Memcached+magent+keepalived high Availability cluster two.
We've already configured memcached+magent for the next
If the magent is down and you can't continue to provide the service, use the keepalived and add an alternate magent proxy server. When a magent is down, it is replaced by another.
Configuration steps:
Two magent proxy Server, a set of one to do from, when the main downtime from the successor to continue to work.
The two magent proxy servers specify the same primary cache node and the backup cache node when they are restarted.
Set priority by Keepalived configuration file to determine who magent or from Magent, when the main magent normal operation of the VIP in the main magent, when the main magent downtime VIP automatically changed from Magent, After the main magent restored VIP automatically returned to the main magent.
Install Magent and keepalived on the master and subordinate Magent respectively
Magent installation is not in the same description as before.
Keepalived Installation
Install dependencies I only have one bag here, so I only have one, and if the difference is much based on the preconfigured Yes error confidence install the response package
[root@www keepalived-1.2.13]# yum-y Install Openssl-devel
Decompression Compile installation keepalived
[Root@www ~]# tar-xf/src/keepalived-1.2.13.tar.gz-c/usr/src/
[Root@www ~]# cd/usr/src/keepalived-1.2.13/
[Root@www keepalived-1.2.13]#./configure--prefix=/&& make && make install
keepalived settings
Main kepalived
From the keepalived setting is basically consistent with the Lord, where the difference is already marked.
Test whether the VIP will be transferred (before the test, the definition script and the part of the calling script are commented out because there is no script keepalived the search does not appear to be error)
The master will have a VIP on top of it.
Main
From
Turn off the main keepalived view from the IP will see VIP transferred from above
Main:
From