Prometheus 是一个开源的监控和告警工具包,旨在提供高可靠性和可扩展性。它最初由 SoundCloud 开发,现已成为云原生计算基金会(CNCF)的一部分。以下是 Prometheus 的一些关键特性和概念:
1. **时间序列数据库**:Prometheus 将所有数据存储为时间序列,这些时间序列通过指标名称和一组键值对(称为标签)进行标识。这种设计允许高维度的数据收集和查询。
2. **数据收集**:Prometheus 使用拉取模型来收集指标。它会在配置的端点上以指定的时间间隔抓取指标,这些端点以简单的文本格式暴露指标。
3. **查询语言**:Prometheus 提供了一种强大的查询语言,称为 PromQL(Prometheus 查询语言),允许用户轻松提取和操作时间序列数据。
4. **告警功能**:Prometheus 内置了告警功能。用户可以基于 PromQL 查询定义告警规则,Prometheus 可以通过 Alertmanager 将告警发送到各种通知渠道(如电子邮件、Slack 等)。
5. **可视化**:虽然 Prometheus 本身不提供高级可视化功能,但可以与 Grafana 等工具集成,以创建仪表板和可视化的指标表现。
6. **可扩展性**:Prometheus 设计用于处理大量数据,可以通过使用多个实例和联邦机制进行水平扩展。
7. **服务发现**:Prometheus 可以使用各种服务发现机制(例如 Kubernetes、Consul 等)自动发现要抓取指标的目标,使其适用于动态环境。
8. **生态系统**:Prometheus 拥有丰富的导出器生态系统,导出器是将各种服务和系统(如数据库、Web 服务器等)的指标以 Prometheus 可抓取的格式暴露出来的组件。
总的来说,Prometheus 被广泛用于监控云原生应用和微服务,为系统性能和可靠性提供了宝贵的洞察。
下载安装包:
Prometheus下载: https://prometheus.io/download/
wget https://github.com/prometheus/prometheus/releases/download/v2.53.3/prometheus-2.53.3.linux-amd64.tar.gz
Grafana下载: https://grafana.com/grafana/download?pg=get&plcmt=selfmanaged-box1-cta1
wget https://dl.grafana.com/enterprise/release/grafana-enterprise-11.3.1.linux-amd64.tar.gz
Alertmanager下载: https://prometheus.io/download/#alertmanager
wget https://github.com/prometheus/alertmanager/releases/download/v0.27.0/alertmanager-0.27.0.linux-amd64.tar.gz
Node_exporter下载: https://prometheus.io/download/#node_exporter
wget https://github.com/prometheus/node_exporter/releases/download/v1.8.2/node_exporter-1.8.2.linux-amd64.tar.gz
解压:
mkdir /usr/local/prometheus
tar xvf prometheus-2.53.3.linux-amd64.tar.gz
mv prometheus-2.53.3.linux-amd64 /usr/local/prometheus/prometheus
tar xvf node_exporter-1.8.2.linux-amd64.tar.gz
mv node_exporter-1.8.2.linux-amd64 /usr/local/prometheus/node_exporter
tar xvf alertmanager-0.27.0.linux-amd64.tar.gz
mv alertmanager-0.27.0.linux-amd64 /usr/local/prometheus/alertmanager
tar xvf grafana-enterprise-11.3.1.linux-amd64.tar.gz
mv grafana-v11.3.1 /usr/local/prometheus/grafana
配置开机Prometheus开机服务
cat > /usr/lib/systemd/system/prometheus.service << 'EOF'
[Unit]
Description=Prometheus Server
Documentation=https://prometheus.io/docs/introduction/overview/
After=network-online.target
[Service]
Type=simple
User=prometheus
Group=prometheus
Restart=on-failure
ExecStart=/usr/local/prometheus/prometheus/prometheus \
--config.file=/usr/local/prometheus/prometheus/prometheus.yml \
--storage.tsdb.path=/usr/local/prometheus/prometheus/data \
--storage.tsdb.retention.time=60d \
--web.enable-lifecycle
[Install]
WantedBy=multi-user.target
EOF
配置grafana开机服务
cat > /usr/lib/systemd/system/grafana.service << 'EOF'
[Unit]
Description=Grafana server
Documentation=http://docs.grafana.org
[Service]
Type=simple
User=prometheus
Group=prometheus
Restart=on-failure
ExecStart=/usr/local/prometheus/grafana/bin/grafana-server \
--config=/usr/local/prometheus/grafana/conf/defaults.ini \
--homepath=/usr/local/prometheus/grafana
[Install]
