安装插件
配置数据源
导入模版
查看

1.安装

wget https://mirrors.tuna.tsinghua.edu.cn/grafana/yum/rpm/Packages/grafana-10.0.0-1.x86_64.rpm
[root@rocky8 apps]# yum install grafana-10.0.0-1.x86_64.rpm

[root@rocky8 apps]# systemctl start grafana-server.service
修改为中文
[root@rocky8 apps]# vi /etc/grafana/grafana.ini
default_language = zh-Hans
[root@rocky8 apps]# systemctl restart grafana-server.service
访问
http://192.168.1.250:3000/

列出插件版本
[root@rocky8 apps]# grafana-cli plugins list-versions alexanderzobnin-zabbix-app
4.4.1
4.4.0

安装插件
[root@rocky8 apps]# grafana-cli plugins install alexanderzobnin-zabbix-app 4.4.1
重启
systemctl restart grafana-server.service
安装插件的目录
ll /var/lib/grafana/plugins/alexanderzobnin-zabbix-app/

开启zabbix插件

添加数据源
在granfana上添加zabbix的解析 192.168.1.150 zabbix.luo.com

http://zabbix.luo.com/api_jsonrpc.php #zabbix是通过nginx安装的不用加zabbix

填写zabbix用户名密码，测试成功
Admin
zabbix

2.导入模版

导入官网的其它 Grafana 模板
https://grafana.com/grafana/dashboards

模板ID 5363

自定义面板

自定义cpu时间

选择仪表盘保存的文件夹

自定义重cpu时间

自定义内存

选择zabbix server ，主机等，选择饼图

命名标题

标签选择

显示百分比和值

自定义变量

组

host

保存

编辑仪表板，group改为$group,host改为$host保存，就可以选择组和主机查看

查看

导出

优化zabbix缓存
CacheSize=128M#调大

修改zabbix进程导致zabbix一直重启
StartDBSyncers=400 #启动多少个线程和数据库连接

StartPollersUnreachable=1000 #用于设置当Zabbix服务器无法连接到被监控主机时，启动的不可达性轮询的数量。

进程加大,可能会导致数据库连接过多,超过最大值,导致Zabbix 无法启动,解决方案,加大数据库的连接数
#注意:如果无法启动,可能是数据库连接数不足
[root@zabbix-server ~]#tail /var/log/zabbix/zabbix_server.log

1367396:20230906:222534.347 [Z3001] connection to database ‘zabbix’ failed: [1040] Too many connections
1367396:20230906:222534.347 Cannot connect to the database. Exiting…
1367399:20230906:222534.350 server #189 started [trapper #4]
1367068:20230906:222534.353 HA manager has been paused
1367400:20230906:222534.354 server #190 started [trapper #5]
1367398:20230906:222534.358 [Z3001] connection to database ‘zabbix’ failed: [1040] Too many connections

查看数据库连接
mysql> show status like ‘Threads%’;

root@server02:~# vi /etc/mysql/mysql.conf.d/mysqld.cnf
max_connections = 1000

ysql> select @@max_connections;
±------------------+
| @@max_connections |
±------------------+
| 1000 |
±------------------+
1 row in set (0.00 sec)

修改数据库参数后数据库变为只读

1372996:20230906:223235.053 [Z3005] query failed: [1290] The MySQL server is running with the --read-only option so it cannot execute this statement [select ha_nodeid,name,status,lastaccess,address,port,ha_sessionid from ha_node order by ha_nodeid for update]
mysql> show global variables like '%read_only%';
+-----------------------+-------+
| Variable_name         | Value |
+-----------------------+-------+
| innodb_read_only      | OFF   |
| read_only             | ON    |
| super_read_only       | OFF   |
| transaction_read_only | OFF   |
+-----------------------+-------+
4 rows in set (0.00 sec)

mysql>  set global read_only = 0;
修改mysql配置文件添加
read_only = 0




打开的文件数太多，优化参考https://www.programminghunter.com/article/88832382383/ 
zabbix_server [1374128]: failed to open log file: [24] Too many open files
zabbix_server [1374128]: failed to write [cannot accept incoming IPC connection: [24] Too many open files] into log file
1374143:20230906:223937.970 executing housekeeper

zabbix不能重启 pkill  -9 zabbix_server

root@server02:~# lsof  -u zabbix|wc -l 
90537

 cat vim /etc/security/limits.conf 
zabbix soft nofile 1024000
zabbix hard nofile 1024000
zabbix soft soft  1024000
zabbix hard soft  1024000

Zabbix出现 0ut Of Memory，将原本2G内存加到8G还是Out 0f Memory
缓存太少

zabbix配置文件

StartIPMIPollers=0 #初始化启动时，启动IPMI子进程轮询器的数量，数量越多，则服务端吞吐能力越强，对系统资源消耗越大 , 主要用于IPMI技术用于获取硬件状态场景。
StartDiscoverers=1 #设置用于自动发现主机的线程数量，若单台代理所管理机器超过500台，可以考虑加大此数值（仅适用于直接AGENT场景）！！！！ 1.8.5以前最大数值为255