Execution Error, return code 1 from org.apache.hadoop.hive.ql.exec.mr.MapredLocalTask

hive 遇到的问题，总结一下：

Query ID = grid_20151208110606_377592ad-5984-4f7b-9cfc-9cb2d6be4b6a
Total jobs = 1
java.io.IOException: Cannot run program "/home/grid/hadoop-2.6.1-64/bin/hadoop" (in directory "/root"): error=13, 权限不够
 at java.lang.ProcessBuilder.start(ProcessBuilder.java:1041)
 at java.lang.Runtime.exec(Runtime.java:617)
 at java.lang.Runtime.exec(Runtime.java:450)
 at org.apache.hadoop.hive.ql.exec.mr.MapredLocalTask.executeInChildVM(MapredLocalTask.java:289)
 at org.apache.hadoop.hive.ql.exec.mr.MapredLocalTask.execute(MapredLocalTask.java:137)
 at org.apache.hadoop.hive.ql.exec.Task.executeTask(Task.java:160)
 at org.apache.hadoop.hive.ql.exec.TaskRunner.runSequential(TaskRunner.java:85)
 at org.apache.hadoop.hive.ql.Driver.launchTask(Driver.java:1604)
 at org.apache.hadoop.hive.ql.Driver.execute(Driver.java:1364)
 at org.apache.hadoop.hive.ql.Driver.runInternal(Driver.java:1177)
 at org.apache.hadoop.hive.ql.Driver.run(Driver.java:1004)
 at org.apache.hadoop.hive.ql.Driver.run(Driver.java:994)
 at org.apache.hadoop.hive.cli.CliDriver.processLocalCmd(CliDriver.java:201)
 at org.apache.hadoop.hive.cli.CliDriver.processCmd(CliDriver.java:153)
 at org.apache.hadoop.hive.cli.CliDriver.processLine(CliDriver.java:364)
 at org.apache.hadoop.hive.cli.CliDriver.executeDriver(CliDriver.java:712)
 at org.apache.hadoop.hive.cli.CliDriver.run(CliDriver.java:631)
 at org.apache.hadoop.hive.cli.CliDriver.main(CliDriver.java:570)
 at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
 at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
 at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
 at java.lang.reflect.Method.invoke(Method.java:606)
 at org.apache.hadoop.util.RunJar.run(RunJar.java:221)
 at org.apache.hadoop.util.RunJar.main(RunJar.java:136)
Caused by: java.io.IOException: error=13, 权限不够
 at java.lang.UNIXProcess.forkAndExec(Native Method)
 at java.lang.UNIXProcess.(UNIXProcess.java:135)
 at java.lang.ProcessImpl.start(ProcessImpl.java:130)
 at java.lang.ProcessBuilder.start(ProcessBuilder.java:1022)
 ... 23 more
FAILED: Execution Error, return code 1 from org.apache.hadoop.hive.ql.exec.mr.MapredLocalTask

从上边错误及下边报错的类MapredLocalTask可以看出跟本地任务有关

hive从0.7版以后，为了提高小数据的计算速度，增加了本地模式，即将hdfs上的数据拉到hiveserver本地进行计算，可以通过以下几个参数对相关行为进行设置 

hive.exec.mode.local.auto=false
hive.exec.mode.local.auto.input.files.max=4
hive.exec.mode.local.auto.inputbytes.max=134217728

其中第一个为不启用本地模式，第二个参数表示文件数小于4时使用本地模式，第三个参数表示文件大小小于128m时采用本地模式

默认为不启用本地模式；在启用的情况下，满足第二、三个条件中的任意一个都会使用本地模式。

        在之前我们用过的0.8.1、0.10版上都未遇到过上述错误，怀疑是现在0.12版本的问题突然导致上述错误。任务是在root用户下通过crontab调用的，进入shell后先启动hiveserver，所以默认工作目录其实是/root；为了能正常读写hdfs上的文件，hiveserver在启动时切换到了hdfs用户，一旦遇到上述两种满足启用本地模式的情况，hdfs用户试图向当前工作目录/root拉取数据，必然没有权限从而导致以上错误：

