[SQOOP-3086] Sqoop export with Parquet data fails with error (parquet.io.ParquetDecodingException: Can not read value at 1 in block 0 in file)

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Component/s: tools
• Assignee: Unassigned
• Reporter: Markus Kemper
• Created: 21/Dec/16 15:29
• Updated: 21/Dec/16 20:04

Description

Test Case

#################
# STEP 01 - Setup Table and Data
#################

export MYCONN=jdbc:oracle:thin:@oracle.cloudera.com:1521/orcl12c;
export MYUSER=sqoop
export MYPSWD=cloudera

sqoop eval --connect $MYCONN --username $MYUSER --password $MYPSWD --query "drop table t1_oracle"
sqoop eval --connect $MYCONN --username $MYUSER --password $MYPSWD --query "create table t1_oracle (c1 int, c2 varchar(10))"
sqoop eval --connect $MYCONN --username $MYUSER --password $MYPSWD --query "insert into t1_oracle values (1, 'data')"
sqoop eval --connect $MYCONN --username $MYUSER --password $MYPSWD --query "select * from t1_oracle"

Output:
-------------------------------------
| C1                   | C2         | 
-------------------------------------
| 1                    | data       | 
-------------------------------------

#################
# STEP 02 - Import Data as Parquet
#################

sqoop import --connect $MYCONN --username $MYUSER --password $MYPSWD --table T1_ORACLE --target-dir /user/user1/t1_oracle_parquet --delete-target-dir --num-mappers 1 --as-parquetfile 

Output:
16/12/21 07:11:47 INFO mapreduce.ImportJobBase: Transferred 1.624 KB in 50.1693 seconds (33.1478 bytes/sec)
16/12/21 07:11:47 INFO mapreduce.ImportJobBase: Retrieved 1 records.

#################
# STEP 03 - Verify Parquet Data
#################

hdfs dfs -ls /user/user1/t1_oracle_parquet/*.parquet
parquet-tools schema -d hdfs://namenode.cloudera.com/user/user1/t1_oracle_parquet/a6ba3dda-b5fc-42d7-9555-5837a12a036b.parquet

Output:
-rw-r--r--   3 user1 user1        597 2016-12-21 07:11 /user/user1/t1_oracle_parquet/a6ba3dda-b5fc-42d7-9555-5837a12a036b.parquet
---
message T1_ORACLE {
  optional binary C1 (UTF8);
  optional binary C2 (UTF8);
}
creator: parquet-mr version 1.5.0-cdh5.8.3 (build ${buildNumber})
extra: parquet.avro.schema = {"type":"record","name":"T1_ORACLE","doc":"Sqoop import of T1_ORACLE","fields":[{"name":"C1","type":["null","string"],"default":null,"columnName":"C1","sqlType":"2"},{"name":"C2","type":["null","string"],"default":null,"columnName":"C2","sqlType":"12"}],"tableName":"T1_ORACLE"}
file schema: T1_ORACLE
------------------------------------------------------------------------------------------------------------------------
C1: OPTIONAL BINARY O:UTF8 R:0 D:1
C2: OPTIONAL BINARY O:UTF8 R:0 D:1
row group 1: RC:1 TS:85
------------------------------------------------------------------------------------------------------------------------
C1:  BINARY SNAPPY DO:0 FPO:4 SZ:40/38/0.95 VC:1 ENC:PLAIN,RLE,BIT_PACKED
C2:  BINARY SNAPPY DO:0 FPO:44 SZ:49/47/0.96 VC:1 ENC:PLAIN,RLE,BIT_PACKED

#################
# STEP 04 - Export Parquet Data
#################

sqoop export --connect $MYCONN --username $MYUSER --password $MYPSWD --table T1_ORACLE --export-dir /user/user1/t1_oracle_parquet --num-mappers 1 --verbose

