黑盒监控blackbox_exporter
前边介绍有很多exporter可以直接将metrics暴露给Prometheus进行监控,这些称为“白盒监控”,那些exporter无法监控到的指标呢?或者未暴露Metrics给Prometheus的一些服务怎么办?这时就要用到 blackbox_exporte “黑盒监控”。
blackbox_exporte支持用户通过:HTTP、HTTPS、DNS、TCP和ICMP的方式对网络进行探测,还可以探测SSL证书过期时间。
部署及使用blackbox_exporter
部署blackbox_exporter
这里以linux二进制部署为例:
- 下载安装包
curl -LO https://github.com/prometheus/blackbox_exporter/releases/download/v0.22.0/blackbox_exporter-0.22.0.linux-amd64.tar.gz
- 展开程序包:
tar xf blackbox_exporter-0.22.0.linux-amd64.tar.gz -C /usr/local/
ln -sv /usr/local/blackbox_exporter-0.22.0.linux-amd64 /usr/local/blackbox_exporter
- 创建用户,或prometheus用户已经存在,可略过该步骤:
useradd -r prometheus
- 创建Systemd Unitfile,保存于/usr/lib/systemd/system/blackbox_exporter.service文件中:
[Unit]
Description=blackbox_exporter
After=network.target
[Service]
Type=simple
User=root
Group=root
ExecStart=/usr/local/blackbox_exporter/blackbox_exporter
--config.file=/usr/local/blackbox_exporter/blackbox.yml
--web.listen-address=:9115
Restart=on-failure
[Install]
WantedBy=multi-user.target
- 启动服务:
systemctl daemon-reload
systemctl start blackbox_exporter.service
systemctl enable blackbox_exporter.service
- 验证监听的端口,并测试访问其暴露的指标
ss -tnlp | grep '9115'
curl localhost:9115/metrics
随后即可访问Blackbox Exporter的Web UI,其使用的URL如下,其中的要替换为节点的实际地址:http://:9115/
icmp监控,监控主机存活状态
prometheus 添加相关监控,Blackbox 使用默认配置启动即可
vi /usr/local/prometheus/prometheus.yml
- job_name: "icmp_ping"
metrics_path: /probe
params:
module: [icmp] # 使用icmp模块
file_sd_configs:
- refresh_interval: 10s #检测时间间隔
files:
- "ping/ping_status*.yml" #具体的配置文件路径
relabel_configs:
- source_labels: [__address__]
regex: (.*)(:80)?
target_label: __param_target
replacement: ${1}
- source_labels: [__param_target]
target_label: instance
- source_labels: [__param_target]
regex: (.*)
target_label: ping
replacement: ${1}
- source_labels: []
regex: .*
target_label: __address__
replacement: 127.0.0.1:9115
这里有很多relabel的操作,下篇博客会详细讲解
创建对应的ping目录
cd /usr/local/prometheus/
mkdir ping
cd ping
vi ping_status.yml
- targets: ['monitor.example.com']
labels:
group: '跳板机'
- targets: ['10.xx.xx.xx','10.xx.xx.xx','10.xx.xx.xx']
labels:
group: 'k8s cluster'
- targets: ['www.baidu.com']
labels:
group: '百度'
配置完成后,可以检查配置文件语法,并让Prometheus重载配置。
./promtool check config prometheus.yml
curl -XPOST monitor.example.com:9090/-/reload
打开Prometheus web UI,可以看到,已经监控到了主机icmp情况:
http监控
编辑prometheus的主配置文件prometheus.yml,添加类似如下内容,即可用户对目标站点的探测。
# Blackbox Exporter
- job_name: 'http_get_status'
metrics_path: /probe
params:
module: [http_2xx] # Look for a HTTP 200 response.
file_sd_configs:
- refresh_interval: 2m
files:
- "httpget/http_get*.yml" #具体的配置文件
relabel_configs:
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: "monitor.example.com:9115" # 指向实际的Blackbox exporter.
