1
2ethfabricanalysis(8)         EFSFFCLIRG (Man Page)        ethfabricanalysis(8)
3
4
5

NAME

7       ethfabricanalysis
8
9
10
11       Performs analysis of the fabric.
12

Syntax

14       ethfabricanalysis  [-b|-e]  [-s]  [-d   dir]  [-c  file] [-E  file] [-p
15       planes] [-T  topology_inputs] [-f  host_files]
16

Options

18       --help
19
20                 Produces full help text.
21
22       -b
23
24                 Specifies the baseline mode. Default is compare/check mode.
25
26       -e
27
28                 Evaluates health only. Default is compare/check mode.
29
30       -s
31
32                 Saves history of failures (errors/differences).
33
34       -d dir
35
36                 Specifies the top-level directory  for  saving  baseline  and
37                 history   of  failed  checks.  Default  is  /var/usr/lib/eth-
38                 tools/analysis
39
40       -c file
41
42                 Specifies  the  error  thresholds  config   file.Default   is
43                 /etc/eth-tools/ethmon.conf
44
45       -E file
46
47                 Specifies  Ethernet  Mgt  configuration  file. The default is
48                 /etc/eth-tools/mgt_config.xml.
49
50
51       -p planes
52
53                 Specifies Fabric planes separated by space.  The  default  is
54                 the  first  enabled plane defined in config file. Value 'ALL'
55                 will use all enabled planes.
56
57
58       -f host_files
59
60                 Hosts files separated by space. It overrides  the  HostsFiles
61                 defined  in  Mgt  config  file  for the corresponding planes.
62                 Value 'DEFAULT' will use the HostFile defined in  Mgt  config
63                 file for the corresponding plane
64
65
66       -T topology_inputs
67
68                 Specifies  the  name of topology input filenames separated by
69                 space. See Details and ethreport for more information.
70

Example

72       ethfabricanalysis
73
74       ethfabricanalysis -p 'p1 p2' -f 'hosts1 DEFAULT'
75
76       The fabric analysis tool checks the following:
77
78       •      Fabric links (both internal to switch and external cables)
79
80       •      Fabric components (nodes, links, systems, and  their  configura‐
81              tion)
82
83       •      Fabric error counters and link speed mismatches
84
85       NOTE:  The comparison includes components on the fabric. Therefore, op‐
86       erations such as shutting down a server cause the server to  no  longer
87       appear  on  the fabric and are flagged as a fabric change or failure by
88       ethfabricanalysis.
89
90

Environment Variables

92       The following environment variables are also used by this command:
93
94       FF_ANALYSIS_DIR
95
96                 Top-level directory for baselines and failed health checks.
97

