如何采用Python zabbix_api 获取性能数据
如何采用Python zabbix_api 获取性能数据
最近领导需要一份数据，OpenStack ，VMware，物理机之间的性能报告，在撰写报告之前需要数据支撑，我们采用的是zabbix 监控，需要采取一周内的历史数据作为对比，那数据如何获取，请看以下章节
一：安装zabbix api 接口，修改zabbix api 调用接口，获取数据、
*****************************************************************************************
easy_install zabbix_api
*****************************************************************************************
#!/usr/bin/python
# The research leading to these results has received funding from the
# European Commission's Seventh Framework Programme (FP7/2007-13)
# under grant agreement no 257386.
# http://www.bonfire-project.eu/
# Copyright 2012 Yahya Al-Hazmi, TU Berlin
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License
# this script fetches resource monitoring information from Zabbix-Server
# through Zabbix-API
#
# To run this script you need to install python-argparse "apt-get install python-argparse"
from zabbix_api import ZabbixAPI
import sys
import datetime
import time
import argparse
def fetch_to_csv(username,password,server,hostname,key,output,datetime1,datetime2,debuglevel):
zapi = ZabbixAPI(server=server, log_level=debuglevel)
try:
zapi.login(username, password)
except:
print "zabbix server is not reachable: %s" % (server)
sys.exit()
host = zapi.host.get({"filter":{"host":hostname}, "output":"extend"})
if(len(host)==0):
print "hostname: %s not found in zabbix server: %s, exit" % (hostname,server)
sys.exit()
else:
hostid=host[0]["hostid"]
print '*' * 100
print key
print '*' * 100
if(key==""):
print '*' * 100
items = zapi.item.get({"filter":{"hostid":hostid} , "output":"extend"})
if(len(items)==0):
print "there is no item in hostname: %s, exit" % (hostname)
sys.exit()
dict={}
for item in items:
dict[str(item['itemid'])]=item['key_']
if (output == ''):
output=hostname+".csv"
f = open(output, 'w')
str1="#key;timestamp;value\n"
if (datetime1=='' and datetime2==''):
for itemid in items:
itemidNr=itemid["itemid"]
str1=str1+itemid["key_"]+";"+itemid["lastclock"]+";"+itemid["lastvalue"]+"\n"
f.write(str1)
print "Only the last value from each key has been fetched, specify t1 or t1 and t2 to fetch more data"
f.close()
elif (datetime1!='' and datetime2==''):
try:
d1=datetime.datetime.strptime(datetime1,'%Y-%m-%d %H:%M:%S')
except:
print "time data %s does not match format Y-m-d H:M:S, exit" % (datetime1)
sys.exit()
timestamp1=time.mktime(d1.timetuple())
timestamp2=int(round(time.time()))
inc=0
history = zapi.history.get({"hostids":[hostid,],"time_from":timestamp1,"time_till":timestamp2, "output":"extend" })
for h in history:
str1=str1+dict[h["itemid"]]+";"+h["clock"]+";"+h["value"]+"\n"
inc=inc+1
f.write(str1)
f.close()
print str(inc) +" records has been fetched and saved into: " + output
elif (datetime1=='' and datetime2!=''):
for itemid in items:
itemidNr=itemid["itemid"]
str1=str1+itemid["key_"]+";"+itemid["lastclock"]+";"+itemid["lastvalue"]+"\n"
f.write(str1)
print "Only the last value from each key has been fetched, specify t1 or t1 and t2 to fetch more data"
f.close()
else:
try:
d1=datetime.datetime.strptime(datetime1,'%Y-%m-%d %H:%M:%S')
except:
print "time data %s does not match format Y-m-d H:M:S, exit" % (datetime1)
sys.exit()
try:
d2=datetime.datetime.strptime(datetime2,'%Y-%m-%d %H:%M:%S')
except:
print "time data %s does not match format Y-m-d H:M:S, exit" % (datetime2)
sys.exit()
timestamp1=time.mktime(d1.timetuple())
timestamp2=time.mktime(d2.timetuple())
inc=0
history = zapi.history.get({"hostids":[hostid,],"time_from":timestamp1,"time_till":timestamp2, "output":"extend" })
for h in history:
str1=str1+dict[h["itemid"]]+";"+h["clock"]+";"+h["value"]+"\n"
inc=inc+1
f.write(str1)
f.close()
print str(inc) +" records has been fetched and saved into: " + output
else:
#print "key is: %s" %(key)