WantedBy=multi-user.target
EOF
配置alertmanager开机服务
cat > /usr/lib/systemd/system/alertmanager.service << 'EOF'
[Unit]
Description=Alert Manager
Wants=network-online.target
After=network-online.target
[Service]
Type=simple
User=prometheus
Group=prometheus
ExecStart=/usr/local/prometheus/alertmanager/alertmanager \
--config.file=/usr/local/prometheus/alertmanager/alertmanager.yml \
--storage.path=/usr/local/prometheus/alertmanager/data
Restart=always
[Install]
WantedBy=multi-user.target
EOF
配置node_exporter开机服务
cat > /usr/lib/systemd/system/node_exporter.service << 'EOF'
[Unit]
Description=node_exporter
Documentation=https://prometheus.io/
After=network.target
[Service]
User=prometheus
Group=prometheus
ExecStart=/usr/local/prometheus/node_exporter/node_exporter
Restart=on-failure
[Install]
WantedBy=multi-user.target
EOF
配置开机自启
useradd --no-create-home --shell /bin/false prometheus
或者
useradd -M -s /usr/sbin/nologin prometheus
chown -R prometheus.prometheus /usr/local/prometheus
systemctl daemon-reload
systemctl enable --now prometheus
systemctl status prometheus
systemctl enable --now alertmanager
systemctl status alertmanager
systemctl enable --now node_exporter
systemctl status node_exporter
systemctl enable --now grafana
systemctl status grafana
端口说明:
root@u22pro:~# netstat -tnlp
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp 0 0 127.0.0.53:53 0.0.0.0:* LISTEN 677/systemd-resolve
tcp 0 0 0.0.0.0:22 0.0.0.0:* LISTEN 36805/sshd: /usr/sb
tcp6 0 0 :::3000 :::* LISTEN 104697/grafana
tcp6 0 0 :::22 :::* LISTEN 36805/sshd: /usr/sb
tcp6 0 0 :::9100 :::* LISTEN 104555/node_exporte
tcp6 0 0 :::9090 :::* LISTEN 104520/prometheus
tcp6 0 0 :::9093 :::* LISTEN 104539/alertmanager
tcp6 0 0 :::9094 :::* LISTEN 104539/alertmanager
root@u22pro:~#
3000 - grafana 管理页面
9100 - node_exporter上报metrics 端口
9090 - prometheus 管理页面
9093 - alertmanager 管理页面
9094 - alertmanager
prometheus.yml配置
root@u22pro:/usr/local/prometheus/prometheus# cat prometheus.yml | grep -v '#'
global:
alerting:
alertmanagers:
- static_configs:
- targets:
rule_files:
- "alert.yml"
scrape_configs:
- job_name: "prometheus"
static_configs:
- targets: ["localhost:9090"]
- job_name: 'node-exporter'
scrape_interval: 15s
static_configs:
- targets: ['localhost:9100']
labels:
instance: Prometheus服务器
- targets: ['192.168.50.5:9100']
labels:
instance: linux-192.168.50.5
- targets: ['192.168.50.6:9100']
labels:
instance: linux-192.168.50.6
alert.yml
root@u22pro:/usr/local/prometheus/prometheus# cat alert.yml
groups:
- name: Prometheus alert
rules:
# 对任何实例超过30s无法联系的情况 发出警报
- alert: 服务告警
expr: up == 0
for: 30s
labels:
severity: critical
annotations:
instance: "{{ $labels.instance }}"
description: "{{ $labels.job }} 服务已关闭"
- name: 服务器资源监控
rules:
- alert: 内存使用率过高
expr: 100 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes) * 100 > 80 #这里的监控参数根据自己实际监控的指标去修改,其他维度的同理
for: 3m
labels:
severity: 严重告警
annotations:
summary: "{{ $labels.instance }} 内存使用率过高, 请尽快处理!"
