1. nagios被動監控傳遞狀態碼
#!/usr/bin/python
import sys,time,os
DATE=time.strftime("%Y-%m-%d",time.localtime())
FILE='/home/OffLineRanking/'+'topqueryNormalize.log'+DATE
SUM=int(os.popen('cat %s |grep ERROR |wc -l' %FILE).read())
if SUM>=100:
print "error log is %s" % SUM
sys.exit(2)
else:
print"ok"
sys.exit(0)
知識點：
1.time.strftime（）格式化輸出時間格式，time.localtime()獲取本地時間，還有一個time.time()獲取秒級別，例如：DATE=time.strftime("%Y-%m-%d",time.localtime(time.time()))
2.sys.exit(),返回命令執行結果的狀態碼，相當於shell中的exit ，這樣可以根據echo $? 的值進行後期的判斷處理
3.python操作命令並獲取返回值。有很多模塊可以完成這個功能，我這裏使用的os.popen(),注意一下int()以及os.popen()內部FILE變量的傳遞
4.字符串的拼加
2.python操作ssh
這個命題使用shell也可以完成很多功能，使用py完全是爲了學習它操作ssh的方式。
一般py操作ssh有兩種密碼認證方式，明文的用戶名和密碼以及ssh-key的形式，我這裏使用明文進行操作，儘管我們線上ssh通道是完全打通的：
#!/usr/bin/python
####just a test script####
import paramiko,sys
x=raw_input("input the log data:")
if x.isdigit()==True:
for i in [xx1.alibaba.com','xx2.alibaba.com','xx3..alibaba.com','xx4.alibaba.com']:
file='access_log_'+x
ssh=paramiko.SSHClient()
ssh.set_missing_host_key_policy(paramiko.AutoAddPolicy())
ssh.connect(i,username='test',password='1234')
stdin, stdout, stderr = ssh.exec_command('cd /home/logs/nginx/;cat %s |grep error |wc -l ' %file)
for j in stdout.readlines():
print j,
知識點：
1.還是注意file的傳遞方式，這個方法並不通用
2.print j後面的逗號
3.操作http
回頭寫吧，我先看完14章
4.監控系統進程腳本
#!/usr/bin/python
import os,sys
def checkserver(svr,msg):
num_cmd="ps -eo comm |grep %s -c"%svr
SERVER_NUM=int(os.popen(num_cmd).read())
pid_cmd="ps -ef |grep %s |grep -v grep |awk '{print $2}'"%svr
strf_pid=os.popen(pid_cmd).read()
format_pid=strf_pid.replace('\n',' ').split(' ')[:-2]
for i in format_pid:
PATH='/proc/'+i+"/status"
STATS=os.popen('grep [Zz]ombie %s'%PATH).read().lower()
if "zombie" not in STATS and SERVER_NUM==21:
stdout="%s is ok!"%msg
exit_code=0
continue
else:
print "%s is error!"%msg
sys.exit(2)
print stdout
sys.exit(exit_code)
if __name__ == '__main__':
if len(sys.argv)<1:
print "input error!"
sys.exit(-1)
else:
checkserver(sys.argv[1],sys.argv[2])
5.系統進程數監控
#!/usr/bin/env python
import os,sys
try:
cmd=int(os.popen('ps -eo comm |grep %s -c'%sys.argv[1]).read().strip("\n"))
if cmd!=int(sys.argv[2]):
print "%s num is abnormal"%sys.argv[1]
sys.exit(2)
else:
print "%s num is correct"%sys.argv[1]
sys.exit(0)
except (IndexError,ValueError,SyntaxError):
print "useage:%s servername servernumber"%sys.argv[0]
sys.exit(2)
6.監控core file
#!/usr/bin/env python
import sys,os
try:
m=[i for i in os.listdir(sys.argv[1]) if "core." in i]
if m:
print "%s core now!"%sys.argv[2]
sys.exit(2)
else:
print "%s status ok!"%sys.argv[2]
sys.exit(0)
except (IndexError,ValueError,SyntaxError,OSError):
print "useage:%s corepath servername"%sys.argv[0]
sys.exit(2)
7.監控tair空間
#!/usr/bin/env python
#coding: UTF-8
import sys,os,socket
def formatdata(k):
hostname=socket.getfqdn(k.split(" ")[0])
data=k.split(" ")[1].strip()
return hostname,data
def tairspace(cmd):
m=os.popen(cmd).readlines()
if not m:
print "tair usespace is ok!"
