[TOC
描述: Grafana是一个用Javascript写的开源的(Dashboard)可视化面板,能齐全的度量仪表盘和图形编辑器和漂亮的布局展示,并且支持Graphite、elasticsearch、zabbix等的数据可视化的实现,可以给你的数据换个皮肤,使你的数据展示更加直观和漂亮。
OpenTSDB、PostgreSQL
,我们的Prometheus就是采用TSDB时序数据库。Grafana官网链接,内含各种官方文档(包括仪表盘设置等): http://docs.grafana.org/installation/rpm/
项目地址: https://github.com/grafana/grafana/ 官方地址: https://grafana.com
安装环境: CentOS Linux release 7.6.1810 (Core) 3.10.0-957.10.1.el7.x86_64
安装实践
1.yum方式安装grafana最新版本6.1.6
wget https://dl.grafana.com/oss/release/grafana-6.1.6-1.x86_64.rpm
sudo yum localinstall grafana-6.1.6-1.x86_64.rpm
#安装好后守护进程重启
$ systemctl daemon-reload
$ systemctl start grafana-server
$ systemctl enable grafana-server
#初始化插件安装
$ yum -y install fontconfig freetype* urw-fonts #服务器端图像是一项可选功能
$ systemctl start grafana-server
2.docker方式安装,注意默认安装最新版本
docker run -d --name=grafana -p 3000:3000 grafana/grafana
tee docker-compose.yml <<'EOF'
version: '3.2'
services:
grafana:
image: grafana/grafana:7.5.5
container_name: grafana
user: "472"
environment:
- TZ=Asia/Shanghai
- GF_SECURITY_ADMIN_PASSWORD=weiyigeek
volumes:
- /nfsdisk-31/monitor/grafana/data:/var/lib/grafana
- /etc/localtime:/etc/localtime
ports:
- '3000:3000'
restart: always
dns:
- 223.6.6.6
- 192.168.12.254
EOF
3.二进制方式安装Grafana,安装完成后访问主机的3000端口即可,
wget https://s3-us-west-2.amazonaws.com/grafana-releases/release/grafana-4.3.2.linux-x64.tar.gz
tar -zxvf grafana-4.3.2.linux-x64.tar.gz
mv grafana-4.3.2 grafana
cd grafana/bin/
#查看可以安装的插件
./grafana-cli plugins list-remote
#安装zabbix插件
./grafana-cli plugins install alexanderzobnin-zabbix-app
#安装饼图插件
./grafana-cli plugins install grafana-piechart-panel
#启动,并后台运行
./grafana-server -homepath /opt/grafana/grafana &
温馨提示: 第一次登陆默认密码为admin/admin
WeiyiGeek.grafana登陆
资源清单方式:
mkdir /app/grafana && chown -R 472:0 /app/grafana
tee grafana.yaml <<'EOF'
apiVersion: apps/v1
kind: Deployment
metadata:
labels:
app: grafana
name: grafana
namespace: default
spec:
selector:
matchLabels:
app: grafana
template:
metadata:
labels:
app: grafana
spec:
securityContext:
fsGroup: 472
supplementalGroups: 0
containers:
- name: grafana
image: grafana/grafana:8.5.4
imagePullPolicy: IfNotPresent
ports:
- containerPort: 3000
name: http-grafana
protocol: TCP
readinessProbe:
failureThreshold: 3
httpGet:
path: /robots.txt
port: 3000
scheme: HTTP
initialDelaySeconds: 10
periodSeconds: 30
successThreshold: 1
timeoutSeconds: 2
livenessProbe:
failureThreshold: 3
initialDelaySeconds: 30
periodSeconds: 10
successThreshold: 1
tcpSocket:
port: 3000
timeoutSeconds: 1
resources:
requests:
cpu: 250m
memory: 750Mi
volumeMounts:
- mountPath: /var/lib/grafana
name: grafana-pv
volumes:
- name: grafana-pv
hostPath:
path: /app/grafana
type: DirectoryOrCreate
---
apiVersion: v1
kind: Service
metadata:
name: grafana
namespace: default
spec:
ports:
- port: 3000
protocol: TCP
targetPort: http-grafana
nodePort: 30002
selector:
app: grafana
sessionAffinity: None
type: NodePort
EOF
kubectl apply -f grafana.yaml
kubectl get pod