- target_label: region
replacement: "local"
vi httpget/http_get.yml
static_configs:
- targets:
- "https://monitor.example.com"
- "http://monitor.example.com:8080"
- "www.google.com"
refresh_interval: 2m
重新加载Prometheus
curl -XPOST monitor.example.com:9090/-/reload
展示:
这里要注意,blackbox下并不是prometheus中State状态为UP就认为是正常状态,其实并不然,这里我们随便写一个不存在的域名,例如:http://www.buzhida2222o.com,这里看也是UP状态:
但实际看其的metrics指标并不正常,探活的指标是失败,这里不确定是BUG还是怎样。
tcp端口监控
大体的步骤都是一致的,这里就直接上配置:
- job_name: 'tcp_port_status'
metrics_path: /probe
params:
module: [tcp_connect]
static_configs:
- targets: ['monitor.example.com:80','monitor.example.com:8080','monitor.example.com:443']
labels:
instance: 'port_status'
group: 'tcp'
relabel_configs:
- source_labels: [__address__]
target_label: __param_target
- source_labels: [__param_target]
target_label: instance
- target_label: __address__
replacement: monitor.example.com:9115
成功监控:
至此,黑盒监控中常用的一些功能介绍完毕,监控项配置完毕后,可以通过导入dashboard到grafana来更直观的查看监控数据。
自定义blackbox.yml
blackbox的默认监控配置也可以进行自定义修改,例如http GET添加一些headers,设置boby_size_limit值或判断一些response body是否符合预期,还有一些TLS的设置等等,我们可以参考官网文档中给出的example来进行自定义:
# github地址 https://github.com/prometheus/blackbox_exporter # github中blackbox.yml各配置项解析 blackbox_exporter/CONFIGURATION.md at master · prometheus/blackbox_exporter · GitHub # github中example文件 blackbox_exporter/example.yml at master · prometheus/blackbox_exporter · GitHub
这里我们做一个演示https及私有TLS证书的演示,监控证书过期时间。
首先需要更改默认的vim blackbox.yml
modules: http_2xx: prober: http http: preferred_ip_protocol: "ip4" valid_http_versions: ["HTTP/1.1", "HTTP/2"] valid_status_codes: [200,301,302,303] tls_config: insecure_skip_verify: true http_ca_example: prober: http http: method: GET preferred_ip_protocol: "ip4" valid_http_versions: ["HTTP/1.1", "HTTP/2"] fail_if_ssl: false fail_if_not_ssl: true tls_config: insecure_skip_verify: false ca_file: /usr/local/blackbox_exporter/certs/ca.crt cert_file: /usr/local/blackbox_exporter/certs/server.crt key_file: /usr/local/blackbox_exporter/certs/server.key
http_2xx:
在此模块的基础上添加了请求状态码的校验,并设置了tls校验直接跳过,这种就比较省事,可以直接跳过证书的校验。
http_ca_example:
此模块为新增,主要配置了tls证书的一些配置,添加ca、证书和私钥的文件位置,使blackbox请求时带着证书。
配置完成后,重启blackbox服务:
systemctl restart blackbox_exporter.service
然后配置prometheus.yml添加对应模块的使用:
- job_name: 'http_get_status' metrics_path: /probe params: module: [http_2xx] # Look for a HTTP 200 response. file_sd_configs: - refresh_interval: 2m files: - "httpget/http_get*.yml" #具体的配置文件 relabel_configs: - source_labels: [__address__] target_label: __param_target - source_labels: [__param_target] target_label: instance - target_label: __address__ replacement: "monitor.example.com:9115" # 指向实际的Blackbox exporter. - target_label: region replacement: "local" - job_name: 'http_get_ca_status' metrics_path: /probe params: module: [http_ca_example] file_sd_configs: - refresh_interval: 2m files: - "httpget/http_ca.yml" relabel_configs: - source_labels: [__address__] target_label: __param_target - source_labels: [__param_target] target_label: instance - target_label: __address__ replacement: "monitor.example.com:9115" # 指向实际的Blackbox exporter. - target_label: region replacement: "beijing"
在对应的发现文件中,配置主机并重新加载prometheus配置,并查看监控状态:
curl -XPOST monitor.example.com:9090/-/reload
搜索指标
probe_http_duration_seconds{phase="tls"}
证书过期时间已经获取成功,这个时候可以导入ID为13230的Dashboard到Grafana,然后再设置一个告警rule,就可以完成TLS证书过期监控啦。
最终效果:
服务器托管,北京服务器托管,服务器租用 http://www.fwqtg.net
机房租用,北京机房租用,IDC机房托管, http://www.fwqtg.net