description: "{{ $labels.instance }}内存使用率超过80%,当前使用率{{ $value }}%."
- alert: 服务器宕机
expr: up == 0
for: 1s
labels:
severity: 严重告警
annotations:
summary: "{{$labels.instance}} 服务器宕机, 请尽快处理!"
description: "{{$labels.instance}} 服务器延时超过3分钟,当前状态{{ $value }}. "
- alert: CPU高负荷
expr: 100 - (avg by (instance,job)(irate(node_cpu_seconds_total{mode="idle"}[5m])) * 100) > 90
for: 5m
labels:
severity: 严重告警
annotations:
summary: "{{$labels.instance}} CPU使用率过高,请尽快处理!"
description: "{{$labels.instance}} CPU使用大于90%,当前使用率{{ $value }}%. "
- alert: 磁盘IO性能
expr: avg(irate(node_disk_io_time_seconds_total[1m])) by(instance,job)* 100 > 90
for: 5m
labels:
severity: 严重告警
annotations:
summary: "{{$labels.instance}} 流入磁盘IO使用率过高,请尽快处理!"
description: "{{$labels.instance}} 流入磁盘IO大于90%,当前使用率{{ $value }}%."
- alert: 网络流入
expr: ((sum(rate (node_network_receive_bytes_total{device!~'tap.*|veth.*|br.*|docker.*|virbr*|lo*'}[5m])) by (instance,job)) / 100) > 102400
for: 5m
labels:
severity: 严重告警
annotations:
summary: "{{$labels.instance}} 流入网络带宽过高,请尽快处理!"
description: "{{$labels.instance}} 流入网络带宽持续5分钟高于100M. RX带宽使用量{{$value}}."
- alert: 网络流出
expr: ((sum(rate (node_network_transmit_bytes_total{device!~'tap.*|veth.*|br.*|docker.*|virbr*|lo*'}[5m])) by (instance,job)) / 100) > 102400
for: 5m
labels:
severity: 严重告警
annotations:
summary: "{{$labels.instance}} 流出网络带宽过高,请尽快处理!"
description: "{{$labels.instance}} 流出网络带宽持续5分钟高于100M. RX带宽使用量{$value}}."
- alert: TCP连接数
expr: node_netstat_Tcp_CurrEstab > 10000
for: 2m
labels:
severity: 严重告警
annotations:
summary: " TCP_ESTABLISHED过高!"
description: "{{$labels.instance}} TCP_ESTABLISHED大于100%,当前使用率{{ $value }}%."
- alert: 磁盘容量
expr: 100-(node_filesystem_free_bytes{fstype=~"ext4|xfs"}/node_filesystem_size_bytes {fstype=~"ext4|xfs"}*100) > 90
for: 1m
labels:
severity: 严重告警
annotations:
summary: "{{$labels.mountpoint}} 磁盘分区使用率过高,请尽快处理!"
description: "{{$labels.instance}} 磁盘分区使用大于90%,当前使用率{{ $value }}%."
修改配置之后,手动重载配置
curl -X POST http://localhost:9090/-/reload
检测Prometheus配置文件是否正确
./promtool check config prometheus.yml
打开gafana管理页面,添加prometheus datasource
添加node_exportor dashboard
grafana服务器监控显示面板: