Josh,

Thanks. Looks like I need to make an enhancement to the AccumuloStorageHandler 
to support columns without CQ.

Regards,
Mike Fagan

From: Josh Elser <[email protected]<mailto:[email protected]>>
Reply-To: "[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>
Date: Tuesday, July 14, 2015 at 2:57 AM
To: "[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>
Subject: Re: Mapping accumulo table to hive - No Column Quantifiers


Try putting in a column qualifier too. The mapping syntax you used implies a 
map with the qualifier as the key and the value as the value on the family foo. 
I'm guessing it doesn't like the empty colqual, but that's just a hunch.

On Jul 13, 2015 9:24 PM, "Fagan, Michael" 
<[email protected]<mailto:[email protected]>> wrote:
Hi,

I am tying to test the integration of Hive14 with Accumulo 1.6.1 using table 
definitions below:

Accumulo

user@dev mike_test> scan

1 foo: []    value;

2 foo: []    value2



Hive

CREATE EXTERNAL TABLE mike_test(row string, foo string)

STORED BY 'org.apache.hadoop.hive.accumulo.AccumuloStorageHandler'

WITH SERDEPROPERTIES ("accumulo.columns.mapping" = ":rowID,foo:*")

TBLPROPERTIES ("accumulo.table.name<http://accumulo.table.name>" = "mike_test");



When selecting from the hive table I receive the following errror:


hive> select * from mike_test;

OK

Failed with exception 
java.io.IOException:org.apache.hadoop.hive.ql.metadata.HiveException: Error 
evaluating row

Time taken: 3.262 seconds


Is this not the correct way to map this or did I miss something?


Regards,

Mike Fagan

Reply via email to