path: root/upstream/fedora-rawhide/man8/traceroute.8

.\" Copyright (c)  2006   Dmitry Butskoy (dmitry@butskoy.name)
.\" License: GPL v2 or any later version
.\" See COPYING for the status of this software
.TH TRACEROUTE 8 "11 October 2006" "Traceroute" "Traceroute For Linux"
.\" .UC 6
.SH NAME
traceroute \- print the route packets trace to network host
.SH SYNOPSIS
.na
.BR traceroute " [" \-46dFITUnreAV "] [" "\-f first_ttl" "] [" "\-g gate,..." ]
.br
.ti +8
.BR "" [ "-i device" "] [" "-m max_ttl" "] [" "-p port" "] [" "-s src_addr" ]
.br
.ti +8
.BR "" [ "-q nqueries" "] [" "-N squeries" "] [" "-t tos" ]
.br
.ti +8
.BR "" [ "-l flow_label" "] [" "-w waittimes" "] [" "-z sendwait" "] [" "-UL" "] [" "-D" ]
.br
.ti +8
.BR "" [ "-P proto" "] [" "--sport=port" "] [" "-M method" "] [" "-O mod_options" ]
.br
.ti +8
.BR "" [ "--mtu" "] [" "--back" ]
.br
.ti +8
.BR host " [" "packet_len" "]"
.br
.BR traceroute6
.RI " [" options ]
.ad
.SH DESCRIPTION
.I traceroute
tracks the route packets taken from an IP network on their
way to a given host. It utilizes the IP protocol's time to live (TTL) field
and attempts to elicit an ICMP TIME_EXCEEDED response from each gateway
along the path to the host.
.P
.I traceroute6
is equivalent to
.I traceroute
.B \-6
.PP
The only required parameter is the name or IP address of the
destination
.BR host \ .
The optional
.B packet_len\fR`gth
is the total size of the probing packet (default 60 bytes
for IPv4 and 80 for IPv6). The specified size can be ignored
in some situations or increased up to a minimal value.
.PP
This program attempts to trace the route an IP packet would follow to some
internet host by launching probe
packets with a small ttl (time to live) then listening for an
ICMP "time exceeded" reply from a gateway.  We start our probes
with a ttl of one and increase by one until we get an ICMP "port
unreachable" (or TCP reset), which means we got to the "host", or hit a max (which
defaults to 30 hops). Three probes (by default) are sent at each ttl setting
and a line is printed showing the ttl, address of the gateway and
round trip time of each probe. The address can be followed by additional
information when requested. If the probe answers come from
different gateways, the address of each responding system will
be printed.  If there is no response within a certain timeout,
an "*" (asterisk) is printed for that probe.
.PP
After the trip time, some additional annotation can be printed:
.BR !H ,
.BR !N ,
or
.B !P
(host, network or protocol unreachable),
.B !S
(source route failed),
.B !F
(fragmentation needed),
.B !X
(communication administratively prohibited),
.B !V
(host precedence violation),
.B !C
(precedence cutoff in effect), or
.B !<num>
(ICMP unreachable code <num>).
If almost all the probes result in some kind of unreachable, traceroute
will give up and exit.
.PP
We don't want the destination host to process the UDP probe packets,
so the destination port is set to an unlikely value (you can change it with the
.B \-p
flag). There is no such a problem for ICMP or TCP tracerouting (for TCP we
use half-open technique, which prevents our probes to be seen by applications
on the destination host).
.PP
In the modern network environment the traditional traceroute methods
can not be always applicable, because of widespread use of firewalls.
Such firewalls filter the "unlikely" UDP ports, or even ICMP echoes.
To solve this, some additional tracerouting methods are implemented
(including tcp), see
.B LIST OF AVAILABLE METHODS
below. Such methods try to use particular protocol
and source/destination port, in order to bypass firewalls (to be seen
by firewalls just as a start of allowed type of a network session).
.SH OPTIONS
.TP
.BI \--help
Print help info and exit.
