marathon-autoscaling功能研究

1.概述:

目前marathon主要支持两种autoscaling:1.基于cpu、内存使用率的autoscaling;2.基于服务请求频率的autoscaling。下面将就这两种autoscaling的原理和用法进行详述。

2.基于CPU和内存使用率的autoscaling

2.1原理

app autoscaling的实现主要依靠一个运行在主机上名为marathon-autoscale.py的Python脚本,不断监测app中cpu和内存的使用率,当使用率达到阈值时自动扩展app中instance的数量。这个脚本启动时主要配置好几个参数:Marathon的地址、要监测并自动扩展的app名字、这个app所有instance中平均内存使用率的上限、这个app所有instance中平均CPU使用率的上限、自动扩展的Instance数量的上限。

通过看脚本的代码可以看到脚本中有一个marathon的class,通过传入marathon的url来初始化它,如下图所示:

 

在main函数中可以看到使用marathon类创建了一个aws_marathon的对象,然后调用这个对象的get_app_detail函数返回一个所有APP的task和所在host的信息。之后主要就是通过这个两个信息来调用get_task_agentstatistics(task,host)函数从http://'+host +':5051/monitor/statistics.json获取每个instance上cpu和内存的使用情况。

以上分析得出结论,其实最终是调用mesos的监控机制,从mesos-slave上获取app中各instance上cpu和内存的使用率的。想了解Mesos的monitoring机制请看:http://mesos.apache.org/documentation/latest/monitoring/

 

 

2.2使用

1.创建app

on 10.134.29.141

•        1.编辑.json文件

vim nginx-autoscaling.json

{

  "id": "nginx-autoscaling”,

  "Container": {

    "type": "Docker",

    "docker": {

      "image":"10.133.19.25:5000/library/nginx:latest",

      "network": "BRIDGE",

      "portMappings": [

        { "hostPort": 0,"containerPort": 80, "servicePort": 10002 }

      ],

      "forcePullImage":true

    }

  },

  "instances": 1,

  "cpus": 0.1,

  "mem": 65,

  "healthChecks": [{

      "protocol": "HTTP",

      "path": "/",

      "portIndex": 0,

      "timeoutSeconds": 10,

      "gracePeriodSeconds": 10,

      "intervalSeconds": 2,

      "maxConsecutiveFailures": 10

  }],

  "labels":{

   "HAPROXY_GROUP":"external,internal"

  }

}

•        2.启动app

dcosmarathon app add nginx-autoscaling.json

 

2.启动autoscaleron 10.134.29.134:

•        1.获取源码

Git clonehttps://github.com/mesosphere/marathon-autoscale.git

cdmarathon-autoscale

•        2.运行autoscale脚本

pythonmarathon-autoscale.py

       Enter the DNS hostname or IP of yourMarathon Instance : 10.134.29.134

       Enter the Marathon Application Name toConfigure Autoscale for from the Marathon UI : nginx-autoscaling

       Enter the Max percent of Mem Usageaveraged across all Application Instances to trigger Autoscale (ie. 80) : 5

       Enter the Max percent of CPU Usageaveraged across all Application Instances to trigger Autoscale (ie. 80) : 5

       Enter which metric(s) to triggerAutoscale ('and', 'or') : or

       Enter Autoscale multiplier for triggeredAutoscale (ie 1.5) : 2

       Enter the Max instances that should everexist for this application (ie. 20) : 10

3.验证:

•        1.进入docker

docker exec-it fb700c6045ad /bin/bash

•        2.加压

ddif=/dev/zero of=test bs=1M count=1000

此时查看会发现多了instance

3. 基于服务请求频率的autoscaling

3.1原理

这个的原理是启动一个marathon-lb-autoscale的app来查看Haproxy监测到的对某一app的请求频率。请求频率超过阈值便自动扩展。详细了解请见:

https://docs.mesosphere.com/administration/admin-tutorials/autoscaling-with-marathon/marathon-app-for-rate-based-autoscaling/

https://mesosphere.com/blog/2015/12/13/service-discovery-and-load-balancing-with-dcos-and-marathon-lb-part-2/

 

3.2使用

1.创建测试扩展的nginx app

2.创建marathon-lb-autoscaleapp

on 10.134.29.141:

•        1.编辑app配置文件.json

vimmarathon-lb-autoscale.json

{

  "id":"marathon-lb-autoscale",

  "args":[

    "--marathon","http://10.134.29.134:8080",

    "--haproxy","http://marathon-lb.marathon.mesos:9090",

    "--target-rps", "100",

    "--apps", "nginx_10002"

  ],

  "cpus": 0.1,

  "mem": 16.0,

  "instances": 1,

  "container": {

    "type": "DOCKER",

    "docker": {

      "image":"docker.io/mesosphere/marathon-lb-autoscale",

      "network": "HOST",

      "forcePullImage": true

    }

  }

}

•        2.启动app

dcosmarathon app add marathon-lb-autoscale.json

 

3.验证

先将siege镜像用docker pull下载并上传到私有仓库中:

docker pull yokogawa/siege

docker tag docker.io/yokogawa/siege:latest10.133.19.25:5000/siege:latest

docker push10.133.19.25:5000/siege:latest(如果上传不成功的话需要上传的主机的docker启动参数加上--insecure-registry=10.133.19.25:5000)

•        1.创建一个siege app用来发送http请求进行测试:

vim siege.json

{

 "id": "siege",

 "args":[

   "-d1",

   "-r1000",

   "-c100",

   "http://10.134.29.142:10000/"

  ],

 "cpus": 0.5,

 "mem": 16.0,

 "instances": 1,

 "container": {

   "type": "DOCKER",

   "volumes": [],

   "docker": {

     "image": "10.133.19.25:5000/siege:latest",

     "network": "HOST",

     "privileged": false,

     "parameters": [],

     "forcePullImage": false

    }

  }

}

 

dcos marathon app add siege.json

•        2.加大request数量:

dcos marathon app update /siege instances=15

此时查看会发现多了nginxinstance

 

http://blog.csdn.net/liukuan73/article/details/50769790

猜你喜欢

转载自m635674608.iteye.com/blog/2345274