xref: /openbsd-src/usr.sbin/mtrace/mtrace.8 (revision f2da64fbbbf1b03f09f390ab01267c93dfd77c4c)
1.\"	$OpenBSD: mtrace.8,v 1.18 2014/09/08 01:27:55 schwarze Exp $
2.\"	$NetBSD: mtrace.8,v 1.4 1995/12/10 10:57:11 mycroft Exp $
3.\"
4.\" Copyright (c) 1993, 1998-2001.
5.\" The University of Southern California/Information Sciences Institute.
6.\" All rights reserved.
7.\"
8.\" Redistribution and use in source and binary forms, with or without
9.\" modification, are permitted provided that the following conditions
10.\" are met:
11.\" 1. Redistributions of source code must retain the above copyright
12.\"    notice, this list of conditions and the following disclaimer.
13.\" 2. Redistributions in binary form must reproduce the above copyright
14.\"    notice, this list of conditions and the following disclaimer in the
15.\"    documentation and/or other materials provided with the distribution.
16.\" 3. Neither the name of the project nor the names of its contributors
17.\"    may be used to endorse or promote products derived from this software
18.\"    without specific prior written permission.
19.\"
20.\" THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND
21.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE
24.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
30.\" SUCH DAMAGE.
31.\"
32.\" Other copyrights might apply to parts of this software and are so
33.\" noted when applicable.
34.\"
35.\" This manual page (but not the software) was derived from the
36.\" manual page for the traceroute program which bears the following
37.\" copyright notice:
38.\"
39.\" Copyright (c) 1988 The Regents of the University of California.
40.\" All rights reserved.
41.\"
42.\" This code is derived from software contributed to Berkeley by
43.\" Van Jacobson.
44.\"
45.\" Redistribution and use in source and binary forms, with or without
46.\" modification, are permitted provided that the following conditions
47.\" are met:
48.\" 1. Redistributions of source code must retain the above copyright
49.\"    notice, this list of conditions and the following disclaimer.
50.\" 2. Redistributions in binary form must reproduce the above copyright
51.\"    notice, this list of conditions and the following disclaimer in the
52.\"    documentation and/or other materials provided with the distribution.
53.\" 3. Neither the name of the University nor the names of its contributors
54.\"    may be used to endorse or promote products derived from this software
55.\"    without specific prior written permission.
56.\"
57.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
58.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
59.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
60.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
61.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
62.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
63.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
64.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
65.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
66.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
67.\" SUCH DAMAGE.
68.\"
69.Dd $Mdocdate: September 8 2014 $
70.Dt MTRACE 8
71.Os
72.Sh NAME
73.Nm mtrace
74.Nd print multicast path from a source to a receiver
75.Sh SYNOPSIS
76.Nm mtrace
77.Op Fl lMnpsv
78.Op Fl g Ar gateway
79.Op Fl i Ar if_addr
80.Op Fl m Ar max_hops
81.Op Fl q Ar nqueries
82.Op Fl r Ar host
83.Op Fl S Ar stat_int
84.Op Fl t Ar ttl
85.Op Fl w Ar waittime
86.Ar source
87.Op Ar receiver
88.Op Ar group
89.Sh DESCRIPTION
90Assessing problems in the distribution of IP multicast traffic
91can be difficult.
92.Nm
93utilizes a tracing feature implemented in multicast routers
94.Pf ( Nm mrouted
95version 3.3 and later) that is
96accessed via an extension to the IGMP protocol.
97A trace query is passed hop-by-hop along the reverse path from the
98.Ar receiver
99to the
100.Ar source ,
101collecting hop addresses, packet counts, and routing error conditions
102along the path, and then the response is returned to the requestor.
103.Pp
104The only required parameter is the
105.Ar source
106host name or address.
107The default
108.Ar receiver
109is the host running mtrace, and the default
110.Ar group
111is "MBone Audio" (224.2.0.1), which is sufficient if packet loss
112statistics for a particular multicast group are not needed.
