hadoop无法正常启动(1)
执行  $ bin/hadoop start-all.sh之后,无法启动.
异常一

Exception in thread "main" java.lang.IllegalArgumentException: Invalid URI for NameNode address (check fs.defaultFS): 
file:/// has no authority.
localhost:      at org.apache.hadoop.hdfs.server.namenode.NameNode.getAddress(NameNode.java:214)
localhost:      at org.apache.hadoop.hdfs.server.namenode.SecondaryNameNode.initialize(SecondaryNameNode.java:135)
localhost:      at org.apache.hadoop.hdfs.server.namenode.SecondaryNameNode.<init>(SecondaryNameNode.java:119)
localhost:      at org.apache.hadoop.hdfs.server.namenode.SecondaryNameNode.main(SecondaryNameNode.java:481)

解决方法：此时是没有配置conf/mapred-site.xml的缘故.  在0.21.0版本上是配置mapred-site.xml,在之前的版本是配置core-site.xml,0.20.2版本中配置mapred-site.xml无效,只能配置core-site.xml文件

<configuration>
  <property>
    <name>fs.default.name</name>
    <value>hdfs://localhost:9000</value>
  </property>
  <property>
    <name>mapred.job.tracker</name>
    <value>hdfs://localhost:9001</value>
  </property>
  <property>
    <name>dfs.replication</name>
    <value>1</value>
  </property>
</configuration>

hadoop无法正常启动(2)

异常二、

starting namenode, logging to /home/xixitie/hadoop/bin/../logs/hadoop-root-namenode-aist.out
localhost: starting datanode, logging to /home/xixitie/hadoop/bin/../logs/hadoop-root-datanode-aist.out
localhost: starting secondarynamenode, logging to /home/xixitie/hadoop/bin/../logs/hadoop-root-secondarynamenode-aist.out
localhost: Exception in thread "main" java.lang.NullPointerException
localhost:      at org.apache.hadoop.net.NetUtils.createSocketAddr(NetUtils.java:134)
localhost:      at org.apache.hadoop.hdfs.server.namenode.NameNode.getAddress(NameNode.java:156)
localhost:      at org.apache.hadoop.hdfs.server.namenode.NameNode.getAddress(NameNode.java:160)
localhost:      at org.apache.hadoop.hdfs.server.namenode.SecondaryNameNode.initialize(SecondaryNameNode.java:131)
localhost:      at org.apache.hadoop.hdfs.server.namenode.SecondaryNameNode.<init>(SecondaryNameNode.java:115)
localhost:      at org.apache.hadoop.hdfs.server.namenode.SecondaryNameNode.main(SecondaryNameNode.java:469)
starting jobtracker, logging to /home/xixitie/hadoop/bin/../logs/hadoop-root-jobtracker-aist.out
localhost: starting tasktracker, logging to /home/xixitie/hadoop/bin/../logs/hadoop-root-tasktracker-aist.out

解决方法：此时是没有配置conf/mapred-site.xml的缘故.  在0.21.0版本上是配置mapred-site.xml,在之前的版本是配置core-site.xml   , 0.20.2版本中配置mapred-site.xml无效,只能配置core-site.xml文件

<configuration>
  <property>
    <name>fs.default.name</name>
    <value>hdfs://localhost:9000</value>
  </property>
  <property>
    <name>mapred.job.tracker</name>
    <value>hdfs://localhost:9001</value>
  </property>
  <property>
    <name>dfs.replication</name>
    <value>1</value>
  </property>
</configuration>

hadoop无法正常启动(3)

异常三、

starting namenode, logging to /home/xixitie/hadoop/bin/../logs/hadoop-root-namenode-aist.out
localhost: starting datanode, logging to /home/xixitie/hadoop/bin/../logs/hadoop-root-datanode-aist.out
localhost: Error: JAVA_HOME is not set.
localhost: starting secondarynamenode, logging to /home/xixitie/hadoop/bin/../logs/hadoop-root-secondarynamenode-aist.out
localhost: Error: JAVA_HOME is not set.
starting jobtracker, logging to /home/xixitie/hadoop/bin/../logs/hadoop-root-jobtracker-aist.out
localhost: starting tasktracker, logging to /home/xixitie/hadoop/bin/../logs/hadoop-root-tasktracker-aist.out
localhost: Error: JAVA_HOME is not set.

