Hi Zoltan, I am saying that hadoop2-stable and hadoop1 are binary compatible. I don't know what version of hadoop is used in cdh4-mr2 but I guess it was hadoop2 alpha, since bigtop was at hadoop 2.0.6 alpha last time I checked, which was last week.
Just try it and let us know if you experience a problem. PS: We're currently working on building mahout against hadoop2 by the way Best > On Dec 11, 2013, at 8:53 PM, Zoltan Prekopcsak <preko1...@gmail.com> wrote: > > Hi Gokhan, > > Thank you for the clarification. > Does it mean that Mahout is using the mapred API everywhere and there is no > mapreduce API left? As far as I know, the mapreduce API needs to be > recompiled and I remember needing to recompile Mahout for CDH4 when it first > came out. > > Thanks, Zoltan > > >> On 12/10/13 10:02 PM, Gokhan Capan wrote: >> I meant that you shouldn't need to modify mahout's dependencies, just mvn >> package and it should work against hadoop 2.2.0 (Yeah, 2.2.0 is not alpha) >> >> Quoting from >> http://hadoop.apache.org/docs/current/hadoop-mapreduce-client/hadoop-mapreduce-client-core/MapReduce_Compatibility_Hadoop1_Hadoop2.html >> "First, we ensure binary compatibility to the applications that use old >> mapred APIs. This means that applications which were built against MRv1 >> mapred APIs can run directly on YARN without recompilation, merely by >> pointing them to an Apache Hadoop 2.x cluster via configuration." >> >> If you encounter with any problems, just let the list know. >> >> Best >> >> >>> On Dec 9, 2013, at 9:40 PM, Hi There <srudamas...@yahoo.com> wrote: >>> >>> Hi Gokhan, >>> >>> My project currently fetches every dependency through Maven--is there any >>> way I can grab the version you mentioned that way? >>> >>> In that vein, I am using the following version of hadoop: >>> <dependency> >>> <groupId>org.apache.hadoop</groupId> >>> <artifactId>hadoop-client</artifactId> >>> <version>2.2.0</version> >>> </dependency> >>> >>> >>> That's not alpha, right? >>> >>> Thanks! >>> >>> >>> >>> >>> >>> On Monday, December 9, 2013 10:05 AM, Gokhan Capan <gkhn...@gmail.com> >>> wrote: >>> >>> Mahout actually should work with hadoop-2 stable without recompiling, >>> not with hadoop-2 alpha though. >>> >>> We're, by the way, currently in the process of adding support to build >>> mahout with hadoop-2. >>> >>> Please see mahout-1354 for the relevant issue >>> >>> Sent from my iPhone >>> >>> >>>> On Dec 9, 2013, at 19:54, Hi There <srudamas...@yahoo.com> wrote: >>>> >>>> Is Dec 2013 still the intended release date of the next mahout release >>>> that will be compatible with Hadoop 2.2.0? >>>> >>>> >>>> >>>> >>>> On Thursday, November 21, 2013 12:36 PM, Suneel Marthi >>>> <suneel_mar...@yahoo.com> wrote: >>>> >>>> Targeted for Dec 2013. >>>> >>>> >>>> >>>> >>>> >>>> >>>> On Thursday, November 21, 2013 3:26 PM, Hi There <srudamas...@yahoo.com> >>>> wrote: >>>> >>>> Thanks for the reply! Is there a timeline for then the next release will >>>> be? >>>> >>>> >>>> Thanks, >>>> Victor >>>> >>>> >>>> >>>> >>>> On Tuesday, November 19, 2013 7:30 PM, Suneel Marthi >>>> <suneel_mar...@yahoo.com> wrote: >>>> >>>> Hi Victor, >>>> >>>> Future releases of Mahout will support Hadoop 2.x, the present codebase >>>> still only supports Hadoop 1.x. >>>> >>>> >>>> >>>> >>>> >>>> >>>> On Tuesday, November 19, 2013 1:42 PM, Hi There <srudamas...@yahoo.com> >>>> wrote: >>>> >>>> >>>> >>>> Hello, >>>> >>>> I recently upgraded to hadoop's >>>> newest release, and it seems one of their interfaces has changed, and >>>> when I try to create sparse vectors from sequence files, I get the >>>> following exception: >>>> >>>> java.lang.IncompatibleClassChangeError: Found interface >>>> org.apache.hadoop.mapreduce.Counter, but class was expected >>>> >>>> I can include more of the stack trace if necessary. >>>> >>>> Are there any plans in the immediate future to upgrade mahout to be >>>> compatible with the newest hadoop release? >>>> >>>> Thanks, >>>> Victor >