# NAME READY STATUS RESTARTS AGE
# grafana-f6799b947-rl47d 1/1 Running 0 45m
1) 资源清单部署方式
步骤 01.资源清单(granana.ini)与 (Deployment)部署资源清单。
tee grafana.ini <<'EOF'
[server]
domain = devops.weiyigeek.top
root_url = %(protocol)s://%(domain)s:%(http_port)s/grafana/
serve_from_sub_path = true
EOF
tee grafana-deploy.yaml <<'EOF'
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: grafana-pvc
namespace: dashboard
spec:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 2Gi
---
apiVersion: apps/v1
kind: Deployment
metadata:
labels:
app: grafana
name: grafana
namespace: dashboard
spec:
selector:
matchLabels:
app: grafana
template:
metadata:
labels:
app: grafana
spec:
securityContext:
fsGroup: 472
supplementalGroups:
- 0
containers:
- name: grafana
image: grafana/grafana:9.0.2
imagePullPolicy: IfNotPresent
ports:
- containerPort: 3000
name: http-grafana
protocol: TCP
readinessProbe:
failureThreshold: 3
httpGet:
path: /robots.txt
port: 3000
scheme: HTTP
initialDelaySeconds: 10
periodSeconds: 30
successThreshold: 1
timeoutSeconds: 2
livenessProbe:
failureThreshold: 3
initialDelaySeconds: 30
periodSeconds: 10
successThreshold: 1
tcpSocket:
port: 3000
timeoutSeconds: 1
resources:
requests:
cpu: 512m
memory: 1024Mi
volumeMounts:
- mountPath: /var/lib/grafana
name: grafana-pv
- mountPath: /etc/grafana/grafana.ini
name: ge-config
subPath: grafana.ini
volumes:
- name: grafana-pv
persistentVolumeClaim:
claimName: grafana-pvc
- name: ge-config
configMap:
name: ge-config
items:
- key: grafana.ini
path: grafana.ini
---
apiVersion: v1
kind: Service
metadata:
name: grafana
namespace: dashboard
spec:
ports:
- port: 3000
protocol: TCP
targetPort: http-grafana
selector:
app: grafana
sessionAffinity: None
type: LoadBalancer
EOF
步骤 02.创建 grafana.ini 的 configmap 以及部署 grafana 到集群中
kubectl create configmap ge-config --from-file=grafana.ini --namespace dashboard
kubectl create -f grafana-deploy.yaml --namespace dashboard
步骤 03.部署情况与日志情况查看。
kubectl logs -f --tail 50 -n dashboard -l app=grafana
# 关键日志:
# logger=http.server t=2022-07-14T07:24:25.409774065Z level=info msg="HTTP Server Listen" address=[::]:3000 protocol=http subUrl=/grafana socket=
步骤 04.配置 grafana ingress 的资源清单我们便可通过域名+grafana目录
的方式进行访问。
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
annotations:
ingressclass.kubernetes.io/is-default-class: "true"
nginx.ingress.kubernetes.io/client-body-buffer-size: 50m
nginx.ingress.kubernetes.io/proxy-body-size: 50m
nginx.ingress.kubernetes.io/proxy-buffer-size: 50m
nginx.ingress.kubernetes.io/proxy-buffering: "on"
nginx.ingress.kubernetes.io/proxy-buffers-number: "4"
nginx.ingress.kubernetes.io/proxy-connect-timeout: 60s
nginx.ingress.kubernetes.io/proxy-read-timeout: 120s
nginx.ingress.kubernetes.io/proxy-send-timeout: 120s
nginx.ingress.kubernetes.io/rewrite-target: /$2
labels:
app: devops
ref: devops
url: devops.weiyigeek.top
name: devops-weiyigeek
namespace: dashboard
spec:
ingressClassName: nginx
rules:
