云原生 | K8S集群helm部署loki

环境配置：K8s集群1.21.1，已安装nfs持久化

 1.安装helm,已经装可忽略

tar -zxvf helm-v3.10.2-linux-amd64.tar.gz
mv linux-amd64/helm /usr/local/bin/helm

 查看版本

[root@k8s-master-1 ~]# helm version
version.BuildInfo{Version:"v3.10.2", GitCommit:"50f003e5ee8704ec937a756c646870227d7c8b58", GitTreeState:"clean", GoVersion:"go1.18.8"}

 2.添加loki源

#添加loki源
helm repo add loki https://grafana.github.io/loki/charts

#查看loki相关包
helm search repo loki 

 3.安装loki

需要先创建ns

#创建lokins
kubectl create ns monitoring

#安装loki
Helm install loki  loki/loki-stack   -n monitoring

 查看安装情况

kubectl get pod  -n monitoring

 4.创建loki的pvc做持久化

vim loki-pvc.yaml

kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: k8s-loki-claim
  namespace: monitoring  #---指定namespace为monitoring
spec:
  storageClassName: nfs-storage #---需要与上面创建的storageclass的名称一致
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 200Gi

生成pvc

kubectl apply -f loki-pvc.yaml

查看pvc绑定情况

kubectl get pvc  -n monitoring

5.创建grafana做可视化看板

#下载grafana
https://github.com/prometheus-operator/kube-prometheus#

 进入grafana目录，创建pvc

vim grafana-pvc.yaml

kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: k8s-grafana-claim
  namespace: monitoring  #---指定namespace为monitoring
spec:
  storageClassName: nfs-storage #---需要与上面创建的storageclass的名称一致
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 100Gi

进去到manifests目录,执行grafana目录

[root@master-tg-1 manifests]# kubectl apply -f grafana/

 查看生成pod情况

kubectl get pod -n monitoring

 查看pvc和svc情况

kubectl get pvc,svc -n monitoring

 6.访问grafana，添加loki做数据源

ip+端口

10.103.11.8:30090

初始账号密码：admin admin 

 添加数据源

 添加域名+3100端口即可，不用使用cluster ip

http://loki:3100

 进行校验，通过即可

 Explore按钮查看日志

 点击Log browser展开相关日志目录

 选中日志master-tg-1，点击Show logs即可查看master-tg-1日志

 相关日志信息

 7.卸载loki

#卸载loki
helm uninstall loki -n monitoring