解决方法:

请在$hadoop/conf/hadoop-env.sh文件中配置JDK的环境变量

JAVA_HOME=/home/xixitie/jdk
CLASSPATH=$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar
export JAVA_HOME CLASSPATH

hadoop无法正常启动(4)

异常四:mapred-site.xml配置中使用hdfs://localhost:9001，而不使用localhost:9001的配置

异常信息如下：

11/04/20 23:33:25 INFO security.Groups: Group mapping impl=org.apache.hadoop.security.ShellBasedUnixGroupsMapping; cacheTimeout=300000
11/04/20 23:33:25 WARN fs.FileSystem: "localhost:9000" is a deprecated filesystem name. Use "hdfs://localhost:9000/" instead.
11/04/20 23:33:25 WARN conf.Configuration: mapred.task.id is deprecated. Instead, use mapreduce.task.attempt.id
11/04/20 23:33:25 WARN fs.FileSystem: "localhost:9000" is a deprecated filesystem name. Use "hdfs://localhost:9000/" instead.
11/04/20 23:33:25 WARN fs.FileSystem: "localhost:9000" is a deprecated filesystem name. Use "hdfs://localhost:9000/" instead.

解决方法：

mapred-site.xml配置中使用hdfs://localhost:9000，而不使用localhost:9000的配置

 <property>
    <name>fs.default.name</name>
    <value>hdfs://localhost:9000</value>
  </property>
  <property>
    <name>mapred.job.tracker</name>
    <value>hdfs://localhost:9001</value>
  </property>

hadoop无法正常启动(5)

异常五、no namenode to stop 问题的解决:
异常信息如下：

11/04/20 21:48:50 INFO ipc.Client: Retrying connect to server: localhost/127.0.0.1:9000. Already tried 0 time(s).
11/04/20 21:48:51 INFO ipc.Client: Retrying connect to server: localhost/127.0.0.1:9000. Already tried 1 time(s).
11/04/20 21:48:52 INFO ipc.Client: Retrying connect to server: localhost/127.0.0.1:9000. Already tried 2 time(s).
11/04/20 21:48:53 INFO ipc.Client: Retrying connect to server: localhost/127.0.0.1:9000. Already tried 3 time(s).
11/04/20 21:48:54 INFO ipc.Client: Retrying connect to server: localhost/127.0.0.1:9000. Already tried 4 time(s).
11/04/20 21:48:55 INFO ipc.Client: Retrying connect to server: localhost/127.0.0.1:9000. Already tried 5 time(s).
11/04/20 21:48:56 INFO ipc.Client: Retrying connect to server: localhost/127.0.0.1:9000. Already tried 6 time(s).
11/04/20 21:48:57 INFO ipc.Client: Retrying connect to server: localhost/127.0.0.1:9000. Already tried 7 time(s).
11/04/20 21:48:58 INFO ipc.Client: Retrying connect to server: localhost/127.0.0.1:9000. Already tried 8 time(s).

解决方法:

这个问题是由namenode没有启动起来引起的,为什么no namenode to stop,可能之前的一些数据对namenode有影响,

你需要执行:

$ bin/hadoop namenode -format

然后

$bin/hadoop start-all.sh

hadoop无法正常启动(6)

异常六、no datanode to stop 问题的解决:

有时数据结构出现问题会产生无法启动datanode的问题。然后用 hadoop namenode -format  重新格式化后仍然无效，/tmp中的文件并没有清楚。

其实还需要清除/tmp/hadoop*里的文件。

执行步骤：

一、先删除hadoop:///tmp

hadoop  fs -rmr /tmp

二、停止 hadoop  

stop-all.sh

三、删除/tmp/hadoop*

 rm -rf /tmp/hadoop*

四、格式化hadoop

 hadoop namenode -format

五、启动hadoop

start-all.sh

之后即可解决这个datanode没法启动的问题

                                                                                                                                                                                                                                            --转自

好贴，顶一个

楼楼好贴

不错诶~~~~

不错不错