You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@carbondata.apache.org by "charles horrell (JIRA)" <ji...@apache.org> on 2019/01/07 15:20:00 UTC
[jira] [Updated] (CARBONDATA-3234) Unable to read data from
carbondata table stored in S3 using Presto running on EMR
[ https://issues.apache.org/jira/browse/CARBONDATA-3234?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
charles horrell updated CARBONDATA-3234:
----------------------------------------
Summary: Unable to read data from carbondata table stored in S3 using Presto running on EMR (was: Unable to read data from carbondata table stored in S3 with Presto running on EMR)
> Unable to read data from carbondata table stored in S3 using Presto running on EMR
> ----------------------------------------------------------------------------------
>
> Key: CARBONDATA-3234
> URL: https://issues.apache.org/jira/browse/CARBONDATA-3234
> Project: CarbonData
> Issue Type: Bug
> Components: presto-integration
> Environment: Amazon EMR 5.19
> Reporter: charles horrell
> Priority: Major
>
> We are unable to use presto to query a carbondata table stored in S3.
> {code:java}
> presto:default> select count(*) from test_table;
> Query 20190107_135333_00026_8r2c8 failed: tried to access method org.apache.hadoop.metrics2.lib.MutableCounterLong.<init>(Lorg/apache/hadoop/metrics2/MetricsInfo;J)V from class org.apache.hadoop.fs.s3a.S3AInstrumentation
>
> presto:default> select * from test_table;
> Query 20190107_135610_00028_8r2c8 failed: tried to access method org.apache.hadoop.metrics2.lib.MutableCounterLong.<init>(Lorg/apache/hadoop/metrics2/MetricsInfo;J)V from class org.apache.hadoop.fs.s3a.S3AInstrumentation
> {code}
> The catalog appears to have been picked up okay as show tables works as expected as does describing the table it is just when actually trying to access the data that we see the error.
> We configured presto as per the examples here:
> [http://carbondata.apache.org/quick-start-guide.html]
> Querying from Spark works okay however it is vital for our use case that presto also works and with S3.
> Amazon EMR version 5.19
> Spark 2.3.2
> Hadoop 2.8.5
> Presto 0.212
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)