You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
6.9 KiB
6.9 KiB
Consul字段设计说明
- 所有数据存在一个名为
blackbox_exporter
的Services项中,每个监控目标为一个子Service。 - 每个Service包含一个Tag,会自动配置为meta中的
module
的值,作为Prometheus自动发现的tags。 - 每个Service使用Meta的kv保存监控目标的明细:
module
,company
,project
,env
,name
,instance
,分别表示:监控类型,公司部门,项目,环境,名称,实例url - 前5个字段合并即为consul的serviceID,作为唯一监控项标识
- 建议监控类型字段:
meta内的module
,blackbox-exporter配置中的module
及Prometheus的job名
使用同一命名。
配置Prometheus
基于Consul实现Prometheus的自动发现功能配置
- 根据Consul每个service的tag来把监控目标关联到Prometheus的JOB。
- 把Consul每个service的Meta的KV关联到Prometheus每个指标的标签。
- 根据每个指标的标签来对监控目标分类,分组,方便管理维护。
以下配置的同一个job的job_name
,module
,tags
使用同一命名,关联job,module与consul的tags
参考配置:2XX,4XX,TCP类型的监控,注意JOB名称不要与已有的重复。
vi prometheus.yml
#####blackbox_exporter#####
- job_name: 'http_2xx'
metrics_path: /probe
params:
module: [http_2xx]
consul_sd_configs:
- server: 'x.x.x.x:8500'
token: 'xxx-xxx-xxx-xxx'
services: ['blackbox_exporter']
tags: ['http_2xx']
relabel_configs:
- source_labels: ["__meta_consul_service_metadata_instance"]
target_label: __param_target
- source_labels: ["__meta_consul_service_metadata_company"]
target_label: company
- source_labels: ["__meta_consul_service_metadata_env"]
target_label: env
- source_labels: ["__meta_consul_service_metadata_name"]
target_label: name
- source_labels: ["__meta_consul_service_metadata_project"]
target_label: project
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: 127.0.0.1:9115
- job_name: 'http_4xx'
metrics_path: /probe
params:
module: [http_4xx]
consul_sd_configs:
- server: 'x.x.x.x:8500'
token: 'xxx-xxx-xxx-xxx'
services: ['blackbox_exporter']
tags: ['http_4xx']
relabel_configs:
- source_labels: ["__meta_consul_service_metadata_instance"]
target_label: __param_target
- source_labels: ["__meta_consul_service_metadata_company"]
target_label: company
- source_labels: ["__meta_consul_service_metadata_env"]
target_label: env
- source_labels: ["__meta_consul_service_metadata_name"]
target_label: name
- source_labels: ["__meta_consul_service_metadata_project"]
target_label: project
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: 127.0.0.1:9115
- job_name: 'tcp_connect'
metrics_path: /probe
params:
module: [tcp_connect]
consul_sd_configs:
- server: 'x.x.x.x:8500'
token: 'xxx-xxx-xxx-xxx'
services: ['blackbox_exporter']
tags: ['tcp_connect']
relabel_configs:
- source_labels: ["__meta_consul_service_metadata_instance"]
target_label: __param_target
- source_labels: ["__meta_consul_service_metadata_company"]
target_label: company
- source_labels: ["__meta_consul_service_metadata_env"]
target_label: env
- source_labels: ["__meta_consul_service_metadata_name"]
target_label: name
- source_labels: ["__meta_consul_service_metadata_project"]
target_label: project
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: 127.0.0.1:9115
配置Blackbox_Exporter
参考配置:2XX,4XX,TCP类型的监控,注意模块名称不要与已有的重复。
cat blackbox.yml
modules:
http_2xx:
prober: http
http:
valid_status_codes:
- 200
- 301
- 302
- 303
no_follow_redirects: true
preferred_ip_protocol: ip4
ip_protocol_fallback: false
http_4xx:
prober: http
http:
valid_status_codes:
- 401
- 403
- 404
preferred_ip_protocol: ip4
ip_protocol_fallback: false
tcp_connect:
prober: tcp
批量导入脚本
在units目录下instance.list
中写入监控目标的信息:JOB名称,公司/部门,项目,环境,名称,实例url,每行一个,空格分隔。
注意:前5个字段组合起来必须唯一,作为一个监控项的ID。
修改units目录下导入脚本中的consul_token和consul_url,保存后执行input.py,即可导入所有监控目标到Consul,并符合Prometheus的自动发现配置。
导入Blackbox Exporter Dashboard
- 支持Grafana 8,基于blackbox_exporter 0.19.0设计
- 采用图表+曲线图方式展示TCP,ICMP,HTTPS的服务状态,各阶段请求延时,HTTPS证书信息等
- 优化展示效果,支持监控目标的分组、分类级联展示,多服务同时对比展示。
导入ID:9965
详细URL:https://grafana.com/grafana/dashboards/9965
Prometheus 站点监控告警规则
- name: Domain
rules:
- alert: 站点可用性
expr: probe_success == 0
for: 1m
labels:
alertype: domain
severity: critical
annotations:
description: "{{$labels.env}}_{{ $labels.name }}({{ $labels.project }}):站点无法访问\n> {{ $labels.instance }}"
- alert: 站点1h可用性低于80%
expr: sum_over_time(probe_success[1h])/count_over_time(probe_success[1h]) * 100 < 80
for: 3m
labels:
alertype: domain
severity: warning
annotations:
description: "{{$labels.env}}_{{ $labels.name }}({{ $labels.project }}):站点1h可用性:{{ $value | humanize }}%\n> {{ $labels.instance }}"
- alert: 站点状态异常
expr: (probe_success == 0 and probe_http_status_code > 499) or probe_http_status_code == 0
for: 1m
labels:
alertype: domain
severity: warning
annotations:
description: "{{$labels.env}}_{{ $labels.name }}({{ $labels.project }}):站点状态异常:{{$value}}\n> {{ $labels.instance }}"
- alert: 站点耗时过高
expr: probe_duration_seconds > 0.5
for: 2m
labels:
alertype: domain
severity: warning
annotations:
description: "{{$labels.env}}_{{ $labels.name }}({{ $labels.project }}):当前站点耗时:{{$value | humanize}}s\n> {{ $labels.instance }}"
- alert: SSL证书有效期
expr: (probe_ssl_earliest_cert_expiry-time()) / 3600 / 24 < 15
for: 2m
labels:
alertype: domain
severity: warning
annotations:
description: "{{$labels.env}}_{{ $labels.name }}({{ $labels.project }}):证书有效期剩余{{ $value | humanize }}天\n> {{ $labels.instance }}"