1 .. This work is licensed under a Creative Commons Attribution 4.0 International License.
2 .. SPDX-License-Identifier: CC-BY-4.0
3 .. (c) Cisco Systems, Inc
9 This section covers a few examples on how to run NFVbench with multiple different settings.
10 Below are shown the most common and useful use-cases and explained some fields from a default config file.
12 How to change any NFVbench run configuration (CLI)
13 --------------------------------------------------
14 NFVbench always starts with a default configuration which can further be refined (overridden) by the user from the CLI or from REST requests.
16 At first have a look at the default config:
20 nfvbench --show-default-config
22 It is sometimes useful derive your own configuration from a copy of the default config:
26 nfvbench --show-default-config > nfvbench.cfg
28 At this point you can edit the copy by:
30 - removing any parameter that is not to be changed (since NFVbench will always load the default configuration, default values are not needed)
31 - edit the parameters that are to be changed changed
33 A run with the new confguration can then simply be requested using the -c option and by using the actual path of the configuration file
34 as seen from inside the container (in this example, we assume the current directory is mapped to /tmp/nfvbench in the container):
38 nfvbench -c /tmp/nfvbench/nfvbench.cfg
40 The same -c option also accepts any valid yaml or json string to override certain parameters without having to create a configuration file.
42 NFVbench provides many configuration options as optional arguments. For example the number of flows can be specified using the --flow-count option.
44 The flow count option can be specified in any of 3 ways:
46 - by providing a confguration file that has the flow_count value to use (-c myconfig.yaml and myconfig.yaml contains 'flow_count: 100k')
47 - by passing that yaml paremeter inline (-c "flow_count: 100k") or (-c "{flow_count: 100k}")
48 - by using the flow count optional argument (--flow-count 100k)
50 Showing the running configuration
51 ---------------------------------
53 Because configuration parameters can be overriden, it is sometimes useful to show the final configuration (after all oevrrides are done) by using the --show-config option.
54 This final configuration is also called the "running" configuration.
56 For example, this will only display the running configuration (without actually running anything):
60 nfvbench -c "{flow_count: 100k, debug: true}" --show-config
63 Connectivity and Configuration Check
64 ------------------------------------
66 NFVbench allows to test connectivity to devices used with the selected packet path.
67 It runs the whole test, but without actually sending any traffic.
68 It is also a good way to check if everything is configured properly in the configuration file and what versions of components are used.
70 To verify everything works without sending any traffic, use the --no-traffic option:
78 * ``--no-traffic`` or ``-0`` : sending traffic from traffic generator is skipped
84 NFVbench allows to restart TRex traffic generator between runs.
85 It runs the whole test, but restart TRex instance before generating new traffic.
87 To force restart, use the --restart option:
95 * ``--restart`` : restart traffic generator (TRex)
100 Parameter ``--rate`` accepts different types of values:
102 * packets per second (pps, kpps, mpps), e.g. ``1000pps`` or ``10kpps``
103 * load percentage (%), e.g. ``50%``
104 * bits per second (bps, kbps, Mbps, Gbps), e.g. ``1Gbps``, ``1000bps``
105 * NDR/PDR (ndr, pdr, ndr_pdr), e.g. ``ndr_pdr``
107 NDR/PDR is the default rate when not specified.
112 Fixed rate run is the most basic type of NFVbench usage. It can be used to measure the drop rate with a fixed transmission rate of packets.
114 This example shows how to run the PVP packet path (which is the default packet path) with multiple different settings:
118 nfvbench -c nfvbench.cfg --no-cleanup --rate 100000pps --duration 30 --interval 15 --json results.json
122 * ``-c nfvbench.cfg`` : path to the config file
123 * ``--no-cleanup`` : resources (networks, VMs, attached ports) are not deleted after test is finished
124 * ``--rate 100000pps`` : defines rate of packets sent by traffic generator
125 * ``--duration 30`` : specifies how long should traffic be running in seconds
126 * ``--interval 15`` : stats are checked and shown periodically (in seconds) in this interval when traffic is flowing
127 * ``--json results.json`` : collected data are stored in this file after run is finished
129 .. note:: It is your responsibility to clean up resources if needed when ``--no-cleanup`` parameter is used. You can use the nfvbench_cleanup helper script for that purpose.