itemid = zapi.item.get({"filter":{"key_":key, "hostid":hostid} , "output":"extend"})
if(len(itemid)==0):
print "item key: %s not found in hostname: %s" % (key,hostname)
sys.exit()
itemidNr=itemid[0]["itemid"]
if (output == ''):
output=hostname+".csv"
f = open(output, 'w')
str1="#key;timestamp;value\n"
if (datetime1=='' and datetime2==''):
str1=str1+key+";"+itemid[0]["lastclock"]+";"+itemid[0]["lastvalue"]+"\n"
#f.write(str1)
f.write(str1)
f.close()
print "Only the last value has been fetched, specify t1 or t1 and t2 to fetch more data"
elif (datetime1!='' and datetime2==''):
d1=datetime.datetime.strptime(datetime1,'%Y-%m-%d %H:%M:%S')
timestamp1=time.mktime(d1.timetuple())
timestamp2=int(round(time.time()))
history = zapi.history.get({"history":itemid[0]["value_type"],"time_from":timestamp1,"time_till":timestamp2, "itemids":[itemidNr,], "output":"extend" })
inc=0
for h in history:
str1 = str1 + key + ";" + h["clock"] +";"+h["value"] + "\n"
inc=inc+1
f.write(str1)
f.close()
print str(inc) +" records has been fetched and saved into: " + output
elif (datetime1=='' and datetime2!=''):
str1=str1+key+";"+itemid[0]["lastclock"]+";"+itemid[0]["lastvalue"]+"\n"
f.write(str1)
f.close()
print "Only the last value has been fetched, specify t1 or t1 and t2 to fetch more data"
else:
d1=datetime.datetime.strptime(datetime1,'%Y-%m-%d %H:%M:%S')
d2=datetime.datetime.strptime(datetime2,'%Y-%m-%d %H:%M:%S')
timestamp1=time.mktime(d1.timetuple())
timestamp2=time.mktime(d2.timetuple())
history = zapi.history.get({"history":itemid[0]["value_type"],"time_from":timestamp1,"time_till":timestamp2, "itemids":[itemidNr,], "output":"extend" })
inc=0
for h in history:
str1 = str1 + key + ";" + h["clock"] +";"+h["value"] + "\n"
inc=inc+1
print str(inc) +" records has been fetched and saved into: " + output
f.write(str1)
f.close()
parser = argparse.ArgumentParser(description='Fetch history from aggregator and save it into CSV file')
parser.add_argument('-s', dest='server_IP', required=True,
help='aggregator IP address')
parser.add_argument('-n', dest='hostname', required=True,
help='name of the monitored host')
parser.add_argument('-k', dest='key',default='',
help='zabbix item key, if not specified the script will fetch all keys for the specified hostname')
parser.add_argument('-u', dest='username', default='Admin',
help='zabbix username, default Admin')
parser.add_argument('-p', dest='password', default='zabbix',
help='zabbix password')
parser.add_argument('-o', dest='output', default='',
help='output file name, default hostname.csv')
parser.add_argument('-t1', dest='datetime1', default='',
help='begin date-time, use this pattern \'2011-11-08 14:49:43\' if only t1 specified then time period will be t1-now ')
parser.add_argument('-t2', dest='datetime2', default='',
help='end date-time, use this pattern \'2011-11-08 14:49:43\'')
parser.add_argument('-v', dest='debuglevel', default=0, type=int,
help='log level, default 0')
args = parser.parse_args()
fetch_to_csv(args.username, args.password, "http://"+args.server_IP+"/zabbix", args.hostname, args.key, args.output, args.datetime1,args.datetime2,args.debuglevel)
二：撰写通过key获取一周内的数据
items : 在zabbix 中搜索主机，选择最新数据，找到项目(items)，点击进入，能看到机器的所有keys，在负载到程序的items 字典中，程序会循环读取items ，获取数据
#/usr/bin/env python
#-*-coding:UTF-8
"""
wget http://doc.bonfire-project.eu/R4.1/_static/scripts/fetch_items_to_csv.py
"""
import os,sys,time
users=u'admin'
pawd = 'zabbix'
exc_py = '/data/zabbix/fetch_items_to_csv.py'
os.system('easy_install zabbix_api')
os.system('mkdir -p /data/zabbix/cvs/')
if not os.path.exists(exc_py):
os.system("mkdir -p /data")
os.system("wget http://doc.bonfire-project.eu/R4.1/_static/scripts/fetch_items_to_csv.py -O /data/zabbix/fetch_items_to_csv.py")
def show_items(moniter, dip):
items = dict()
