Zabbix 高可用架構部署方案(MySQL + 雙 VIP+HAProxy+Nginx)
前景提要:使用 MySQL 作為數據庫,兩個虛擬 IP(10.0.0.100 和 10.0.0.200),HAProxy 作為數據庫負載均衡,Nginx 作為 Web 訪問入口。
1. 架構規劃
Server1(10.0.0.12):主 Zabbix Server + MySQL 主庫 + HAProxy(主) + Keepalived
Server2(10.0.0.15):備 Zabbix Server + MySQL 從庫 + HAProxy(備) + Keepalived
Server3(10.0.0.18):Nginx 負載均衡器
2.環境準備
在所有服務器上執行:
# 更新系統 時間可能會有點兒長(可選)
yum update -y# 關閉防火墻和SELinux(生產環境需配置規則)
systemctl disable --now firewalld
setenforce 0
sed -i 's/SELINUX=enforcing/SELINUX=disabled/g' /etc/selinux/config# 配置主機名解析
cat > /etc/hosts << EOF
127.0.0.1 localhost localhost.localdomain
10.0.0.12 server1 zabbix-master
10.0.0.15 server2 zabbix-backup
10.0.0.18 server3 zabbix-lb
10.0.0.100 zabbix-web
10.0.0.200 zabbix-db
EOF# 安裝基礎工具(可選)
yum install -y vim wget net-tools
3.安裝 MySQL
在Server2(10.0.0.15)上執行: 建議server1在安裝zabbix時安裝mysql
# 安裝mysql
yum install mysql-server -y# 啟動并設置開機自啟
systemctl enable --now mysql# 安全初始化
mysql_secure_installation
4. 配置 MySQL主從復制(挫折重重)
主庫(server1)配置? 此步驟建議在zabbix官網安裝zabbix完成之后再進行配置
cat > /etc/my.cnf.d/mysql-server.cnf << EOF
[mysqld]
server-id=1
log-bin=mysql-bin
binlog-do-db=zabbix
expire-logs-days=10
max-binlog-size=100M
binlog-format=ROW
innodb_flush_log_at_trx_commit=1
sync_binlog=1
EOF# 重啟MySQL
systemctl restart mysqld# 創建復制用戶 zabbix已經在安裝zabbix 的時候創建好,這里就不再贅述
CREATE USER 'repl'@'%' IDENTIFIED BY 'ReplicationPassword';
GRANT REPLICATION SLAVE ON *.* TO 'repl'@'%';
SHOW MASTER STATUS;
記錄SHOW MASTER STATUS輸出的File和Position值。? 并將值填寫到從庫配置里
從庫(server2)配置
cat > /etc/my.cnf.d/mysql-server.cnf << EOF
[mysqld]
server-id=2
log-bin=mysql-bin
binlog-do-db=zabbix
expire-logs-days=10
max-binlog-size=100M
binlog-format=ROW
relay-log=mysql-relay-bin
read-only=1
innodb_flush_log_at_trx_commit=1
sync_binlog=1
EOF# 重啟MySQL
systemctl restart mysqld# 配置從庫連接主庫(替換FILE和POSITION值)CHANGE MASTER TOMASTER_HOST='10.0.0.12',MASTER_USER='repl',MASTER_PASSWORD='ReplicationPassword',MASTER_LOG_FILE='mysql-bin.000006',MASTER_LOG_POS=1117065;START SLAVE;
SHOW SLAVE STATUS\G;
確保Slave_IO_Running和Slave_SQL_Running均為Yes。
實際挫折1 好幾次出現下面的報錯
解決方法:
主節點操作:
- 修改復制用戶的認證插件為?
