Prometheus监控系统Docker容器化部署实战

2,733

1 部署 Docker 服务

curl https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo -o /etc/yum.repos.d/docker.repo
yum list docker-ce --showduplicates | sort -r    # 显示所有版本
yum install -y docker-ce-20.10.5    # 指定 docker 版本
systemctl start docker    # 启动 docker 服务
systemctl status docker    # 查看 docker 服务状态
systemctl enable docker    # 设置 docker 开机自启动

2 部署 Prometheus 服务

创建 mon 用户,创建目录:

groupadd -g 2000 mon
useradd -u 2000 -g mon mon
mkdir -p /home/mon/prometheus/{etc,data,rules}

创建配置文件:

vim /home/mon/prometheus/etc/prometheus.yml
# my global config
global:
  scrape_interval:     15s # Set the scrape interval to every 15 seconds. Default is every 1 minute.
  evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is every 1 minute.
  # scrape_timeout is set to the global default (10s).

# Alertmanager configuration
alerting:
  alertmanagers:
  - static_configs:
    - targets:
      # - alertmanager:9093

# Load rules once and periodically evaluate them according to the global 'evaluation_interval'.
rule_files:
  # - "first_rules.yml"
  # - "second_rules.yml"

# A scrape configuration containing exactly one endpoint to scrape:
# Here it's Prometheus itself.
scrape_configs:
  # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
  - job_name: 'prometheus'

    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.

    static_configs:
    - targets: ['localhost:9090']

启动容器服务:

docker pull prom/prometheus
cd /home/mon/
chown mon. -R prometheus
docker run -d --user root -p 9090:9090 --name prometheus \
    -v /home/mon/prometheus/etc/prometheus.yml:/etc/prometheus/prometheus.yml \
    -v /home/mon/prometheus/rules:/etc/prometheus/rules \
    -v /home/mon/prometheus/data:/data/prometheus \
    prom/prometheus \
    --config.file="/etc/prometheus/prometheus.yml" \
    --storage.tsdb.path="/data/prometheus" \
    --web.listen-address="0.0.0.0:9090"

3 部署 Grafana 服务

创建数据目录:

mkdir -p /home/mon/grafana/plugins

安装插件:下载Grafana插件

tar zxf /tmp/grafana-plugins.tar.gz -C /home/mon/grafana/plugins/
chown -R mon. /home/mon/grafana
chmod 777 -R /home/mon/grafana

启动容器服务:

docker pull grafana/grafana:latest
docker run -d -p 3000:3000 -v /home/mon/grafana:/var/lib/grafana --name=grafana grafana/grafana:latest

4 配置 Grafana 对接 Prometheus

访问 http://ip:3000,初始账号密码为 admin/admin,会要求更改密码。

按照如下截图顺序配置 Prometheus Dashboard:

5 部署 Node_Exporter 服务

我以监控一台阿里云ECS为例。

安装配置Node_Exporter:

curl https://github.com/prometheus/node_exporter/releases/download/v1.1.1/node_exporter-1.1.1.linux-amd64.tar.gz > /opt/node_exporter-1.1.1.linux-amd64.tar.gz
cd /opt
tar zxf node_exporter-1.1.1.linux-amd64.tar.gz
mv node_exporter-1.1.1.linux-amd64 node_exporter

配置服务启动脚本:

vim /usr/lib/systemd/system/node_exporter.service

[Unit]
Description=node_exporter service
 
[Service]
User=root
ExecStart=/opt/node_exporter/node_exporter
 
TimeoutStopSec=10
Restart=on-failure
RestartSec=5
 
[Install]
WantedBy=multi-user.target
systemctl daemon-reload
systemctl start node_exporter
systemctl status node_exporter
systemctl enable node_exporter

在这台ECS服务器的Nginx下配置反向代理:

vim /usr/local/nginx/conf/conf.d/www.conf

    # prometheus monitor node exporter
    location /node/exporter {
        proxy_pass   http://127.0.0.1:9100/metrics;
        proxy_set_header Host $http_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;
    }
/usr/local/nginx/sbin/nginx -s reload

在Prometheus服务器端修改 配置文件:

vim /home/mon/prometheus/etc/prometheus.yml
        
  - job_name: 'node'
    static_configs:
    - targets: ['www.test.com']
      labels:
        instance: node
    scheme: https
    metrics_path: /node/exporter

