nexmon – Blame information for rev 1

Subversion Repositories:
Rev:
Rev Author Line No. Line
1 office 1 This file is a HOWTO for Wireshark developers. It describes how Wireshark
2 heuristic protocol dissectors work and how to write them.
3  
4 This file is compiled to give in depth information on Wireshark.
5 It is by no means all inclusive and complete. Please feel free to send
6 remarks and patches to the developer mailing list.
7  
8  
9 Prerequisites
10 -------------
11 As this file is an addition to README.dissector, it is essential to read
12 and understand that document first.
13  
14  
15 Why heuristic dissectors?
16 -------------------------
17 When Wireshark "receives" a packet, it has to find the right dissector to
18 start decoding the packet data. Often this can be done by known conventions,
19 e.g. the Ethernet type 0x0800 means "IP on top of Ethernet" - an easy and
20 reliable match for Wireshark.
21  
22 Unfortunately, these conventions are not always available, or (accidentally
23 or knowingly) some protocols don't care about those conventions and "reuse"
24 existing "magic numbers / tokens".
25  
26 For example TCP defines port 80 only for the use of HTTP traffic. But, this
27 convention doesn't prevent anyone from using TCP port 80 for some different
28 protocol, or on the other hand using HTTP on a port number different than 80.
29  
30 To solve this problem, Wireshark introduced the so called heuristic dissector
31 mechanism to try to deal with these problems.
32  
33  
34 How Wireshark uses heuristic dissectors?
35 ----------------------------------------
36 While Wireshark starts, heuristic dissectors (HD) register themselves slightly
37 different than "normal" dissectors, e.g. a HD can ask for any TCP packet, as
38 it *may* contain interesting packet data for this dissector. In reality more
39 than one HD will exist for e.g. TCP packet data.
40  
41 So if Wireshark has to decode TCP packet data, it will first try to find a
42 dissector registered directly for the TCP port used in that packet. If it
43 finds such a registered dissector it will just hand over the packet data to it.
44  
45 In case there is no such "normal" dissector, WS will hand over the packet data
46 to the first matching HD. Now the HD will look into the data and decide if that
47 data looks like something the dissector "is interested in". The return value
48 signals WS if the HD processed the data (so WS can stop working on that packet)
49 or if the heuristic didn't match (so WS tries the next HD until one matches -
50 or the data simply can't be processed).
51  
52 Note that it is possible to configure WS through preference settings so that it
53 hands off a packet to the heuristic dissectors before the "normal" dissectors
54 are called. This allows the HD the chance to receive packets and process them
55 differently than they otherwise would be. Of course if no HD is interested in
56 the packet, then the packet will ultimately get handed off to the "normal"
57 dissector as if the HD wasn't involved at all. As of this writing, the DCCP,
58 SCTP, TCP, TIPC and UDP dissectors all provide this capability via their
59 "Try heuristic sub-dissectors first" preference, but none of them have this
60 option enabled by default.
61  
62 Once a packet for a particular "connection" has been identified as belonging
63 to a particular protocol, Wireshark should then be set up to always directly
64 call the dissector for that protocol. This removes the overhead of having
65 to identify each packet of the connection heuristically.
66  
67  
68 How do these heuristics work?
69 -----------------------------
70 It's difficult to give a general answer here. The usual heuristic works as follows:
71  
72 A HD looks into the first few packet bytes and searches for common patterns that
73 are specific to the protocol in question. Most protocols starts with a
74 specific header, so a specific pattern may look like (synthetic example):
75  
76 1) first byte must be 0x42
77 2) second byte is a type field and can only contain values between 0x20 - 0x33
78 3) third byte is a flag field, where the lower 4 bits always contain the value 0
79 4) fourth and fifth bytes contain a 16 bit length field, where the value can't
80 be larger than 10000 bytes
81  
82 So the heuristic dissector will check incoming packet data for all of the
83 4 above conditions, and only if all of the four conditions are true there is a
84 good chance that the packet really contains the expected protocol - and the
85 dissector continues to decode the packet data. If one condition fails, it's
86 very certainly not the protocol in question and the dissector returns to WS
87 immediately "this is not my protocol" - maybe some other heuristic dissector
88 is interested!
89  
90 Obviously, this is *not* 100% bullet proof, but it's the best WS can offer to
91 its users here - and improving the heuristic is always possible if it turns out
92 that it's not good enough to distinguish between two given protocols.
93  
94 Note: The heuristic code in a dissector *must not* cause an exception
95 (before returning FALSE) as this will prevent following
96 heuristic dissector handoffs. In practice, this normally means
97 that a test should be done to verify that the required data is
98 available in the tvb before fetching from the tvb. (See the
99 example below).
100  
101  
102 Heuristic Code Example
103 ----------------------
104 You can find a lot of code examples in the Wireshark sources, e.g.:
105 grep -l heur_dissector_add epan/dissectors/*.c
106 returns 177 files (October 2015).
107  
108 For the above example criteria, the following code example might do the work
109 (combine this with the dissector skeleton in README.developer):
110  
111 XXX - please note: The following code examples were not tried in reality,
112 please report problems to the dev-list!
113  
114 --------------------------------------------------------------------------------------------
115  
116 static dissector_handle_t PROTOABBREV_tcp_handle;
117 static dissector_handle_t PROTOABBREV_pdu_handle;
118  
119 /* Heuristics test */
120 static gboolean
121 test_PROTOABBREV(packet_info *pinfo _U_, tvbuff_t *tvb, int offset _U_, void *data _U_)
122 {
123 /* 0) Verify needed bytes available in tvb so tvb_get...() doesn't cause exception.
