1 .\" Copyright (c) 1989, 1995, 1996, 1997, 1999, 2000
2 .\" The Regents of the University of California. All rights reserved.
4 .\" Redistribution and use in source and binary forms are permitted
5 .\" provided that the above copyright notice and this paragraph are
6 .\" duplicated in all such forms and that any documentation,
7 .\" advertising materials, and other materials related to such
8 .\" distribution and use acknowledge that the software was developed
9 .\" by the University of California, Berkeley. The name of the
10 .\" University may not be used to endorse or promote products derived
11 .\" from this software without specific prior written permission.
12 .\" THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR
13 .\" IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
14 .\" WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
16 .\" $Id: traceroute.8,v 1.19 2000/09/21 08:44:19 leres Exp $
24 .Nd "print the route packets take to network host"
40 .Op Fl z Ar pausemsecs
45 The Internet is a large and complex aggregation of
46 network hardware, connected together by gateways.
47 Tracking the route one's packets follow (or finding the miscreant
48 gateway that's discarding your packets) can be difficult.
50 utilizes the IP protocol `time to live' field and attempts to elicit an
51 ICMP TIME_EXCEEDED response from each gateway along the path to some
54 The only mandatory parameter is the destination host name or IP number.
55 The default probe datagram length is 40 bytes, but this may be increased
56 by specifying a packet length (in bytes) after the destination host
62 Turn on AS# lookups for each hop encountered.
64 Turn on AS# lookups and use the given server instead of the
67 Firewall evasion mode.
68 Use fixed destination ports for UDP and TCP probes.
69 The destination port does NOT increment with each packet sent.
71 Set the initial time-to-live used in the first outgoing probe packet.
73 Set the "don't fragment" bit.
75 Enable socket level debugging.
77 When an ICMP response to our probe datagram is received,
78 print the differences between the transmitted packet and
79 the packet quoted by the ICMP response.
80 A key showing the location of fields within the transmitted packet is printed,
81 followed by the original packet in hex,
82 followed by the quoted packet in hex.
83 Bytes that are unchanged in the quoted packet are shown as underscores.
85 the IP checksum and the TTL of the quoted packet are not expected to match.
86 By default, only one probe per hop is sent with this option.
88 Specify a loose source route gateway (8 maximum).
90 Specify a network interface to obtain the source IP address for
91 outgoing probe packets. This is normally only useful on a multi-homed
94 flag for another way to do this.)
96 Use ICMP ECHO instead of UDP datagrams. (A synonym for "-P icmp").
98 Set the initial time-to-live value used in outgoing probe packets.
99 The default is 1, i.e., start with the first hop.
101 Set the max time-to-live (max number of hops) used in outgoing probe
102 packets. The default is the value of the
105 (the same default used for TCP
108 Print hop addresses numerically rather than symbolically and numerically
109 (saves a nameserver address-to-name lookup for each gateway found on the
112 Send packets of specified IP protocol. The currently supported protocols
113 are: UDP, TCP, GRE and ICMP. Other protocols may also be specified (either by
114 name or by number), though
116 does not implement any special knowledge of their packet formats. This
117 option is useful for determining which router along a path may be
118 blocking packets based on IP protocol number. But see BUGS below.
120 Protocol specific. For UDP and TCP, sets
123 number used in probes (default is 33434).
124 Traceroute hopes that nothing is listening on UDP ports
127 .Em base + nhops * nprobes - 1
128 at the destination host (so an ICMP PORT_UNREACHABLE message will
129 be returned to terminate the route tracing). If something is
130 listening on a port in the default range, this option can be used
131 to pick an unused port range.
133 Set the number of probes per hop (default is 3,
139 Bypass the normal routing tables and send directly to a host on an attached
141 If the host is not on a directly-attached network,
142 an error is returned.
143 This option can be used to ping a local host through an interface
144 that has no route through it (e.g., after the interface was dropped by
147 Use the following IP address (which usually is given as an IP number, not
148 a hostname) as the source address in outgoing probe packets. On
149 multi-homed hosts (those with more than one IP
150 address), this option can be used to
151 force the source address to be something other than the IP address
152 of the interface the probe packet is sent on. If the IP address
153 is not one of this machine's interface addresses, an error is
154 returned and nothing is sent. (See the
156 flag for another way to do this.)
158 Print a summary of how many probes were not answered for each hop.
162 in probe packets to the following value (default zero). The value must be
163 a decimal integer in the range 0 to 255. This option can be used to
164 see if different types-of-service result in different paths. (If you
165 are not running 4.4bsd, this may be academic since the normal network
166 services like telnet and ftp don't let you control the TOS).
