我正在编写一个Python脚本来执行“ TCP Traceroute”。我了解到scapy是执行此操作的有用库，但是我没有得到所需的结果。谁能帮我解决这个问题？我希望python脚本生成与命令行类似的结果。
我正在使用Linux，python 2.7和scapy 2.4。我不确定为什么所有跃点都显示相同的IP地址。
from scapy.layers.inet import traceroute
result, unans = traceroute('172.217.17.46', maxttl=30)
for snd, rcv in result:
print snd.ttl, rcv.src, snd.sent_time, rcv.time
运行此代码时，我得到以下结果：
1 10.0.2.2 1541113255.58 1541113255.6
2 172.217.17.46 1541113255.58 1541113255.72
3 172.217.17.46 1541113255.58 1541113255.72
4 172.217.17.46 1541113255.58 1541113255.72
5 172.217.17.46 1541113255.59 1541113255.73
6 172.217.17.46 1541113255.59 1541113255.73
7 172.217.17.46 1541113255.6 1541113255.74
8 172.217.17.46 1541113255.6 1541113255.74
9 172.217.17.46 1541113255.6 1541113255.74
10 172.217.17.46 1541113255.61 1541113255.75
11 172.217.17.46 1541113255.61 1541113255.75
12 172.217.17.46 1541113255.61 1541113255.75
13 172.217.17.46 1541113255.62 1541113255.76
14 172.217.17.46 1541113255.62 1541113255.76
15 172.217.17.46 1541113255.62 1541113255.76
16 172.217.17.46 1541113255.62 1541113255.77
17 172.217.17.46 1541113255.63 1541113255.77
18 172.217.17.46 1541113255.63 1541113255.77
19 172.217.17.46 1541113255.63 1541113255.77
20 172.217.17.46 1541113255.63 1541113255.77
21 172.217.17.46 1541113255.64 1541113255.78
22 172.217.17.46 1541113255.64 1541113255.78
23 172.217.17.46 1541113255.64 1541113255.78
24 172.217.17.46 1541113255.64 1541113255.78
25 172.217.17.46 1541113255.65 1541113255.79
26 172.217.17.46 1541113255.65 1541113255.79
27 172.217.17.46 1541113255.65 1541113255.79
28 172.217.17.46 1541113255.66 1541113255.8
29 172.217.17.46 1541113255.66 1541113255.8
30 172.217.17.46 1541113255.66 1541113255.8
我想获得与从命令行运行tcptraceroute时得到的结果相同的结果：tcptraceroute 172.217.17.46
命令行结果：
Selected device en0, address 192.168.86.24, port 49618 for outgoing packets
Tracing the path to 172.217.17.46 on TCP port 80 (http), 30 hops max
1 192.168.86.1 2.848 ms 1.224 ms 1.330 ms
2 96.120.101.53 10.423 ms 13.646 ms 12.221 ms
3 po-115-rur102.bellevue.wa.seattle.comcast.net (68.87.205.245) 18.877 ms 18.818 ms 12.593 ms
4 be-103-ar01.seattle.wa.seattle.comcast.net (69.139.164.77) 15.188 ms 14.272 ms 14.005 ms
5 be-33650-cr01.seattle.wa.ibone.comcast.net (68.86.93.165) 14.547 ms 15.273 ms 19.750 ms
6 be-10846-pe01.seattle.wa.ibone.comcast.net (68.86.86.90) 14.546 ms 14.266 ms 13.521 ms
7 50.242.150.242 14.159 ms 15.791 ms 14.037 ms
8 74.125.243.195 14.635 ms 22.377 ms 13.558 ms
9 72.14.236.174 15.051 ms 27.454 ms 14.312 ms
10 108.170.235.60 66.430 ms 69.762 ms 68.606 ms
11 216.239.58.255 85.531 ms 84.354 ms 85.303 ms
12 172.253.51.157 153.310 ms 154.710 ms 153.375 ms
13 209.85.142.166 157.376 ms 166.552 ms 157.562 ms