重启 Prometheus 容器:

docker restart prometheus

验证是否获取到数据,在浏览器输入:https://www.test.com/node/exporter:

# HELP go_gc_duration_seconds A summary of the pause duration of garbage collection cycles.
# TYPE go_gc_duration_seconds summary
go_gc_duration_seconds{quantile="0"} 0
go_gc_duration_seconds{quantile="0.25"} 0
go_gc_duration_seconds{quantile="0.5"} 0
go_gc_duration_seconds{quantile="0.75"} 0
go_gc_duration_seconds{quantile="1"} 0
go_gc_duration_seconds_sum 0
go_gc_duration_seconds_count 0
# HELP go_goroutines Number of goroutines that currently exist.
# TYPE go_goroutines gauge
go_goroutines 7
# HELP go_info Information about the Go environment.
# TYPE go_info gauge
go_info{version="go1.15.8"} 1
# HELP go_memstats_alloc_bytes Number of bytes allocated and still in use.
# TYPE go_memstats_alloc_bytes gauge

······

创建Dashboard:

6 部署 Alertmanager 服务

创建目录:

mkdir -p /home/mon/alertmanager
chmod 777 -R /home/mon/alertmanager

创建配置文件:

vim  /home/mon/alertmanager/alertmanager.yml

global:
  resolve_timeout: 5m
  smtp_from: '319981932@qq.com'
  smtp_smarthost: 'smtp.qq.com:465'
  smtp_auth_username: '319981932@qq.com'
  # 注意这里需要配置QQ邮箱的授权码,不是登录密码,授权码在账户配置中查看
  smtp_auth_password: 'abcdefghijklmnop'
  smtp_require_tls: false

route:
  group_by: ['alert_node']
  group_wait: 5s
  group_interval: 5s
  repeat_interval: 5m
  receiver: 'email'

receivers:
- name: 'email'
  email_configs:
  - to: '319981932@qq.com'
    send_resolved: true

inhibit_rules:
  - source_match:
      severity: 'critical'
    target_match:
      severity: 'warning'
    equal: ['alert_node', 'dev', 'instance']

拉取镜像并启动容器:

docker pull prom/alertmanager:latest
chown -R mon. alertmanager/
docker run -d --user root -p 9093:9093 --name alertmanager \
    -v /home/mon/alertmanager/alertmanager.yml:/etc/alertmanager/alertmanager.yml \
    prom/alertmanager:latest \
    --config.file="/etc/alertmanager/alertmanager.yml" \
    --web.listen-address="0.0.0.0:9093"

查看alertmanager容器IP地址,用于配置prometheus对接接口:

docker exec -it alertmanager /bin/sh -c "ip a"

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
10: eth0@if11: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1500 qdisc noqueue
    link/ether 02:42:ac:11:00:04 brd ff:ff:ff:ff:ff:ff
    inet 172.17.0.4/16 brd 172.17.255.255 scope global eth0
       valid_lft forever preferred_lft forever

修改prometheus配置文件对接alertmanager:

vim /home/mon/prometheus/etc/prometheus.yml

alerting:
  alertmanagers:
  - static_configs:
    - targets:
      - 172.17.0.4:9093

rule_files:
  - "/etc/prometheus/rules/*rules.yml"

配置告警规则:

vim /home/mon/prometheus/rules/alert-node-rules.yml
groups:
  - name: alert-node
    rules:
    - alert: NodeDown
      # 注意:这里的job_name一定要跟prometheus配置文件中配置的相匹配
      expr: up{job="node-service"} == 0
      for: 1m
      labels:
        severity: critical
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} is down"
        description: "Instance: {{ $labels.instance }} 已经宕机 1分钟"
        value: "{{ $value }}"

    - alert: NodeCpuHigh
      expr: (1 - avg by (instance) (irate(node_cpu_seconds_total{job="node-service",mode="idle"}[5m]))) * 100 > 80
      for: 5m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} cpu使用率过高"
        description: "CPU 使用率超过 80%"
        value: "{{ $value }}"

