Archive
Highlighted

How to configure Hunk for ORC format data in Hive?

Explorer

Hi Team,

I am using Apache Hadoop 2.6.0 and Hive 0.14. I have configured Hunk for text data in hive and HDFS.

but I am not able to configure ORC format data in Hive. I tried properties discussed in (http://answers.splunk.com/answers/210194/settings-to-configure-hunk-for-hive-orc-table.html ) but it is not working. I am getting a garbage values in Hunk.

Please help me by providing settings.

Thanks & Regards
Abhisek

0 Karma
Highlighted

Re: How to configure Hunk for ORC format data in Hive?

Splunk Employee
Splunk Employee

Can you please post how your existing settings look like (provider + virtual index).

0 Karma
Highlighted

Re: How to configure Hunk for ORC format data in Hive?

Explorer

Hi please find below current settings:

Provider:
vix.splunk.search.splitter=HiveSplitGenerator
and all other properties are set to default

Virtual Index:
Path to data in HDFS =
vix.input.1.splitter.hive.dbname = default
vix.input.1.splitter.hive.tablename = hivetable_orc

the above setting is working fine with text files but not working with ORC format.

please help...

0 Karma
Highlighted

Re: How to configure Hunk for ORC format data in Hive?

Splunk Employee
Splunk Employee

For any hive formats other than text, you need to specify fileformat property. So in your case:

vix.input.1.splitter.hive.fileformat = orc

Highlighted

Re: How to configure Hunk for ORC format data in Hive?

Splunk Employee
Splunk Employee

Also, it seems like the provider may be missing this flag
vix.hive.metastore.uris = thrift://metastore.example.com:9083

0 Karma
Highlighted

Re: How to configure Hunk for ORC format data in Hive?

New Member

Hi

I'm using Splunk and Hunk on AWS. We have a Hive external table with textfile format. It's field delimiter is comma. I'm able to setup a index provider and virtual index for the hive table with the following setup. However, the field values are not parsed properly. It seems Splunk put the entire row to the 1st field. Is there a vix.input.1.hive.??? parameter for hive field delimiter?

vix.input.1.splitter.hive.columnnames=id,source,type,amount,status,frompin,regionid,ipaddress,dateclosing,datecreated,dateevent,userseller,accountseller,advertiserid,userid,accountid,saletype,saleitem,saleid,saleamount,request,orderid,fromco,clientid,productprice,utcdate,bonus
vix.input.1.splitter.hive.columntypes=string:string:string:double:string:string:string:string:string:string:string:string:string:string:string:string:string:string:string:string:string:string:string:string:string:string:string
vix.input.1.splitter.hive.dbname=rs
analytics
vix.input.1.splitter.hive.tablename=transaction
vix.input.1.splitter.hive.fileformat=textfile

0 Karma
Highlighted

Re: How to configure Hunk for ORC format data in Hive?

Splunk Employee
Splunk Employee

Yes.
vix.input.1.splitter.hive.rowformat.fields.terminated=,

0 Karma