4 editcap - Edit and/or translate the format of capture files
9 S<[ B<-A> E<lt>start timeE<gt> ]>
10 S<[ B<-B> E<lt>stop timeE<gt> ]>
11 S<[ B<-c> E<lt>packets per fileE<gt> ]>
12 S<[ B<-C> E<lt>choplenE<gt> ]>
13 S<[ B<-E> E<lt>error probabilityE<gt> ]>
14 S<[ B<-F> E<lt>file formatE<gt> ]>
16 S<[ B<-i> E<lt>seconds per fileE<gt> ]>
19 S<[ B<-s> E<lt>snaplenE<gt> ]>
20 S<[ B<-S> E<lt>strict time adjustmentE<gt> ]>
21 S<[ B<-t> E<lt>time adjustmentE<gt> ]>
22 S<[ B<-T> E<lt>encapsulation typeE<gt> ]>
26 S<[ I<packet#>[-I<packet#>] ... ]>
30 S< B<-D> E<lt>dup windowE<gt> > |
31 S< B<-w> E<lt>dup time windowE<gt> >
38 B<Editcap> is a program that reads some or all of the captured packets from the
39 I<infile>, optionally converts them in various ways and writes the
40 resulting packets to the capture I<outfile> (or outfiles).
42 By default, it reads all packets from the I<infile> and writes them to the
43 I<outfile> in pcap file format.
45 An optional list of packet numbers can be specified on the command tail;
46 individual packet numbers separated by whitespace and/or ranges of packet
47 numbers can be specified as I<start>-I<end>, referring to all packets from
48 I<start> to I<end>. By default the selected packets with those numbers will
49 I<not> be written to the capture file. If the B<-r> flag is specified, the
50 whole packet selection is reversed; in that case I<only> the selected packets
51 will be written to the capture file.
53 B<Editcap> can also be used to remove duplicate packets. Several different
54 options (B<-d>, B<-D> and B<-w>) are used to control the packet window
55 or relative time window to be used for duplicate comparison.
57 B<Editcap> is able to detect, read and write the same capture files that
58 are supported by B<Wireshark>.
59 The input file doesn't need a specific filename extension; the file
60 format and an optional gzip compression will be automatically detected.
61 Near the beginning of the DESCRIPTION section of wireshark(1) or
62 L<http://www.wireshark.org/docs/man-pages/wireshark.html>
63 is a detailed description of the way B<Wireshark> handles this, which is
64 the same way B<Editcap> handles this.
66 B<Editcap> can write the file in several output formats. The B<-F>
67 flag can be used to specify the format in which to write the capture
68 file; B<editcap -F> provides a list of the available output formats.
74 =item -A E<lt>start timeE<gt>
76 Saves only the packets whose timestamp is on or after start time.
77 The time is given in the following format YYYY-MM-DD HH:MM:SS
79 =item -B E<lt>stop timeE<gt>
81 Saves only the packets whose timestamp is before stop time.
82 The time is given in the following format YYYY-MM-DD HH:MM:SS
84 =item -c E<lt>packets per fileE<gt>
86 Splits the packet output to different files based on uniform packet counts
87 with a maximum of <packets per file> each. Each output file will
88 be created with a suffix -nnnnn, starting with 00000. If the specified
89 number of packets is written to the output file, the next output file is
90 opened. The default is to use a single output file.
92 =item -C E<lt>choplenE<gt>
94 Sets the chop length to use when writing the packet data. Each packet is
95 chopped by <choplen> bytes of data. Positive values chop at the packet
96 beginning while negative values chop at the packet end.
98 This is useful for chopping headers for decapsulation of an entire capture or
99 in the rare case that the conversion between two file formats leaves some random
100 bytes at the end of each packet.
102 NOTE: This option can be used more than once, effectively allowing you to chop
103 bytes from the beginning of a packet as well as from the end of a packet in a
108 Attempts to remove duplicate packets. The length and MD5 hash of the
109 current packet are compared to the previous four (4) packets. If a
110 match is found, the current packet is skipped. This option is equivalent
111 to using the option B<-D 5>.
113 =item -D E<lt>dup windowE<gt>
115 Attempts to remove duplicate packets. The length and MD5 hash of the
116 current packet are compared to the previous <dup window> - 1 packets.