    - alert: NodeCpuIowaitHigh
      expr: avg by (instance) (irate(node_cpu_seconds_total{job="node-service",mode="iowait"}[5m])) * 100 > 50
      for: 5m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} cpu iowait 使用率过高"
        description: "CPU iowait 使用率超过 50%"
        value: "{{ $value }}"

    - alert: NodeLoad5High
      expr: node_load5 > (count by (instance) (node_cpu_seconds_total{job="node-service",mode='system'})) * 1.2
      for: 5m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} load(5m) 过高"
        description: "Load(5m) 过高,超出cpu核数 1.2倍"
        value: "{{ $value }}"

    - alert: NodeMemoryHigh
      expr: (1 - node_memory_MemAvailable_bytes{job="node-service"} / node_memory_MemTotal_bytes{job="node-service"}) * 100 > 60
      for: 5m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} memory 使用率过高"
        description: "Memory 使用率超过 90%"
        value: "{{ $value }}"

    - alert: NodeDiskRootHigh
      expr: (1 - node_filesystem_avail_bytes{job="node-service",fstype=~"ext.*|xfs",mountpoint ="/"} / node_filesystem_size_bytes{job="node-service",fstype=~"ext.*|xfs",mountpoint ="/"}) * 100 > 90
      for: 10m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} disk(/ 分区) 使用率过高"
        description: "Disk(/ 分区) 使用率超过 90%"
        value: "{{ $value }}"

    - alert: NodeDiskBootHigh
      expr: (1 - node_filesystem_avail_bytes{job="node-service",fstype=~"ext.*|xfs",mountpoint ="/boot"} / node_filesystem_size_bytes{job="node-service",fstype=~"ext.*|xfs",mountpoint ="/boot"}) * 100 > 80
      for: 10m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} disk(/boot 分区) 使用率过高"
        description: "Disk(/boot 分区) 使用率超过 80%"
        value: "{{ $value }}"

    - alert: NodeDiskReadHigh
      expr: irate(node_disk_read_bytes_total{job="node-service"}[5m]) > 20 * (1024 ^ 2)
      for: 5m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} disk 读取字节数 速率过高"
        description: "Disk 读取字节数 速率超过 20 MB/s"
        value: "{{ $value }}"

    - alert: NodeDiskWriteHigh
      expr: irate(node_disk_written_bytes_total{job="node-service"}[5m]) > 20 * (1024 ^ 2)
      for: 5m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} disk 写入字节数 速率过高"
        description: "Disk 写入字节数 速率超过 20 MB/s"
        value: "{{ $value }}"

    - alert: NodeDiskReadRateCountHigh
      expr: irate(node_disk_reads_completed_total{job="node-service"}[5m]) > 3000
      for: 5m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} disk iops 每秒读取速率过高"
        description: "Disk iops 每秒读取速率超过 3000 iops"
        value: "{{ $value }}"

    - alert: NodeDiskWriteRateCountHigh
      expr: irate(node_disk_writes_completed_total{job="node-service"}[5m]) > 3000
      for: 5m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} disk iops 每秒写入速率过高"
        description: "Disk iops 每秒写入速率超过 3000 iops"
        value: "{{ $value }}"

    - alert: NodeInodeRootUsedPercentHigh
      expr: (1 - node_filesystem_files_free{job="node-service",fstype=~"ext4|xfs",mountpoint="/"} / node_filesystem_files{job="node-service",fstype=~"ext4|xfs",mountpoint="/"}) * 100 > 80
      for: 10m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} disk(/ 分区) inode 使用率过高"
        description: "Disk (/ 分区) inode 使用率超过 80%"
        value: "{{ $value }}"

    - alert: NodeInodeBootUsedPercentHigh
      expr: (1 - node_filesystem_files_free{job="node-service",fstype=~"ext4|xfs",mountpoint="/boot"} / node_filesystem_files{job="node-service",fstype=~"ext4|xfs",mountpoint="/boot"}) * 100 > 80
      for: 10m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} disk(/boot 分区) inode 使用率过高"
        description: "Disk (/boot 分区) inode 使用率超过 80%"
        value: "{{ $value }}"

    - alert: NodeFilefdAllocatedPercentHigh
      expr: node_filefd_allocated{job="node-service"} / node_filefd_maximum{job="node-service"} * 100 > 80
      for: 10m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} filefd 打开百分比过高"
        description: "Filefd 打开百分比 超过 80%"
        value: "{{ $value }}"