sys.exit(0)
elif len(m)==1:
j=formatdata(m[0])
print "%s tairspace is %s%%"%(j[0],(int(j[1])/(monitor_size/100)))
sys.exit(2)
else:
k=open('/home/a/logs/spaceerror','w')
for i in m:
j=formatdata(i)
k.writelines("%s %s %s%%\n"%(j[0],j[1],(int(j[1])/(monitor_size/100))))
k.close()
print "%s nods tairspace exceed 80%%,check log for detail!"%len(m)
sys.exit(2)
if __name__=='__main__':
try:
if len(sys.argv)>3:
print "use %s cs_ip:port group_name"%sys.argv[0]
sys.exit(2)
else:
slab_mem=int(os.popen('grep slab_mem `rpm -ql tair |grep dataserver`').read().split("=")[1])
monitor_size=slab_mem*1024*1024
cmd='/opt/csr/tair-2.3/sbin/tairclient -c %s -g %s -l stat 2>&1|grep 172|grep use |awk -F "[: ]" \'{a[$1]+=$7}END{for(i in a){if(a[i]>%s){print i\" \"a[i]}}}\''%(sys.argv[1],sys.argv[2],monitor_size*0.8)
tairspace(cmd)
except IndexError:
print "use %s cs_ip:port group_name"%sys.argv[0]
sys.exit(2)
8.python表示時間日期
昨天.今天.明天
from datetime import *
today=date.today()
yesterday=date.today()-timedelta(days=1)
tommorrow = date.today()+timedelta(days=1)
如下輸出20120101的格式：
print (date.today()-timedelta(days=1)).strftime('%Y%m%d')
9.監控tt2
#!/usr/bin/python
import sys,os,time
def checktt2(path):
cu_date=time.strftime("%Y%m%d%H",time.localtime())
if os.path.isdir('%s'%path):
rpath=os.getcwd()
files=os.listdir(rpath)
for i in files:
# if "tmp" not in i:
if "taobao_ie_" in i:
z=open(i,'r').readlines()
r=z[0].strip('\n')
k=int(z[-1].strip('\n'))
y=os.path.getsize(r)
if y-k<52428800:
print "tt2 status is ok!"
sys.exit(0)
else:
print "tt2 status is error!"
sys.exit(2)
else:
print "tt2 path is not exist!"
sys.exit(2)
if __name__=='__main__':
path='/home/server/tt2'
checktt2(path)
10.socket 簡單通信
server：
if __name__ == '__main__':
import socket,sys
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.bind(('111.4.23.23', 8001))
sock.listen(6)
i=5
all=[]
while True:
try:
connection,address = sock.accept()
connection.settimeout(5)
buf = connection.recv(1024)
if buf !="":
all.append(buf.split(","))
else:
connection.send('please go out!')
sys.exit(2)
except Exception,e:
print str(e)
connection.close()
i=i-1
if i ==0:
print sorted(dict(all).items(),key=lambda d:int(d[0]))
break
client：
#!/usr/bin/python
import time,socket,os,sys
def sum(path):
if os.path.isdir(path):
os.chdir(path)
curr_date=time.strftime('%Y%m%d%H',time.localtime())
file=path+curr_date
if os.path.isfile(file):
lines=os.popen('wc -l %s'%file).read()
lines_format=lines.strip("\n").split(" ")[0]
return lines_format
else:
print "file is not exist!"
sys.exit(2)
else:
print "path is not exist!"
sys.exit(2)
def content(lines_format):
hostname=socket.gethostname()
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.connect(('119.42.236.236',8001))
time.sleep(2)
#lines=lines_format+","+"\'%s\'"%hostname
lines=lines_format+","+hostname
sock.send('%s'%lines)
print sock.recv(1024)
sock.close()
if __name__=='__main__':
path="/home/guizhong/"
date=sum(path)
content(date)