LVS as a load balancer, all requests are sent to LVS first. Therefore, LVS is in a very important position. If the LVS server breaks down, the backend Web server cannot provide services, causing serious impact.
To prevent the load balancing server from going down, you need to set up a backup machine. Both the master server and the backup machine run high availability monitors that monitor each other’s health by sending messages such as “I am alive”. When the backup machine cannot receive the message within a certain period of time, it takes over the service IP address of the primary server and continues to provide load balancing services. When the secondary server receives the message “I am alive” from the primary server, it releases the service IP address and the primary server starts to provide load balancing services again.
### What is Keepalived?
Keepalived is a service software for cluster management to ensure high availability of a cluster and to prevent a single point of failure.
The purpose of Keepalived is to detect the state of the web server. If a web server fails or fails, Keepalived will detect it and remove the faulty web server from the system. When the web server works properly, Keepalived automatically adds the Web server to the server farm. All the work is done automatically, without human intervention. All you need to do is to repair the faulty Web server.
### Keepalived works
Keepalived is based on Virtual Router Redundancy Protocol (VRRP), which stands for Virtual Router Redundancy Protocol.
Virtual route redundancy protocol can be regarded as a protocol to realize router high availability. That is, N routers providing the same function form a router group, which has a master and multiple backup. The master has a VIP that provides services (the default route is this VIP for other machines in the LAN where the router is located). The master sends multicast packets. If the backup fails to receive VRRP packets, the master considers that it has broken down. In this case, you need to elect a backup as master based on the VRRP priority to ensure high availability of the router.
Keepalived consists of three modules, namely core, Check and VRRP. The core module is the core of Keepalived, which is responsible for the startup and maintenance of the main process as well as the loading and parsing of the global configuration file. Check is responsible for health checks, including common checks. VRRP mode is used to implement the VRRP protocol.
###LVS+Keepalived to implement the active/standby process
The initial state
The host is down, and the standby machine provides services
The host to restore
To configure keepalived for 192.168.12.4 and 192.168.12.8, see my previous blog [Diary Report 2017-12-28] about FastDFS * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
Keepalived After the basic configuration is complete, check whether the system service exists
chkconfig –list
Configuring log Files
1. Will keepalived log output to local0 vim/etc/sysconfig/keepalived, KEEPALIVED_OPTIONS = “0” – D – D – S
2. Add local0.* /var/log/keepalive.log to /etc/rsyslog.conf
3. Restart the Keepalived and Rsyslog services: service rsyslog restart service keepalived restart 4. Go under /var/log/ and find keepalive.log
5. Check keepalived.log. tail -f keepalived.log
Chmod +x /etc/init.d/keepalived
Keepalived # # # # configuration
Modify/etc/keepalived/keepalived. Conf, before we look at what wrote in it.
We refer to the official configuration and make corresponding modifications to the keepalive. conf of the main LVS, as follows:
! Configuration File for keepalived
global_defs {
notification_email {
# Email address sent when failure occurs#[email protected]
}
# Which email address to use#notification_email_from [email protected]# Sender server#smtp_server xxx.com
smtp_connect_timeout 30
router_id LVS_DEVEL
}
vrrp_script check_lvs {
script "/etc/keepalived/lvs_check.sh"## Monitor script
interval 2 Time interval, 2 seconds
weight -20 # # the weight
}
vrrp_instance VI_1 {
state MASTER # indicate primary LVS
interface eth0 # HA check port
virtual_router_id 51 The virtual_router_id of the active and standby devices must be the same
priority 100 The secondary LVS are slightly smaller than the primary LVS
advert_int 1 # VRRP Multicast Number of broadcast cycles in seconds
authentication { # define authentication
auth_type PASS The authentication mode is password authentication
auth_pass 1111 # define password
}
track_script {
check_lvs # monitor script
}
virtual_ipaddress { # define the VIP
192.168.12.100 # Multiple VIPs can be added line feed}} virtual_server 192.168.12.100 80 {delay_loop 6Check the status of the realServer every 6 seconds
lb_algo wlc # The scheduling algorithm is weighted minimum number of connections
lb_kind DR LVS works in DR(direct routing) modeNat_mask 255.255.255.0 persistence_timeout 50# Connections with the same IP are allocated to the same realserver within 50 seconds.
protocol TCP Use TCP to monitor the status of the realServerReal_server 192.168.12.2 80 {# define the realservers
weight 3 # define weight
TCP_CHECK { # Note the Spaces between TCP_CHECK and {, if not, only the first realserver will be added
connect_timeout 3 No response for three seconds timed outNb_get_retry 3 delay_before_retry 3 connect_port 80}} real_server 192.168.12.380 {weight 3 TCP_CHECK { connect_timeout 3 nb_get_retry 3 delay_before_retry 3 connect_port 80 } } }Copy the code
Modify the standby under the LVS/etc/keepalived/keepalived conf file. Note the following: Change state to BACKUP, priority to lower than MASTER, and virtual_router_id to the same as MASTER.
In the /etc/keepalived/ directory of the active and standby LVS, create the lvs_check.sh script as follows:
#! /bin/sh
aa=`ipvsadm -ln`
str="Route"
bb=`echo$aa|grep $str|wc -l`
if [ $bb= 0];then
service lvsdr start
sleep 3
aa=`ipvsadm -ln`
bb=`echo$aa|grep $str|wc -l`
if [ $bb= 0];then
killall keepalived
fifiCopy the code
Disable keepalived and LVSDR services in active and standby LVS first.
service lvsdr stop service keepalived stop
Then enable the Keepalived service in the active and standby LVS.
We then checked with ipvsadm to see if LVS was enabled and found that LVS was enabled!
Enable the nginx and LVSDR services of 192.168.12.2 and 192.16812.3.
Enable tomcat for 192.168.12.6 and 192.168.12.7
Then we look at the virtual IP through the IP a command
Virtual IP address of the active LVS device
Virtual IP address of the standby LVS device
This does not seem to be what we expected. We can turn off the firewall first. We first turned off keepalived for the master LVS and then turned off Keepalived for the standby LVS. Then restart the Keepalived service for the primary and standby LVS. After rebooting, we found that the eth0 node of the standby LVS had no virtual ip192.168.12.100, as I expected!
Log in to 192.168.12.100 for the first time after the active and standby LVS are successfully started.
Second access to 192.168.12.100
When we stopped the main LVS, 192.168.12.100 was still accessible. In the standby LVS, IP A is used to check the virtual IP address 192.168.12.100 on node eth0
Then the master LVS is unlocked, and the control is back to the master LVS.
###Summary
Tomorrow we will summarize the problem and then implement LVS+Keepalived dual main mode.
