docker高级应用之资源监控

最近忙着开发docker平台，所以挺久没有更新博客了，今天给大家分享一下，我开发docker平台里如何监控资源与进行图像展示的。
默认docker 1.5版本有stats命令查看容器的cpu使用率、内存使用量与网络流量，但此功能有2个必须：
1、必须是docker 1.5版本
2、必须使用默认docker0的网桥（如果你使用ovs这样非原生的网桥无法获取数据的）
我开发的监控里docker是1.5版本，并且通过使用ifconfig来获取容器rx或rx量来获取容器流量，解决了必须使用docker默认网桥才可以获取流量数据。
下面是容器资源监控效果图
1、平台里资源监控界面



2、查看容器yangjing-test的cpu使用率资源监控



3、查看内存使用量资源监控



4、查看容器网络流量信息



下面是监控数据收集脚本信息
使用python写的,由于需要往mysql里写入数据，所以需要安装MySQLdb模块以及服务端mysql开启账号

[root@ip-10-10-29-201 code]# cat collect_docker_monitor_data_multi.py
#!/usr/bin/env python
#-*- coding: utf-8 -*-
#author:Deng Lei
#email: dl528888@gmail.com
from docker import Client
import os
import socket, struct, fcntl
import etcd
import MySQLdb
import re
import multiprocessing
import subprocess
import time
def get_local_ip(iface = 'em1'):
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sockfd = sock.fileno()
SIOCGIFADDR = 0x8915
ifreq = struct.pack('16sH14s', iface, socket.AF_INET, '\x00'*14)
try:
res = fcntl.ioctl(sockfd, SIOCGIFADDR, ifreq)
except:
return None
ip = struct.unpack('16sH2x4s8x', res)[2]
return socket.inet_ntoa(ip)
def docker_container_all():
docker_container=docker_client.containers(all=True)
container_name=[]
container_stop_name=[]
for i in docker_container:
container_name.append(i['Names'])
for b in container_name:
for c in b:
container_stop_name.append(c[1::])
return container_stop_name
def docker_container_run():
docker_container=docker_client.containers(all=True)
container_name=[]
container_stop_name=[]
for i in docker_container:
if re.match('Up',i['Status']):
container_name.append(i['Names'])
for b in container_name:
for c in b:
container_stop_name.append(c[1::])
return container_stop_name
def check_container_stats(name):
container_collect=docker_client.stats(name)
old_result=eval(container_collect.next())
new_result=eval(container_collect.next())
container_collect.close()
cpu_total_usage=new_result['cpu_stats']['cpu_usage']['total_usage'] - old_result['cpu_stats']['cpu_usage']['total_usage']
cpu_system_uasge=new_result['cpu_stats']['system_cpu_usage'] - old_result['cpu_stats']['system_cpu_usage']
cpu_num=len(old_result['cpu_stats']['cpu_usage']['percpu_usage'])
cpu_percent=round((float(cpu_total_usage)/float(cpu_system_uasge))*cpu_num*100.0,2)
mem_usage=new_result['memory_stats']['usage']
mem_limit=new_result['memory_stats']['limit']
mem_percent=round(float(mem_usage)/float(mem_limit)*100.0,2)
#network_rx_packets=new_result['network']['rx_packets']
#network_tx_packets=new_result['network']['tx_packets']
network_check_command="""docker exec %s ifconfig eth1|grep bytes|awk -F ':' '{print $2,$3}'|awk -F '(' '{print $1,$2}'|awk -F ')' '{print $1}'|awk '{print "{\\"rx\\":"$1",\\"tx\\":"$2"}"}'"""%name
network_old_result=eval(((subprocess.Popen(network_check_command,shell=True,stdout=subprocess.PIPE)).stdout.readlines()[0]).strip('\n'))
time.sleep(1)
network_new_result=eval(((subprocess.Popen(network_check_command,shell=True,stdout=subprocess.PIPE)).stdout.readlines()[0]).strip('\n'))
#unit KB
network_rx_packets=(int(network_new_result['rx']) - int(network_old_result['rx']))/1024
network_tx_packets=(int(network_new_result['tx']) - int(network_old_result['tx']))/1024
collect_time=str(new_result['read'].split('.')[0].split('T')[0])+' '+str(new_result['read'].split('.')[0].split('T')[1])
msg={'Container_name':name,'Cpu_percent':cpu_percent,'Memory_usage':mem_usage,'Memory_limit':mem_limit,'Memory_percent':mem_percent,'Network_rx_packets':network_rx_packets,'Network_tx_packets':network_tx_packets,'Collect_time':collect_time}
#write_mysql(msg)
return msg
def write_mysql(msg):
container_name=msg['Container_name']
search_sql="select dc.id from docker_containers dc,docker_physics dp where dc.container_name='%s' and dp.physics_internal_ip='%s';"%(container_name,local_ip)
n=mysql_cur.execute(search_sql)
container_id=[int(i[0]) for i in mysql_cur.fetchall()][0]
insert_sql="insert into docker_monitor(container_id,cpu_percent,memory_usage,memory_limit,memory_percent,network_rx_packets,network_tx_packets,collect_time) values('%s','%s','%s','%s','%s','%s','%s','%s');"%(container_id,msg['Cpu_percent'],msg['Memory_usage'],msg['Memory_limit'],msg['Memory_percent'],msg['Network_rx_packets'],msg['Network_tx_packets'],msg['Collect_time'])
n=mysql_cur.execute(insert_sql)
if __name__ == "__main__":
local_ip=get_local_ip('ovs1')
if local_ip is None:
local_ip=get_local_ip('em1')
etcd_client=etcd.Client(host='127.0.0.1', port=4001)
docker_client = Client(base_url='unix://var/run/docker.sock', version='1.17')
mysql_conn=MySQLdb.connect(host='10.10.27.10',user='ops',passwd='1FE@!#@NVE',port=3306,charset="utf8")
mysql_cur=mysql_conn.cursor()
mysql_conn.select_db('devops')
#docker_container_all_name=docker_container_all()
docker_container_run_name=docker_container_run()
if len(docker_container_run_name) == 1:
num=1
elif len(docker_container_run_name) >= 4 and len(docker_container_run_name) <=8:
num=4
elif len(docker_container_run_name) >8 and len(docker_container_run_name) <=15:
num=8
elif len(docker_container_run_name) >15 and len(docker_container_run_name) <=30:
num=20
else:
num=40
pool = multiprocessing.Pool(processes=num)
scan_result=[]
#collect container monitor data
for i in docker_container_run_name:
pool.apply_async(check_container_stats, (i,))
scan_result.append(pool.apply_async(check_container_stats, (i,)))
pool.close()
pool.join()
result=[]
for res in scan_result:
if res.get() is not None:
write_mysql(res.get())
else:
print 'fail is %s'%res.get()
mysql_conn.commit()
mysql_cur.close()
mysql_conn.close()

下面是把此脚本放入crontab里每分钟收集一下

*/1 * * * * python /root/collect_docker_monitor_data_multi.py >>/root/docker_log/docker_monitor.log 2>&1

另外说明一下，上面的监控数据图形化使用highstock，使用ajax进行动态加载数据，每次获取容器所有时间监控数据。有问题请留言。

本文出自 “吟―技术交流” 博客，请务必保留此出处http://dl528888.blog.51cto.com/2382721/1635951