113These two optional parameters may be specified to test the path to some other
114receiver in a particular group, subject to some constraints as
115detailed below.
116The two parameters can be distinguished because the
117.Ar receiver
118is a unicast address and the
119.Ar group
120is a multicast address.
121.Pp
122The options are as follows:
123.Bl -tag -width addr_xy
124.It Fl g Ar gateway
125Send the trace query via unicast directly to the multicast router
126.Ar gateway
127rather than multicasting the query.
128This must be the last-hop router on the path from the intended
129.Ar source
130to the
131.Ar receiver .
132.Em NOTE: Read the BUGS section below.
133.It Fl i Ar if_addr
134Use
135.Ar if_addr
136as the local interface address (on a multi-homed host) for sending the
137trace query and as the default for the
138.Ar receiver
139and the response destination.
140.It Fl l
141Loop indefinitely printing packet rate and loss statistics for the
142multicast path every 10 seconds (see
143.Fl S Ar stat_int ) .
144.It Fl M
145Always send the response using multicast rather than attempting
146unicast first.
147.It Fl m Ar max_hops
148Set to
149the maximum number of hops that will be traced from the
150.Ar receiver
151back toward the
152.Ar source .
153The default is 32 hops (infinity for the DVMRP routing protocol).
154.It Fl n
155Print hop addresses numerically rather than symbolically and numerically
156(saves a nameserver address-to-name lookup for each router found on the
157path).
158.It Fl p
159Listen passively for multicast responses from traces initiated by others.
160This works best when run on a multicast router.
161.It Fl q Ar nqueries
162Set the maximum number of query attempts for any hop to
163.Ar nqueries .
164The default is 3.
165.It Fl r Ar host
166Send the trace response to
167.Ar host
168rather than to the host on which
169.Nm
170is being run, or to a multicast address other than the one registered
171for this purpose (224.0.1.32).
172.It Fl S Ar stat_int
173Change the interval between statistics gathering traces to
174.Ar stat_int
175seconds (default 10 seconds).
176.It Fl s
177Print a short form output including only the multicast path and not
178the packet rate and loss statistics.
179.It Fl t Ar ttl
180Set the
181.Ar ttl
182(time-to-live, or number of hops) for multicast trace queries and
183responses.
184The default is 64, except for local queries to the
185"all routers" multicast group which use ttl 1.
186.It Fl v
187Verbose mode; show hop times on the initial trace and statistics display.
188.It Fl w Ar waittime
189Set the time to wait for a trace response to
190.Ar waittime
191seconds (default 3 seconds).
192.El
193.Ss How \&It Works
194The technique used by the
195.Nm traceroute
196tool to trace unicast network paths will not work for IP multicast
197because ICMP responses are specifically forbidden for multicast traffic.
198Instead, a tracing feature has been built into the multicast routers.
199This technique has the advantage that additional information about
200packet rates and losses can be accumulated while the number of packets
201sent is minimized.
202.Pp
203Since multicast uses
204reverse path forwarding, the trace is run backwards from the
205.Ar receiver
206to the
207.Ar source .
208A trace query packet is sent to the last
209hop multicast router (the leaf router for the desired
210.Ar receiver
211address).
212The last hop router builds a trace response packet, fills in
213a report for its hop, and forwards the trace packet using unicast to
214the router it believes is the previous hop for packets originating
215from the specified
216.Ar source .
217Each router along the path adds its report and forwards the packet.
218When the trace response packet reaches the first hop router (the router
219that is directly connected to the source's net), that router sends the
220completed response to the response destination address specified in
221the trace query.
222.Pp
223If some multicast router along the path does not implement the
224multicast traceroute feature or if there is some outage, then no
225response will be returned.
226To solve this problem, the trace query includes a maximum hop count field
227to limit the number of hops traced before the response is returned.
228That allows a partial path to be traced.