124 if (tvb_captured_length(tvb) < 5)
125 return FALSE;
126  
127 /* 1) first byte must be 0x42 */
128 if ( tvb_get_guint8(tvb, 0) != 0x42 )
129 return FALSE;
130  
131 /* 2) second byte is a type field and only can contain values between 0x20-0x33 */
132 if ( tvb_get_guint8(tvb, 1) < 0x20 || tvb_get_guint8(tvb, 1) > 0x33 )
133 return FALSE;
134  
135 /* 3) third byte is a flag field, where the lower 4 bits always contain the value 0 */
136 if ( tvb_get_guint8(tvb, 2) & 0x0f )
137 return FALSE;
138  
139 /* 4) fourth and fifth bytes contains a 16 bit length field, where the value can't be longer than 10000 bytes */
140 /* Assumes network byte order */
141 if ( tvb_get_ntohs(tvb, 3) > 10000 )
142 return FALSE;
143  
144 /* Assume it's your packet ... */
145 return TRUE;
146 }
147  
148 /* Dissect the complete PROTOABBREV pdu */
149 static int
150 dissect_PROTOABBREV_pdu(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data _U_)
151 {
152 /* Dissection ... */
153  
154 return tvb_reported_length(tvb);
155 }
156  
157 /* For tcp_dissect_pdus() */
158 static guint
159 get_PROTOABBREV_len(packet_info *pinfo _U_, tvbuff_t *tvb, int offset, void *data _U_)
160 {
161 return (guint) tvb_get_ntohs(tvb, offset+3);
162 }
163  
164 static int
165 dissect_PROTOABBREV_tcp(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
166 {
167 tcp_dissect_pdus(tvb, pinfo, tree, TRUE, 5,
168 get_PROTOABBREV_len, dissect_PROTOABBREV_pdu, data);
169 return tvb_reported_length(tvb);
170 }
171  
172 static gboolean
173 dissect_PROTOABBREV_heur_tcp(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
174 {
175 if (!test_PROTOABBREV(pinfo, tvb, 0, data))
176 return FALSE;
177  
178 /* specify that dissect_PROTOABBREV is to be called directly from now on for
179 * packets for this "connection" ... but only do this if your heuristic sits directly
180 * on top of (was called by) a dissector which established a conversation for the
181 * protocol "port type". In other words: only directly over TCP, UDP, DCCP, ...
182 * otherwise you'll be overriding the dissector that called your heuristic dissector.
183 */
184 conversation = find_or_create_conversation(pinfo);
185 conversation_set_dissector(conversation, PROTOABBREV_tcp_handle);
186  
187 /* and do the dissection */
188 dissect_PROTOABBREV_tcp(tvb, pinfo, tree, data);
189  
190 return (TRUE);
191 }
192  
193 static int
194 dissect_PROTOABBREV_udp(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
195 {
196 udp_dissect_pdus(tvb, pinfo, tree, TRUE, 5, NULL,
197 get_PROTOABBREV_len, dissect_PROTOABBREV_pdu, data);
198 return tvb_reported_length(tvb);
199 }
200  
201 static gboolean
202 dissect_PROTOABBREV_heur_udp(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
203 {
204 ...
205 /* and do the dissection */
206 return (udp_dissect_pdus(tvb, pinfo, tree, TRUE, 5, test_PROTOABBREV,
207 get_PROTOABBREV_len, dissect_PROTOABBREV_pdu, data) != 0);
208 }
209  
210 void
211 proto_reg_handoff_PROTOABBREV(void)
212 {
213 PROTOABBREV_tcp_handle = create_dissector_handle(dissect_PROTOABBREV_tcp,
214 proto_PROTOABBREV);
215 PROTOABBREV_pdu_handle = create_dissector_handle(dissect_PROTOABBREV_pdu,
216 proto_PROTOABBREV);
217  
218 /* register as heuristic dissector for both TCP and UDP */
219 heur_dissector_add("tcp", dissect_PROTOABBREV_heur_tcp, "PROTOABBREV over TCP",
220 "PROTOABBREV_tcp", proto_PROTOABBREV, HEURISTIC_ENABLE);
221 heur_dissector_add("udp", dissect_PROTOABBREV_heur_udp, "PROTOABBREV over UDP",
222 "PROTOABBREV_udp", proto_PROTOABBREV, HEURISTIC_ENABLE);
223  
224 #ifdef OPTIONAL
225 /* It's possible to write a dissector to be a dual heuristic/normal dissector */
226 /* by also registering the dissector "normally". */
227 dissector_add_uint("ip.proto", IP_PROTO_PROTOABBREV, PROTOABBREV_pdu_handle);
228 #endif
229 }
230  
231  
232 Please note, that registering a heuristic dissector is only possible for a
233 small variety of protocols. In most cases a heuristic is not needed, and
234 adding the support would only add unused code to the dissector.
235  
236 TCP and UDP are prominent examples that support HDs, as there seems to be a
237 tendency to re-use known port numbers for new protocols. But TCP and UDP are
238 not the only dissectors that provide support for HDs. You can find more
239 examples by searching the Wireshark sources as follows:
240 grep -l register_heur_dissector_list epan/dissectors/packet-*.c
241 returns 45 files (November 2014).