[ 
https://issues.apache.org/jira/browse/DRILL-4426?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Jason Altekruse updated DRILL-4426:
-----------------------------------
    Description: 
A number of these issues have been fixed in the past in individual instances, 
but we should review any remaining cases where a failure does not produce an 
error message with as much useful context information as possible. Filename 
should always be possible, column or record/line number where possible would be 
good.

One such case with a low level parquet failure was reported here.

http://search-hadoop.com/m/qRVAX48ao4xTDne/drill+Query+Return+Error+because+of+a+single+file&subj=Query+Return+Error+because+of+a+single+file

  was:
A number of these issues have been fixed in the past in individual instances. 
but we should review any remaining cases where a failure does not produce an 
error message with as much useful context information as possible. Filename 
should always be possible, column or record/line number where possible would be 
good.

One such case with a low level parquet failure was reported here.

http://search-hadoop.com/m/qRVAX48ao4xTDne/drill+Query+Return+Error+because+of+a+single+file&subj=Query+Return+Error+because+of+a+single+file


> Review storage and format plugins like parquet, JSON, Avro, Hive, etc. to 
> ensure they fail with useful error messages including filename, column, etc.
> ------------------------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: DRILL-4426
>                 URL: https://issues.apache.org/jira/browse/DRILL-4426
>             Project: Apache Drill
>          Issue Type: Improvement
>            Reporter: Jason Altekruse
>            Assignee: Jason Altekruse
>
> A number of these issues have been fixed in the past in individual instances, 
> but we should review any remaining cases where a failure does not produce an 
> error message with as much useful context information as possible. Filename 
> should always be possible, column or record/line number where possible would 
> be good.
> One such case with a low level parquet failure was reported here.
> http://search-hadoop.com/m/qRVAX48ao4xTDne/drill+Query+Return+Error+because+of+a+single+file&subj=Query+Return+Error+because+of+a+single+file



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

Reply via email to