Hadoop HDFS NFS GateWay部署深入具体解释
目的：通过挂载的方式，能够相似訪问本地磁盘的方式一样的訪问Hadoop文件。简单、方便、快捷。0.系统版本号&hadoop版本号
1）系统版本号
# cat /etc/issue
CentOS>
2）Hadoop版本号
# hadoop version
Hadoop 2.7.2
Subversion https://git-wip-us.apache.org/repos/asf/hadoop.git -r b165c4fe8a74265c792ce23f546c64604acf0e41
Compiled by jenkins on 2016-01-26T00:08Z
Compiled with protoc 2.5.0
From source with checksum d0fda26633fa762bff87ec759ebe689c
This command was run using /home/hadoop/hadoop/share/hadoop/common/hadoop-common-2.7.2.jar
You have new mail in /var/spool/mail/root
1、配置Core-site.xml文件
<property> <name>hadoop.proxyuser.hadoop.groups</name>
<value>*</value>
<description>
The 'nfsserver' user is allowed to proxy all members of the 'users-group1' and
'users-group2' groups. Note that in most cases you will need to include the
group "root" because the user "root" (which usually belonges to "root" group) will
generally be the user that initially executes the mount on the NFS client system.
Set this to '*' to allow nfsserver user to proxy any group.
</description>
</property>
<property>
<name>hadoop.proxyuser.hadoop.hosts</name>
<value>*</value>
<description>
This is the host where the nfs gateway is running. Set this to '*' to allow
requests from any hosts to be proxied.
</description>
</property>
含义解读：
1）hadoop.proxyuser.hadoop.groups：第二个和运行Hadoop的username一致，本測试用的名称为Hadoop。
2）hadoop.proxyuser.hadoop.hosts：第二个和运行Hadoop的username一致。本測试用的名称为Hadoop。
2、配置hdfs-site.xml文件
<property> <name>dfs.namenode.accesstime.precision</name>
<value>3600000</value>
<description>The access time for HDFS file is precise upto this value.
The default value is 1 hour. Setting a value of 0 disables
access times for HDFS.
</description>
</property>
<property>
<name>nfs.dump.dir</name>
<value>/tmp/.hdfs-nfs</value>
</property>
<property>
<name>nfs.exports.allowed.hosts</name>
<value>* rw</value>
</property>
<property>
<name>nfs.rtmax</name>
<value>1048576</value>
<description>This is the maximum>
</property>
<property>
<name>nfs.wtmax</name>
<value>65536</value>
<description>This is the maximum>
</property>
3、配置*/etc/hadoop/log4j.properties
log4j.logger.org.apache.hadoop.hdfs.nfs=DEBUG log4j.logger.org.apache.hadoop.oncrpc=DEBUG
4、root用户下停止portmap(Centos 5.X系统）。停止nfs
# service portmap stop
Stopping portmap: [ OK ]
# service nfs stop
注意：5.X的portmap 等价于 6.X上的rpcbind
并通过:service nfs status 验证是否已经结束。
并通过：service portmap status 验证是否已经结束。
5、开启Hadoop的Portmap服务（须要root权限）
> /sbin/hadoop-daemon.sh –script /bin/hdfs start portmap
6、开启Hadoop的Nfs3服务（须要运行Hadoop的用户权限，本系统为Hadoop）
$ /sbin/hadoop-daemon.sh –script /bin/hdfs start nfs3
7、开启Hadoop并验证(须要运行Hadoop的用户权限，本系统为Hadoop）
# ./start-dfs.sh
# jps
14728 ResourceManager
27478 Jps
5228 Portmap
14104 NameNode
14373 SecondaryNameNode
16762 org.eclipse.equinox.launcher_1.3.0.v20130327-1440.jar
13512 Nfs3
8、验证NFS相关服务是否启动成功
1）验证各服务是否成功运行
# rpcinfo -p $nfs_server_ip
program vers proto port
100005 2 tcp 4242 mountd
100000 2 udp 111 portmapper
100000 2 tcp 111 portmapper
100005 1 tcp 4242 mountd
100003 3 tcp 2049 nfs
100005 1 udp 4242 mountd
100005 3 udp 4242 mountd
100005 3 tcp 4242 mountd
2）显示挂载路径
# showmount -e $nfs_server_ip
Export list for WEB-W031:
/ *
9、实施挂载
>mount -t nfs -o vers=3,proto=tcp,nolock,noacl,sync $10.0.1.31://hdfsnew
10、查看挂载是否成功
# ls -al
total 10
drwxr-xr-x5 root 2584148964160 Mar 11 09:43 .
drwxr-xr-x 31 root root 4096 Mar 11 10:40 ..
drwxr-xr-x3 root root 96 Mar 11 09:58 testing
drwxrwx---3 hadoop 2584148964 96 Mar9 11:14 tmp
drwxr-xr-x3 hadoop 2584148964 96 Mar9 13:24 user
# cd testing/
# ls -al
total 2
drwxr-xr-x 3 root root 96 Mar 11 09:58 .
drwxr-xr-x 5 root 2584148964 160 Mar 11 09:43 ..
-rw-rw-r-- 1 hadoop hadoop 18 Mar 11 10:08 dffafdadafasdfaf
# echo"testinglaoyang" > testlaoyang
# ls -al
total 2
drwxr-xr-x 4 root root 128 Mar 11 18:05 .
drwxr-xr-x 5 root 2584148964 160 Mar 11 09:43 ..
-rw-rw-r-- 1 hadoop hadoop 18 Mar 11 10:08 dffafdadafasdfaf
-rw-r--r-- 1 root root 15 Mar 11 18:05 testlaoyang
11、訪问效果
12、结束服务
1）NFS3服务结束（必须和启动hadoop用户一致）
$ */sbin/hadoop-daemon.sh --script */bin/hdfs stop nfs3
2）PortMap服务结束（必须为root用户）
> */sbin/hadoop-daemon.sh --script */bin/hdfs start portmap
3）Hadoop结束（必须和启动hadoop用户一致）
“`
# ./stop-dfs.sh
13、常见问题
java 进程 占用了111 端口
# lsof -i:111
COMMAND PID USER FD TYPE DEVICE>
java 31206 root 474u IPv4 93384812 0t0 TCP *:sunrpc (LISTEN)
java 31206 root 476u IPv4 93384815 0t0 UDP *:sunrpc
“
解决方式：kill -9 31206
错误明细
# mount -t nfs -o vers=3,proto=tcp,nolock,noacl,sync localhost:/ /hdfs_y
mount.nfs: Input/output error
注意：一定要注意配置文件的细节含义。
解决方式：
http://stackoverflow.com/questions/35910751/hdfs-nfs-gateway-mount-nfs-input-output-error/35935229#35935229
參考：
https://hadoop.apache.org/docs/r2.7.2/hadoop-project-dist/hadoop-hdfs/HdfsNfsGateway.html#Allow_mounts_from_unprivileged_clients
http://tianxingzhe.blog.51cto.com/3390077/1650182
作者：铭毅天下
转载请标明出处，原文地址：http://blog.csdn.net/laoyang360/article/details/50865025
假设感觉本文对您有帮助，请点击‘顶’支持一下，您的支持是我坚持写作最大的动力。谢谢！
页:
[1]