131 The ``--json`` parameter makes it easy to store NFVbench results. The --show-summary (or -ss) option can be used to display the results in a json results file in a text tabular format:
135 nfvbench --show-summary results.json
138 This example shows how to specify a different packet path:
142 nfvbench -c nfvbench.cfg --rate 1Mbps --inter-node --service-chain PVVP
146 * ``-c nfvbench.cfg`` : path to the config file
147 * ``--rate 1Mbps`` : defines rate of packets sent by traffic generator
148 * ``--inter-node`` : VMs are created on different compute nodes, works only with PVVP flow
149 * ``--service-chain PVVP`` or ``-sc PVVP`` : specifies the type of service chain (or packet path) to use
151 .. note:: When parameter ``--inter-node`` is not used or there aren't enough compute nodes, VMs are on the same compute node.
153 A fixed rate run can also be used to check the running drop rate while traffic is being generated. In that case the --interval option can be used
154 to specify the reporting interval in seconds (minimum is 1 second). This can be useful for example to see how packet drop rate
155 evolves over time. One common use case is to see the drop rate when there is a network degradation (e.g. one of the 2 links in a bond
157 The console output will show at every reporting interval the number of packets transmitted, received and estimated drop rate for the last reporting interval.
158 The smaller is the interval the more precise is the drop rate.
160 Example of output where the reporting interval is set to 1 (second):
164 2020-04-25 12:59:16,618 INFO TX: 1,878,719,266; RX: 1,666,641,890; (Est.) Dropped: 2; Drop rate: 0.0000%
165 2020-04-25 12:59:17,625 INFO TX: 1,883,740,078; RX: 1,671,662,706; (Est.) Dropped: -4; Drop rate: -0.0001%
166 2020-04-25 12:59:18,632 INFO TX: 1,888,764,404; RX: 1,676,686,993; (Est.) Dropped: 39; Drop rate: 0.0008%
167 2020-04-25 12:59:19,639 INFO TX: 1,893,785,063; RX: 1,681,276,714; (Est.) Dropped: 430,938; Drop rate: 8.5833%
168 2020-04-25 12:59:20,645 INFO TX: 1,898,805,769; RX: 1,683,782,636; (Est.) Dropped: 2,514,784; Drop rate: 50.0883%
169 2020-04-25 12:59:21,652 INFO TX: 1,903,829,191; RX: 1,686,289,860; (Est.) Dropped: 2,516,198; Drop rate: 50.0893%
170 2020-04-25 12:59:22,658 INFO TX: 1,908,850,478; RX: 1,691,283,008; (Est.) Dropped: 28,139; Drop rate: 0.5604%
171 2020-04-25 12:59:23,665 INFO TX: 1,913,870,692; RX: 1,696,301,242; (Est.) Dropped: 1,980; Drop rate: 0.0394%
172 2020-04-25 12:59:24,672 INFO TX: 1,918,889,696; RX: 1,698,806,224; (Est.) Dropped: 2,514,022; Drop rate: 50.0901%
173 2020-04-25 12:59:25,680 INFO TX: 1,923,915,470; RX: 1,701,314,663; (Est.) Dropped: 2,517,335; Drop rate: 50.0885%
174 2020-04-25 12:59:26,687 INFO TX: 1,928,944,879; RX: 1,705,886,869; (Est.) Dropped: 457,203; Drop rate: 9.0906%
175 2020-04-25 12:59:27,696 INFO TX: 1,933,969,377; RX: 1,710,911,346; (Est.) Dropped: 21; Drop rate: 0.0004%
176 2020-04-25 12:59:28,702 INFO TX: 1,938,998,536; RX: 1,713,843,740; (Est.) Dropped: 2,096,765; Drop rate: 41.6922%
177 2020-04-25 12:59:29,710 INFO TX: 1,944,019,920; RX: 1,718,226,356; (Est.) Dropped: 638,768; Drop rate: 12.7210%