 接下来

[grid@h1 /]$ ls -l
总计 162
drwxr-xr-x   2 root root  4096 09-30 04:07 bin
drwxr-xr-x   4 root root  1024 09-29 11:42 boot
drwxr-xr-x   3 root root  4096 10-16 10:38 data0
drwxr-xr-x  11 root root  4000 11-19 09:42 dev
drwxr-xr-x 105 root root 12288 12-05 04:03 etc
drwxr-xr-x   5 root root  4096 12-03 18:15 home
drwxr-xr-x  11 root root  4096 09-30 04:07 lib
drwxr-xr-x   8 root root 12288 09-30 04:07 lib64
drwx------   2 root root 16384 2012-09-17 lost+found
drwxr-xr-x   2 root root  4096 11-19 09:39 media
drwxr-xr-x   2 root root     0 11-19 09:39 misc
drwxr-xr-x   2 root root  4096 2011-05-11 mnt
drwxr-xr-x   2 root root     0 11-19 09:39 net
drwxr-xr-x   3 root root  4096 12-03 18:16 opt
dr-xr-xr-x 143 root root     0 11-19 09:37 proc
drwxr-x---  20 root root  4096 12-08 14:18 root
drwxr-xr-x   2 root root 12288 09-30 04:07 sbin
drwxr-xr-x   4 root root     0 11-19 09:37 selinux
drwxr-xr-x   2 root root  4096 2011-05-11 srv
drwxr-xr-x  11 root root     0 11-19 09:37 sys
drwxrwxrwt  15 root root  4096 12-08 14:19 tmp
drwxr-xr-x  22 root root  4096 11-24 14:16 usr
drwxr-xr-x  24 root root  4096 09-29 11:30 var

看到root   其它用户无  -x 和-r权限 

[grid@h1 /]$ sudo chmod -R 755 /root
[sudo] password for grid:
[grid@h1 /]$ ls -l
总计 162
drwxr-xr-x   2 root root  4096 09-30 04:07 bin
drwxr-xr-x   4 root root  1024 09-29 11:42 boot
drwxr-xr-x   3 root root  4096 10-16 10:38 data0
drwxr-xr-x  11 root root  4000 11-19 09:42 dev
drwxr-xr-x 105 root root 12288 12-05 04:03 etc
drwxr-xr-x   5 root root  4096 12-03 18:15 home
drwxr-xr-x  11 root root  4096 09-30 04:07 lib
drwxr-xr-x   8 root root 12288 09-30 04:07 lib64
drwx------   2 root root 16384 2012-09-17 lost+found
drwxr-xr-x   2 root root  4096 11-19 09:39 media
drwxr-xr-x   2 root root     0 11-19 09:39 misc
drwxr-xr-x   2 root root  4096 2011-05-11 mnt
drwxr-xr-x   2 root root     0 11-19 09:39 net
drwxr-xr-x   3 root root  4096 12-03 18:16 opt
dr-xr-xr-x 143 root root     0 11-19 09:37 proc
drwxr-xr-x  20 root root  4096 12-08 14:18 root
drwxr-xr-x   2 root root 12288 09-30 04:07 sbin
drwxr-xr-x   4 root root     0 11-19 09:37 selinux
drwxr-xr-x   2 root root  4096 2011-05-11 srv
drwxr-xr-x  11 root root     0 11-19 09:37 sys
drwxrwxrwt  15 root root  4096 12-08 14:19 tmp
drwxr-xr-x  22 root root  4096 11-24 14:16 usr
drwxr-xr-x  24 root root  4096 09-29 11:30 var
[grid@h1 /]$ hive

15/12/08 14:45:32 WARN conf.HiveConf: DEPRECATED: Configuration property hive.metastore.local no longer has any effect. Make sure to provide a valid value for hive.metastore.uris if you are connecting to a remote metastore.
15/12/08 14:45:32 WARN conf.HiveConf: HiveConf of name hive.metastore.local does not exist

