Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Hadoop >> mail # dev >> Hadoop in Fedora updated to 2.2.0

Copy link to this message
Re: Hadoop in Fedora updated to 2.2.0
>> https://issues.apache.org/**jira/browse/HADOOP-9594<https://issues.apache.org/jira/browse/HADOOP-9594>
>> https://issues.apache.org/**jira/browse/MAPREDUCE-5431<https://issues.apache.org/jira/browse/MAPREDUCE-5431>
>> https://issues.apache.org/**jira/browse/HADOOP-9611<https://issues.apache.org/jira/browse/HADOOP-9611>
>> https://issues.apache.org/**jira/browse/HADOOP-9613<https://issues.apache.org/jira/browse/HADOOP-9613>
>> https://issues.apache.org/**jira/browse/HADOOP-9623<https://issues.apache.org/jira/browse/HADOOP-9623>
>> https://issues.apache.org/**jira/browse/HDFS-5411<https://issues.apache.org/jira/browse/HDFS-5411>
>> https://issues.apache.org/**jira/browse/HADOOP-10067<https://issues.apache.org/jira/browse/HADOOP-10067>
>> https://issues.apache.org/**jira/browse/HDFS-5075<https://issues.apache.org/jira/browse/HDFS-5075>
>> https://issues.apache.org/**jira/browse/HADOOP-10068<https://issues.apache.org/jira/browse/HADOOP-10068>
>> https://issues.apache.org/**jira/browse/HADOOP-10075<https://issues.apache.org/jira/browse/HADOOP-10075>
>> https://issues.apache.org/**jira/browse/HADOOP-10076<https://issues.apache.org/jira/browse/HADOOP-10076>
>> https://issues.apache.org/**jira/browse/HADOOP-9849<https://issues.apache.org/jira/browse/HADOOP-9849>
> most (all?) of these are  pom changes

A good number are basically pom changes to update to newer versions of
dependencies.  A few, such as commons-math3, required code changes as
well because of a namespace change.  Some are minor code changes to
enhance compatibility with newer dependencies.  Even tomcat is mostly
changes in pom files.

>> Most of the changes are minor.  There are 2 big updates though: Jetty 9
>> (which requires java 7) and tomcat 7.  These are also the most difficult
>> patches to rebase when hadoop produces a new release.
> that's not going to go in the 2.x branch. Java 6 is still a common platform
> that people are using, because historically java7 (or any leading edge java
> version) is buggy.
> that said, our QA team did test hadoop 2 & HDP-2 at scale on java7 and
> openjdk 7, so it all works -it's just the commit "java7 only" is a big
> decision that

I realize moving to java 7 is a big decision and wasn't trying to imply
this should happen without discussion and planning, just that it would
be nice to have the discussion and see where things land.  It can also
help minimize work.  There is an open bz for updating jetty to jetty 8
(the last version that would work on java 6), but if there are plans to
move to java7, maybe it makes sense to just to jetty 9 and not test a
new version of jetty twice.

With Hadoop in Fedora running on these newer deps there is a test bed to
play with to give some level of confidence before taking the plunge on
any major change.