.TP
.BR \-4 ", " \-6
Explicitly force IPv4 or IPv6 tracerouting. By default, the program
will try to resolve the name given, and choose the appropriate
protocol automatically. If resolving a host name returns both
IPv4 and IPv6 addresses,
.I traceroute
will use IPv4.
.TP
.B \-I, \-\-icmp
Use ICMP ECHO for probes
.TP
.B \-T, \-\-tcp
Use TCP SYN for probes
.TP
.B \-d, --debug
Enable socket level debugging (when the Linux kernel supports it)
.TP
.B \-F, --dont-fragment
Do not fragment probe packets. (For IPv4 it also sets DF bit, which tells
intermediate routers not to fragment remotely as well).
.br

.br
Varying the size of the probing packet by the
.B packet_len
command line parameter, you can manually obtain information
about the MTU of individual network hops. The
.B \--mtu
option (see below) tries to do this automatically.
.br

.br
Note, that non-fragmented features (like
.B \-F
or
.B \--mtu\fR)
work properly since the Linux kernel 2.6.22 only.
Before that version, IPv6 was always fragmented, IPv4 could use
the once the discovered final mtu only (from the route cache), which can be
less than the actual mtu of a device.
.TP
.BI \-f " first_ttl" ", --first=" first_ttl
Specifies with what TTL to start. Defaults to 1.
.TP
.BI \-g " gateway" ", --gateway=" gateway
Tells traceroute to add an IP source routing option to the outgoing
packet that tells the network to route the packet through the
specified
.IR gateway
(most routers have disabled source routing for security reasons).
In general, several
.IR gateway\fR's
is allowed (comma separated). For IPv6, the form of
.IR num\fB,\fIaddr\fB,\fIaddr...
is allowed, where
.IR num
is a route header type (default is type 2). Note the type 0 route header
is now deprecated (rfc5095).
.TP
.BI \-i " interface" ", --interface=" interface
Specifies the interface through which
.I traceroute
should send packets. By default, the interface is selected
according to the routing table.
.TP
.BI \-m " max_ttl" ", --max-hops=" max_ttl
Specifies the maximum number of hops (max time-to-live value)
.I traceroute
will probe. The default is 30.
.TP
.BI \-N " squeries" ", --sim-queries=" squeries
Specifies the number of probe packets sent out simultaneously.
Sending several probes concurrently can speed up
.I traceroute
considerably. The default value is 16.
.br
Note that some routers and hosts can use ICMP rate throttling. In such
a situation specifying too large number can lead to loss of some responses.
.TP
.BI \-n
Do not try to map IP addresses to host names when displaying them.
.TP
.BI \-p " port" ", --port=" port
For UDP tracing, specifies the destination port base
.I traceroute
will use (the destination port number will be incremented by each probe).
.br
For ICMP tracing, specifies the initial ICMP sequence value (incremented
by each probe too).
.br
For TCP and others specifies just the (constant) destination
port to connect.
.TP
.BI \-t " tos" ", --tos=" tos
For IPv4, set the Type of Service (TOS) and Precedence value. Useful values
are 16 (low delay) and 8 (high throughput). Note that in order to use
some TOS precedence values, you have to be super user.
.br
For IPv6, set the Traffic Control value.
.TP
.BI \-l " flow_label" ", --flowlabel=" flow_label
Use specified flow_label for IPv6 packets.
.TP
.BI \-w " max\fR[\fB,\fIhere\fB,\fInear\fR]" ", --wait=" max\fR[\fB,\fIhere\fB,\fInear\fR]
Determines how long to wait for a response to a probe.
.br

.br
There are three (in general) float values separated by a comma
(or a slash).
.IR Max
specifies the maximum time (in seconds, default 5.0) to wait, in any case.
.br

.br
Traditional traceroute implementation always waited whole
.IR max
seconds for any probe. But if we already have some replies from the
.B same
hop, or even from some
.B next
hop, we can use the round trip time of such a reply as a hint
to determine the actual reasonable amount of time to wait.
