Support Questions

Find answers, ask questions, and share your expertise

Install Accumulo 1.6 on Cloudera quickstart

avatar
Explorer

I am trying to install accumulo 1.6 in the current quickstart  cdh5.2.x distribution (downloaded 12/13/2014) to train folks.

 

I am suprised it is not pre-installed.

 

I want to be able to compile and run accumulo client programs in java and get to apprpriate java docs for accumulo.

 

I tried the instructions at http://www.cloudera.com/content/cloudera/en/documentation/Accumulo/latest/PDF/Apache-Accumulo-Instal...

 

Install Apache Accumulo from Distribution Packages
This section describes how to install Cloudera's packaging of Accumulo from packages (RPM or DEB)
instead of using Cloudera Manager.

 

I am not sure to use rpms, repo, .targz. 

 

I find the instructions somewhat confusing and would appreciate your assistance.

 

thanks in advance.

 

 

1 ACCEPTED SOLUTION

avatar
Guru
The link you posted appears to be broken, but given the snippet, it sounds like that is the approach I would recommend you take in the QuickStart VM - so keep following the docs you already found. CDH is installed in the VM using RPMs, and by default Cloudera Manager is not used, so it would be simplest to use the same method to install Accumulo.

I will look to have the Accumulo repository installed by default in a future QuickStart VM so you can install Accumulo by just running "yum install accumulo*", but the software itself is not installed as we try to keep the VM as small as possible - so stuff is excluded unless we think a majority of users would use it.

View solution in original post

8 REPLIES 8

avatar
Guru
The link you posted appears to be broken, but given the snippet, it sounds like that is the approach I would recommend you take in the QuickStart VM - so keep following the docs you already found. CDH is installed in the VM using RPMs, and by default Cloudera Manager is not used, so it would be simplest to use the same method to install Accumulo.

I will look to have the Accumulo repository installed by default in a future QuickStart VM so you can install Accumulo by just running "yum install accumulo*", but the software itself is not installed as we try to keep the VM as small as possible - so stuff is excluded unless we think a majority of users would use it.

avatar
Explorer

If you could set it up to do so with both Cloudera and Spark QuickStart with yum to install 1.6 that would be awsome. I spend so much time gettign things installed rather than concentrating on solving algorithmic issues. You would so rock!!!!

 

thanks!

avatar
Explorer

Ps HortonWorks already does the yum accumulo, but I use cloudera products.

avatar
Guru
FWIW, this should basically be the procedure to install and run from
packages in the current QuickStart VM:

REPO=
http://archive-primary.cloudera.com/accumulo-c5/redhat/6/x86_64/cdh/cloudera-accumulo.repo
(cd /etc/yum.repos.d && sudo wget ${REPO}) # install yum repository
sudo yum install accumulo-*
sudo service accumulo-master init # follow prompts
for role in (cd /etc/init.d && ls accumulo\*); do
sudo service ${role} start
done

avatar
New Contributor

I tried the instruction above however it looks like I am running into classpath issues.

 

[cloudera@quickstart ~]$ sudo service accumulo-master init
NOTE: it is strongly recommended that you override the following defaults in /etc/accumulo/conf/accumulo-site.xml:
Set logger.dir.walog to a directory on a partition with sufficient space for write-ahead logs
Set tracer.user and tracer.password to values of your choosing
Initializing Accumulo: [ OK ]
Thread "init" died org/apache/htrace/core/Tracer$Builder
java.lang.NoClassDefFoundError: org/apache/htrace/core/Tracer$Builder
at org.apache.hadoop.fs.FsTracer.get(FsTracer.java:42)
at org.apache.hadoop.fs.FileSystem.createFileSystem(FileSystem.java:2683)
at org.apache.hadoop.fs.FileSystem.access$200(FileSystem.java:94)
at org.apache.hadoop.fs.FileSystem$Cache.getInternal(FileSystem.java:2733)
at org.apache.hadoop.fs.FileSystem$Cache.get(FileSystem.java:2715)
at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:382)
at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:181)
at org.apache.accumulo.core.volume.VolumeConfiguration.getConfiguredBaseDir(VolumeConfiguration.java:74)
at org.apache.accumulo.core.volume.VolumeConfiguration.getVolumeUris(VolumeConfiguration.java:96)
at org.apache.accumulo.server.fs.VolumeManagerImpl.get(VolumeManagerImpl.java:407)
at org.apache.accumulo.server.init.Initialize.main(Initialize.java:614)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:606)
at org.apache.accumulo.start.Main$1.run(Main.java:141)
at java.lang.Thread.run(Thread.java:745)
Caused by: java.lang.ClassNotFoundException: org.apache.htrace.core.Tracer$Builder

 

Not sure why I am getting class errors for htrace, didn't think that was required for Accumulo?  Or is that because the CDH repos build accumulo with htrace in the class path?

 

Any ideas?

avatar
Explorer

Hi, did you ever get past this error? we're getting the exact same thing.

avatar
New Contributor

Worked perfectly

avatar
New Contributor

yum install accumulo hasn't worked well.

It prompts that "You need to be root to perform this action"