Copied the below information from Apache JIRA.
--------------------------------------
Hi Lionel
Global dictionary is generated successfully but data loading graph is not started because it seems that kettle home at executor size is not set properly as displayed in logs.
NFO 23-12 16:58:47,461 - Executor task launch worker-4 {carbon.graph.rowset.size=100000, carbon.enable.quick.filter=false, carbon.number.of.cores=4, carbon.sort.file.buffer.size=20, carbon.kettle.home=$<SPARK_HOME>/carbonlib/carbonplugins,
Carbon property is referred from /opt/cloudera/parcels/CDH-5.8.0-1.cdh5.8.0.p0.42/lib/spark/conf/carbon.properties path .
Below suggested solution can work.
a.. Correct the Kettle home path in carbon.properties and try again .
b. use without Kettle flow (please refer examples) .
Thanks
Babu
Reply
Cao, Lionel added a comment - 13 hours ago
Thank you Babu! It works!
Lu Cao wrote
Hi team,
Could you help look into this issue?
I have attached the log in the Jira ticket.
Thanks & Best Regards,
Lionel
On Fri, Dec 23, 2016 at 5:47 PM, Cao, Lionel (JIRA) <
[hidden email]> wrote:
> Cao, Lionel created CARBONDATA-559:
> --------------------------------------
>
> Summary: Job failed at last step
> Key: CARBONDATA-559
> URL:
https://issues.apache.org/jira/browse/CARBONDATA-559> Project: CarbonData
> Issue Type: Bug
> Components: core
> Affects Versions: 0.2.0-incubating
> Environment: carbon version: branch-0.2
> hadoop 2.4.0
> spark 1.6.0
> OS centOS
> Reporter: Cao, Lionel
>
>
> Hi team,
> My job alwasy failed at last step:
> it said 'yarn' user don't have write access to target data
> path(storeLocation).
> But I tested twice with 10000 rows data, both successed. could you help
> look into the log? Please refer to the attachment.
> Search 'access=WRITE' you can see the exception.
> Search 'Exception' for other exceptions.
>
> thanks,
> Lionel
>
>
>
>
> --
> This message was sent by Atlassian JIRA
> (v6.3.4#6332)
>
Locked
