mysql - 使用 '--direct' 选项的 Sqoop 失败，mysqldump 退出代码为 2 和 3

Question

我在 AWS EMR 中运行 Sqoop。我正在尝试将 ~10 GB 的表从 MySQL 复制到 HDFS。

我得到以下异常

15/07/06 12:19:07 INFO mapreduce.Job: Task Id : attempt_1435664372091_0048_m_000000_2, Status : FAILED
Error: java.io.IOException: mysqldump terminated with status 3
at org.apache.sqoop.mapreduce.MySQLDumpMapper.map(MySQLDumpMapper.java:485)
at org.apache.sqoop.mapreduce.MySQLDumpMapper.map(MySQLDumpMapper.java:49)
at org.apache.hadoop.mapreduce.Mapper.run(Mapper.java:152)
at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:773)
at org.apache.hadoop.mapred.MapTask.run(MapTask.java:341)
at org.apache.hadoop.mapred.YarnChild$2.run(YarnChild.java:175)
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:1548)
at org.apache.hadoop.mapred.YarnChild.main(YarnChild.java:170)

15/07/06 12:19:07 INFO mapreduce.Job: Task Id : attempt_1435664372091_0048_m_000005_2, Status : FAILED
Error: java.io.IOException: mysqldump terminated with status 2
at org.apache.sqoop.mapreduce.MySQLDumpMapper.map(MySQLDumpMapper.java:485)
at org.apache.sqoop.mapreduce.MySQLDumpMapper.map(MySQLDumpMapper.java:49)
at org.apache.hadoop.mapreduce.Mapper.run(Mapper.java:152)
at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:773)
at org.apache.hadoop.mapred.MapTask.run(MapTask.java:341)
at org.apache.hadoop.mapred.YarnChild$2.run(YarnChild.java:175)
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:1548)
at org.apache.hadoop.mapred.YarnChild.main(YarnChild.java:170)

15/07/06 12:19:08 INFO mapreduce.Job:  map 0% reduce 0%
15/07/06 12:19:20 INFO mapreduce.Job:  map 25% reduce 0%
15/07/06 12:19:22 INFO mapreduce.Job:  map 38% reduce 0%
15/07/06 12:19:23 INFO mapreduce.Job:  map 50% reduce 0%
15/07/06 12:19:24 INFO mapreduce.Job:  map 75% reduce 0%
15/07/06 12:19:25 INFO mapreduce.Job:  map 100% reduce 0%

15/07/06 12:23:11 INFO mapreduce.Job: Job job_1435664372091_0048 failed with state FAILED due to: Task failed task_1435664372091_0048_m_000000
Job failed as tasks failed. failedMaps:1 failedReduces:0

15/07/06 12:23:11 INFO mapreduce.Job: Counters: 8
        Job Counters 
        Failed map tasks=28
        Launched map tasks=28
        Other local map tasks=28
        Total time spent by all maps in occupied slots (ms)=34760760
        Total time spent by all reduces in occupied slots (ms)=0
        Total time spent by all map tasks (ms)=5793460
        Total vcore-seconds taken by all map tasks=5793460
        Total megabyte-seconds taken by all map tasks=8342582400
15/07/06 12:23:11 WARN mapreduce.Counters: Group FileSystemCounters is deprecated. Use org.apache.hadoop.mapreduce.FileSystemCounter instead
15/07/06 12:23:11 INFO mapreduce.ImportJobBase: Transferred 0 bytes in 829.8697 seconds (0 bytes/sec)
15/07/06 12:23:11 WARN mapreduce.Counters: Group   org.apache.hadoop.mapred.Task$Counter is deprecated. Use org.apache.hadoop.mapreduce.TaskCounter instead
15/07/06 12:23:11 INFO mapreduce.ImportJobBase: Retrieved 0 records.
15/07/06 12:23:11 ERROR tool.ImportTool: Error during import: Import job failed!

如果我在没有“--direct”选项的情况下运行，我会得到通信异常，如 https://issues.cloudera.org/browse/SQOOP-186

我已将 MySQL 中的“net-write-timeout”和“net-read-timeout”值设置为 6000。

我的 Sqoop 命令如下所示

sqoop import -D mapred.task.timeout=0 --fields-terminated-by '\t' --escaped-by '\\' --optionally-enclosed-by '\"' --bindir ./ --connect jdbc:mysql://<remote ip>/<mysql db> --username tuser --password tuser --table table1 --target-dir=/base/table1 --split-by id -m 8 --direct

如何解决？我错过了什么吗？

我还创建了 SQOOP JIRA - https://issues.apache.org/jira/browse/SQOOP-2411

Answer 1

我曾看到当 Sqoop 无法平均划分键空间并且其中一个映射任务处理零行数据时会发生此错误。可能的解决方法是更改​​映射器的数量 (-n) 或指定具有均匀分布值的不同键列 (--split-by)。