117 If a match is found, the current packet is skipped.
119 The use of the option B<-D 0> combined with the B<-v> option is useful
120 in that each packet's Packet number, Len and MD5 Hash will be printed
121 to standard out. This verbose output (specifically the MD5 hash strings)
122 can be useful in scripts to identify duplicate packets across trace
125 The <dup window> is specified as an integer value between 0 and 1000000 (inclusive).
127 NOTE: Specifying large <dup window> values with large tracefiles can
128 result in very long processing times for B<editcap>.
130 =item -E E<lt>error probabilityE<gt>
132 Sets the probability that bytes in the output file are randomly changed.
133 B<Editcap> uses that probability (between 0.0 and 1.0 inclusive)
134 to apply errors to each data byte in the file. For instance, a
135 probability of 0.02 means that each byte has a 2% chance of having an error.
137 This option is meant to be used for fuzz-testing protocol dissectors.
139 =item -F E<lt>file formatE<gt>
141 Sets the file format of the output capture file.
142 B<Editcap> can write the file in several formats, B<editcap -F>
143 provides a list of the available output formats. The default
144 is the B<pcap> format.
148 Prints the version and options and exits.
150 =item -i E<lt>seconds per fileE<gt>
152 Splits the packet output to different files based on uniform time intervals
153 using a maximum interval of <seconds per file> each. Each output file will
154 be created with a suffix -nnnnn, starting with 00000. If packets for the specified
155 time interval are written to the output file, the next output file is
156 opened. The default is to use a single output file.
160 Adjust the original frame length accordingly when chopping and/or snapping
161 (in addition to the captured length, which is always adjusted regardless of
162 whether B<-L> is specified or not). See also B<-C <choplen>> and B<-s <snaplen>>.
166 Reverse the packet selection.
167 Causes the packets whose packet numbers are specified on the command
168 line to be written to the output capture file, instead of discarding them.
170 =item -s E<lt>snaplenE<gt>
172 Sets the snapshot length to use when writing the data.
173 If the B<-s> flag is used to specify a snapshot length, packets in the
174 input file with more captured data than the specified snapshot length
175 will have only the amount of data specified by the snapshot length
176 written to the output file.
178 This may be useful if the program that is
179 to read the output file cannot handle packets larger than a certain size
180 (for example, the versions of snoop in Solaris 2.5.1 and Solaris 2.6
181 appear to reject Ethernet packets larger than the standard Ethernet MTU,
182 making them incapable of handling gigabit Ethernet captures if jumbo
185 =item -S E<lt>strict time adjustmentE<gt>
187 Time adjust selected packets to insure strict chronological order.
189 The <strict time adjustment> value represents relative seconds
190 specified as [-]I<seconds>[I<.fractional seconds>].
192 As the capture file is processed each packet's absolute time is
193 I<possibly> adjusted to be equal to or greater than the previous
194 packet's absolute timestamp depending on the <strict time
197 If <strict time adjustment> value is 0 or greater (e.g. 0.000001)
198 then B<only> packets with a timestamp less than the previous packet
199 will adjusted. The adjusted timestamp value will be set to be
200 equal to the timestamp value of the previous packet plus the value
201 of the <strict time adjustment> value. A <strict time adjustment>
202 value of 0 will adjust the minimum number of timestamp values
203 necessary to insure that the resulting capture file is in
204 strict chronological order.
206 If <strict time adjustment> value is specified as a
207 negative value, then the timestamp values of B<all>
208 packets will be adjusted to be equal to the timestamp value
209 of the previous packet plus the absolute value of the
210 <lt>strict time adjustment<gt> value. A <strict time
211 adjustment> value of -0 will result in all packets
212 having the timestamp value of the first packet.
214 This feature is useful when the trace file has an occasional
215 packet with a negative delta time relative to the previous
218 =item -t E<lt>time adjustmentE<gt>
220 Sets the time adjustment to use on selected packets.
221 If the B<-t> flag is used to specify a time adjustment, the specified
222 adjustment will be applied to all selected packets in the capture file.
223 The adjustment is specified as [-]I<seconds>[I<.fractional seconds>].
224 For example, B<-t> 3600 advances the timestamp on selected packets by one
225 hour while B<-t> -0.5 reduces the timestamp on selected packets by
228 This feature is useful when synchronizing dumps
229 collected on different machines where the time difference between the
230 two machines is known or can be estimated.
