[
https://issues.apache.org/jira/browse/TAJO-1796?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14706419#comment-14706419
]
Jongyoung Park commented on TAJO-1796:
--------------------------------------
While Parquet version is updated, verification code for empty schema is added.
And this exception is from the part, because Tajo doesn't assign any projection
schema when count all is done.
> Count all query for Parquet is crashed
> --------------------------------------
>
> Key: TAJO-1796
> URL: https://issues.apache.org/jira/browse/TAJO-1796
> Project: Tajo
> Issue Type: Bug
> Components: Storage
> Reporter: Jongyoung Park
> Assignee: Jongyoung Park
> Priority: Critical
> Fix For: 0.11.0
>
>
> When 'select count(*)' is excuted, it fails with following error log:
> {quote}
> 2015-08-21 17:17:31,371 ERROR
> org.apache.tajo.engine.planner.physical.HashShuffleFileWriteExec: A group
> type can not be empty. Parquet does not support empty group without leaves.
> Empty group: table_schema
> org.apache.parquet.schema.InvalidSchemaException: A group type can not be
> empty. Parquet does not support empty group without leaves. Empty group:
> table_schema
> at org.apache.parquet.schema.GroupType.<init>(GroupType.java:92)
> at org.apache.parquet.schema.GroupType.<init>(GroupType.java:48)
> at org.apache.parquet.schema.MessageType.<init>(MessageType.java:50)
> at
> org.apache.tajo.storage.parquet.TajoSchemaConverter.convert(TajoSchemaConverter.java:152)
> at
> org.apache.tajo.storage.parquet.TajoReadSupport.init(TajoReadSupport.java:76)
> at
> org.apache.parquet.hadoop.InternalParquetRecordReader.initialize(InternalParquetRecordReader.java:172)
> at
> org.apache.parquet.hadoop.ParquetReader.initReader(ParquetReader.java:152)
> at org.apache.parquet.hadoop.ParquetReader.read(ParquetReader.java:128)
> at
> org.apache.tajo.storage.parquet.ParquetScanner.next(ParquetScanner.java:73)
> ...
> {quote}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
