See https://builds.apache.org/job/Hadoop-Mapreduce-trunk/1961/
###
## LAST 60 LINES OF THE CONSOLE
###
[...truncated 30621 lines...]
Tests run: 519, Failures: 0, Errors
Hi,
I also found that a following command fails with hadoop-2.6.0-rc1-src.tar.gz:
$ mvn package -Pdocs -DskipTests -Dtar
I filed it as HADOOP-11316. We should fix it if possible. Or, should
we fix it for 2.6.1?
Thanks,
- Tsuyoshi
On Tue, Nov 18, 2014 at 3:04 PM, Tsuyoshi OZAWA
wrote:
> Suresh
Thanks Ravi!
How to update and build the code ? I want to have two versions of Hadoop
the old one and the updated one inside my machine -so I can test the
performance of both,How to do that?
My regards,
Nada
On Mon, Nov 17, 2014 at 10:17 PM, Ravi Prakash wrote:
> Nada!
> You can look at MergeM
Tsuyoshi,
I think you have to pass along -Pdist too… afaik that is how the Jenkins we
use for building release artifacts operates.
For now, I've marked HADOOP-11316 for inclusion into 2.6.1 if required.
thanks,
Arun
On Nov 18, 2014, at 7:26 AM, Tsuyoshi OZAWA wrote:
> Hi,
>
> I also foun
+1 . Thanks Arun, for producing this release. I downloaded and built the
source. I started local cluster and ran wordcount, sleep, and streaming jobs.
- I ran a distributed shell job which tested preserving containers across AM
restart by setting the -keep_containers_across_application_attempts
Hi Nada!
Please visit https://github.com/apache/hadoop to clone the source code. Please
familiarize yourself with git. You should be able to create branches in your
local repository.
HTHRavi
On Tuesday, November 18, 2014 7:47 AM, Nada Saif
wrote:
Thanks Ravi!
How to update and bu
+1,
Built from source.
Deployed a single node cluster.
Ran sample MapReduce jobs while restarting RM successfully.
Jian
On Tue, Nov 18, 2014 at 8:47 AM, Eric Payne
wrote:
> +1 . Thanks Arun, for producing this release. I downloaded and built
> the source. I started local cluster and ran wordco
Wei Yan created MAPREDUCE-6165:
--
Summary: [JDK8] TestCombineFileInputFormat failed on JDK8
Key: MAPREDUCE-6165
URL: https://issues.apache.org/jira/browse/MAPREDUCE-6165
Project: Hadoop Map/Reduce
+1 (binding) on the source tarball, would be nice to redo the binary
tarball.
- Stood up a pseudo-dist cluster, and ran some HDFS and MR jobs.
- The binary is about 40 MB larger than the previous release; it appears
the docs are copied over twice - share/doc/hadoop and share/hadoop. The
binary tar
+1 if once discussion about HDFS-6833 is done(non-binding).
* Verified checksum.
* Deployed cluster and ran some mr examples.
* Checked RM restart while running jobs.
> I think you have to pass along -Pdist too… afaik that is how the Jenkins we
> use for building release artifacts operates.
Th
+1 (non-binding)
+ built from source tarball
+ ran some example jobs in HiBench such as
nutchindexing, wordcount, dfsioe, hivebench, kmeans, pagerank, bayes, sort,
terasort
on 3 node cluster.
I'm on the same side with Tsuyoshi and would like HDFS-6833 to be fixed in
2.6.1 at least.
Masata
11 matches
Mail list logo
