KEEPALIVED dual computer hot standby
Overview of "1"
The function of KEEPALIVED is to detect the status of the back-end TCP service. If a back-end node providing TCP service goes down or fails, KEEPALIVED detects it in time and removes the faulty node from the system. When the node providing TCP service is restored and the service is provided normally, KEEPALIVED automatically sends the node providing TCP service to the cluster. All this work is done automatically by KEEPALIVED without human intervention. All that needs to be done manually is to repair the failed server.
"2" installation
Visit the official website http://www.keepalived.org to get the software package. Version 1.2.16 here
The function implemented this time is that when accessing the WEB service of 192.168.3.118, the real servers 192.168.3.1 and 192.168.3.2 are automatically proxied to the back end, and the host computer is 192.168.3.87 and the standby machine is 192.168.3.88.
(1) tar xvf keeplived-1.2.16.tar.gz
(2) cd keepalived-1.2.16
(3) yum install-y openssl openssl-devel
(4). / configure-prefix=/usr/local/keepalived
(5) make & make install
(6) ln-s / usr/local/keepalived/etc/keepalived
"3" configuration
/ etc/keepalived/keepalived.conf master configuration file (master node)
one! Cconfigure File for keepalived
two
3 vrrp_instance VI_1 {
4 # specify that the node should be set to BACKUP on the standby node of the primary node
5 state MASTER
6 # Network interface of kidnapping virtual IP
7 interface eno1677736
8 # VRRP group name, both nodes need to be set the same to indicate that each node belongs to the same VRRP group
9 virtual_router_id 51
10 # the priority of the master node. The value is 1-254. Note that the slave node must have a lower priority than the master node.
11 priority 50
12 # Multicast message sending interval, the two nodes need to set the same
13 advert_int 1
14 # set verification information, the two nodes need to be consistent
15 authentication {
16 auth_type PASS
17 auth_pass 1234
18}
19 # specify a virtual IP. The two nodes need to be set the same.
20 virtual_ipaddress {
21 192.168.3.118
22}
23}
24 # Virtual IP Service
25 virtual_server 192.168.3.118 80 {
26 # set the interval between checking the actual server
27 delay_loop 6
28 # specify LVS algorithm
29 lb_algo rr
30 # specify LVS mode
31 lb_kind nat
32 nat_mask 255.255.255.255
33 # persistent connection settings, session persistence time
34 persistence_timeout 50
35 # forwarding protocol is TCP
36 protocol TCP
37 # actual backend TCP service configuration (1)
38 real_server 192.168.3.1 80 {
39 # weight
40 weight 1
41 TCP_CHECK {
42 # connection Port
43 connect_port port number
44 # connection timeout
45 connect_timeout 3
46 # retry count
47 nb_get_retry 3
48 # retry interval (seconds)
49 delay_before_retry 4
50}
51}
52 # actual backend TCP service configuration (2) the configuration content is the same as above
53 real_server 192.168.3.2 80 {
54 weight 1
55 TCP_CHECK {
56 connect_port port number
57 connect_timeout 3
58 nb_get_retry 3
59 delay_before_retry 4
60}
61}
62}
(backup node configuration)
Unlike the master node, the standby state is set to BACKUP
State BACKUP
The priority is lower than the primary node.
Priority 50
Other configurations are the same as the primary node
"4" starts KEEPALIVED
Primary node:
Export PATH=/usr/local/keepalived/sbin:$PATH.
Keepalived-D-f / etc/keepalived/keepalived.conf
Use the ip addr list command to view service status
Backup node:
/ usr/local/keepalived/sbin/keepalived-D-f / etc/keepalived/keepalived.conf
Also use the ip addr list command to view the service status
Direct access via VIP: curl http://192.168.3.118 (HELLO 192.168.3.1 / HELLO 192.168.3.2 is successful)
At this point, the installation, configuration and startup of KEEPALIVED are basically completed.
