linux-2019全新課程參考筆記容器云prometheus監(jiān)控ceph

1、Prometheus CephPrometheus Server創(chuàng)建運(yùn)Prometheus Server進(jìn)程的系統(tǒng)，并為其創(chuàng)建家錄/var/lib/prometheus作為數(shù)據(jù) 錄。# useradd -r -m -d /var/lib/prometheus prometheus #并安裝Prometheus Server，以2.7.2版為例： # wget /prometheus/prometheu s/releases/download/v2.7.2/prometheus-2.7.2.linux-amd64.tar.gz # tar xf prometheus-2.7.2.linux- a

2、md64.tar.gz -C /usr/local/# ln /usr/local/prometheus-2.7.2.linux-amd64 /usr/local/prometheus創(chuàng)建Unit File 創(chuàng)建Prometheus專(zhuān)的Unit File，件路徑為/usr/lib/systemd/system/prometheus.serviceUnitDescription=The Prometheus 2 monitoring system and time series database. ation=After=network.ServiceEnvironmentFile=-/etc/

3、sysconfig/prometheus Us rometheus ExecStart=/usr/local/prometheus/prometheus -storage.tsdb.path=/var/lib/prometheus -config.file=/usr/local/prometheus/prometheus.yml -web.listen-address=:9090 -web.external-url= Restart=on-failure StartLimit erval=1 RestartSec=3Install WantedBy=multi-user.4. 編輯配置件 Pr

4、ometheus的主配置件為prometheus.yml，它主要由global、rule_les、 scr _congs、alerting、remote_write和remote_read個(gè)配置段組成：global：全局配置段；scr _congs：scr 配置集合，于定義 的標(biāo)對(duì)象（ ）的集合，以及描述如何抓?。╯cr ）相關(guān)指標(biāo)數(shù)據(jù)的配置參數(shù)；通常，每個(gè)scr 配置對(duì)應(yīng)于個(gè)單獨(dú)的作業(yè)（job），每個(gè)s可通過(guò)靜態(tài)配置（s ic_congs）直接給出定義，也可基于Prometheus持的服務(wù)發(fā)現(xiàn)機(jī)制進(jìn)動(dòng)配置；alertmanager_congs：可由Prometheus使的Alertmanag

5、er實(shí)例的集合，以及如何同這些 Alertmanager交互的配置參數(shù)；每個(gè)Alertmanager可通過(guò)靜態(tài)配置（s ic_congs）直接給出定義，也可基于Prometheus 的服務(wù)發(fā)現(xiàn)機(jī)制進(jìn)動(dòng)配置；remote_write：配置“ 寫(xiě)”機(jī)制，Prometheus需要將數(shù)據(jù)保存于外部的 系統(tǒng)（例如InuxDB）時(shí)定義此配置段，隨后Prometheus將樣本數(shù)據(jù)通過(guò)HTTP協(xié)議發(fā)送給由URL指定適配器(Adaptor)；remote_read：配置“ 讀”機(jī)制，Prometheus到的查詢請(qǐng)求交給由URL指定適配器（ ater）執(zhí)，Adapter將請(qǐng)求條件轉(zhuǎn)換為 Prometheus可的格

6、式；配置件組成格式及常的全局配置參數(shù)如下所：服務(wù)中的查詢請(qǐng)求，并將獲取的響應(yīng)數(shù)據(jù)轉(zhuǎn)換為global:# How frequently to scrs by default. scr _ erval: | default = 1m # How long until a scr request times out. scr _timeout: | default = 10s # How frequently to evaluate rules. evaluation_ erval: | default = 1m # The labels to add to any time series or a

7、lerts when communicating with # external systems (federation, remote storage, Alertmanager). external_labels: : . # Rule files specifies a list of globs. Rules and alerts are read from # all matching files.rule_files: - . # A list of scr configurations. scr _configs: - . # Alerting specifies setting

8、s related to the Alertmanager. alerting:alert_relabel_configs: - . alertmanagers: - . # Settings related to the remote write feature. remote_write: - . # Settings related to the remote read feature. remote_read: - . scr 配置段中，使s ic_congs配置Job的語(yǔ)法格式：# Thes specified by the s ic config.s: - # Labels ass

9、igned to all metrics scr d from the labels: : . s.使le_sd_congs配置Job的語(yǔ)法格式：s: , . ,labels: : , .,.配置件中的默認(rèn)配置僅持以靜態(tài)式通過(guò)node_exporter Prometheus Server本機(jī)的系統(tǒng)指標(biāo)，其配置如下所。scr _configs:- job_name: prometheus# metrics_path defaults to /metrics # scheme defaults to http.s ic_configs:-s: localhost:90905. 啟動(dòng)服務(wù) # sys

10、temctl daemon-reload # systemctl start prometheus.service 6.node_exporter安裝程序包 # wget /prometheus/node exporter/releases/download/v0.17.0/node exporter-0.17.0.linux-amd64.tar.gz # tar xf node_exporter-0.17.0.linux-amd64.tar.gz -C /usr/local/ # ln -sv /usr/local/node_exporter-0.17.0.linux-amd64 /usr/

11、local/node_exporter #創(chuàng)建運(yùn)Prometheus Server進(jìn)程的系統(tǒng)，并為其創(chuàng)建家錄/var/lib/prometheus作為數(shù)據(jù) 錄。# useradd -r -m -d /var/lib/prometheus prometheus #創(chuàng)建Unit File件路徑為/usr/lib/systemd/system/node_exporter.serviceUnitDescription=Prometheus exporter for machine metrics, written in Go with pluggable metric collectors.ation

12、= After=network./prometheus/node_exporterServiceEnvironmentFile=-/etc/sysconfig/node_exporter UsrometheusExecStart=/usr/local/node_exporter/node_exporter $NODE_EXPORTER_OPTSRestart=on-failure StartLimit erval=1 RestartSec=3Install WantedBy=multi-user.4. 啟動(dòng)服務(wù) # systemctl daemon-reload # systemctl sta

13、rt node_exporter.service # systemctl enable node_exporter.service設(shè)定Ceph MgrCeph Manager內(nèi)置了眾多模塊，包括prometheus模塊，于直接輸出Prometheus格的指標(biāo)數(shù)據(jù)。# ceph mgr module enable prometheus Prometheus模塊默認(rèn) 于TCP協(xié)議的9283端。配置Prometheus Job修改Prometheus的配置件，添加與Ceph相關(guān)的Job。scr _configs:# The job name is added as a label job= to any timeseries scr d from this config.- job_name: prometheus# metrics_path default