229.Pp
230The reports inserted by each router contain not only the address of
231the hop, but also the ttl required to forward and some flags to indicate
232routing errors, plus counts of the total number of packets on the
233incoming and outgoing interfaces and those forwarded for the specified
234.Ar group .
235Taking differences in these counts for two traces separated in time
236and comparing the output packet counts from one hop with the input
237packet counts of the next hop allows the calculation of packet rate
238and packet loss statistics for each hop to isolate congestion
239problems.
240.Ss Finding the Last-Hop Router
241The trace query must be sent to the multicast router which is the
242last hop on the path from the
243.Ar source
244to the
245.Ar receiver .
246If the
247.Ar receiver
248is on the local subnet (as determined using the subnet
249mask), then the default method is to multicast the trace query to
250all-routers.mcast.net (224.0.0.2) with a ttl of 1.
251Otherwise, the trace query is multicast to the
252.Ar group
253address since the last hop router will be a member of that group if
254the
255.Ar receiver
256is.
257Therefore it is necessary to specify a
258.Ar group
259that the intended
260.Ar receiver
261is joined.
262This multicast is sent with a default ttl of 64, which may not be sufficient
263for all cases (changed with the
264.Fl t
265option).
266If the last hop router is known, it may also be addressed directly
267using the
268.Fl g
269option).
270Alternatively, if it is desired to trace a group that the
271.Ar receiver
272has not joined, but it is known that the last-hop router is a
273member of another group, the
274.Fl g
275option may also be used to specify a different multicast address for the
276trace query.
277.Pp
278When tracing from a multihomed host or router, the default
279.Ar receiver
280address may not be the desired interface for the path from the
281.Ar source .
282In that case, the desired interface should be specified explicitly as
283the
284.Ar receiver .
285.Ss Directing the Response
286By default,
287.Nm
288first attempts to trace the full reverse path, unless the number of
289hops to trace is explicitly set with the
290.Fl m
291option.
292If there is no response within a 3 second timeout interval
293(changed with the
294.Fl m
295option), a "*" is printed and the probing switches to hop-by-hop mode.
296Trace queries are issued starting with a maximum hop count of one and
297increasing by one until the full path is traced or no response is
298received.
299At each hop, multiple probes are sent (default is three, changed with
300.Fl q
301option).
302The first half of the attempts (default is one) are made with
303the unicast address of the host running
304.Nm
305as the destination for the response.
306Since the unicast route may be blocked, the remainder of attempts request
307that the response be multicast to mtrace.mcast.net (224.0.1.32) with the
308ttl set to 32 more than what's needed to pass the thresholds seen so far
309along the path to the
310.Ar receiver .
311For the last quarter of the attempts (default is
312one), the ttl is increased by another 32 each time up to a maximum of 192.
313Alternatively, the ttl may be set explicitly with the
314.Fl t
315option and/or the initial unicast attempts can be forced to use
316multicast instead with the
317.Fl m
318option.
319For each attempt, if no response is received within the timeout,
320a "*" is printed.
321After the specified number of attempts have failed,
322.Nm
323will try to query the next hop router with a DVMRP_ASK_NEIGHBORS2
324request (as used by the
325.Nm mrinfo
326program) to see what kind of router it is.
327.Sh EXAMPLES
328The output of
329.Nm
330is in two sections.
331The first section is a short listing of the hops in the order they are
332queried, that is, in the reverse of the order from the
333.Ar source
334to the
335.Ar receiver .
336For each hop, a line is printed showing the hop number (counted
337negatively to indicate that this is the reverse path); the multicast
338routing protocol (DVMRP, MOSPF, PIM, etc.); the threshold required to
339forward data (to the previous hop in the listing as indicated by the
340up-arrow character); and the cumulative delay for the query to reach
341that hop (valid only if the clocks are synchronized).
342This first section ends with a line showing the round-trip time which measures
343the interval from when the query is issued until the response is
344received, both derived from the local system clock.
345A sample use and output might be:
346.Bd -literal
347oak.isi.edu 80# mtrace -l caraway.lcs.mit.edu 224.2.0.3
348Mtrace from 18.26.0.170 to 128.9.160.100 via group 224.2.0.3
349Querying full reverse path...
