High Availability (HA) Zabbix Server on CentOs 7

Prerequisite

First time, we need two virtual machines with Ubuntu 18.04 and 3 IP for make high availability for Zabbix Server 3.4

• 1st server :  zabbix-node01 – 10.0.0.1
• 2nd server : zabbix-node02 – 10.0.0.2
• Virtual IP* for HA – 10.0.0.3

What’s Virtual IP :

A virtual IP address (VIP or VIPA) is an IP address that doesn’t correspond to an actual physical network interface. Uses for VIPs include network address translation (especially, one-to-many NAT), fault-tolerance, and mobility. From Wikipedia

Step 1: Configure DNS

To optimize dns queries and avoid false positives, we need to configure our /etc/hosts on both servers.

1 2	10.0.0.1  zabbix-node01  zabbix-node01.example.com 10.0.0.2  zabbix-node02  zabbix-node02.example.com

Step 2: Install Zabbix Server

Add repository Zabbix on both servers

1 2 3

rpm -ivh http://repo.zabbix.com/zabbix/3.0/rhel/7/x86_64/zabbix-release-3.0-1.el7.noarch.rpm yum update -y

Install zabbix-server-mysql on both servers

1	yum install zabbix-server-mysql -y

We will separate the mysql server and frontend

Edit file /etc/zabbix/zabbix_server.conf on both servers

1 2 3 4 5 6

SourceIP=10.0.0.3 <--- cluster_ip ListenIP=10.0.0.3 <--- cluster_ip DBHost=10.0.0.4 <---- Server SQL DBName=zabbixdb DBUser=zabbix DBPassword=password

Step 3: Configure Pacemaker

Install pacemaker  and pcs on both servers

1 2 3 4

yum install pacemaker pcs -y passwd hacluster systemctl enable pcsd.service systemctl start pcsd.service

Now only zabbix-node01, configure authentication

1 2	pcs cluster auth zabbix-node01  zabbix-node02 Username: hacluster

Then create cluster zabbixserver

1	sudo pcs cluster setup --name zabbixserver zabbix-node01 zabbix-node02

Start cluster

1	sudo pcs cluster start --all

if you have a problem please unmask the service pacemaker

1	systemctl unmask pacemaker

On both servers, make sure that Pacemaker and Corosync are started at boot

1 2	systemctl enable corosync.service systemctl enable pacemaker.service

Check cluster and nodes

1 2	sudo pcs status cluster sudo pcs status nodes

Disable stonith and ignore low quorum

1 2	sudo pcs property set stonith-enabled=false pcs property set no-quorum-policy=ignore

We need to create VIP for zabbixserver

1	pcs resource create cluster_vip ocf:heartbeat:IPaddr2 ip=10.0.0.3 cidr_netmask=255.255.255.0 op monitor interval=20s

Test your VIP and status of pcs

1 2	ping 10.0.0.3 pcs status

Disable zabbix-server on both servers, because Pacemaker will actually control the process

1	sudo systemctl disable zabbix-server

Create a new ressource Zabbix, to verify that the zabbix service still works in cluster

1	sudo pcs resource create zabbix_server lsb:zabbix-server op monitor interval=10s

Check if ressource appear when you launch the command

1	sudo pcs status

Ensuring resources run on the same host

1 2	sudo pcs constraint colocation add zabbix_server cluster_vip sudo pcs constraint order cluster_vip then zabbix_server

Test your cluster

Test cluster

Shell

1	sudo pcs cluster stop zabbix-node01