You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hive.apache.org by "Marcell Szabo (JIRA)" <ji...@apache.org> on 2016/04/11 14:22:25 UTC

[jira] [Created] (HIVE-13474) avro partitions with common prefix get mixed up

Marcell Szabo created HIVE-13474:
------------------------------------

             Summary: avro partitions with common prefix get mixed up
                 Key: HIVE-13474
                 URL: https://issues.apache.org/jira/browse/HIVE-13474
             Project: Hive
          Issue Type: Bug
          Components: Query Processor
    Affects Versions: 1.1.0
            Reporter: Marcell Szabo


If we have two Avro tables in the same query, and the path of one table startsWith the other table's path (e.g. tableA, tableAbc), then the table processing get mixed up (slices of tableAbc are read when processing tableA).

Symptom, e.g.
 INFO [main] org.apache.hadoop.hive.ql.io.avro.AvroGenericRecordReader: Matching partition hdfs://nameservice1/foo/BAR.tableA with input split hdfs://nameservice1/foo/BAR.tableAbc/000008_0:0+134217728 

The problem seems to be in
https://github.com/apache/hive/blob/26b5c7b56a4f28ce3eabc0207566cce46b29b558/ql/src/java/org/apache/hadoop/hive/ql/io/avro/AvroGenericRecordReader.java#L114

pathIsInPartition() does a false match because partitionPath does not  contain a slash at the end.



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