Logging initialized using configuration in jar:file:/home/grid/apache-hive-1.0.1-bin/lib/hive-common-1.0.1.jar!/hive-log4j.properties
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/home/grid/hadoop-2.6.1-64/share/hadoop/common/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/home/grid/apache-hive-1.0.1-bin/lib/hive-jdbc-1.0.1-standalone.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]
hive> select a.UUID, a.TASK_NAME, a.START_TIME, b.ORG_NAME,b.ORG_ID FROM NISMP_E_YEAR_WORK_PLAN_ETL a join NISMP_R_ORG_YEAR_WORK_PLAN_ETL b on (a.UUID = b.TASK_UUID);
Query ID = grid_20151208144545_cf044720-85e5-4c04-9a84-e75d465041a0
Total jobs = 1
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/home/grid/hadoop-2.6.1-64/share/hadoop/common/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/home/grid/apache-hive-1.0.1-bin/lib/hive-jdbc-1.0.1-standalone.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]
15/12/08 14:46:00 WARN conf.HiveConf: DEPRECATED: Configuration property hive.metastore.local no longer has any effect. Make sure to provide a valid value for hive.metastore.uris if you are connecting to a remote metastore.
15/12/08 14:46:00 WARN conf.HiveConf: HiveConf of name hive.metastore.local does not exist
Execution log at: /tmp/grid/grid_20151208144545_cf044720-85e5-4c04-9a84-e75d465041a0.log
2015-12-08 02:46:00 Starting to launch local task to process map join; maximum memory = 477102080
2015-12-08 02:46:02 Dump the side-table for tag: 0 with group count: 11 into file: file:/tmp/hive/local/e4a94925-1b3e-400a-9d0b-5a75cb757c84/hive_2015-12-08_14-45-53_214_6791763111790754109-1/-local-10003/HashTable-Stage-3/MapJoin-mapfile00--.hashtable
2015-12-08 02:46:02 Uploaded 1 File to: file:/tmp/hive/local/e4a94925-1b3e-400a-9d0b-5a75cb757c84/hive_2015-12-08_14-45-53_214_6791763111790754109-1/-local-10003/HashTable-Stage-3/MapJoin-mapfile00--.hashtable (1050 bytes)
2015-12-08 02:46:02 End of local task; Time Taken: 1.641 sec.
Execution completed successfully
MapredLocal task succeeded
Launching Job 1 out of 1
Number of reduce tasks is set to 0 since there's no reduce operator
Starting Job = job_1447925899672_0008, Tracking URL = http://h1:8088/proxy/application_1447925899672_0008/
Kill Command = /home/grid/hadoop-2.6.1-64/bin/hadoop job  -kill job_1447925899672_0008
Hadoop job information for Stage-3: number of mappers: 2; number of reducers: 0
2015-12-08 14:46:12,591 Stage-3 map = 0%,  reduce = 0%
2015-12-08 14:46:23,123 Stage-3 map = 50%,  reduce = 0%, Cumulative CPU 1.73 sec
2015-12-08 14:46:24,174 Stage-3 map = 100%,  reduce = 0%, Cumulative CPU 3.72 sec
MapReduce Total cumulative CPU time: 3 seconds 720 msec
Ended Job = job_1447925899672_0008
MapReduce Jobs Launched:
Stage-Stage-3: Map: 2   Cumulative CPU: 3.72 sec   HDFS Read: 16591 HDFS Write: 8819 SUCCESS
Total MapReduce CPU Time Spent: 3 seconds 720 msec
OK
9368588218954e30b49d125a7c8602cf 测试 2015-11-12 测试部 ceshi
20aa5494ee9248f6b6a6b282729b56d9 大家好 2015-11-13 南通 320600
c8a269c18e6c49229e0a8c10473a0657 你好 2015-11-13 南通 320600
c8a269c18e6c49229e0a8c10473a0657 你好 2015-11-13 扬州 321000
c8a269c18e6c49229e0a8c10473a0657 你好 2015-11-13 淮安 320800

已经ok了