服務發現
基於文件的服務發現
現有配置：
[root@mcw03 ~]# cat /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: - "rules/node_rules.yml" # - "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'] - job_name: 'agent1' static_configs: - targets: ['10.0.0.14:9100','10.0.0.12:9100'] - job_name: 'promserver' static_configs: - targets: ['10.0.0.13:9100'] - job_name: 'server_mariadb' static_configs: - targets: ['10.0.0.13:9104'] - job_name: 'docker' static_configs: - targets: ['10.0.0.12:8080'] metric_relabel_configs: - regex: 'kernelVersion' action: labeldrop [root@mcw03 ~]#
把static_configs 替換成file_sd_configs
配置刷新重載文件配置的時間。可以不用手動刷新
創建目錄並修改配置，指定使用的文件配置
下面紅色配置錯了，直接指定文件路徑就可以，不需要targets鍵
[root@mcw03 ~]# ls /etc/prometheus.yml /etc/prometheus.yml [root@mcw03 ~]# mkdir -p /etc/targets/{nodes,docker} [root@mcw03 ~]# vim /etc/prometheus.yml [root@mcw03 ~]# cat /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: - "rules/node_rules.yml" # - "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'] - job_name: 'agent1' file_sd_configs: - files: - targets: targets/nodes/*.json refresh_interval: 5m - job_name: 'promserver' static_configs: - targets: ['10.0.0.13:9100'] - job_name: 'server_mariadb' static_configs: - targets: ['10.0.0.13:9104'] - job_name: 'docker' file_sd_configs: - files: - targets: targets/docker/*.json refresh_interval: 5m # metric_relabel_configs: # - regex: 'kernelVersion' # action: labeldrop [root@mcw03 ~]#
創建配置文件
[root@mcw03 ~]# touch /etc/targets/nodes/nodes.json [root@mcw03 ~]# touch /etc/targets/docker/daemons.json [root@mcw03 ~]#
修改到json文件配置中
[root@mcw03 ~]# vim /etc/targets/nodes/nodes.json [root@mcw03 ~]# vim /etc/targets/docker/daemons.json [root@mcw03 ~]# cat /etc/targets/nodes/nodes.json [{ "targets": [ "10.0.0.14:9100", "10.0.0.12:9100" ] }] [root@mcw03 ~]# cat /etc/targets/docker/daemons.json [{ "targets": [ "10.0.0.12:8080" ] }] [root@mcw03 ~]#
報錯了
[root@mcw03 ~]# curl -X POST http://localhost:9090/-/reload failed to reload config: couldn't load configuration (--config.file="/etc/prometheus.yml"): parsing YAML file /etc/prometheus.yml: yaml: unmarshal errors: line 34: cannot unmarshal !!map into string line 45: cannot unmarshal !!map into string [root@mcw03 ~]#
上面配置寫錯了
[root@mcw03 ~]# vim /etc/prometheus.yml [root@mcw03 ~]# cat /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: - "rules/node_rules.yml" # - "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'] - job_name: 'agent1' file_sd_configs: - files: - targets/nodes/*.json refresh_interval: 5m - job_name: 'promserver' static_configs: - targets: ['10.0.0.13:9100'] - job_name: 'server_mariadb' static_configs: - targets: ['10.0.0.13:9104'] - job_name: 'docker' file_sd_configs: - files: - targets/docker/*.json refresh_interval: 5m # metric_relabel_configs: # - regex: 'kernelVersion' # action: labeldrop [root@mcw03 ~]# curl -X POST http://localhost:9090/-/reload [root@mcw03 ~]#
此時看，可以看到服務發現的客戶端
http://10.0.0.13:9090/service-discovery
改爲yml格式
[root@mcw03 ~]# vim /etc/prometheus.yml [root@mcw03 ~]# cat /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: - "rules/node_rules.yml" # - "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'] - job_name: 'agent1' file_sd_configs: - files: - targets/nodes/*.json refresh_interval: 5m - job_name: 'promserver' static_configs: - targets: ['10.0.0.13:9100'] - job_name: 'server_mariadb' static_configs: - targets: ['10.0.0.13:9104'] - job_name: 'docker' file_sd_configs: - files: - targets/docker/*.yml refresh_interval: 5m # metric_relabel_configs: # - regex: 'kernelVersion' # action: labeldrop [root@mcw03 ~]# cp /etc/targets/docker/daemons.json /etc/targets/docker/daemons.yml [root@mcw03 ~]# vim /etc/targets/docker/daemons.yml [root@mcw03 ~]# cat /etc/targets/docker/daemons.yml - targets: - "10.0.0.12:8080" [root@mcw03 ~]#
[root@mcw03 ~]# curl -X POST http://localhost:9090/-/reload [root@mcw03 ~]#
重載之後正常，
從標籤裏可以看到，服務自動發現來自哪裏
因爲target是yml或者json數據，所以可以用salt，cmdb等等各種，進行配置集中管理，實現監控
基於文件的自動發現，添加標籤的實現
修改配置
[root@mcw03 ~]# vim /etc/targets/nodes/nodes.json [root@mcw03 ~]# cat /etc/targets/nodes/nodes.json [{ "targets": [ "10.0.0.14:9100", "10.0.0.12:9100" ], "labels": { "datacenter": "mcwhome" } }] [root@mcw03 ~]# vim /etc/targets/docker/daemons.yml [root@mcw03 ~]# cat /etc/targets/docker/daemons.yml - targets: - "10.0.0.12:8080" - labels: "datacenter": "mcwymlhome" [root@mcw03 ~]#
不需要重啓服務，這個標籤自動就有了。不過yml格式的，添加標籤，沒有生效。不清楚咋添加
基於api的服務發現
基於dns的服務發現