Details

99       You can specify the topology_input file to be used with one of the fol‐
100       lowing methods:
101
102       •      On the command line using the -T option.
103
104       •      Using the TopologyFile specified in Ethernet Mgt config file.
105
106       If the specified file does not exist, no topology_input file is used.
107
108       For more information on topology_input, refer to ethreport
109
110       By  default,  the error analysis includes counters and slow links (that
111       is, links running below enabled speeds). You can change this using  the
112       FF_FABRIC_HEALTH  configuration  parameter  in ethfastfabric.conf. This
113       parameter specifies the ethreport options and reports to  be  used  for
114       the health analysis.
115
116       When  a  topology_input  file  is used, it can also be useful to extend
117       FF_FABRIC_HEALTH to include fabric topology verification  options  such
118       as -o verifylinks.
119
120       The  thresholds  for  counter  analysis  default to /etc/eth-tools/eth‐
121       mon.conf. However, you can specify an alternate configuration file  for
122       thresholds  using  the  -c  option. The ethmon.si.conf file can also be
123       used to check for any non-zero values for signal integrity  (SI)  coun‐
124       ters.
125
126       All  files  generated  by  ethfabricanalysis start with fabric in their
127       file name.
128
129       The ethfabricanalysis tool generates files such as the following within
130       FF_ANALYSIS_DIR :
131
132       Health Check
133
134
135       •      latest/fabric.<plane_name>.errors stdout of ethreport for errors
136              encountered during fabric error analysis.
137
138
139       •      latest/fabric.<plane_name>.errors.stderr  stderr  of   ethreport
140              during fabric error analysis.
141
142
143       Baseline
144
145
146       During  a  baseline  run,  the  following  files  are  also  created in
147       FF_ANALYSIS_DIR/latest.
148
149       •      baseline/fabric.<plane_name>.snapshot.xml ethreport snapshot  of
150              complete fabric components and configuration.
151
152
153       •      baseline/fabric.<plane_name>.comps  ethreport  summary of fabric
154              components and basic configuration.
155
156
157       •      baseline/fabric.<plane_name>.links ethreport summary of internal
158              and external links.
159
160
161       Full Analysis
162
163
164       •      latest/fabric.<plane_name>.snapshot.xml  ethreport  snapshot  of
165              complete fabric components and configuration.
166
167
168       •      latest/fabric.<plane_name>.snapshot.stderr stderr  of  ethreport
169              during snapshot.
170
171
172       •      latest/fabric.<plane_name>.errors stdout of ethreport for errors
173              encountered during fabric error analysis.
174
175
176       •      latest/fabric.<plane_name>.errors.stderr  stderr  of   ethreport
177              during fabric error analysis.
178
179
180       •      latest/fabric.<plane_name>.comps  stdout of ethreport for fabric
181              components and configuration.
182
183
184       •      latest/fabric.<plane_name>.comps.stderr stderr of ethreport  for
185              fabric components.
186
187
188       •      latest/fabric.<plane_name>.comps.diff  diff of baseline and lat‐
189              est fabric components.
190
191
192       •      latest/fabric.<plane_name>.links stdout of ethreport summary  of
193              internal and external links.
194
195
196       •      latest/fabric.<plane_name>.links.stderr stderr of ethreport sum‐
197              mary of internal and external links.
198
199
200       •      latest/fabric.<plane_name>.links.diff diff of baseline and  lat‐
201              est fabric internal and external links.
202
203
204       •      latest/fabric.<plane_name>.links.changes.stderr stderr of ethre‐
205              port comparison of links.
206
207
208       •      latest/fabric.<plane_name>.links.changes ethreport comparison of
209              links  against  baseline.  This is typically easier to read than
210              the links.diff file and contains the same information.
211
212
213       •      latest/fabric.<plane_name>.comps.changes.stderr stderr of ethre‐
214              port comparison of components.
215
216
217       •      latest/fabric.<plane_name>.comps.changes ethreport comparison of
218              components against baseline. This is typically  easier  to  read
219              than the comps.diff file and contains the same information.
220
221
222       The  .diff  and  .changes files are only created if differences are de‐
223       tected.
224
225       If the -s option is used and failures are detected,  files  related  to
226       the  checks  that  failed are also copied to the time-stamped directory
227       name under FF_ANALYSIS_DIR.
228

Fabric Items Checked Against the Baseline

230       Based on ethreport -o links:
231
232       •      Unconnected/down/missing cables
233
234       •      Added/moved cables
235
236       •      Changes in link width and speed
237
238       •      Changes to IfAddr in fabric (replacement of NIC or Switch  hard‐
239              ware)
240
241       •      Adding/Removing  Nodes  [NIC,  Virtual  NICs,  Virtual Switches,
242              Physical Switches,  Physical  Switch  internal  switching  cards
243              (leaf/spine)]
244
245       •      Changes to server or switch names
246
247       Based on ethreport -o comps:
248
249       •      Overlap with items from links report
250
251       •      Changes in port MTU
252
253       •      Changes in port speed/width enabled or supported
254
255       •      Changes  in NIC or switch device IDs/revisions/VendorID (for ex‐
256              ample, ASIC hardware changes)
257
258       •      Changes in port Capability mask (which  features/agents  run  on
259              port/server)
260
261       •      Changes to IOUs/IOCs/IOC Services provided
262
263
264

Fabric Items Also Checked During Health Check

266       Based on ethreport -s -o errors -o slowlinks:
267
268       •      error  counters  on  all  Intel(R)  Ethernet  Fabric ports (NIC,
269              switch external and switch  internal)  checked  against  config‐
270              urable thresholds.
271
272       •      Typically identifies potential fabric errors, such as symbol er‐
273              rors.
274
275       •      May also identify transient congestion, depending on  the  coun‐
276              ters that are monitored.
277
278       •      Link active speed/width as compared to Enabled speed.
279
280       •      Identifies  links  whose  active  speed/width  is < min (enabled
281              speed/width on each side of link).
282
283       •      This typically reflects bad cables or bad ports or poor  connec‐
284              tions.
285
286       •      Side effect is the verification of fabric health.
287
288
289
290Copyright(C) 2020-2022         Intel Corporation          ethfabricanalysis(8)