.br

.br
The optional
.IR here
(default 3.0) specifies a factor to multiply the round trip time of an already
received response from the
.B same
hop. The resulting value is used as a timeout for the probe, instead of 
(but no more than)
.IR max\fR.
The optional
.IR near
(default 10.0) specifies a similar factor for a response from some
.B next
hop.
(The time of the first found result is used in both cases).
.br

.br
First, we look for the
.B same
hop (of the probe which will be printed first from now).
If nothing found, then look for some
.B next
hop. If nothing found, use
.IR max\fR.
If
.IR here
and/or
.IR near
have zero values, the corresponding computation is skipped.
.br
.IR Here
and
.IR near
are always set to zero if only
.IR max
is specified (for compatibility with previous versions).
.TP
.BI \-q " nqueries" ", --queries=" nqueries
Sets the number of probe packets per hop. The default is 3.
.TP
.BI \-r
Bypass the normal routing tables and send directly to a host on
an attached network.  If the host is not on a directly-attached
network, an error is returned.  This option can be used to ping a
local host through an interface that has no route through it.
.TP
.BI \-s " source_addr" ", --source=" source_addr
Chooses an alternative source address. Note that you must select the
address of one of the interfaces.
By default, the address of the outgoing interface is used.
.TP
.BI \-z " sendwait" ", --sendwait=" sendwait
Minimal time interval between probes (default 0).
If the value is more than 10, then it specifies a number in milliseconds,
else it is a number of seconds (float point values allowed too).
Useful when some routers use rate-limit for ICMP messages.
.TP
.B \-e, \-\-extensions
Show ICMP extensions (rfc4884). The general form is
.I CLASS\fB/\fITYPE\fB:
followed by a hexadecimal dump.
The MPLS (rfc4950) is shown parsed, in a form:
.B MPLS:L=\fIlabel\fB,E=\fIexp_use\fB,S=\fIstack_bottom\fB,T=\fITTL
(more objects separated by
.B /
). The Interface Information (rfc5837) is shown parsed as well, in a following form:
.B \fR{\fBINC\fR|\fBSUB\fR|\fBOUT\fR|\fBNXT\fR}\fB:\fIindex\fB,\fIIP_addr\fB,"\fIname\fB",mtu=\fIMTU\fB
(all four fields may be missing).
.TP
.B \-A, \-\-as\-path\-lookups
Perform AS path lookups in routing registries and print results
directly after the corresponding addresses.
.TP
.B \-V, \-\-version
Print the version and exit.
.br
.P
There are additional options intended for advanced usage
(such as alternate trace methods etc.):
.TP
.B \--sport\fR=\fIport
Chooses the source port to use. Implies
.B \-N\ 1\fR\ -w\ 5 .
Normally source ports (if applicable) are chosen by the system.
.TP
.B \--fwmark\fR=\fImark
Set the firewall mark for outgoing packets (since the Linux kernel 2.6.25).
.TP
.BI \-M " method" ", --module=" name
Use specified method for traceroute operations. Default traditional udp method
has name
.IR default ,
icmp
.BR "" ( "-I" ) "
and tcp
.BR "" ( "-T" ) "
have names
.I icmp
and
.I tcp
respectively.
.br
Method-specific options can be passed by
.BR \-O\  .
Most methods have their simple shortcuts,
.BR "" ( "-I " means " -M icmp" ,
etc).
.TP
.BI \-O " option" ", --options=" options
Specifies some method-specific option. Several options are separated by comma (or use several
.B \-O
on cmdline).
Each method may have its own specific options, or many not have them at all.
To print information about available options, use
.BR \-O\ help .
.TP
.B \-U, \-\-udp
Use UDP to particular destination port for tracerouting (instead of increasing
the port per each probe). Default port is 53 (dns).
.TP
.BI \-UL
Use UDPLITE for tracerouting (default port is 53).
.TP
.B \-D, \-\-dccp
Use DCCP Requests for probes.
