In addition to Weibo, there is also WeChat
Please pay attention
WeChat public account
Shulou
2025-01-19 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > Internet Technology >
Share
Shulou(Shulou.com)06/03 Report--
The application of NFS+SSH is mainly to solve the trouble of changing public key in Hadoop cluster.
I. NFS configuration
1. Installation
[root@kry132 ~] # yum-y install rpcbind nfs-utils
Rpcbind: for communication
Nfs-utils:nfs program
two。 Create a shared directory
[root@kry132 ~] # mkdir / data/nfsdata
[root@kry132 ~] # chown hadoop.hadoop / data/nfsdata/
3. Configure nfs
[root@kry132 ~] # vim / etc/exports
Add the following:
/ data/nfsdata 192.168.0.135 (rw,sync)
If you need to specify a user, you also need to join all_squash,anonuid=502,anongid=502.
Note that the UID and GID of the server and client users must be the same. For example, if a server user is abc,uid and gid is 502, then the client should also have an abc user, and uid and gid should also be 502.
4. Start the service
[root@kry132 ~] # / etc/init.d/rpcbind start
[root@kry132 ~] # / etc/init.d/nfs start
5. Set up boot boot
[root@kry132] # chkconfig-- level 35 rpcbind on
[root@kry132] # chkconfig-- level 35 nfs on
6. View shared directory (client)
[root@Kry135] # showmount-e 192.168.0.132
Export list for 192.168.0.132:
/ data/nfsdata 192.168.0.135
II. SSH configuration
1. Install ssh
[root@kry132 ~] # yum-y install openssh-server openssh-clients
two。 Configuration
[root@kry132 ~] # vim / etc/ssh/sshd_config
RSAAuthentication yes
PubkeyAuthentication yes
AuthorizedKeysFile .ssh / authorized_keys
3. Generate key
[root@kry132] # ssh-keygen-t rsa
4. Modify public key
[root@kry132 ~] # cd .ssh
[root@kry132 .ssh] # cat id_rsa.pub > > authorized_keys
[root@kry132 .ssh] # chown hadoop.hadoop authorized_keys
[root@kry132 .ssh] # chmod 600 authorized_keys
5. Put the public key in the NFS shared directory
[root@kry132 .ssh] # cp-p authorized_keys / data/nfsdata/
III. Client configuration
1. Install ssh
[root@Kry135 ~] # yum-y install openssh-server openssh-clients
two。 Configuration
[root@Kry135 ~] # vim / etc/ssh/sshd_config
RSAAuthentication yes
PubkeyAuthentication yes
AuthorizedKeysFile .ssh / authorized_keys
3. Create a nfs shared directory
[root@Kry135 ~] # mkdir / data/nfsdata/
[root@Kry135 ~] # chown hadoop.hadoop / data/nfsdata/
4. Mount the NFS shared directory
[root@Kry135] # mount-t nfs 192.168.0.132:/data/nfsdata / data/nfsdata
5. Create a public key storage directory
[root@Kry135 ~] # mkdir / home/hadoop/.ssh
[root@Kry135 ~] # chown hadoop.hadoop / home/hadoop/.ssh/
[root@Kry135 ~] # chmod 700 / home/hadoop/.ssh/
Note: if you switch to hadoop user action, you do not need to change the owner of the main group to which the user belongs.
6. Create a soft connection
[root@Kry135] # ln-s / data/nfsdata/authorized_keys / home/hadoop/.ssh/authorized_keys
IV. Testing
1. Close selinux
[root@Kry135 ~] # setenfore 0
[root@Kry135 ~] # vim / etc/selinux/config
SELINUX=permissive or disabled
Note: the client also needs to be closed. Do the same thing above.
two。 Configure NFS fixed port
[root@Kry135 ~] # vim / etc/service (the port number must be below 1024 and not occupied)
Mountd 1011/tcp # rpc.mountd
Mountd 1011/udp # rpc.mountd
Rquotad 1012/tcp # rpc.rquotad
Rquotad 1012/udp # rpc.rquotad
When the nfs service is restarted, the port is fixed.
The ports associated with nfs are:
Portmap 111
Nfsd 2049
Mountd 1011
Rquotad 1012
Note: then configure the iptables release port with the client on the server side, if you just test to shut down iptabls directly.
3. Test secret-free login
[hadoop@kry132 ~] $ssh hadoop@192.168.0.135
Last login: Thu Aug 4 18:08:31 2016 from slave.hadoop
4. Test secret-free transfer files
[root@kry132 ~] # su-hadoop
[hadoop@kry132 ~] $scp test.txt hadoop@192.168.0.135:/home/hadoop/
Test.txt 100% 13 0.0KB/s 00:00
Test login and file transfer are not required to enter a password, indicating that the NFS+SSH is built successfully!
Expand content
1.nfs optimization
[root@kry132 ~] # vim / etc/init.d/nfs
[- z "$RPCNFSDCOUNT"] & & RPCNFSDCOUNT=8
Modify to
[- z "$RPCNFSDCOUNT"] & & RPCNFSDCOUNT=32
The default kernel parameters are:
Cat / proc/sys/net/core/rmem_default
229376
Cat / proc/sys/net/core/rmem_max
229376
Cat / proc/sys/net/core/wmem_default
229376
Cat / proc/sys/net/core/wmem_max
229376
Modified to:
Echo 262144 > / proc/sys/net/core/rmem_default-maximum TCP data reception buffer
Echo 262144 > / proc/sys/net/core/rmem_max-maximum TCP data send buffer
Echo 262144 > / proc/sys/net/core/wmem_default-default send window size
Echo 262144 > / proc/sys/net/core/wmem_max-maximum size of the send window
Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.
Views: 0
*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.
Continue with the installation of the previous hadoop.First, install zookooper1. Decompress zookoope
"Every 5-10 years, there's a rare product, a really special, very unusual product that's the most un
© 2024 shulou.com SLNews company. All rights reserved.