350  0  oak.isi.edu (128.9.160.100)
351 -1  cub.isi.edu (128.9.160.153)  DVMRP  thresh^ 1  3 ms
352 -2  la.dart.net (140.173.128.1)  DVMRP  thresh^ 1  14 ms
353 -3  dc.dart.net (140.173.64.1)  DVMRP  thresh^ 1  50 ms
354 -4  bbn.dart.net (140.173.32.1)  DVMRP  thresh^ 1  63 ms
355 -5  mit.dart.net (140.173.48.2)  DVMRP  thresh^ 1  71 ms
356 -6  caraway.lcs.mit.edu (18.26.0.170)
357Round trip time 124 ms
358.Ed
359.Pp
360The second section provides a pictorial view of the path in the
361forward direction with data flow indicated by arrows pointing downward
362and the query path indicated by arrows pointing upward.
363For each hop, both the entry and exit addresses of the router are shown if
364different, along with the initial ttl required on the packet in order
365to be forwarded at this hop and the propagation delay across the hop
366assuming that the routers at both ends have synchronized clocks.
367The right half of this section is composed of several columns of
368statistics in two groups.
369Within each group, the columns are the number of packets lost, the number
370of packets sent, the percentage lost, and the average packet rate at each hop.
371These statistics are calculated from differences between traces and from
372hop to hop as explained above.
373The first group shows the statistics for all traffic flowing out the interface
374at one hop and in the interface at the next hop.
375The second group shows the statistics only for traffic forwarded
376from the specified
377.Ar source
378to the specified
379.Ar group .
380.Pp
381These statistics are shown on one or two lines for each hop.
382Without any options, this second section of the output is printed only once,
383approximately 10 seconds after the initial trace.
384One line is shown for each hop showing the statistics over that 10-second
385period.
386If the
387.Fl l
388option is given, the second section is repeated every 10 seconds and
389two lines are shown for each hop.
390The first line shows the statistics for the last 10 seconds, and the second
391line shows the cumulative statistics over the period since the initial trace,
392which is 101 seconds in the example below.
393The second section of the output is omitted if the
394.Fl s .
395option is set.
396.Bd -literal
397Waiting to accumulate statistics... Results after 101 seconds:
398
399  Source       Response Dest  Packet Statistics For  Only For Traffic
40018.26.0.170    128.9.160.100  All Multicast Traffic  From 18.26.0.170
401     |       __/ rtt  125 ms  Lost/Sent = Pct  Rate    To 224.2.0.3
402     v      /    hop   65 ms  ---------------------  ------------------
40318.26.0.144
404140.173.48.2   mit.dart.net
405     |     ^     ttl    1      0/6    = --%   0 pps   0/2  = --%  0 pps
406     v     |     hop    8 ms   1/52   =  2%   0 pps   0/18 =  0%  0 pps
407140.173.48.1
408140.173.32.1   bbn.dart.net
409     |     ^     ttl    2      0/6    = --%   0 pps   0/2  = --%  0 pps
410     v     |     hop   12 ms   1/52   =  2%   0 pps   0/18 =  0%  0 pps
411140.173.32.2
412140.173.64.1   dc.dart.net
413     |     ^     ttl    3      0/271  =  0%  27 pps   0/2  = --%  0 pps
414     v     |     hop   34 ms  -1/2652 =  0%  26 pps   0/18 =  0%  0 pps
415140.173.64.2
416140.173.128.1  la.dart.net
417     |     ^     ttl    4     -2/831  =  0%  83 pps   0/2  = --%  0 pps
418     v     |     hop   11 ms  -3/8072 =  0%  79 pps   0/18 =  0%  0 pps
419140.173.128.2
420128.9.160.153  cub.isi.edu
421     |      \e__  ttl    5        833         83 pps     2         0 pps
422     v         \e hop   -8 ms     8075        79 pps     18        0 pps
423128.9.160.100  128.9.160.100
424  Receiver     Query Source
425.Ed
426.Pp
427Because the packet counts may be changing as the trace query is
428propagating, there may be small errors (off by 1 or 2) in these
429statistics.