mysql_native_password
(兼容性好,MySQL 5.7 及之前常用 ):
ALTER USER 'repl'@'%' IDENTIFIED WITH mysql_native_password BY 'ReplicationPassword'; FLUSH PRIVILEGES;
從節點操作:
重新配置主從連接(無需 SSL ,簡單場景 ):
CHANGE MASTER TO MASTER_HOST='10.0.0.12', MASTER_USER='repl', MASTER_PASSWORD='ReplicationPassword', MASTER_LOG_FILE='mysql-bin.xxxxxx', MASTER_LOG_POS=xxxxxx;
START SLAVE;
SHOW SLAVE STATUS\G; # 查看是否恢復
實際挫折2? 修改完之后又出現下面的報錯
錯誤分析:主從節點都為1
解決方法:
在主節點或者從節點更改一下server_id=xxx 使兩個值不一樣即可(/etc/my.cnf.d/mysql-server.cnf)
實際挫折3 改完又遇到下面的錯誤
查 performance_schema: 登錄 MySQL,查詢 performance_schema.replication_applier_status_by_worker 表,獲取 Worker線程的詳細錯誤
SELECT * FROM performance_schema.replication_applier_status_by_worker\G;
重點看?LAST_ERROR_MESSAGE
?字段,能看到事務執行失敗的具體 SQL 或原因。
解決辦法:
(1)主庫導出數據
mysqldump -u root -p zabbix > zabbix_db.sql
(2)將主庫里面的zabbix 轉到從庫里
scp zabbix_db.sql 從庫用戶@從庫IP:/tmp/
(3)在從庫導入 zabbix 庫:
CREATE DATABASE zabbix;
mysql zabbix < zabbix_db.sql
(4)停止從庫
STOP SLAVE;
非 GTID 模式:
#跳過錯誤事務
CHANGE MASTER TO MASTER_LOG_FILE='mysql - bin.000006', MASTER_LOG_POS=75510; -- 錯誤位置 +1GTID 模式:先查當前 GTID 集合,找到對應事務的 GTID 并跳過(假設 GTID 為 xxx:123 ):SET GLOBAL sql_slave_skip_counter = 1;啟動從庫復制:START SLAVE;
驗證復制狀態:SHOW SLAVE STATUS\G;
實際挫折4? 做完上述的 又遇到新的錯誤
解決方法:
stop slave; reset slave; CHANGE MASTER TO ...(此項又操作一遍) 后再次查看恢復正常
5. 導入 Zabbix 數據庫架構
在主庫(Server1)上執行:? ?下載Zabbix? ?官網安裝參考??
# 添加Zabbix倉庫
rpm -Uvh https://repo.zabbix.com/zabbix/7.0/rocky/9/x86_64/zabbix-release-latest-7.0.el9.noarch.rpm
dnf clean all# 導入Zabbix數據庫架構
zcat /usr/share/zabbix-sql-scripts/mysql/server.sql.gz | mysql --default-character-set=utf8mb4 -uzabbix -p zabbix
6. 安裝 Zabbix Server
在主從庫上分別執行:
# 安裝Zabbix Server、Web前端和Agent
dnf install -y zabbix-server-mysql zabbix-web-mysql zabbix-nginx-conf zabbix-sql-scripts zabbix-agent# 配置Zabbix Server連接數據庫(最好是將原來的文件備份然后重新再建一個)
cat > /etc/zabbix/zabbix_server.conf << EOF
LogFile=/var/log/zabbix/zabbix_server.log
LogFileSize=0
PidFile=/var/run/zabbix/zabbix_server.pid
DBHost=10.0.0.200
DBName=zabbix
DBUser=zabbix
DBPassword=ZabbixPassword
DBPort=3306
SNMPTrapperFile=/var/log/snmptrap/snmptrap.log
Timeout=4
AlertScriptsPath=/usr/lib/zabbix/alertscripts
ExternalScripts=/usr/lib/zabbix/externalscripts
LogSlowQueries=3000
StartPollers=15
StartPollersUnreachable=5
StartTrappers=5
StartPingers=1
StartDiscoverers=1
CacheSize=128M
HistoryCacheSize=64M
TrendCacheSize=64M
ValueCacheSize=256M
EOF# 配置Web前端時區
sed -i 's/;date.timezone =/date.timezone = Asia\/Shanghai/' /etc/php.ini# 啟動服務
systemctl enable --now zabbix-server zabbix-agent nginx php-fpm
7. 配置 HAProxy(數據庫負載均衡)
在 Server1 和 Server2 上分別執行:
# 安裝HAProxy
dnf install -y haproxy# 配置HAProxy
cat > /etc/haproxy/haproxy.cfg << EOF
globallog /dev/log local0log /dev/log local1 noticechroot /var/lib/haproxystats socket /var/lib/haproxy/statsuser haproxygroup haproxydaemondefaultslog globalmode tcpoption tcplogoption dontlognulltimeout connect 5000timeout client 50000timeout server 50000listen mysql-clusterbind 10.0.0.200:3306mode tcpbalance sourceoption mysql-check user haproxy_checkserver mysql-master 10.0.0.12:3306 check weight 100server mysql-slave 10.0.0.15:3306 check weight 50 backuplisten statsbind *:9000mode httpstats enablestats uri /statsstats realm HAProxy\ Statisticsstats auth admin:password
EOF# 創建監控用戶
mysql -u root -p << EOF
CREATE USER 'haproxy_check'@'%' IDENTIFIED BY 'CheckPassword';
GRANT PROCESS ON *.* TO 'haproxy_check'@'%';
FLUSH PRIVILEGES;
EOF# 啟動HAProxy
systemctl enable --now haproxy
遇到的問題 haproxy 重啟失敗
配置文件的問題?