232 =item -T E<lt>encapsulation typeE<gt>
234 Sets the packet encapsulation type of the output capture file.
235 If the B<-T> flag is used to specify an encapsulation type, the
236 encapsulation type of the output capture file will be forced to the
238 B<editcap -T> provides a list of the available types. The default
239 type is the one appropriate to the encapsulation type of the input
243 forces the encapsulation type of the output file to be the specified
244 type; the packet headers of the packets will not be translated from the
245 encapsulation type of the input capture file to the specified
246 encapsulation type (for example, it will not translate an Ethernet
247 capture to an FDDI capture if an Ethernet capture is read and 'B<-T
248 fddi>' is specified). If you need to remove/add headers from/to a
249 packet, you will need od(1)/text2pcap(1).
253 Causes B<editcap> to print verbose messages while it's working.
255 Use of B<-v> with the de-duplication switches of B<-d>, B<-D> or B<-w>
256 will cause all MD5 hashes to be printed whether the packet is skipped
259 =item -w E<lt>dup time windowE<gt>
261 Attempts to remove duplicate packets. The current packet's arrival time
262 is compared with up to 1000000 previous packets. If the packet's relative
263 arrival time is I<less than or equal to> the <dup time window> of a previous packet
264 and the packet length and MD5 hash of the current packet are the same then
265 the packet to skipped. The duplicate comparison test stops when
266 the current packet's relative arrival time is greater than <dup time window>.
268 The <dup time window> is specified as I<seconds>[I<.fractional seconds>].
270 The [.fractional seconds] component can be specified to nine (9) decimal
271 places (billionths of a second) but most typical trace files have resolution
272 to six (6) decimal places (millionths of a second).
274 NOTE: Specifying large <dup time window> values with large tracefiles can
275 result in very long processing times for B<editcap>.
277 NOTE: The B<-w> option assumes that the packets are in chronological order.
278 If the packets are NOT in chronological order then the B<-w> duplication
279 removal option may not identify some duplicates.
285 To see more detailed description of the options use:
289 To shrink the capture file by truncating the packets at 64 bytes and writing it as Sun snoop file use:
291 editcap -s 64 -F snoop capture.pcap shortcapture.snoop
293 To delete packet 1000 from the capture file use:
295 editcap capture.pcap sans1000.pcap 1000
297 To limit a capture file to packets from number 200 to 750 (inclusive) use:
299 editcap -r capture.pcap small.pcap 200-750
301 To get all packets from number 1-500 (inclusive) use:
303 editcap -r capture.pcap first500.pcap 1-500
307 editcap capture.pcap first500.pcap 501-9999999
309 To exclude packets 1, 5, 10 to 20 and 30 to 40 from the new file use:
311 editcap capture.pcap exclude.pcap 1 5 10-20 30-40
313 To select just packets 1, 5, 10 to 20 and 30 to 40 for the new file use:
315 editcap -r capture.pcap select.pcap 1 5 10-20 30-40
317 To remove duplicate packets seen within the prior four frames use:
319 editcap -d capture.pcap dedup.pcap
321 To remove duplicate packets seen within the prior 100 frames use:
323 editcap -D 101 capture.pcap dedup.pcap
325 To remove duplicate packets seen I<equal to or less than> 1/10th of a second:
327 editcap -w 0.1 capture.pcap dedup.pcap
329 To display the MD5 hash for all of the packets (and NOT generate any
332 editcap -v -D 0 capture.pcap /dev/null
334 or on Windows systems
336 editcap -v -D 0 capture.pcap NUL
338 To advance the timestamps of each packet forward by 3.0827 seconds:
340 editcap -t 3.0827 capture.pcap adjusted.pcap
342 To insure all timestamps are in strict chronological order:
344 editcap -S 0 capture.pcap adjusted.pcap
346 To introduce 5% random errors in a capture file use:
348 editcap -E 0.05 capture.pcap capture_error.pcap
352 pcap(3), wireshark(1), tshark(1), mergecap(1), dumpcap(1), capinfos(1),
353 text2pcap(1), od(1), pcap-filter(7) or tcpdump(8)
357 B<Editcap> is part of the B<Wireshark> distribution. The latest version
358 of B<Wireshark> can be found at L<http://www.wireshark.org>.
360 HTML versions of the Wireshark project man pages are available at:
361 L<http://www.wireshark.org/docs/man-pages>.
367 Richard Sharpe <sharpe[AT]ns.aus.com>
372 Guy Harris <guy[AT]alum.mit.edu>
373 Ulf Lamping <ulf.lamping[AT]web.de>