178 2020-04-25 12:59:30,718 INFO TX: 1,949,050,206; RX: 1,723,256,639; (Est.) Dropped: 3; Drop rate: 0.0001%
179 2020-04-25 12:59:31,725 INFO TX: 1,954,075,270; RX: 1,728,281,726; (Est.) Dropped: -23; Drop rate: -0.0005%
180 2020-04-25 12:59:32,732 INFO TX: 1,959,094,908; RX: 1,733,301,290; (Est.) Dropped: 74; Drop rate: 0.0015%
181 2020-04-25 12:59:33,739 INFO TX: 1,964,118,902; RX: 1,738,325,357; (Est.) Dropped: -73; Drop rate: -0.0015%
182 2020-04-25 12:59:34,746 INFO TX: 1,969,143,790; RX: 1,743,350,230; (Est.) Dropped: 15; Drop rate: 0.0003%
183 2020-04-25 12:59:35,753 INFO TX: 1,974,165,773; RX: 1,748,372,291; (Est.) Dropped: -78; Drop rate: -0.0016%
184 2020-04-25 12:59:36,759 INFO TX: 1,979,188,496; RX: 1,753,394,957; (Est.) Dropped: 57; Drop rate: 0.0011%
185 2020-04-25 12:59:37,767 INFO TX: 1,984,208,956; RX: 1,757,183,844; (Est.) Dropped: 1,231,573; Drop rate: 24.5311%
186 2020-04-25 12:59:38,773 INFO TX: 1,989,233,595; RX: 1,761,729,705; (Est.) Dropped: 478,778; Drop rate: 9.5286%
187 2020-04-25 12:59:39,780 INFO TX: 1,994,253,350; RX: 1,766,749,467; (Est.) Dropped: -7; Drop rate: -0.0001%
188 2020-04-25 12:59:40,787 INFO TX: 1,999,276,622; RX: 1,771,772,738; (Est.) Dropped: 1; Drop rate: 0.0000%
189 2020-04-25 12:59:41,794 INFO TX: 2,004,299,940; RX: 1,776,796,065; (Est.) Dropped: -9; Drop rate: -0.0002%
190 2020-04-25 12:59:42,800 INFO TX: 2,009,320,453; RX: 1,781,816,583; (Est.) Dropped: -5; Drop rate: -0.0001%
191 2020-04-25 12:59:43,807 INFO TX: 2,014,340,581; RX: 1,786,503,172; (Est.) Dropped: 333,539; Drop rate: 6.6440%
192 2020-04-25 12:59:44,814 INFO TX: 2,019,362,996; RX: 1,789,009,857; (Est.) Dropped: 2,515,730; Drop rate: 50.0900%
193 2020-04-25 12:59:45,821 INFO TX: 2,024,386,346; RX: 1,791,517,070; (Est.) Dropped: 2,516,137; Drop rate: 50.0888%
196 How to read each line:
200 2020-04-25 10:46:41,276 INFO TX: 4,004,436; RX: 4,004,381; (Est.) Dropped: 55; Drop rate: 0.0014%
202 At this poing in time, NFvbench has sent 4,004,436 and received 4,004,381 since the start of the run.
203 There is deficit of 55 packets on reception which corresponds to 0.0014% of all packets sent during that reporting window interval (last 1 second)
204 A negative value means that the RX count is higher than the tx count in that window – this is possible since the RX and TX reads are not atomic.
210 The NDR and PDR test is used to determine the maximum throughput performance of the system under test
211 following guidelines defined in RFC-2544:
213 * NDR (No Drop Rate): maximum packet rate sent without dropping any packet
214 * PDR (Partial Drop Rate): maximum packet rate sent while allowing a given maximum drop rate
216 The NDR search can also be relaxed to allow some very small amount of drop rate (lower than the PDR maximum drop rate).
217 NFVbench will measure the NDR and PDR values by driving the traffic generator through multiple iterations
218 at different transmission rates using a binary search algorithm.
220 The configuration file contains section where settings for NDR/PDR can be set.
224 # NDR/PDR configuration
226 # Drop rates represent the ratio of dropped packet to the total number of packets sent.
