Debian Codegraph @

rfc1369.html

raw
1 <pre>Network Working Group F. Kastenholz
2 Request for Comments: 1369 FTP Software
3 October 1992
4
5
6 <span class="h1">Implementation Notes and Experience for</span>
7 <span class="h1">The Internet Ethernet MIB</span>
8
9 Status of this Memo
10
11 This memo provides information for the Internet community. It does
12 not specify an Internet standard. Distribution of this memo is
13 unlimited.
14
15 Table of Contents
16
17 <a href="#section-1">1</a>. Introduction ................................................ <a href="#page-1">1</a>
18 <a href="#section-2">2</a>. Observations ................................................ <a href="#page-2">2</a>
19 <a href="#section-3">3</a>. Conclusions ................................................. <a href="#page-3">3</a>
20 <a href="#section-4">4</a>. Final Action ................................................ <a href="#page-4">4</a>
21 <a href="#section-5">5</a>. Implementation Data ......................................... <a href="#page-5">5</a>
22 <a href="#section-6">6</a>. Security Considerations ..................................... <a href="#page-7">7</a>
23 <a href="#section-7">7</a>. Author's Address ............................................ <a href="#page-7">7</a>
24
25 <span class="h2"><a class="selflink" id="section-1" href="#section-1">1</a>. Introduction</span>
26
27 The Ethernet MIB Working group has been tasked with the following two
28 work items:
29
30 1) Develop a document explaining the rationale for assigning
31 MANDATORY status to MIB variables which are optional in
32 the relevant IEEE 802.3 specification (the technical
33 basis for the Internet Ethernet MIB). This shall not be a
34 standards-track document.
35
36 (2) Develop an implementation report on the Ethernet MIB.
37 This report shall cover MIB variables which are
38 implemented in both Ethernet interface chips, and in
39 software (i.e., drivers), and discuss the issues
40 pertaining to both. This report shall also summarize
41 field experience with the MIB variables, especially
42 concentrating on those variables which are in dispute.
43 This document shall not be a standards-track document.
44 While the Ethernet MIB is progressing through the
45 standardization process, this document shall be
46 periodically updated to reflect the latest implementation
47 and operational experience.
48
49
50
51
52 <span class="grey">Kastenholz [Page 1]</span></pre>
53 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-2" ></span>
54 <span class="grey"><a href="./rfc1369">RFC 1369</a> Ethernet MIB Implementations October 1992</span>
55
56
57 This document reflects the currently known status of 11 different
58 implementations of the MIB by 7 different vendors on 7 different
59 Ethernet interface chips.
60
61 <span class="h2"><a class="selflink" id="section-2" href="#section-2">2</a>. Observations</span>
62
63 There are some interesting points to be noted from this information:
64
65 1) Only 4 variables are actually implemented in all
66 implementations: AlignmentErrors, FCSErrors,
67 ExcessiveCollisions and InternalMacTransmitErrors.
68
69 2) There were another five variables implemented in all but
70 one of the reported implementations,
71 SingleCollisionFrames, MultipleCollisionFrames,
72 LateCollisions, FrameTooLongs, and CarrierSenseErrors.
73
74 Three of these variables exist in implementations that
75 use the same chip as the implementation that does not
76 contain the variable. Specifically:
77
78 A) SingleCollisionFrames is not implemented in
79 implementation number 3, which uses the AMD LANCE.
80 However, other AMD LANCE implementations (7, 8, and 10)
81 do implement the variable, implying that it is
82 available on the LANCE.
83
84 B) MultipleCollisionFrames is not implemented in
85 implementation number 3, which uses the AMD LANCE.
86 However, other AMD LANCE implementations (7, 8, and 10)
87 do implement the variable, implying that it is
88 available on the LANCE.
89
90 C) LateCollisions is not implemented in implementation
91 number 1, which uses the Intel 82586. However, another
92 Intel 82586 based implementation (11) does implement
93 the variable, implying that it is available on the
94 Intel 82586.
95
96 D) CarrierSenseErrors is not implemented on implementation
97 number 2, which is based on the Fujitsu 86950 chip.
98 However, there is only one implementation based on this
99 chip and I have not been able to locate a data sheet on
100 this part so no conclusion can be drawn at this time.
101
102 E) FrameTooLongs is not implemented on implementation
103 number 5, which is based on the National NIC 8390 chip.
104 However, there is only one implementation based on this
105
106
107
108 <span class="grey">Kastenholz [Page 2]</span></pre>
109 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-3" ></span>
110 <span class="grey"><a href="./rfc1369">RFC 1369</a> Ethernet MIB Implementations October 1992</span>
111
112
113 chip and I have not been able to locate a data sheet on
114 this part. It should also be noted that this variable
115 is easily maintained by software as a "driver-level"
116 function.
