首页 > 代码库 > python 增量同步远程文件夹
python 增量同步远程文件夹
因为发的时候,有些敏感信息没去除,所以大家发文的时候,切记要过滤。不然抓取的很厉害。
这个脚本的作用是同步远程机器的少量log到一台机器上,然后通过logstash过滤显示
犯了好几个错误
local写错locals,locals是python特有定义变量的,不能重名
sftp老报错,IOError: [Errno 2] No such file,这个是因为没有创建本地目录
第一次全量同步,第二次同步五分钟之内修改的文件
特别对列表的处理,用strip去除\n
read()和readlines的区别,其实python基础编程类书籍有讲到
演示
cat filelist
x.x.x.x::password::tomcat::/home/python/::/opt/logs/::
#!/usr/bin/python import pexpect import paramiko import os import sys import time import multiprocessing import datetime import crash_on_ipy from stat import S_ISDIR ip_list = [] #room_id = sys.argv[1] class run_cmd(): def __init__(self,hostname=None,password=None,username=None,port=None,echo_cmd=None): #threading.Thread.__init__(self) self.hostname=hostname self.password=password self.username=username self.port=port self.echo_cmd=echo_cmd #self.thread_stop=False def run(self): paramiko.util.log_to_file(‘paramiko.log‘) s=paramiko.SSHClient() s.set_missing_host_key_policy(paramiko.AutoAddPolicy()) s.connect(hostname = self.hostname,username=self.username, password=self.password) stdin,stdout,stderr=s.exec_command(self.echo_cmd) return stdout.readlines() s.close() def stop(self): self.thread_stop=True class get_thread(): def __init__(self,hostname,password,username,port=None): #def __init__(self,hostname,username=‘root‘,key_file=None,password=None): #def __init__(self,hostname=None,password=None,username=None,port=None,local_dir=None,remote_dir=None): self.hostname = hostname self.username = username self.password = password self.scp = paramiko.SSHClient() self.scp.set_missing_host_key_policy(paramiko.AutoAddPolicy()) self.scp = paramiko.Transport((hostname, 22)) self.scp.connect(username=username, password=password) self.sftp = paramiko.SFTPClient.from_transport(self.scp) def _walk_remote(self, dirpath): dirnames = [] filenames = [] for fd in self.sftp.listdir_attr(dirpath): if S_ISDIR(fd.st_mode): dirnames.append(fd.filename) else: filenames.append(fd.filename) yield dirpath, dirnames, filenames for dirname in dirnames: new_dirpath = os.path.join(dirpath, dirname) # yield from self._walk_remote(new_dirpath) for walk in self._walk_remote(new_dirpath): yield walk def getall(self,local,remote): st_mode = self.sftp.stat(remote).st_mode if not S_ISDIR(st_mode): filename = os.path.basename(remote) self.sftp.get(remote, os.path.join(local, filename)) else: parent, child = os.path.split(remote) for dirpath, dirnames, filenames in self._walk_remote(remote): dirpath = dirpath.replace(parent, ‘.‘) parentc = os.path.join(local,dirpath) if not os.path.exists(parentc): os.makedirs(parentc) for dirname in dirnames: try: os.makedirs(os.path.join(local, dirpath, dirname)) except: pass for filename in filenames: localpath = os.path.join(local, dirpath, filename) remotepath = os.path.join(parent, dirpath, filename) self.sftp.get(remotepath, localpath) self.scp.close() if __name__==‘__main__‘: port = 22 now = datetime.datetime.now() strdatetime = now.strftime("%Y-%m-%d") year=strdatetime.split(‘-‘)[0] mon=strdatetime.split(‘-‘)[1] day=strdatetime.split(‘-‘)[2] Datenow1= year + "/" + mon + "/" + day + "/" Datenow= year + "/" + mon print "-"*50 f = file(‘/home/python/filelist‘,‘r‘) c = f.readlines() for x in c: hostname = x.split(‘::‘)[0] password = x.split(‘::‘)[1] username = x.split(‘::‘)[2] local= x.split(‘::‘)[3].strip(‘\n‘) remotes = x.split(‘::‘)[4].strip(‘\n‘) localz=local + "/" + mon + "/" + day if remotes.endswith(‘/‘): remote1 = remotes + Datenow remote2 = remotes + Datenow1 else: remote3 = remotes if not os.path.exists(localz): remote = remote1 getthread=get_thread(hostname,password,username) getthread.getall(local,remote) else: remote = remote2 echo_cmd=‘/bin/find %s -maxdepth 1 -type d -mmin -5‘ % (remote) cmd_thread=run_cmd(hostname,password,username,port,echo_cmd) result=cmd_thread.run() del result[0] for item in result: print str(item) items = item.strip(‘\n‘) getthread=get_thread(hostname,password,username) getthread.getall(localz,items) f.close()
执行time python test.py
最主要可以增量五分钟修改的,节约时间。
本文出自 “人,要有自己的想法” 博客,谢绝转载!
python 增量同步远程文件夹
声明:以上内容来自用户投稿及互联网公开渠道收集整理发布,本网站不拥有所有权,未作人工编辑处理,也不承担相关法律责任,若内容有误或涉及侵权可进行投诉: 投诉/举报 工作人员会在5个工作日内联系你,一经查实,本站将立刻删除涉嫌侵权内容。