helm安装配置mysql exporter
2021-07-07 本文已影响0人
戈羽殇雪
过程1波三折,实际上需要做的并不多
mysql exporter 使用helm安装,只需要将mysql的设定传入到chart下的values.yaml中
具体步骤如下:
#查看当前的helm repo
[root@master ~]# helm repo list
NAME URL
stable https://charts.helm.sh/stable
拉取相关的chart,不直接install的原因是将mysql 连接需要的值传入到values.yaml中
helm pull prometheus-mysql-exporter stable/prometheus-mysql-exporter
cd $HOME/.cache/helm/repository/ #helm 文件下载的目录
tar zxvf prometheus-mysql-exporter-0.7.1.tgz #解压mysql exporter 的文件
cd prometheus-mysql-exporter/
#修改values.yaml的值,主要是将连接数据库的用户名,host信息等传入进去,同时在建立容器的同时建立servicemonitor
...
serviceMonitor:
enabled: true
...
mysql:
db: ""
host: "你的ip"
param: ""
pass: "你的密码"
port: 3306
protocol: ""
user: "exporter"
existingSecret: false
...
当然exporter在数据库中首先定义好相关的权限,以及密码,访问限制等
CREATE USER 'exporter'@'localhost' IDENTIFIED BY 'XXXXXXXX' WITH MAX_USER_CONNECTIONS 3;
GRANT PROCESS, REPLICATION CLIENT, SELECT ON *.* TO 'exporter'@'localhost';
操作完以上之后,需要重新打包helm文件
rm -f prometheus-mysql-exporter-0.7.1.tgz
tar zcvf prometheus-mysql-exporter-0.7.1.tgz prometheus-mysql-exporter
#之后执行helm install
helm install mysql prometheus-mysql-exporter-0.7.1.tgz --namespace monitoring
#观察容器是否正常启动
kubectl -n monitoring get pods
NAME READY STATUS RESTARTS AGE
alertmanager-prometheus-prometheus-oper-alertmanager-0 2/2 Running 0 69d
mysql-prometheus-mysql-exporter-bfc59c7b7-dnhhm 1/1 Running 0 4h54m
prometheus-grafana-5558455c66-52dsg 2/2 Running 0 69d
prometheus-kube-state-metrics-6b46f67bf6-h5nhs 1/1 Running 0 46d
prometheus-prometheus-node-exporter-2jrvp 1/1 Running 0 69d
prometheus-prometheus-node-exporter-ct96h 1/1 Running 0 69d
prometheus-prometheus-node-exporter-lbsnm 1/1 Running 0 69d
prometheus-prometheus-node-exporter-m6jd6 1/1 Running 0 14d
prometheus-prometheus-node-exporter-nlfwt 1/1 Running 0 69d
prometheus-prometheus-oper-operator-7c75ff5864-g2lkq 2/2 Running 0 23h
prometheus-prometheus-prometheus-oper-prometheus-0 3/3 Running 2 45h
#测试与数据库的连通性
[root@master ~]# kubectl -n monitoring get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
alertmanager-operated ClusterIP None <none> 9093/TCP,9094/TCP,9094/UDP 69d
mysql-prometheus-mysql-exporter ClusterIP 10.105.203.102 <none> 9104/TCP 4h55m
prometheus-grafana NodePort 10.104.181.79 <none> 80:32669/TCP 69d
prometheus-kube-state-metrics ClusterIP 10.102.76.12 <none> 8080/TCP 69d
prometheus-operated ClusterIP None <none> 9090/TCP 69d
prometheus-prometheus-node-exporter ClusterIP 10.103.134.197 <none> 9100/TCP 69d
prometheus-prometheus-oper-alertmanager ClusterIP 10.103.19.123 <none> 9093/TCP 69d
prometheus-prometheus-oper-operator ClusterIP 10.97.236.198 <none> 8080/TCP,443/TCP 69d
prometheus-prometheus-oper-prometheus NodePort 10.97.65.121 <none> 9090:30658/TCP 69d
curl 10.105.203.102:9104/metrics |grep mysql_up
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 347k 100 347k 0 0 2079k 0 --:--:-- --:--:-- --:--:-- 2081k
# HELP mysql_up Whether the MySQL server is up.
# TYPE mysql_up gauge
mysql_up 1
mysql_up是1 证明与数据库连接正常,接下来要看与prometheus 是否连接正常
#确认service monitor已创建
[root@master ~]# kubectl -n monitoring get servicemonitors.monitoring.coreos.com
NAME AGE
mysql-prometheus-mysql-exporter 5h4m
浏览器确认状态:
image.png
上图表明,prometheus已经成功加入了mysql exporter。
在这个过程中,我开始创建的mysql exporter 并未在网页端targets选项中找到,最后发现问题的所在是要
image.png
这里默认生成的release 是mysql ,但是需要指定为prometheus,修改完之后,需要等一段时间重新查看就能看到了