通过使用telegraf+influx+grafana监控VMware虚拟化平台

一：Telegraf安装配置

1、安装Telegraf

rpm -ivh wget https://dl.influxdata.com/telegraf/releases/telegraf-1.15.1-1.x86_64.rpm

2、备份配置文件

mv /etc/telegraf/telegraf.conf /etc/telegraf/telegraf.conf.bak

3、新建配置文件

vi /etc/telegraf/telegraf.conf

# Configuration for sending metrics to InfluxDB
[[outputs.influxdb]]
    urls = ["http://127.0.0.1:8086"]
    database = "vmware"
    timeout = "0s"
    username = "vmware"
    password = "P@ssw0rd"

# Read metrics from VMware vCenter
 [[inputs.vsphere]]
 ## List of vCenter URLs to be monitored. These three lines must be uncommented
 ## and edited for the plugin to work.
 vcenters = [ "https://192.168.230.10/sdk" ]
    username = "administrator@vsphere.local"
    password = "P@ssw0rd"
 #
 ## VMs
 ## Typical VM metrics (if omitted or empty, all metrics are collected)
 vm_metric_include = [
      "cpu.demand.average",
      "cpu.idle.summation",
      "cpu.latency.average",
      "cpu.readiness.average",
      "cpu.ready.summation",
      "cpu.run.summation",
      "cpu.usagemhz.average",
      "cpu.used.summation",
      "cpu.wait.summation",
      "mem.active.average",
      "mem.granted.average",
      "mem.latency.average",
      "mem.swapin.average",
      "mem.swapinRate.average",
      "mem.swapout.average",
      "mem.swapoutRate.average",
      "mem.usage.average",
      "mem.vmmemctl.average",
      "net.bytesRx.average",
      "net.bytesTx.average",
      "net.droppedRx.summation",
      "net.droppedTx.summation",
      "net.usage.average",
      "power.power.average",
      "virtualDisk.numberReadAveraged.average",
      "virtualDisk.numberWriteAveraged.average",
      "virtualDisk.read.average",
      "virtualDisk.readOIO.latest",
      "virtualDisk.throughput.usage.average",
      "virtualDisk.totalReadLatency.average",
      "virtualDisk.totalWriteLatency.average",
      "virtualDisk.write.average",
      "virtualDisk.writeOIO.latest",
      "sys.uptime.latest",
    ]
 # vm_metric_exclude = [] ## Nothing is excluded by default
 # vm_instances = true ## true by default
 #
 ## Hosts
 ## Typical host metrics (if omitted or empty, all metrics are collected)
 host_metric_include = [
      "cpu.coreUtilization.average",
      "cpu.costop.summation",
      "cpu.demand.average",
      "cpu.idle.summation",
      "cpu.latency.average",
      "cpu.readiness.average",
      "cpu.ready.summation",
      "cpu.swapwait.summation",
      "cpu.usage.average",
      "cpu.usagemhz.average",
      "cpu.used.summation",
      "cpu.utilization.average",
      "cpu.wait.summation",
      "disk.deviceReadLatency.average",
      "disk.deviceWriteLatency.average",
      "disk.kernelReadLatency.average",
      "disk.kernelWriteLatency.average",
      "disk.numberReadAveraged.average",
      "disk.numberWriteAveraged.average",
      "disk.read.average",
      "disk.totalReadLatency.average",
      "disk.totalWriteLatency.average",
      "disk.write.average",
      "mem.active.average",
      "mem.latency.average",
      "mem.state.latest",
      "mem.swapin.average",
      "mem.swapinRate.average",
      "mem.swapout.average",
      "mem.swapoutRate.average",
      "mem.totalCapacity.average",
      "mem.usage.average",
      "mem.vmmemctl.average",
      "net.bytesRx.average",
      "net.bytesTx.average",
      "net.droppedRx.summation",
      "net.droppedTx.summation",
      "net.errorsRx.summation",
      "net.errorsTx.summation",
      "net.usage.average",
      "power.power.average",
      "storageAdapter.numberReadAveraged.average",
      "storageAdapter.numberWriteAveraged.average",
      "storageAdapter.read.average",
      "storageAdapter.write.average",
      "sys.uptime.latest",
    ]
 # host_metric_exclude = [] ## Nothing excluded by default
 # host_instances = true ## true by default
 #
 ## Clusters
 cluster_metric_include = [] ## if omitted or empty, all metrics are collected
 # cluster_metric_exclude = [] ## Nothing excluded by default
 # cluster_instances = false ## false by default
 #
 ## Datastores
 datastore_metric_include = [] ## if omitted or empty, all metrics are collected
 # datastore_metric_exclude = [] ## Nothing excluded by default
 # datastore_instances = false ## false by default for Datastores only
 #
 ## Datacenters
 datacenter_metric_include = [] ## if omitted or empty, all metrics are collected
# datacenter_metric_exclude = [ "*" ] ## Datacenters are not collected by default.
 # datacenter_instances = false ## false by default for Datastores only
 #
 ## Plugin Settings
 ## separator character to use for measurement and field names (default: "_")
 # separator = "_"
 #
 ## number of objects to retreive per query for realtime resources (vms and hosts)
 ## set to 64 for vCenter 5.5 and 6.0 (default: 256)
 # max_query_objects = 256
 #
 ## number of metrics to retreive per query for non-realtime resources (clusters and datastores)
 ## set to 64 for vCenter 5.5 and 6.0 (default: 256)
 # max_query_metrics = 256
 #
 ## number of go routines to use for collection and discovery of objects and metrics
 # collect_concurrency = 1
 # discover_concurrency = 1
 #
 ## whether or not to force discovery of new objects on initial gather call before collecting metrics
 ## when true for large environments this may cause errors for time elapsed while collecting metrics
 ## when false (default) the first collection cycle may result in no or limited metrics while objects are discovered
 # force_discover_on_init = false
 #
 ## the interval before (re)discovering objects subject to metrics collection (default: 300s)
 # object_discovery_interval = "300s"
 #
 ## timeout applies to any of the api request made to vcenter
 # timeout = "60s"
 #
 ## Optional SSL Config
 # ssl_ca = "/path/to/cafile"
 # ssl_cert = "/path/to/certfile"
 # ssl_key = "/path/to/keyfile"
 ## Use SSL but skip chain & host verification
 insecure_skip_verify = true
 

