[jira] [Created] (PHOENIX-2889) No exception in Phoenix-Spark when column does not exist

["Ian Hellstrom (JIRA)" <ji...@apache.org>]

Ian Hellstrom created PHOENIX-2889:
--------------------------------------

             Summary: No exception in Phoenix-Spark when column does not exist
                 Key: PHOENIX-2889
                 URL: https://issues.apache.org/jira/browse/PHOENIX-2889
             Project: Phoenix
          Issue Type: Bug
    Affects Versions: 4.4.0
         Environment: Phoenix: 4.4
HBase: 1.1.2.2.3.2.0-2950 (Hortonworks HDP)
Spark: 1.4.1 (compiled with Scala 2.10)
            Reporter: Ian Hellstrom


When using {{phoenixTableAsDataFrame}} on a table with auto-capitalized qualifiers where the user has erroneously specified these with lower caps, no exception is returned. Ideally, a {{org.apache.phoenix.schema.ColumnNotFoundException}} is thrown but instead lines like the following show up in the log

{code}
INFO RpcRetryingCaller: Call exception, tries=10, retries=35, started=48168 ms ago, cancelled=false, msg=
{code}

A minimal example:

{code}
CREATE TABLE test (foo INT, bar VARCHAR);
UPSERT INTO test VALUES (1, 'hello');
UPSERT INTO test VALUES (2, 'bye');
{code}

In Spark (shell):

{code}
import org.apache.hadoop.conf.Configuration
import org.apache.hadoop.hbase.HBaseConfiguration
import org.apache.phoenix.spark._
@transient lazy val hadoopConf: Configuration = new Configuration()
@transient lazy val hbConf: Configuration = HBaseConfiguration.create(hadoopConf)
val df = sqlContext.phoenixTableAsDataFrame("TEST", Array("foo", "bar"), conf = hbConf)
{code}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)