hive问题汇总

1. Container killed by the ApplicationMaster. Container killed on request. Exit code is 143 Container exited with a non-zero exit code 143

Cause:

MapReduce -GC overhead limit exceeded Container killed by the ApplicationMaster

Solution:

you can fix this issue by either increase the number of reducers

set mapreduce.job.reduces=15;

or by increasing the reduce heap size

set mapreduce.reduce.java.opts=-Xmx2048m

2. java.net.ConnectException: Call From slave81/10.151.11.81 to slave81:8032 failed on connection exception: java.net.ConnectException: 拒绝连接; For more details see:  http://wiki.apache.org/hadoop/ConnectionRefused

Solution:

start-yarn.sh

3. Call From slave81/10.151.11.81 to 0.0.0.0:10020 failed on connection exception

Solution:

mapred.xml

+<property>
                <name>mapreduce.jobhistory.address</name>
                <value>Master.Hadoop:10020</value>
        </property>

Master.Hadoop为master主机IP

补充：
（1） ping 10.151.11.81能通,telnet 10.151.11.81 9000不能通，说明开启了防火墙
（2）关闭主机10.151.11.81防火墙/etc/init.d/iptables stop，显示:
iptables：清除防火墙规则：[确定]
iptables：将链设置为政策 ACCEPT：filter [确定]
iptables：正在卸载模块：[确定]
（3）重启

（4）hadoop fs -ls hdfs://namenode(ip):8020

4. org.apache.hadoop.ipc.RpcNoSuchProtocolException: Unknown protocol: org.apache.hadoop.mapreduce.v2.api.HSClientProtocolPB

mapred-site.xml

+

1. <property>
   
2.     <name>mapreduce.jobhistory.done-dir</name>
   
3.     <value>${yarn.app.mapreduce.am.staging-dir}/history/done</value>
   
4. </property>
   
5. 
   
6. <property>
   
7.     <name>mapreduce.jobhistory.intermediate-done-dir</name>
   
8.     <value>${yarn.app.mapreduce.am.staging-dir}
   
9.                          /history/done_intermediate</value>
   
10. </property>
    
11. 
    
12. <property>
    
13.     <name>yarn.app.mapreduce.am.staging-dir</name>
    
14.     <value>/tmp/hadoop-yarn/staging</value>
    
15. </property>

./mr-jobhistory-daemon.sh   start historyserver