430However, those errors should not accumulate, so the cumulative statistics
431line should increase in accuracy as a new trace is run every 10 seconds.
432There are two sources of larger errors,
433both of which show up as negative losses:
434.Bl -bullet -offset abcd
435.It
436If the input to a node is from a multi-access network with more than
437one other node attached, then the input count will be (close to) the
438sum of the output counts from all the attached nodes, but the output
439count from the previous hop on the traced path will be only part of
440that.
441Hence the output count minus the input count will be negative.
442.It
443In release 3.3 of the DVMRP multicast forwarding software for SunOS
444and other systems, a multicast packet generated on a router will be
445counted as having come in an interface even though it did not.
446This creates the negative loss that can be seen in the example above.
447.El
448.Pp
449Note that these negative losses may mask positive losses.
450.Pp
451In the example, there is also one negative hop time.
452This simply indicates a lack of synchronization between the system clocks
453across that hop.
454This example also illustrates how the percentage loss is
455shown as two dashes when the number of packets sent is less than 10
456because the percentage would not be statistically valid.
457.Pp
458A second example shows a trace to a
459.Ar receiver
460that is not local; the query is sent to the last-hop router with the
461.Fl g
462option.
463In this example, the trace of the full reverse path resulted
464in no response because there was a node running an old version of
465.Nm mrouted
466that did not implement the multicast traceroute function, so
467.Nm
468switched to hop-by-hop mode.
469The "Route pruned" error code indicates that traffic for group 224.2.143.24
470would not be forwarded.
471.Bd -literal
472oak.isi.edu 108# mtrace -g 140.173.48.2 204.62.246.73 \e
473                           butter.lcs.mit.edu 224.2.143.24
474Mtrace from 204.62.246.73 to 18.26.0.151 via group 224.2.143.24
475Querying full reverse path... * switching to hop-by-hop:
476  0  butter.lcs.mit.edu (18.26.0.151)
477 -1  jam.lcs.mit.edu (18.26.0.144)  DVMRP  thresh^ 1  33 ms  Route pruned
478 -2  bbn.dart.net (140.173.48.1)  DVMRP  thresh^ 1  36 ms
479 -3  dc.dart.net (140.173.32.2)  DVMRP  thresh^ 1  44 ms
480 -4  darpa.dart.net (140.173.240.2)  DVMRP  thresh^ 16  47 ms
481 -5  * * * noc.hpc.org (192.187.8.2) [mrouted 2.2] didn't respond
482Round trip time 95 ms
483.Ed
484.Sh SEE ALSO
485.Xr map-mbone 8 ,
486.Xr mrinfo 8 ,
487.Xr mrouted 8 ,
488.Xr traceroute 8
489.Sh AUTHORS
490.An -nosplit
491Implemented by
492.An Steve Casner
493based on an initial prototype written by
494.An Ajit Thyagarajan .
495The multicast traceroute mechanism was designed by
496.An Van Jacobson
497with help from
498.An Steve Casner ,
499.An Steve Deering ,
500.An Dino Farinacci ,
501and
502.An Deb Agrawal ;
503it was implemented in
504.Nm mrouted
505by
506.An Ajit Thyagarajan
507and
508.An Bill Fenner .
509The option syntax and the output format of
510.Nm
511are modeled after the unicast
512.Xr traceroute 8
513program written by
514.An Van Jacobson .
515.Sh BUGS
516Versions 3.3 and 3.5 of
517.Nm mrouted
518will crash if a trace query is received via a
519unicast packet and
520.Nm mrouted
521has no route for the
522.Ar source
523address.
524Therefore, do not use the
525.Fl g
526option unless the target
527.Nm mrouted
528has been verified to be 3.4 or newer than 3.5.
529