8. 配置 Keepalived 實現雙 VIP
主 Server(10.0.0.12)配置:
# 安裝Keepalived
dnf install -y keepalived# 配置Keepalived
cat > /etc/keepalived/keepalived.conf << EOF
! Configuration File for keepalivedglobal_defs {router_id ZABBIX_MASTER
}# Web VIP (10.0.0.100)
vrrp_instance VI_WEB {state MASTERinterface eth0virtual_router_id 101priority 101advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {10.0.0.100/24}track_script {chk_httpd}
}# DB VIP (10.0.0.200)
vrrp_instance VI_DB {state MASTERinterface eth0virtual_router_id 201priority 101advert_int 1authentication {auth_type PASSauth_pass 2222}virtual_ipaddress {10.0.0.200/24}track_script {chk_haproxy}
}# 監控腳本
vrrp_script chk_httpd {script "systemctl is-active httpd"interval 2weight -20
}vrrp_script chk_haproxy {script "systemctl is-active haproxy"interval 2weight -20
}
EOF# 啟動Keepalived
systemctl enable --now keepalived
備 Server(10.0.0.15)配置:
# 安裝Keepalived
dnf install -y keepalived# 配置Keepalived
cat > /etc/keepalived/keepalived.conf << EOF
! Configuration File for keepalivedglobal_defs {router_id ZABBIX_BACKUP
}# Web VIP (10.0.0.100)
vrrp_instance VI_WEB {state BACKUPinterface eth0virtual_router_id 101priority 100advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {10.0.0.100/24}track_script {chk_httpd}
}# DB VIP (10.0.0.200)
vrrp_instance VI_DB {state BACKUPinterface eth0virtual_router_id 201priority 100advert_int 1authentication {auth_type PASSauth_pass 2222}virtual_ipaddress {10.0.0.200/24}track_script {chk_haproxy}
}# 監控腳本
vrrp_script chk_httpd {script "systemctl is-active httpd"interval 2weight -20
}vrrp_script chk_haproxy {script "systemctl is-active haproxy"interval 2weight -20
}
EOF# 啟動Keepalived
systemctl enable --now keepalived
9. 配置 Nginx 負載均衡(Server3)
# 安裝Nginx
dnf install -y nginx# 配置Nginx代理Zabbix Web
cat > /etc/nginx/conf.d/zabbix.conf << EOF
upstream zabbix_backend {server 10.0.0.100:80 weight=10 max_fails=3 fail_timeout=30s;
}server {listen 80;server_name zabbix.example.com;location / {proxy_pass http://zabbix_backend;proxy_set_header Host $host;proxy_set_header X-Real-IP $remote_addr;proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;proxy_set_header X-Forwarded-Proto $scheme;proxy_connect_timeout 150;proxy_send_timeout 100;proxy_read_timeout 100;proxy_buffers 4 32k;client_max_body_size 8m;client_body_buffer_size 128k;# Zabbix Web優化proxy_http_version 1.1;proxy_set_header Connection "";proxy_cache_bypass $http_upgrade;}
}
EOF# 啟動Nginx
systemctl enable --now nginx
10. 驗證高可用性
訪問 http://10.0.0.18/zabbix 完成 Web 界面初始化配置
驗證 MySQL 主從復制:
bash
mysql -uzabbix -ppassword -h 10.0.0.200 -e "SHOW SLAVE STATUS\G"測試故障轉移:
停止 Server1 的 Keepalived 服務,驗證 VIP 是否自動切換到 Server2
訪問 http://10.0.0.18/zabbix 確認服務正常
恢復 Server1 的 Keepalived 服務,驗證 VIP 是否自動切回
11. 防火墻配置(生產環境)
# Server1和Server2
firewall-cmd --permanent --add-service=http
firewall-cmd --permanent --add-port=10051/tcp
firewall-cmd --permanent --add-port=3306/tcp
firewall-cmd --permanent --add-port=9000/tcp # HAProxy統計頁面
firewall-cmd --permanent --add-protocol=vrrp # Keepalived
firewall-cmd --reload# Server3
firewall-cmd --permanent --add-service=http
firewall-cmd --reload
12. 監控與維護
MySQL 主從狀態:定期檢查復制延遲
HAProxy 狀態:訪問 http://10.0.0.18:9000/stats
Keepalived 狀態:檢查 VIP 是否正常工作
Zabbix 自監控:配置 Zabbix 監控自身組件狀態