- host: devops.weiyigeek.top
http:
paths:
- backend:
service:
name: grafana
port:
number: 3000
path: /grafana(/|$)(.*)
pathType: ImplementationSpecific
tls:
- hosts:
- devops.weiyigeek.top
secretName: weiyigeek.top
温馨提示: 在进行前面操作完毕后我们便可通过https://devops.weiyigeek.top/grafana/login
域名地址进行访问,默认密码 admin。
helm3 方式安装
helm repo add grafana https://grafana.github.io/helm-charts
helm repo update
helm search repo grafana
helm show values grafana/grafana --version 6.29.5 > granfa-values.yaml # 指定chat版本的granfa相关部署参数
helm install grafana grafana/grafana --version 6.29.5 --set nodeSelector="kubernetes\.io/hostname=node-223",service.type=NodePort,persistence.enabled=True,persistence.storageClassName="managed-nfs-storage",persistence.size=5Gi,persistence.accessModes=ReadWriteOnce -n devops --debug # 安装指定chat版本以及部署参数
helm upgrade grafana grafana/grafana --version 6.29.5 --set nodeSelector="kubernetes\.io/hostname=node-223",service.type=NodePort,persistence.enabled=True,persistence.storageClassName="managed-nfs-storage",persistence.size=5Gi,persistence.accessModes=ReadWriteOnce -n devops --debug # 更新指定chat版本以及部署参数
helm list -n devops
# NAME NAMESPACE REVISION UPDATED STATUS CHART APP VERSION
# grafana devops 1 2022-06-01 13:53:51.751609203 +0800 CST deployed grafana-6.29.5 8.5.3
执行结果:
NOTES:
1. Get your 'admin' user password by running:
kubectl get secret --namespace devops grafana -o jsonpath="{.data.admin-password}" | base64 --decode ; echo
# 例如: Y0GW80b6tsMhqO223zRfh0ONx2O8BovtYg3JAKHa
2. The Grafana server can be accessed via port 80 on the following DNS name from within your cluster: grafana.devops.svc.cluster.local
Get the Grafana URL to visit by running these commands in the same shell:
export POD_NAME=$(kubectl get pods --namespace devops -l "app.kubernetes.io/name=grafana,app.kubernetes.io/instance=grafana" -o jsonpath="{.items[0].metadata.name}")
kubectl --namespace devops port-forward $POD_NAME --address 0.0.0.0 30002:3000
3. Login with the password from step 1 and the username: admin
描述: 例如此次添加一个Prometheus数据源到Grafana,注意最后一定要点击Save&Test
来保证Grafana可以正常访问我们的Prometheus。
操作步骤: 侧边栏 -> Congiguration -> Data Sources
WeiyiGeek.添加数据源
WeiyiGeek.Prometheus监控数据源
描述: 我们可以使用模板变量来切换到详细监控主机的相关信息,这个功能非常Nice也非常实用。
node_uname_info
-> 正则表达式来获取标签中的内核版本 -> .*release="(.*?)".*
-> 点击update,我们可以仿照添加一个节点名称的变量以及网卡名称node_network_receive_bytes_total
-> .*device="(docker.*?|eno.*?|ens.*?)".*
。
WeiyiGeek.模板变量
rate(node_network_receive_bytes_total{device="$Device"}[5m])
-> Legend 设置为 Title 设置 Bytes/Received ,其中数据速率下的Unit 设置为bytes/sec,最后点击保存。WeiyiGeek.使用模板变量
描述: 如其名称其主要用于显示图表。
Panel Title -> 点击 Edit (快捷键e)
-> 在metrics中输入查询表达式node_memory_Active_bytes
.
WeiyiGeek.简单图表
描述: 显示单个时间序列的值。
描述: 它可以选择时间范围和刷新频率,它适用于整个仪表盘但你也可以在每个Panel上覆盖配置。
Tips: 当图表刷新时即使基础数据没有改变,但你也会注意到形状可能发生变化,这个情况被称为混叠的信号处理效果。
描述: 在 Grafana 中提供了多种告警方式,其中最常用的就是邮箱告警, 下面我们来简单实践在Grafana中实现邮件告警。
步骤 01.在 grafana 的主配置文件 grafana.ini 中开启 smtp 并配置相关参数(此处我采用了企业邮箱):
$ cat grafana.ini
....