227 # Values provided here are percentages. A value of 0.01 means that at most 0.01% of all
228 # packets sent are dropped (or 1 packet every 10,000 packets sent)
230 # No Drop Rate; Default to 0.001%
232 # Partial Drop Rate; NDR should always be less than PDR
234 # The accuracy of NDR and PDR load percentiles; The actual load percentile that match NDR
235 # or PDR should be within `load_epsilon` difference than the one calculated.
238 Because NDR/PDR is the default ``--rate`` value, it is possible to run NFVbench simply like this:
242 nfvbench -c nfvbench.cfg
244 Other possible run options:
248 nfvbench -c nfvbench.cfg --duration 120 --json results.json
252 * ``-c nfvbench.cfg`` : path to the config file
253 * ``--duration 120`` : specifies how long should be traffic running in each iteration
254 * ``--json results.json`` : collected data are stored in this file after run is finished
260 NFVbench allows to run multiple chains at the same time. For example it is possible to stage the PVP service chain N-times,
261 where N can be as much as your compute power can scale. With N = 10, NFVbench will spawn 10 VMs as a part of 10 simultaneous PVP chains.
263 The number of chains is specified by ``--service-chain-count`` or ``-scc`` flag with a default value of 1.
264 For example to run NFVbench with 3 PVP chains:
268 nfvbench -c nfvbench.cfg --rate 10000pps -scc 3
270 It is not necessary to specify the service chain type (-sc) because PVP is set as default. The PVP service chains will have 3 VMs in 3 chains with this configuration.
271 If ``-sc PVVP`` is specified instead, there would be 6 VMs in 3 chains as this service chain has 2 VMs per chain.
272 Both **single run** or **NDR/PDR** can be run as multichain. Running multichain is a scenario closer to a real life situation than runs with a single chain.
278 NFVbench always generates L3 packets from the traffic generator but allows the user to specify how many flows to generate.
279 A flow is identified by a unique src/dest MAC IP and port tuple that is sent by the traffic generator. Flows are
280 generated by ranging the IP adresses but using a small fixed number of MAC addresses.
282 The number of flows will be spread roughly even between chains when more than 1 chain is being tested.
283 For example, for 11 flows and 3 chains, number of flows that will run for each chain will be 3, 4, and 4 flows respectively.
285 The number of flows is specified by ``--flow-count`` or ``-fc`` flag, the default value is 2 (1 flow in each direction).
286 To run NFVbench with 3 chains and 100 flows, use the following command:
290 nfvbench -c nfvbench.cfg --rate 10000pps -scc 3 -fc 100
292 Note that from a vswitch point of view, the
293 number of flows seen will be higher as it will be at least 4 times the number of flows sent by the traffic generator
294 (add flow to VM and flow from VM).
296 IP addresses generated can be controlled with the following NFVbench configuration options:
300 ip_addrs: ['10.0.0.0/8', '20.0.0.0/8']
301 ip_addrs_step: 0.0.0.1
302 tg_gateway_ip_addrs: ['1.1.0.100', '2.2.0.100']
303 tg_gateway_ip_addrs_step: 0.0.0.1
304 gateway_ip_addrs: ['1.1.0.2', '2.2.0.2']
305 gateway_ip_addrs_step: 0.0.0.1
307 ``ip_addrs`` are the start of the 2 ip address ranges used by the traffic generators as the packets source and destination packets
308 where each range is associated to virtual devices simulated behind 1 physical interface of the traffic generator.
309 These can also be written in CIDR notation to represent the subnet.
311 ``tg_gateway_ip_addrs`` are the traffic generator gateway (virtual) ip addresses, all traffic to/from the virtual devices go through them.
313 ``gateway_ip_addrs`` are the 2 gateway ip address ranges of the VMs used in the external chains. They are only used with external chains and must correspond to their public IP address.
315 The corresponding ``step`` is used for ranging the IP addresses from the `ip_addrs``, ``tg_gateway_ip_addrs`` and ``gateway_ip_addrs`` base addresses.
316 0.0.0.1 is the default step for all IP ranges. In ``ip_addrs``, 'random' can be configured which tells NFVBench to generate random src/dst IP pairs in the traffic stream.