4、启动服务

systemctl restart telegraf

5、设置开机启动

systemctl enable telegraf

 

二：Influx安装配置

1、安装influxdb

rpm -ivh https://dl.influxdata.com/influxdb/releases/influxdb-1.8.1.x86_64.rpm

2、启动服务&开机自启动

systemctl start influxdb

systemctl enable influxdb

3、登录数据库，创建用户和授权

[root@Ansible ~]# influx

> create database vmware;

> use vmware;

> create user "vmware" with password 'P@ssw0rd' with all privileges;

> show users;

user admin

---- -----

vmware true

4、开启http登录授权

修改/etc/influxdb/influxdb.conf 文件中的[http]部分，将以下两个参数设置为true

[http]

# Determines whether HTTP endpoint is enabled.

enabled = true

 

# Determines whether user authentication is enabled over HTTP/HTTPS.

auth-enabled = true

重启生效

systemctl restart influxdb

5、 使用用户名，密码登录

[root@aly ~]# influx -username vmware -password 'P@ssw0rd'

Connected to http://localhost:8086 version 1.8.1

InfluxDB shell version: 1.8.1

> show databases;

name: databases

name

----

_internal

vmware

三：安装grafana

1、安装grafana

rpm -ivh https://dl.grafana.com/oss/release/grafana-7.4.3-1.x86_64.rpm

2、启动服务&开机自启动

systemctl start grafana

systemctl enable grafana

3、登录grafana

http://grafana_ip:3000/，默认用户名和密码都是admin，第一次登录需要更改密码。

4、配置数据源

点击左侧齿轮图标，开始添加数据源

在下拉菜单中选择InfluxDB

配置InfluxDB数据源名称和HTTP URL信息

配置InfluxDB的数据库和账号密码信息

点击下方的测试按钮，出现绿色说明数据源连接成功

四、导入模板

1、登录grafana，点击左侧的Manage

2、点击新建文件夹，这里命名为 VMware

3、点击右侧的Import，开始导入模板

4、导入模板编号

四：最终查看效果图

①集群视图

②主机视图

③存储视图

④虚拟机视图