I am working on getting a mapr cluster up and running, but I have realy tailored my data for HIve 0.8. Is it possible to get 0.8 (or 0.81) working with mapr? The insert append, the binary data field, and other improvements make it a must for me.


asked 15 Feb '12, 16:48

mandoskippy's gravatar image

mandoskippy ♦
accept rate: 23%

Well, without answering this, Hive 0.9 now works with mapr out of the box. No patches needed. That solves this answer because hive 0.9 has all the features I am looking for.


answered 01 Jul '12, 09:52

mandoskippy's gravatar image

mandoskippy ♦
accept rate: 23%

Hello mandoskippy, are you saying that you've successfully tested hive 0.9 from apache on mapr's distro? If so, which version of MapR are you running?

At my company, we are attempting to use analytical functions on Hive, but we are getting errors. We suspect it may be because these functions require hive > 0.7.1, but that is the version packaged in MapR's latest release 7/13/2012 v1.2.9 as of this writing. However, we have tried only on MapR 1.2.0 so far. https://github.com/hbutani/SQLWindowing/wiki

$ hive --service windowingCli 
Hive history file=/tmp/user/hive_job_log_user_201207131711_1150261077.txt 
Exception in thread "main" groovy.lang.MissingMethodException: No signature of method: org.apache.hadoop.hive.cli.CliSessionState.getHost() is applicable for argument types: () values: [] 
Possible solutions: getConf(), getAt(java.lang.String), getClass(), get(), getCmd(), setConf(org.apache.hadoop.hive.conf.HiveConf) 
at org.codehaus.groovy.runtime.ScriptBytecodeAdapter.unwrap(ScriptBytecodeAdapter.java:55) 
at org.codehaus.groovy.runtime.callsite.PojoMetaClassSite.call(PojoMetaClassSite.java:46) 
at org.codehaus.groovy.runtime.callsite.CallSiteArray.defaultCall(CallSiteArray.java:42) 
at org.codehaus.groovy.runtime.callsite.AbstractCallSite.call(AbstractCallSite.java:108) 
at org.codehaus.groovy.runtime.callsite.AbstractCallSite.call(AbstractCallSite.java:112) 
at com.sap.hadoop.windowing.WindowingHiveCliDriver.main(WindowingHiveCliDriver.groovy:187) 
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) 
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39) 
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25) 
at java.lang.reflect.Method.invoke(Method.java:597) 
at org.apache.hadoop.util.RunJar.main(RunJar.java:186)

answered 16 Jul '12, 16:20

mfuery's gravatar image

accept rate: 0%

edited 16 Jul '12, 16:21


For what is worth, here is my two cents..

I was able to run all versions (0.7.1, 0.8.0, 0.8.1, 0.9.0) of hive straight from apache repos (for example http://www.reverse.net/pub/apache/hive/).

I just downloaded the hive-x.x.x.tar.gz file, un-tarred it in the /opt/mapr/hive folder (where original 0.7.1 version of mapr-hive was installed), and changed PATH to point to "/opt/mapr/hive/hive-x.x.x/bin" version I wanted.

For comparisons reasons, I wanted to use the apache version 0.7.1, along side the original mapr-hive 0.7.1, so I did some folder renaming (see below).

To summarize, I was able to keep ALL versions of hive installed, and able to switch back-and-forth via the PATH; did not notice any "strange behavior" etc :)

$ ll /opt/mapr/hive/
(...) Jun 26 16:35 hive-0.7.1A <== apache version
(...) Apr 26 17:48 hive-0.7.1 <== mapr-hive
(...) Jun 26 17:42 hive-0.8.0A
(...) Jun 26 15:10 hive-0.8.1A
(...) Jun 21 11:39 hive-0.9.0A

$ hadoop version
Hadoop 0.20.2-dev


(19 Jul '12, 14:44) Marek
Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here



Answers and Comments

Markdown Basics

  • *italic* or __italic__
  • **bold** or __bold__
  • link:[text](http://url.com/ "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported



Asked: 15 Feb '12, 16:48

Seen: 1,677 times

Last updated: 19 Jul '12, 14:48

powered by OSQA