[smtp]
enabled = true
host = smtp.exmail.qq.com:465
user = monitor@weiyigeek.top
password = weiyigeek.top
from_address = monitor@weiyigeek.top
from_name = Grafana
ehlo_identity = devops.weiyigeek.top
步骤 02.修改 grafana.ini 后便可重启 grafana 登陆其Dashboard ,点左菜单栏中🔔 (Alerting) -> Contact points (联络人) -> 编辑 grafana-default-email
告警对象 -> 选择触发对象为email
-> 输入测试邮箱 -> 测试发信
WeiyiGeek.测试Email发信
步骤 03.发信后你将会在其输入指定邮箱中,看到 Grafana 发送到你邮箱的告警信息。
WeiyiGeek.Email告警配置验证
在 grafana 的主配置文件 grafana.ini 中开启 LDAP 认证:
[auth.ldap]
enabled = true
config_file = /etc/grafana/ldap.toml
config_file
指定grafana的ldap配置文件ldap.toml的具体位置 ldap.toml 配置文件的内容如下[[servers]]
host = "127.0.0.1"
port = 389
use_ssl = false
start_tls = false
ssl_skip_verify = false
bind_dn = "cn=Manager,dc=frognew,dc=com"
bind_password = 'thepassword'
search_filter = "(cn=%s)"
search_base_dns = ["ou=People,dc=frognew,dc=com"]
group_search_base_dns = ["ou=Group,dc=frognew,dc=com"]
group_search_filter = "(&(objectClass=posixGroup)(memberUid=%s))"
[servers.attributes]
name = "givenName"
surname = "sn"
username = "cn"
member_of = "cn"
email = "email"
[[servers.group_mappings]]
group_dn = "grafana-admin"
org_role = "Admin"
[[servers.group_mappings]]
group_dn = "grafana-editor"
org_role = "Editor"
[[servers.group_mappings]]
group_dn = "grafana-viewer"
org_role = "Viewer"
host
和port
指定LDAP Server的地址和端口group_search_base_dns
和group_search_filter
,同事需要设置member_of = "cn"
org_role
的对应关系,注意这里需要和配置成LDAP中Group的cn,这里配置的是grafana-admin
、grafana-editor
、grafana-viewer
,这点和官方文档Grafana LDAP Authentication中给的例子是不一样的。常用命令:
# 查看可以安装的插件
grafana-cli plugins list-remote
grafana-cli plugins install grafana-piechart-panel alexanderzobnin-zabbix-app #安装zabbix插件和饼图插件
错误信息:
docker logs --tail grafana
# grafana | GF_PATHS_DATA='/var/lib/grafana' is not writable.
# grafana | You may have issues with file permissions, more information here: http://docs.grafana.org/installation/docker/#migrate-to-v51-or-later
# grafana | mkdir: can't create directory '/var/lib/grafana/plugins': I/O error
# grafana | GF_PATHS_DATA='/var/lib/grafana' is not writable.
问题原因: 从以前的Docker容器版本迁移问题针对/var/lib/grafana
目录权限有所变化,所以在我们持久化该目录时应设置对应的uid、gid。
解决办法: https://grafana.com/docs/grafana/latest/installation/docker/#migrate-to-v51-or-later
# 用户ID更改与Grafana版本对应
Version User User ID Group Group ID
< 5.1 grafana 104 grafana 107
>= 5.1 grafana 472 grafana 472
>= 7.3 grafana 472 root 0
解决流程
# 1.映射的持久化目录的权限
chown -R 472:0 /data/grafana
# 2.以其他用户身份运行Docker
docker run --user 472 --volume "/data/grafana:/var/lib/grafana" grafana/grafana
错误信息:
$ kubectl get pod -n devops
NAME READY STATUS RESTARTS AGE
grafana-6fb4d77ddc-s7xl5 0/1 Running 4 (18s ago) 11m
# 不断执行 Starting DB migrations
$ kubectl logs -f --tail 50 -n devops grafana-6fb4d77ddc-s7xl5
logger=migrator t=2022-06-01T03:11:30.93+0000 lvl=info msg="Starting DB migrations"
logger=migrator t=2022-06-01T03:11:35.85+0000 lvl=info msg="Executing migration" id="delete stars for deleted dashboards"
logger=migrator t=2022-06-01T03:11:42+0000 lvl=info msg="Executing migration" id="Add index for dashboard_is_folder"
logger=migrator t=2022-06-01T03:11:46.9+0000 lvl=info msg="Executing migration" id="create data_source table"
logger=migrator t=2022-06-01T03:11:52.87+0000 lvl=info msg="Executing migration" id="add index data_source.account_id"
logger=migrator t=2022-06-01T03:11:58.46+0000 lvl=info msg="Executing migration" id="add unique index data_source.account_id_name"
问题可能原因:
1.提高限制内存减少触发OOM, 即 resource 字段设置
resources: {}
。 2.持久化数据persistence卷性能问题导致写过慢。
问题错误: 在 docker 运行 grafana:7.0.3 版本时,grafana不断重启日志如下
$ docker run -d -p 3000:3000 -e TZ="Asia/Shanghai" --link=prometheus -v /data/grafana:/var/lib/grafana -v /data/grafana.ini:/etc/grafana/grafana.ini --name=grafana grafana/grafana:7.0.3
$ docker logs -f --tail 50 grafana
t=2020-06-04T11:26:22+0800 lvl=info msg="HTTP Server Listen" logger=http.server address=[::]:3000 protocol=http subUrl= socket=
t=2020-06-04T11:27:02+0800 lvl=info msg="Shutdown started" logger=server reason="System signal: terminated"
t=2020-06-04T11:27:02+0800 lvl=info msg="Stopped Stream Manager"
解决办法: 尝试docker修改各种命令启动后发现,当命令中 –name=grafana 时才会出现此问题,修改成其他name即可正常启动