117
118 (3) Of the 22 variables in the MIB, 11, or 1/2 of the
119 variables, were implemented in about 1/2 or less of the
120 implementations.
121
122 4) The number of variables implemented per implementation
123 ranges from a low of 11 to a high of 16. The average
124 number of variables truly implemented is 12.8.
125
126 5) The IEEE 802.3 encapsulation-specific variables
127 (InRangeLengthErrors, and OutOfRangeLengthFields) are in
128 2 and 0 implementations respectively.
129
130 <span class="h2"><a class="selflink" id="section-3" href="#section-3">3</a>. Conclusions</span>
131
132 From this, the author concludes that:
133
134 The control variables (IntializeMAC, etc.) are not widely
135 implemented, but this may be due to an aversion to implementing
136 writable variables until security is in place.
137
138 One vendor has stated that the reason that these variables were not
139 implemented was that the vendor did not believe the variables to be
140 useful, and that they were hard to implement. Furthermore, this
141 vendor has recommended dropping the variables entirely.
142
143 The two IEEE 802.3 encapsulation variables (InRangeLengthErrors and
144 OutOfRangeLengthFields) are barely implemented. In Santa Fe, the
145 Working group discussed moving them to an optional, 802.3 specific,
146 group. The author believes that this is justified by this
147 implementation data.
148
149 The collision histogram variables are also barely implemented. They
150 should be in their own optional group -- and they are.
151
152 Of the remaining 13 statistical variables, 9 of them are in 10 or 11
153 implementations. This is good.
154
155 Two of them (SQETestErrors and ExcessiveDeferrals) are in 3 and 1
156 implementations, respectively. This is bad.
157
158 The remaining variables (DeferredTransmissions and
159 InternalMacReceiveErrors) are in 8 or 9 implementations.
160
161
162
163
164 <span class="grey">Kastenholz [Page 3]</span></pre>
165 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-4" ></span>
166 <span class="grey"><a href="./rfc1369">RFC 1369</a> Ethernet MIB Implementations October 1992</span>
167
168
169 It should be noted that one of the two systems that do not implement
170 DeferredTransmissions is based on the AMD LANCE, and other AMD LANCE
171 based systems do implement this counter, leading to the conclusion
172 that DeferredTransmissions could easily be on all but one of the
173 implementations.
174
175 The other such variable, InternalMacReceiveErrors, is a general
176 catchall for all other errors. If no other errors are detected by the
177 hardware or software then returning 0 for the counter is perfectly
178 acceptable.
179
180 This all seems to imply either:
181
182 1) Splitting the statistics group into two groups, one of
183 which is optional and contains SQETestErrors and
184 ExcessiveDeferrals, or
185
186 2) Eliminating SQETestErrors and ExcessiveDeferrals) from
187 the MIB.
188
189 The variables with 8 or 9 implementations are a bit more problematic.
190 They are implemented in more than 2/3s of the implementations, but it
191 may not be appropriate to call this widespread implementation.
192 However, it seems to be safe to conclude that the non-implementations
193 of these variables is due to local implementation considerations
194 rather than a fundamental lack of support for the variable.
195
196 <span class="h2"><a class="selflink" id="section-4" href="#section-4">4</a>. Final Action</span>
197
198 After consideration at the San Diego IETF Meeting on 17 March 1992,
199 the Ethernet MIB Working Group made the following recommendations:
200
201 1) The dot3TestTdrValue object will be deprecated from the
202 standard mib. There are effectively no implementations
203 of this object, and some chips were reported to return an
204 incorrect value for the TDR count.
205
206 2) The dot3StatsInRangeLengthErrors object and the
207 dot3StatsOutOfRangeLengthFields object will be deprecated
208 from the MIB. These objects were not widely implemented
209 and their utility in diagnosing network problems was
210 strongly questioned.
211
212 3) The dot3InitializeMac object, the dot3MacSubLayerStatus
213 object, the dot3MulticastReceiveStatus object, and the
214 dot3TxEnabled object will be deprecated from the MIB.
215 These objects were not widely implemented and their
216 utility in diagnosing network problems was strongly
217
218
219
220 <span class="grey">Kastenholz [Page 4]</span></pre>
221 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-5" ></span>
222 <span class="grey"><a href="./rfc1369">RFC 1369</a> Ethernet MIB Implementations October 1992</span>
223
224
225 questioned.
226
227 4) The dot3StatsExcessiveDeferrals object will be deprecated
228 from the MIB. Only one system implemented this object.
229 Furthermore, its exact definition was called into question.
230
231 5) The dot3StatsSQETestErrors object received few
232 implementations. However, the working group strongly
233 supported its retention in the MIB on the basis that
234 certain forms of transceiver and cable errors that are
235 not uncommon can only be detected with this counter.