.TP
.BI \-P " protocol" ", --protocol=" protocol
Use raw packet of specified protocol for tracerouting. Default protocol is
253 (rfc3692).
.TP
.BI \--mtu
Discover MTU along the path being traced. Implies
.BR \-F\ \-N\ 1 .
New
.I mtu
is printed once in a form of
.B F=\fINUM
at the first probe of a hop which requires such
.I mtu
to be reached. (Actually, the correspond "frag needed" icmp message
normally is sent by the previous hop).
.br

.br
Note, that some routers might cache once the seen information
on a fragmentation. Thus you can receive the final mtu from a closer hop.
Try to specify an unusual
.I tos
by
.B \-t
, this can help for one attempt (then it can be cached there as well).
.br
See
.B \-F
option for more info.
.TP
.BI \--back
Print the number of backward hops when it seems different with the forward
direction. This number is guessed in assumption that remote hops send reply
packets with initial ttl set to either 64, or 128 or 255 (which seems
a common practice). It is printed as a negate value in a form of '-NUM' .
.SH LIST OF AVAILABLE METHODS
In general, a particular traceroute method may have to be chosen by
.BR \-M\ name ,
but most of the methods have their simple cmdline switches
(you can see them after the method name, if present).
.SS default
The traditional, ancient method of tracerouting. Used by default.
.P
Probe packets are udp datagrams with so-called "unlikely" destination ports.
The "unlikely" port of the first probe is 33434, then for each next probe
it is incremented by one. Since the ports are expected to be unused,
the destination host normally returns "icmp unreach port" as a final response.
(Nobody knows what happens when some application listens for such ports,
though).
.P
This method is allowed for unprivileged users.
.SS icmp \  \  \  \-I
Most usual method for now, which uses icmp echo packets for probes.
.br
If you can ping(8) the destination host, icmp tracerouting is applicable
as well.
.P
This method may be allowed for unprivileged users
since the kernel 3.0 (IPv4, for IPv6 since 3.11), which supports new
.I dgram icmp
(or
.IR \fR"\fIping\fR")
sockets. To allow such sockets, sysadmin should provide
.I net/ipv4/ping_group_range
sysctl range to match any group of the user.
.br
Options:
.TP
.B raw
Use only raw sockets (the traditional way).
.br
This way is tried first by default (for compatibility reasons),
then new dgram icmp sockets as fallback.
.TP
.B dgram
Use only dgram icmp sockets.
.SS tcp \  \  \  \ \-T
Well-known modern method, intended to bypass firewalls.
.br
Uses the constant destination port (default is 80, http).
.P
If some filters are present in the network path, then most probably
any "unlikely" udp ports (as for
.I default
method) or even icmp echoes (as for
.IR icmp )
are filtered, and whole tracerouting will just stop at such a firewall.
To bypass a network filter, we have to use only allowed protocol/port
combinations. If we trace for some, say, mailserver, then more likely
.B \-T \-p 25
can reach it, even when
.B \-I
can not.
.P
This method uses well-known "half-open technique", which prevents
applications on the destination host from seeing our probes at all.
Normally, a tcp syn is sent. For non-listened ports we receive tcp reset,
and all is done. For active listening ports we receive tcp syn+ack, but
answer by tcp reset (instead of expected tcp ack), this way the remote tcp
session is dropped even without the application ever taking notice.
.P
There is a couple of options for
.I tcp
method:
.TP
.B syn,ack,fin,rst,psh,urg,ece,cwr
Sets specified tcp flags for probe packet, in any combination.
.TP
.B flags\fR=\fInum
Sets the flags field in the tcp header exactly to
.IR num .
.TP
.B ecn
Send syn packet with tcp flags ECE and CWR (for Explicit Congestion
Notification, rfc3168).
.TP
.B sack,timestamps,window_scaling
Use the corresponding tcp header option in the outgoing probe packet.
.TP
.B sysctl
Use current sysctl
.IR "" ( "/proc/sys/net/*" )
setting for the tcp header options above and
.BR ecn .
Always set by default, if nothing else specified.