167 Not all values of TOS are legal or
168 meaningful \- see the IP spec for definitions. Useful values are
175 Verbose output. Received ICMP packets other than
181 Set the time (in seconds) to wait for a response to a probe (default 5
184 Toggle ip checksums. Normally, this prevents traceroute from calculating
185 ip checksums. In some cases, the operating system can overwrite parts of
186 the outgoing packet but not recalculate the checksum (so in some cases
187 the default is to not calculate checksums and using
189 causes them to be calculated). Note that checksums are usually required
190 for the last hop when using ICMP ECHO probes
192 So they are always calculated when using ICMP.
193 .It Fl z Ar pausemsecs
194 Set the time (in milliseconds) to pause between probes (default 0).
195 Some systems such as Solaris and routers such as Ciscos rate limit
196 icmp messages. A good value to use with this this is 500 (e.g. 1/2 second).
199 This program attempts to trace the route an IP packet would follow to some
200 internet host by launching UDP probe
201 packets with a small ttl (time to live) then listening for an
202 ICMP "time exceeded" reply from a gateway. We start our probes
203 with a ttl of one and increase by one until we get an ICMP "port
204 unreachable" (which means we got to "host") or hit a max (which
205 defaults to the amount of hops specified by the
208 and can be changed with the
213 flag) are sent at each ttl setting and a
214 line is printed showing the ttl, address of the gateway and
215 round trip time of each probe. If the probe answers come from
216 different gateways, the address of each responding system will
217 be printed. If there is no response within a 5 sec. timeout
218 interval (changed with the
220 flag), a "*" is printed for that
223 We don't want the destination
224 host to process the UDP probe packets so the destination port is set to an
225 unlikely value (if some clod on the destination is using that
226 value, it can be changed with the
230 A sample use and output might be:
231 .Bd -literal -offset 4n
232 % traceroute nis.nsf.net.
233 traceroute to nis.nsf.net (35.1.1.48), 64 hops max, 38 byte packet
234 1 helios.ee.lbl.gov (128.3.112.1) 19 ms 19 ms 0 ms
235 2 lilac-dmc.Berkeley.EDU (128.32.216.1) 39 ms 39 ms 19 ms
236 3 lilac-dmc.Berkeley.EDU (128.32.216.1) 39 ms 39 ms 19 ms
237 4 ccngw-ner-cc.Berkeley.EDU (128.32.136.23) 39 ms 40 ms 39 ms
238 5 ccn-nerif22.Berkeley.EDU (128.32.168.22) 39 ms 39 ms 39 ms
239 6 128.32.197.4 (128.32.197.4) 40 ms 59 ms 59 ms
240 7 131.119.2.5 (131.119.2.5) 59 ms 59 ms 59 ms
241 8 129.140.70.13 (129.140.70.13) 99 ms 99 ms 80 ms
242 9 129.140.71.6 (129.140.71.6) 139 ms 239 ms 319 ms
243 10 129.140.81.7 (129.140.81.7) 220 ms 199 ms 199 ms
244 11 nic.merit.edu (35.1.1.48) 239 ms 239 ms 239 ms
247 Note that lines 2 & 3 are the same. This is due to a buggy
248 kernel on the 2nd hop system \- lilac-dmc.Berkeley.EDU \- that forwards
249 packets with a zero ttl (a bug in the distributed version
250 of 4.3BSD). Note that you have to guess what path
251 the packets are taking cross-country since the NSFNet (129.140)
252 doesn't supply address-to-name translations for its NSSes.
254 A more interesting example is:
255 .Bd -literal -offset 4n
256 % traceroute allspice.lcs.mit.edu.