14 216.239.43.37 170.523 ms 168.040 ms 158.182 ms
15 108.170.241.225 158.953 ms 161.418 ms 169.103 ms
16 108.170.236.137 158.561 ms 161.635 ms 157.510 ms
17 ams16s29-in-f46.1e100.net (172.217.17.46) [open] 165.981 ms 160.451 ms 166.120 ms
问题1：Scapy traceroute功能是否真的是TCP traceroute？ Question2：我对scapy和traceroute是陌生的，代码中明显缺少什么吗？如果scapy不适合，我还可以使用其他库吗？我真的很感谢您的帮助和任何指导。
注意：我想对IPV6和IPV4都执行TCP跟踪路由。
答案 0 :(得分：7)
scapy traceroute函数确实是TCP traceroute吗？
是的，它确实执行TCP跟踪路由（以及其他功能）。看看Scapy source-code：
@conf.commands.register def traceroute(target, dport=80, minttl=1, maxttl=30, sport=RandShort(), l4 = None, filter=None, timeout=2, verbose=None, **kargs): """Instant TCP traceroute traceroute(target, [maxttl=30,] [dport=80,] [sport=80,] [verbose=conf.verb]) -> None""" if verbose is None: verbose = conf.verb if filter is None: # we only consider ICMP error packets and TCP packets with at # least the ACK flag set *and* either the SYN or the RST flag # set filter="(icmp and (icmp[0]=3 or icmp[0]=4 or icmp[0]=5 or icmp[0]=11 or icmp[0]=12)) or (tcp and (tcp[13] & 0x16 > 0x10))" if l4 is None: a,b = sr(IP(dst=target, id=RandShort(), ttl=(minttl,maxttl))/TCP(seq=RandInt(),sport=sport, dport=dport), timeout=timeout, filter=filter, verbose=verbose, **kargs) else: # this should always work filter="ip" a,b = sr(IP(dst=target, id=RandShort(), ttl=(minttl,maxttl))/l4, timeout=timeout, filter=filter, verbose=verbose, **kargs) a = TracerouteResult(a.res) if verbose: a.show() return a,b
tcp和ipv6都执行ipv4跟踪路由 Scapy源代码，traceroot6可以完成这项工作：
def traceroute6(target, dport=80, minttl=1, maxttl=30, sport=RandShort(), l4 = None, timeout=2, verbose=None, **kargs): """ Instant TCP traceroute using IPv6 : traceroute6(target, [maxttl=30], [dport=80], [sport=80]) -> None """ if verbose is None: verbose = conf.verb if l4 is None: a,b = sr(IPv6(dst=target, hlim=(minttl,maxttl))/TCP(seq=RandInt(),sport=sport, dport=dport), timeout=timeout, filter="icmp6 or tcp", verbose=verbose, **kargs) else: a,b = sr(IPv6(dst=target, hlim=(minttl,maxttl))/l4, timeout=timeout, verbose=verbose, **kargs) a = TracerouteResult6(a.res) if verbose: a.display() return a,b
traceroute ipv6的示例代码：
from scapy.all import * waypoint = "2001:301:0:8002:203:47ff:fea5:3085" target = "2001:5f9:4:7:2e0:81ff:fe52:9a6b" traceroute6(waypoint, minttl=10, maxttl=40, l4=IPv6ExtHdrRouting(addresses=[target])/ICMPv6EchoRequest(data=RandString(7)))
使用 DNS traceroute ，通过在traceroute（）函数的l4参数中指定完整的数据包，您不会为所有跃点获得相同的IP地址
from scapy.all import * target = ["172.217.17.46"] result, unans = traceroute(target, l4=UDP(sport=RandShort())/DNS(qd=DNSQR(qname="www.google.com")))
或者，您可以使用 TCP SYN traceroute 获得类似的结果：
