thread about two weeks ago announcing the release
> candidate and subsequent release.
> Regarding Cloudera's version, the most recent one available for download is
> based on 0.20.0, but we've included many of the bug fixes that are in
> 0.20.1. We're currently working on updating our patch series to include the
> entirety of 0.20.1 and should have that out next week.
> On Tue, Sep 15, 2009 at 7:41 AM, stephen mulcahy
> <[EMAIL PROTECTED]>wrote:
>> I'm about to install a new hadoop cluster here and I'm wondering which
>> Hadoop to install.
>> Looking at http://hadoop.apache.org/common/releases.html suggests its
>> either 0.19.2 or 0.20.0 but looking in
>> http://ftp.heanet.ie/mirrors/www.apache.org/dist/hadoop/core/ I see 0.20.1
>> has been sitting there since 01-Sep-2009.
>> A quick google suggests Cloudera are going to base their next release on
>> 0.20.1 so I'm taking this as an indication that 0.20.1 is ready for
>> Did I miss the announcement on 0.20.1 or is there some reason why it's
>> being kept quiet so far (if so, apologies for publicising it ;)?
>> Stephen Mulcahy, DI2, Digital Enterprise Research Institute,
>> NUI Galway, IDA Business Park, Lower Dangan, Galway, Ireland
>> http://di2.deri.ie http://webstar.deri.ie http://sindice.com
Hadoop has branches and for the most part, major.minor.release
versioning. If you look at the svn-trunk you are always going to be
latest branch. At the current moment stable points to 18.X.
It seems the trend community is really trying to come together around 20.
Here is some quick notes:
Hadoop updgrade is easy.
Downgrade between majors can be a major pita.
I would suggest playing with the technology for a while and get a feel
for the version that is best for you.