257 traceroute to allspice.lcs.mit.edu (18.26.0.115), 64 hops max
258 1 helios.ee.lbl.gov (128.3.112.1) 0 ms 0 ms 0 ms
259 2 lilac-dmc.Berkeley.EDU (128.32.216.1) 19 ms 19 ms 19 ms
260 3 lilac-dmc.Berkeley.EDU (128.32.216.1) 39 ms 19 ms 19 ms
261 4 ccngw-ner-cc.Berkeley.EDU (128.32.136.23) 19 ms 39 ms 39 ms
262 5 ccn-nerif22.Berkeley.EDU (128.32.168.22) 20 ms 39 ms 39 ms
263 6 128.32.197.4 (128.32.197.4) 59 ms 119 ms 39 ms
264 7 131.119.2.5 (131.119.2.5) 59 ms 59 ms 39 ms
265 8 129.140.70.13 (129.140.70.13) 80 ms 79 ms 99 ms
266 9 129.140.71.6 (129.140.71.6) 139 ms 139 ms 159 ms
267 10 129.140.81.7 (129.140.81.7) 199 ms 180 ms 300 ms
268 11 129.140.72.17 (129.140.72.17) 300 ms 239 ms 239 ms
270 13 128.121.54.72 (128.121.54.72) 259 ms 499 ms 279 ms
275 18 ALLSPICE.LCS.MIT.EDU (18.26.0.115) 339 ms 279 ms 279 ms
278 Note that the gateways 12, 14, 15, 16 & 17 hops away
279 either don't send ICMP "time exceeded" messages or send them
280 with a ttl too small to reach us. 14 \- 17 are running the
281 MIT C Gateway code that doesn't send "time exceeded"s. God
282 only knows what's going on with 12.
284 The silent gateway 12 in the above may be the result of a bug in
285 the 4.[23]BSD network code (and its derivatives): 4.x (x <= 3)
286 sends an unreachable message using whatever ttl remains in the
287 original datagram. Since, for gateways, the remaining ttl is
288 zero, the ICMP "time exceeded" is guaranteed to not make it back
289 to us. The behavior of this bug is slightly more interesting
290 when it appears on the destination system:
291 .Bd -literal -offset 4n
292 1 helios.ee.lbl.gov (128.3.112.1) 0 ms 0 ms 0 ms
293 2 lilac-dmc.Berkeley.EDU (128.32.216.1) 39 ms 19 ms 39 ms
294 3 lilac-dmc.Berkeley.EDU (128.32.216.1) 19 ms 39 ms 19 ms
295 4 ccngw-ner-cc.Berkeley.EDU (128.32.136.23) 39 ms 40 ms 19 ms
296 5 ccn-nerif35.Berkeley.EDU (128.32.168.35) 39 ms 39 ms 39 ms
297 6 csgw.Berkeley.EDU (128.32.133.254) 39 ms 59 ms 39 ms
304 13 rip.Berkeley.EDU (128.32.131.22) 59 ms ! 39 ms ! 39 ms !
307 Notice that there are 12 "gateways" (13 is the final
308 destination) and exactly the last half of them are "missing".
309 What's really happening is that rip (a Sun-3 running Sun OS3.5)
310 is using the ttl from our arriving datagram as the ttl in its
311 ICMP reply. So, the reply will time out on the return path
312 (with no notice sent to anyone since ICMP's aren't sent for
313 ICMP's) until we probe with a ttl that's at least twice the path
314 length. I.e., rip is really only 7 hops away. A reply that
315 returns with a ttl of 1 is a clue this problem exists.
316 Traceroute prints a "!" after the time if the ttl is <= 1.
317 Since vendors ship a lot of obsolete
322 software, expect to see this problem
323 frequently and/or take care picking the target host of your
326 Other possible annotations after the time are:
327 .Bl -hang -offset indent -width 12n
333 Protocol unreachable.
337 Fragmentation needed.
338 The RFC1191 Path MTU Discovery value is displayed.
340 Destination network unknown.
342 Destination host unknown.
344 Source host is isolated.
346 Communication with destination network administratively prohibited.
348 Communication with destination host administratively prohibited.
350 For this ToS the destination network is unreachable.
352 For this ToS the destination host is unreachable.
354 Communication administratively prohibited.
356 Host precedence violation.
358 Precedence cutoff in effect.
360 ICMP unreachable code <num>.
363 These are defined by RFC1812 (which supersedes RFC1716).
364 If almost all the probes result in some kind of unreachable,
366 will give up and exit.
368 This program is intended for use in network testing, measurement
370 It should be used primarily for manual fault isolation.
371 Because of the load it could impose on the network, it is unwise to use
373 during normal operations or from automated scripts.
380 Implemented by Van Jacobson from a suggestion by Steve Deering. Debugged
381 by a cast of thousands with particularly cogent suggestions or fixes from
382 C. Philip Wood, Tim Seaver and Ken Adelman.
384 When using protocols other than UDP, functionality is reduced.
385 In particular, the last packet will often appear to be lost, because
386 even though it reaches the destination host, there's no way to know
387 that because no ICMP message is sent back.
390 should listen for a RST from the destination host (or an intermediate
391 router that's filtering packets), but this is not implemented yet.
393 The AS number capability reports information that may sometimes be
394 inaccurate due to discrepancies between the contents of the
395 routing database server and the current state of the Internet.