Yes you can install the hadoop cluster by yourself; would you please run
$KYLIN_HOME/bin/find-hive-dependency.sh, and then copy the output here?
BTW, where you saw this error, in kylin.log or in a map-red job’s log?

On 5/18/15, 2:48 PM, "190184501" <[email protected]> wrote:

>hi,i use apache hive0.14, and i donnot install hcatalog ,  hcatalog is
>integrate in hive since hive0.11
>
>
>and i use apahce hive0.14 so i think hcatalog is installed already .
> and the hcatalog jar  location is:
>apache-hive-0.14.0-bin\hcatalog\share\hcatalog\hive-hcatalog-core-0.14.0.j
>ar ,so i think hcatalog version is 0.14
>
>
>here is my question :
>1 if i use apache environmen  eg: apache hive0.14,  not  hotonworks ,
>not cloudera,  
>  shoule i do something more configure with hcatalog??
>
>
>2 how do i know the version of Hcatalo?
>
>
>
>
>
>
>
>
>
>
>------------------ Original ------------------
>From:  "Shi, Shaofeng";<[email protected]>;
>Date:  Mon, May 18, 2015 02:34 PM
>To:  "[email protected]"<[email protected]>;
>
>Subject:  Re: build kylin with a sample Cube error  for help
>
>
>
>what¹s the version of Hcatalog that you installed? This indicates the
>hcatalog-core jar is old, suggest upgrade to v0.14;
>
>On 5/18/15, 2:21 PM, "190184501" <[email protected]> wrote:
>
>>can someone help me ??   this problem trouble me for serveral days
>>
>>
>>
>>
>>------------------ Original ------------------
>>From:  "190184501";<[email protected]>;
>>Date:  Mon, May 18, 2015 01:56 PM
>>To:  "dev"<[email protected]>;
>>
>>Subject:  build kylin with a sample Cube error  for help
>>
>>
>>
>>hi all, 
>>
>>
>>I build  a sample Cube and the error is :e
>>
>>
>>java.lang.IncompatibleClassChangeError: Found interface
>>org.apache.mapreduce.JobContext, but class was expected:
>>   at 
>>org.apache,hive.hcatalog.mapreduce.HCatBaseInputFormat.getSplits(HcataogI
>>n
>>putFormat.java 102)
>>
>>
>>
>>
>>is my environment wrong ??
>>
>>
>>the environment is :
>>
>>
>>apache hive0.14
>>apache hadoop2.4
>>apache hbase0.98
>>apache zookeeper3.4.5

Reply via email to