Question

格式化hadoop时出现的错误：in thread "main" java.lang.NoClassDefFoundError:

好多博客说是CLASSPATH设置的问题，但是更改之后仍然如此，求解答
Exception in thread "main" java.lang.NoClassDefFoundError: org/apache/hadoop/security/authorize/RefreshAuthorizationPolicyProtocol
at java.lang.ClassLoader.defineClass1(Native Method)
at java.lang.ClassLoader.defineClass(ClassLoader.java:800)
at java.security.SecureClassLoader.defineClass(SecureClassLoader.java:142)
at java.net.URLClassLoader.defineClass(URLClassLoader.java:449)
at java.net.URLClassLoader.access$100(URLClassLoader.java:71)

Answer 1

您好，
user6@das0 Hadoop-0.20.203.0]$ bin/hadoop namenode -format
12/02/20 14:05:17 INFO namenode.NameNode: STARTUP_MSG:
Re-format filesystem in /data0/hadoop-name ? (Y or N) y
Format aborted in /data0/hadoop-name
12/02/20 14:05:20 INFO namenode.NameNode: SHUTDOWN_MSG:

随后启动hadoop, 发现http://das0:5007无法显示。
将/data0/hadoop-name文件夹整个删除。然后再格，成功！！！

[zhangpeng6@das0 hadoop-0.20.203.0]$ bin/hadoop namenode -format
12/02/20 14:09:57 INFO namenode.NameNode: STARTUP_MSG:
12/02/20 14:09:57 INFO util.GSet: VM type = 64-bit
12/02/20 14:09:57 INFO util.GSet: 2% max memory = 177.77875 MB
12/02/20 14:09:57 INFO util.GSet: capacity = 2^24 = 16777216 entries
12/02/20 14:09:57 INFO util.GSet: recommended=16777216, actual=16777216
12/02/20 14:09:57 INFO namenode.FSNamesystem: fsOwner=zhangpeng6
12/02/20 14:09:57 INFO namenode.FSNamesystem: supergroup=supergroup
12/02/20 14:09:57 INFO namenode.FSNamesystem: isPermissionEnabled=true
12/02/20 14:09:57 INFO namenode.FSNamesystem: dfs.block.invalidate.limit=100
12/02/20 14:09:57 INFO namenode.FSNamesystem: isAccessTokenEnabled=false accessKeyUpdateInterval=0 min(s), accessTokenLifetime=0 min(s)
12/02/20 14:09:57 INFO namenode.NameNode: Caching file names occuring more than 10 times
12/02/20 14:09:57 INFO common.Storage: Image file of size 116 saved in 0 seconds.
12/02/20 14:09:57 INFO common.Storage: Storage directory /data0/hadoop-name/namenode has been successfully formatted.
12/02/20 14:09:57 INFO namenode.NameNode: SHUTDOWN_MSG:
在对namenode格式化之前，要确保dfs.name.dir参数指定的目录不存在。
Hadoop这样做的目的是防止错误地将已存在的集群格式化了。

追问

你在回答我的问题？

Answer 2

Java错误~

追问

嗯，是环境变量的问题么？

追答

是啊，

还有什么问题么？？