from scapy.all import * target = ["172.217.17.46"] result, unans = sr(IP(dst=target, ttl=(1, 10)) / TCP(dport=53, flags="S")) for snd, rcv in result: print(snd.ttl, rcv.src, snd.sent_time, rcv.time)
或者您可以创建 IP and UDP packets 来使用traceroute来实现Scapy，如下所示：
from scapy.all import * hostname = "172.217.17.46" for i in range(1, 28): pkt = IP(dst=hostname, ttl=i) / UDP(dport=33434) reply = sr1(pkt, verbose=0) if reply is None: break elif reply.type == 3: print("Done!", reply.src) break else: print("%d hops away: " % i, reply.src, reply.time)
Scapy不适合使用其他库吗？
import webb webb.traceroute("www.google.com") webb.traceroute("www.google.com",'file-name.txt')
还有check out this tcptraceroute，作者是Thomas Guettler。
或者Addy Yeow（Ayeowch）的此Multi-source traceroute with geolocation实现。
除其他外，使用-j参数（JSON_FILE），它将以JSON文件格式列出源。
或Christian Kreibich的this implementation。它可以将traceroute信息解析为一系列的跃点对象，每个跃点对象都包含一个或多个探测结果，同样是对象实例。同样，字符串格式会产生熟悉的traceroute输出。
（为了使用Python 3，需要将cStringIO更改为from io import StringIO
没有 scapy （使用Windows控制台）：
创建一个名为
output.py的脚本，其中包含以下内容：
import sys
from subprocess import Popen
if len(sys.argv) < 2:
print('Usage: output.py "command to watch"')
sys.exit(1)
cmd_line = sys.argv[1:]
p = Popen(cmd_line)
p.communicate()[0]
示例用法： python output.py ping google.com
ping的示例输出：Pinging google.com [216.58.209.14] with 32 bytes of data: Reply from 216.58.209.14: bytes=32 time=50ms TTL=56 Reply from 216.58.209.14: bytes=32 time=45ms TTL=56 Reply from 216.58.209.14: bytes=32 time=45ms TTL=56 Reply from 216.58.209.14: bytes=32 time=45ms TTL=56 Ping statistics for 216.58.209.14: Packets: Sent = 4, Received = 4, Lost = 0 (0% loss), Approximate round trip times in milli-seconds: Minimum = 45ms, Maximum = 50ms, Average = 46ms
用法示例： python output.py tracert google.com
tracert的示例输出：Tracing route to google.com [172.217.18.174] over a maximum of 30 hops: 1 <1 ms 1 ms 1 ms 192.168.0.1 2 6 ms 8 ms 8 ms xx.xx.xx.xx 3 8 ms 8 ms 8 ms [xx.xx.xxx.xxx] 4 17 ms 16 ms 16 ms be3549.ccr31.sof02.atlas.cogentco.com [154.54.59.138] 5 18 ms 17 ms 20 ms be3421.ccr51.beg03.atlas.cogentco.com [130.117.0.94] 6 32 ms 31 ms 30 ms be3464.ccr52.vie01.atlas.cogentco.com [154.54.59.189] 7 39 ms 37 ms 44 ms be3462.ccr22.muc03.atlas.cogentco.com [154.54.59.182] 8 42 ms 48 ms 44 ms be2960.ccr42.fra03.atlas.cogentco.com [154.54.36.253] 9 44 ms 50 ms 50 ms be3187.agr41.fra03.atlas.cogentco.com [130.117.1.117] 10 43 ms 45 ms 46 ms tata.fra03.atlas.cogentco.com [130.117.15.86] 11 45 ms 45 ms 44 ms 72.14.196.162 12 43 ms 41 ms 46 ms 108.170.251.129 13 46 ms 46 ms 45 ms 74.125.37.167 14 45 ms 52 ms 48 ms fra15s29-in-f14.1e100.net [172.217.18.174]
如果您不希望解析名称，则可以使用tracert -d。
OSX和Linux以及Python 2.7，您可以使用此parser。