.TP
.B fastopen
Use fastopen tcp option (when
.BR syn ),
for initial cookie negotiation only.
.TP
.B mss\fR=[\fInum\fR]
Use value of
.I num
(or unchanged) for maxseg tcp header option (when
.BR syn ),
and discover its clamping along the path being traced.
New changed
.I mss
is printed once in a form of
.B M=\fINUM
at the first probe on which it was detected.
Note, some routers may return too short original fragment
in the time exceeded message, making the check impossible.
Besides that the responses may come in a different order.
All this can lead to a later place of the report
(using 
.BR \-N\ 1
can help for the order).
.TP
.B info
Print tcp flags and supported options
of final tcp replies when the target host is reached.
Allows to determine whether an application listens the port and
other useful things.
Supported tcp options are all that can be set by
.BR \-T\ \-O ,
ie. 
.IR mss ,
.IR sack ,
.IR timestamps ,
.IR window_scaling
and
.IR fastopen ,
with the similar output format (a value for
.IR mss
and just presence for others).
.P
Default options is
.BR syn,sysctl .
.SS tcpconn
An initial implementation of tcp method, simple using connect(2) call,
which does full tcp session opening. Not recommended for normal use, because
a destination application is always affected (and can be confused).
.SS udp \  \  \  \ \-U
Use udp datagram with constant destination port (default 53, dns).
.br
Intended to bypass firewall as well. 
.P
Note, that unlike in
.I tcp
method, the correspond application on the destination host
.B always
receive our probes (with random data), and most can easily be confused
by them. Most cases it will not respond to our packets though, so we will never
see the final hop in the trace. (Fortunately, it seems that at least
dns servers replies with something angry).
.P
This method is allowed for unprivileged users.
.SS udplite \  \ \-UL
Use udplite datagram for probes (with constant destination port,
default 53).
.P
This method is allowed for unprivileged users.
.br
Options:
.TP
.B coverage\fR=\fInum
Set udplite send coverage to
.IR num .
.SS dccp \  \ \-D
Use DCCP Request packets for probes (rfc4340).
.P
This method uses the same "half-open technique" as used for TCP.
The default destination port is 33434.
.P
Options:
.TP
.B service\fR=\fInum
Set DCCP service code to
.IR num
(default is 1885957735).
.SS raw \  \  \  \ \-P proto
Send raw packet of protocol
.IR proto .
.br
No protocol-specific headers are used, just IP header only.
.br
Implies
.B \-N\ 1\fR\ -w\ 5 .
.br
Options:
.TP
.B protocol\fR=\fIproto
Use IP protocol
.I proto
(default 253).
.SH NOTES
.PP
To speed up work, normally several probes are sent simultaneously.
On the other hand, it creates a "storm of packages", especially
in the reply direction. Routers can throttle the rate of icmp responses,
and some of replies can be lost. To avoid this, decrease the number
of simultaneous probes, or even set it to 1 (like in initial traceroute
implementation), i.e.
.B \-N 1
.PP
The final (target) host can drop some of the simultaneous probes,
and might even answer only the latest ones. It can lead to extra
"looks like expired" hops near the final hop. We use a smart algorithm
to auto-detect such a situation, but if it cannot help in your case, just use
.B \-N 1
too.
.PP
For even greater stability you can slow down the program's work by
.B \-z
option, for example use
.B \-z 0.5
for half-second pause between probes.
.PP
To avoid an extra waiting, we use adaptive algorithm for timeouts (see
.B \-w
option for more info). It can lead to premature expiry
(especially when response times differ at times) and printing "*"
instead of a time. In such a case, switch this algorithm off, by specifying
.B \-w
with the desired timeout only (for example,
.B \-w 5\fR).
.PP
If some hops report nothing for every method, the last chance to obtain
something is to use
.B ping -R
command (IPv4, and for nearest 8 hops only).
.SH SEE ALSO
.BR ping (8),
.BR ping6 (8),
.BR tcpdump (8),
.BR netstat (8)