考慮到要明智地使用標籤,咱們須要給事物從新命名。在一個集中的、複雜的監視環境中,咱們有時沒法控制正在監視的全部資源以及它們公開的監視數據。從新標記容許在本身的環境中控制、管理和潛在地標準化度量。node
常見的用例是 :docker
scrape_configs: - job_name: 'prometheus' static_configs: - targets: ['localhost:9090'] metric_relabel_configs: - source_labels: [__name__] separator: ',' regex: '(process_cpu_seconds_total|process_open_fds)' action: drop - job_name: 'node_exporter' static_configs: - targets: ['localhost:9100'] params: collect[]: - cpu - meminfo - diskstats - job_name: 'docker' static_configs: - targets: ['192.168.31.151:8080']
備註:若是咱們指定了多個源標籤,須要用 ; 分開:bash
regex1;regex2;regex3spa
其實是根據已有的標籤,生成一個新標籤3d
scrape_configs: - job_name: 'prometheus' static_configs: - targets: ['localhost:9090'] metric_relabel_configs: - source_labels: [__name__] separator: ',' regex: '(process_cpu_seconds_total|process_open_fds)' action: drop - source_labels: [version] regex: 'go1.*' replacement: 'go_unknown' target_label: go_version - job_name: 'node_exporter' static_configs: - targets: ['localhost:9100'] params: collect[]: - cpu - meminfo - diskstats - job_name: 'docker' static_configs: - targets: ['192.168.31.151:8080']
scrape_configs: - job_name: 'prometheus' static_configs: - targets: ['localhost:9090'] metric_relabel_configs: - source_labels: [__name__] separator: ',' regex: '(process_cpu_seconds_total|process_open_fds)' action: drop - source_labels: [version] regex: 'go1.*' replacement: 'go_unknown' target_label: go_version - regex: 'version' action: labeldrop - job_name: 'node_exporter' static_configs: - targets: ['localhost:9100'] params: collect[]: - cpu - meminfo - diskstats - job_name: 'docker' static_configs: - targets: ['192.168.31.151:8080']