236
237 6) The collision histogram table (dot3CollTable) will be
238 kept as an optional group, even though the objects are
239 not widely implemented nor is there hardware support on
240 all reported chips.
241
242 <span class="h2"><a class="selflink" id="section-5" href="#section-5">5</a>. Implementation Data</span>
243
244 The following raw data has been provided by vendors, each developing
245 an implementation of the Ethernet MIB. Each reported implementation
246 has a separate column in the following table. For each
247 implementation/MIB Variable, a single character code has been entered
248 indicating the rough implementation status of the variable. These
249 codes are:
250
251 Y Fully implemented, reports a truthful count, or
252 indication of state. All values may be written to the
253 variable with the expected action occurring.
254
255 N Not implemented at all. Would return a noSuchName error
256 if accessed.
257
258 C Implemented but returns a constant value for gets and
259 returns a badValue error for any set attempt to set the
260 variable to a value other than this constant (writable
261 variables only).
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276 <span class="grey">Kastenholz [Page 5]</span></pre>
277 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-6" ></span>
278 <span class="grey"><a href="./rfc1369">RFC 1369</a> Ethernet MIB Implementations October 1992</span>
279
280
281 MIB Implementation
282 Variable 1 2 3 4 5 6 7 8 9 10 11 Yesses
283
284 InitializeMac C C Y Y Y Y Y C7 C7 N Y 6
285 MacSubLayerStatus C C Y Y Y Y Y C7 C7 N C 5
286 MulticastReceiveStatus C C Y C3 Y C C C7 C7 N C 2
287 TxEnabled C C Y Y Y Y Y C7 C7 N C 5
288 TestTdrValue C 1 C C4 C C C C4 C4 N C 1
289
290 AlignmentErrors Y Y Y Y Y Y Y Y Y Y Y 11
291 FCSErrors Y Y Y Y Y Y Y Y Y Y Y 11
292 SingleCollisionFrames Y Y Y N Y Y Y Y Y Y Y 10
293 MultipleCollisionFrames Y Y Y N Y Y Y Y Y Y Y 10
294 SQETestErrors Y C C C Y C C C C Y C 3
295 DeferredTransmissions Y C Y N Y Y Y Y Y Y Y 9
296 LateCollisions C Y Y Y Y Y Y Y Y Y Y 10
297 ExcessiveCollisions Y Y Y Y Y Y Y Y Y Y Y 11
298 InternalMacTransmitErrors Y Y Y Y Y Y Y Y Y Y Y 11
299 CarrierSenseErrors Y C Y Y Y Y Y Y Y Y Y 10
300 ExcessiveDeferrals C C Y C C C C C C N C 1
301 FrameTooLongs Y Y2 Y Y C Y Y Y Y Y Y 10
302 InRangeLengthErrors C C C N5 C Y Y C C N C 2
303 OutOfRangeLengthFields C C C C6 C C C C C N C 0
304 InternalMacReceiveErrors Y Y Y Y Y C C Y Y Y C 8
305
306 CollCount Y Y C N N N N C C N Y 3
307 CollFrequencies Y Y C N N N N C C N Y 3
308 Yesses 13 11 16 11 15 14 14 11 11 12 13
309
310
311 Notes:
312
313 1 does not implement TDR test, but reports TDR from last
314 collision!
315
316 2 Not supported by the chip, detected solely in software.
317
318 3 But set to disabled(2) -&gt; badValue
319
320 4 Underlying TDR function not implemented on this chip.
321
322 5 Only counts frames too short though.
323
324 6 Due to Ethernet encapsulation
325
326 7 Implementation does not support set operations but
327 reports the correct value for these.
328
329
330
331
332 <span class="grey">Kastenholz [Page 6]</span></pre>
333 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-7" ></span>
334 <span class="grey"><a href="./rfc1369">RFC 1369</a> Ethernet MIB Implementations October 1992</span>
335
336
337 The implementations are:
338
339 Implementation Vendor Chip
340 1 1 Intel 82586
341 2 1 Fujitsu 86950
342 3 2 Sonic
343 4 3 AMD Lance
344 5 4 National NIC 8390C
345 6 4 Intel 82596
346 7 4 AMD Lance
347 8 5 AMD Lance
348 9 5 AMD ILACC
349 10 6 AMD Lance
350 11 7 Intel 82586
351
352 <span class="h2"><a class="selflink" id="section-6" href="#section-6">6</a>. Security Considerations</span>
353
354 Security issues are not discussed in this memo.
355
356 <span class="h2"><a class="selflink" id="section-7" href="#section-7">7</a>. Author's Address</span>
357
358 Frank J. Kastenholz
359 FTP Software
360 2 High Street
361 North Andover Mass 01845
362
363 Phone: 508-685-4000
364 EMail: [email protected]
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388 Kastenholz [Page 7]
389 </pre>