    - alert: NodeNetworkNetinBitRateHigh
      expr: avg by (instance) (irate(node_network_receive_bytes_total{device=~"eth0|eth1|ens33|ens37"}[1m]) * 8) > 20 * (1024 ^ 2) * 8
      for: 3m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} network 接收比特数 速率过高"
        description: "Network 接收比特数 速率超过 20MB/s"
        value: "{{ $value }}"

    - alert: NodeNetworkNetoutBitRateHigh
      expr: avg by (instance) (irate(node_network_transmit_bytes_total{device=~"eth0|eth1|ens33|ens37"}[1m]) * 8) > 20 * (1024 ^ 2) * 8
      for: 3m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} network 发送比特数 速率过高"
        description: "Network 发送比特数 速率超过 20MB/s"
        value: "{{ $value }}"

    - alert: NodeNetworkNetinPacketErrorRateHigh
      expr: avg by (instance) (irate(node_network_receive_errs_total{device=~"eth0|eth1|ens33|ens37"}[1m])) > 15
      for: 3m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} 接收错误包 速率过高"
        description: "Network 接收错误包 速率超过 15个/秒"
        value: "{{ $value }}"

    - alert: NodeNetworkNetoutPacketErrorRateHigh
      expr: avg by (instance) (irate(node_network_transmit_packets_total{device=~"eth0|eth1|ens33|ens37"}[1m])) > 15
      for: 3m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} 发送错误包 速率过高"
        description: "Network 发送错误包 速率超过 15个/秒"
        value: "{{ $value }}"

    - alert: NodeProcessBlockedHigh
      expr: node_procs_blocked{job="node-service"} > 10
      for: 10m
      labels:
        severity: warning
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} 当前被阻塞的任务的数量过多"
        description: "Process 当前被阻塞的任务的数量超过 10个"
        value: "{{ $value }}"

    - alert: NodeTimeOffsetHigh
      expr: abs(node_timex_offset_seconds{job="node-service"}) > 3 * 60
      for: 2m
      labels:
        severity: info
        instance: "{{ $labels.instance }}"
      annotations:
        summary: "instance: {{ $labels.instance }} 时间偏差过大"
        description: "Time 节点的时间偏差超过 3m"
        value: "{{ $value }}"

重启prometheus容器:

docker restart prometheus

验证是否收到告警邮件,我们将node_exporter关闭,在被监控的ECS服务器操作:

systemctl stop node_exporter

然后刷新prometheus的页面,查看Alerts菜单,我们发现NodeDown规则处于PENDING状态:

等待一分钟后再次刷新查看,已经变成了FIRING状态:

这时候我们去查看下邮箱:

说明已经收到了告警邮件。现在我们把它恢复:

systemctl start node_exporter

然后我们就收到了服务恢复的告警邮件:

7 多容器启动管理

多容器配置,需要修改端口、数据存储路径等信息,例如:

Prometheus

docker run -d --user root -p 9091:9090 --name prometheus-poc \
    -v /home/mon/prometheus-poc/etc/prometheus.yml:/etc/prometheus/prometheus.yml \
    -v /home/mon/prometheus-poc/rules:/etc/prometheus/rules \
    -v /home/mon/prometheus-poc/data:/data/prometheus \
    prom/prometheus \
    --config.file="/etc/prometheus/prometheus.yml" \
    --storage.tsdb.path="/data/prometheus" \
    --web.listen-address="0.0.0.0:9090"

不同之处:

  • -p 9091:9090
  • --name prometheus-poc
  • -v /home/mon/prometheus-poc/etc/prometheus.yml:/etc/prometheus/prometheus.yml
  • -v /home/mon/prometheus-poc/rules:/etc/prometheus/rules
  • -v /home/mon/prometheus-poc/data:/data/prometheus

Grafana

docker run -d -p 3001:3000 -v /home/mon/grafana-poc:/var/lib/grafana --name=grafana-poc grafana/grafana:latest

不同之处:

  • -p 3001:3000
  • --name=grafana-poc
  • -v /home/mon/grafana-poc:/var/lib/grafana

8 参考资料

Docker 部署 Prometheus+Grafana

Dashboard Download