items['io_read_win'] = "perf_counter[\\2\\16]"
items['io_write_win'] = "perf_counter[\\2\\18]"
items['io_read_lin'] = "iostat[,rkB/s]"
items['io_write_lin'] = "iostat[,wkB/s]"
items['io_read_lin_sda'] = "iostat[sda,rkB/s]"
items['io_write_lin_sda'] = "iostat[sda,wkB/s]"
items['io_read_lin_sdb'] = "iostat[sdb,rkB/s]"
items['io_write_lin_sdb'] = "iostat[sdb,wkB/s]"
# Add items, iostate vdb,vdb
items['io_read_lin_vda'] = "iostat[vda,rkB/s]"
items['io_write_lin_vda'] = "iostat[vda,wkB/s]"
items['io_read_lin_vdb'] = "iostat[vdb,rkB/s]"
items['io_write_lin_vdb'] = "iostat[vdb,wkB/s]"
items['cpu_util'] = "system.cpu.util"
items['net_in_linu_vm_web'] = "net.if.in[eth0]"
items['net_out_lin_vm_web'] = "net.if.out[eth0]"
items['net_in_linu_vm_db'] = "net.if.in[eth1]"
items['net_out_lin_vm_db'] = "net.if.out[eth1]"
items['net_in_win_vm'] = "net.if.in[Red Hat VirtIO Ethernet Adapter]"
items['net_in_win_vm_2'] = "net.if.in[Red Hat VirtIO Ethernet Adapter #2]"
items['net_in_win_vm_3'] = "net.if.in[Red Hat VirtIO Ethernet Adapter #3]"
items['net_out_win_vm'] = "net.if.out[Red Hat VirtIO Ethernet Adapter]"
items['net_out_win_vm_2'] = "net.if.out[Red Hat VirtIO Ethernet Adapter #2]"
items['net_out_win_vm_3'] = "net.if.out[Red Hat VirtIO Ethernet Adapter #3]"
items['net_in_phy_web_lin'] = "net.if.in[bond0]"
items['net_out_phy_web_lin'] = "net.if.out[bond0]"
items['net_in_phy_db_lin'] = "net.if.in[bond1]"
items['net_out_phy_db_lin'] = "net.if.out[bond1]"
items['net_in_phy_web_win'] = "net.if.in[TEAM : WEB-TEAM]"
items['net_out_phy_web_win'] = "net.if.in[TEAM : WEB-TEAM]"
items['net_in_phy_db_win'] = "net.if.in[TEAM : DB Team]"
items['net_out_phy_db_win'] = "net.if.out[TEAM : DB Team]"
items['net_in_phy_web_win_1'] = "net.if.in[TEAM : web]"
items['net_out_phy_web_win_1'] = "net.if.out[TEAM : web]"
items['net_in_phy_db_win_1'] = "net.if.in[TEAM : DB]"
items['net_out_phy_db_win_1'] = "net.if.out[TEAM : DB]"
items['net_in_win_pro'] = "net.if.in[Intel(R) PRO/1000 MT Network Connection]"
items['net_out_win_pro'] = "net.if.out[Intel(R) PRO/1000 MT Network Connection]"
items['net_in_phy_web_hp'] = "net.if.in[HP Network Team #1]"
items['net_out_phy_web_hp'] = "net.if.out[HP Network Team #1]"
items['iis_conntion'] = "perf_counter[\\Web Service(_Total)\\Current Connections]"
items['tcp_conntion'] = "k.tcp.conn[ESTABLISHED]"
for x,y in items.items():
os.system('mkdir -p /data/zabbix/cvs/%s' % dip)
cmds = """
python /data/zabbix/fetch_items_to_csv.py -s '%s' -n '%s' -k '%s' -u 'admin' -p '%s' -t1 '2015-06-23 00:00:01' -t2 '2015-06-30 00:00:01' -o /data/zabbix/cvs/%s/%s_%s.cvs""" %(moniter,dip,y,pawd,dip,dip,x)
os.system(cmds)
print "*" * 100
print cmds
print "*" * 100
def work():
moniter='192.168.1.1'
ip_list = ['192.168.1.15','192.168.1.13','192.168.1.66','192.168.1.5','192.168.1.7','192.168.1.16','192.168.1.38','192.168.1.2','192.168.1.13','192.168.1.10']
for ip in ip_list:
show_items(moniter,ip )
if __name__ == "__main__":
sc = work()
三：数据采集完毕，进行格式化输出
#!/usr/bin/env python
#-*-coding:utf8-*-
import os,sys,time
workfile = '/home/zabbix/zabbix/sjz/'
def collect_info():
dict_doc = dict()
for i in os.listdir(workfile):
dict_doc[i] = list()
for v in os.listdir('%s%s' %(workfile,i)):
dict_doc[i].append(v)
count = 0
for x,y in dict_doc.items():
for p in y:
fp = '%s/%s/%s' %(workfile,x,p)
op = open(fp,'r').readlines()
np = '%s.txt' %p
os.system( """ cat %s|awk -F";" '{print $3}' > %s """ %(fp,np))
count += 1
print count
if __name__ == "__main__":
sc = collect_info()
四，整理数据，汇报成图形，撰写技术报告
一些Zabbix相关教程集合：
Ubuntu 14.04下Zabbix2.4.5 源码编译安装
安装部署分布式监控系统Zabbix 2.06
《安装部署分布式监控系统Zabbix 2.06》
CentOS 6.3下Zabbix安装部署
Zabbix分布式监控系统实践
CentOS 6.3下Zabbix监控apache server-status
CentOS 6.3下Zabbix监控MySQL数据库参数
64位CentOS 6.2下安装Zabbix 2.0.6
ZABBIX 的详细介绍：请点这里
ZABBIX 的下载地址：请点这里
本文永久更新链接地址：
评论暂时关闭