Output:
[sqoop debug]
16/12/21 07:15:06 INFO mapreduce.Job:  map 0% reduce 0%
16/12/21 07:15:40 INFO mapreduce.Job:  map 100% reduce 0%
16/12/21 07:15:40 INFO mapreduce.Job: Job job_1481911879790_0026 failed with state FAILED due to: Task failed task_1481911879790_0026_m_000000
Job failed as tasks failed. failedMaps:1 failedReduces:0
16/12/21 07:15:40 INFO mapreduce.Job: Counters: 8
	Job Counters 
		Failed map tasks=1
		Launched map tasks=1
		Data-local map tasks=1
		Total time spent by all maps in occupied slots (ms)=32125
		Total time spent by all reduces in occupied slots (ms)=0
		Total time spent by all map tasks (ms)=32125
		Total vcore-seconds taken by all map tasks=32125
		Total megabyte-seconds taken by all map tasks=32896000
16/12/21 07:15:40 WARN mapreduce.Counters: Group FileSystemCounters is deprecated. Use org.apache.hadoop.mapreduce.FileSystemCounter instead
16/12/21 07:15:40 INFO mapreduce.ExportJobBase: Transferred 0 bytes in 46.8304 seconds (0 bytes/sec)
16/12/21 07:15:40 WARN mapreduce.Counters: Group org.apache.hadoop.mapred.Task$Counter is deprecated. Use org.apache.hadoop.mapreduce.TaskCounter instead
16/12/21 07:15:40 INFO mapreduce.ExportJobBase: Exported 0 records.
16/12/21 07:15:40 DEBUG util.ClassLoaderStack: Restoring classloader: java.net.FactoryURLClassLoader@577cfae6
16/12/21 07:15:40 ERROR tool.ExportTool: Error during export: Export job failed!

[yarn debug]
2016-12-21 07:15:38,911 DEBUG [Thread-11] org.apache.sqoop.mapreduce.AsyncSqlOutputFormat: Committing transaction of 0 statements
2016-12-21 07:15:38,914 WARN [main] org.apache.hadoop.mapred.YarnChild: Exception running child : parquet.io.ParquetDecodingException: Can not read value at 1 in block 0 in file hdfs://nameservice1/user/user1/t1_oracle_parquet/a6ba3dda-b5fc-42d7-9555-5837a12a036b.parquet
	at parquet.hadoop.InternalParquetRecordReader.nextKeyValue(InternalParquetRecordReader.java:241)
	at parquet.hadoop.ParquetRecordReader.nextKeyValue(ParquetRecordReader.java:227)
	at org.kitesdk.data.spi.filesystem.AbstractCombineFileRecordReader.nextKeyValue(AbstractCombineFileRecordReader.java:68)
	at org.apache.hadoop.mapreduce.lib.input.CombineFileRecordReader.nextKeyValue(CombineFileRecordReader.java:69)
	at org.kitesdk.data.spi.AbstractKeyRecordReaderWrapper.nextKeyValue(AbstractKeyRecordReaderWrapper.java:55)
	at org.apache.hadoop.mapred.MapTask$NewTrackingRecordReader.nextKeyValue(MapTask.java:556)
	at org.apache.hadoop.mapreduce.task.MapContextImpl.nextKeyValue(MapContextImpl.java:80)
	at org.apache.hadoop.mapreduce.lib.map.WrappedMapper$Context.nextKeyValue(WrappedMapper.java:91)
	at org.apache.hadoop.mapreduce.Mapper.run(Mapper.java:144)
	at org.apache.sqoop.mapreduce.AutoProgressMapper.run(AutoProgressMapper.java:64)
	at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:787)
	at org.apache.hadoop.mapred.MapTask.run(MapTask.java:341)
	at org.apache.hadoop.mapred.YarnChild$2.run(YarnChild.java:164)
	at java.security.AccessController.doPrivileged(Native Method)
	at javax.security.auth.Subject.doAs(Subject.java:415)
	at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1709)
	at org.apache.hadoop.mapred.YarnChild.main(YarnChild.java:158)
Caused by: java.lang.ClassCastException: T1_ORACLE cannot be cast to org.apache.avro.generic.IndexedRecord
	at parquet.avro.AvroIndexedRecordConverter.start(AvroIndexedRecordConverter.java:185)
	at parquet.io.RecordReaderImplementation.read(RecordReaderImplementation.java:391)
	at parquet.hadoop.InternalParquetRecordReader.nextKeyValue(InternalParquetRecordReader.java:216)

Issue Links

• is cloned by: SQOOP-3088 Sqoop export with Parquet data failure does not contain the MapTask error