Redis-exporter部署及集群监控

Redis-exporter
服务端口:9121

部署

安装包：redis_exporter-v1.33.0.linux-amd64.tar.gz
运行条件：go环境

安装go：

yum install -y epel-release
yum install -y golang

下载安装包

https://github.com/oliver006/redis_exporter/releases

解压

解压重命名redis_exporter-v1.33.0.linux-amd64.tar.gz文件夹后，配置redis_exporter.service，就能用。

配置redis_exporter.service

vim /etc/systemd/system/redis_exporter.service

[Unit]
Description=redis_exporter
Documentation=https://prometheus.io/
After=network.target
[Service]
Type=simple
User=prometheus
ExecStart=/usr/local/redis_exporter-v1.33.0.linux-amd64/redis_exporter -redis.addr {部署predixy代理的ip}:7617  -redis.password XXXXX  -web.listen-address {部署redis_exporter服务的ip}:9121
Restart=always

启动

加入prometheus

vim /etc/prometheus/prometheus.yml

  - job_name: 'GT_WORK_REDIS_CULSTER'
    static_configs:
      - targets:
        - redis://192.168.10.1:7001
        - redis://192.168.10.2:7002
        - redis://192.168.10.3:7003
        - redis://192.168.10.4:7004
        - redis://192.168.10.5:7005
        - redis://192.168.10.6:7006
    metrics_path: /scrape
    relabel_configs:
      - source_labels: [__address__]
        target_label: __param_target
      - source_labels: [__param_target]
        target_label: instance
      - target_label: __address__
        replacement: xxx.xxx.xxx.xxx:9121

配置告警规则

vim /etc/prometheus/rules/redis.yml

groups:
- name: redis
  rules:
  - alert: RedisDown
    annotations:
      description: |-
        Redis instance is down
          VALUE = {{ $value }}
          LABELS: {{ $labels }}
      summary: Redis down (instance {{ $labels.instance }})
    expr: redis_up == 0
    for: 5m
    labels:
      severity: critical
  - alert: RedisOutOfMemory
    annotations:
      description: |-
        Redis is running out of memory (> 85%)
          VALUE = {{ $value }}
          LABELS: {{ $labels }}
      summary: Redis out of memory (instance {{ $labels.instance }})
    expr: redis_memory_used_bytes / redis_total_system_memory_bytes * 100 > 85
    for: 5m
    labels:
      severity: warning
  - alert: RedisTooManyConnections
    annotations:
      description: |-
        Redis instance has too many connections
          VALUE = {{ $value }}
          LABELS: {{ $labels }}
      summary: Redis too many connections (instance {{ $labels.instance }})
    expr: redis_connected_clients > 100
    for: 5m
    labels:
      severity: warning

记得重启prometheus

Grafana导入模板

两种方法：
第一种是本地有redis的模板，直接导入

第二种是搜索编号并直接下载

优化模板

思考：该模板只能看单个集群的每一个实例，无法显示出那些是master，那些是slave，因此对现有的模板进行修改。

label_values(redis_up,job)

label_values(redis_instance_info,role) 

label_values(redis_instance_info{job=~"$cluster",role=~"$role"},instance) 

最后看效果