318 UDP ports can be controlled with the following NFVbench configuration options:
322 udp_src_port: ['1024', '65000']
326 ``udp_src_port`` and ``udp_dst_port`` are the UDP port value used by the traffic generators.
327 These can be written for unique port or range ports for all flow.
329 The corresponding ``udp_port_step`` is used for ranging the UDP port.
330 1 is the default step for all UDP ranges, 'random' can be configured which tells NFVBench to generate random src/dst UDP pairs in the traffic stream.
333 Use of UDP range will increase possible values of flows (based on ip src/dst and port src/dst tuple).
334 NFVBench will calculate the least common multiple for this tuple to adapt flows generation to ``flow_count`` parameter.
336 Examples of multiflow
337 ^^^^^^^^^^^^^^^^^^^^^
339 1. Source IP is static and one UDP port used (default configuration)
341 NFVbench configuration options:
345 ip_addrs: ['110.0.0.0/8', '120.0.0.0/8']
346 ip_addrs_step: 0.0.0.1
347 tg_gateway_ip_addrs: ['1.1.0.100', '2.2.0.100']
348 tg_gateway_ip_addrs_step: 0.0.0.1
349 gateway_ip_addrs: ['1.1.0.2', '2.2.0.2']
350 gateway_ip_addrs_step: 0.0.0.1
352 To run NFVbench with 3 chains and 100 flows, use the following command:
356 nfvbench -c nfvbench.cfg --rate 10000pps -scc 3 -fc 100
358 The least common multiple for this configuration is lcm(16 777 216, 16 777 216, 1, 1) = 16 777 216.
359 .. note:: LCM method used IP pools sizes and UDP source and destination range sizes
361 Requested flow count is lower than configuration capacity. So, NFVbench will limit IP range to generate accurate flows:
365 2020-06-17 07:37:47,012 INFO Port 0, chain 0: IP src range [110.0.0.0,110.0.0.0]
366 2020-06-17 07:37:47,012 INFO Port 0, chain 0: IP dst range [120.0.0.0,120.0.0.15]
367 2020-06-17 07:37:47,012 INFO Port 0, chain 0: UDP src range [53,53]
368 2020-06-17 07:37:47,012 INFO Port 0, chain 0: UDP dst range [53,53]
369 2020-06-17 07:37:47,015 INFO Port 1, chain 0: IP src range [120.0.0.0,120.0.0.0]
370 2020-06-17 07:37:47,015 INFO Port 1, chain 0: IP dst range [110.0.0.0,110.0.0.15]
371 2020-06-17 07:37:47,015 INFO Port 1, chain 0: UDP src range [53,53]
372 2020-06-17 07:37:47,015 INFO Port 1, chain 0: UDP dst range [53,53]
374 2020-06-17 07:38:47,012 INFO Port 0, chain 1: IP src range [110.0.0.1,110.0.0.1]
375 2020-06-17 07:38:47,012 INFO Port 0, chain 1: IP dst range [120.0.0.16,120.0.0.32]
376 2020-06-17 07:38:47,012 INFO Port 0, chain 1: UDP src range [53,53]
377 2020-06-17 07:38:47,012 INFO Port 0, chain 1: UDP dst range [53,53]
378 2020-06-17 07:38:47,015 INFO Port 1, chain 1: IP src range [120.0.0.1,120.0.0.1]
379 2020-06-17 07:38:47,015 INFO Port 1, chain 1: IP dst range [110.0.0.16,110.0.0.32]
380 2020-06-17 07:38:47,015 INFO Port 1, chain 1: UDP src range [53,53]
381 2020-06-17 07:38:47,015 INFO Port 1, chain 1: UDP dst range [53,53]
383 2020-06-17 07:39:47,012 INFO Port 0, chain 2: IP src range [110.0.0.2,110.0.0.2]
384 2020-06-17 07:39:47,012 INFO Port 0, chain 2: IP dst range [120.0.0.33,120.0.0.49]
385 2020-06-17 07:39:47,012 INFO Port 0, chain 2: UDP src range [53,53]
386 2020-06-17 07:39:47,012 INFO Port 0, chain 2: UDP dst range [53,53]
387 2020-06-17 07:39:47,015 INFO Port 1, chain 2: IP src range [120.0.0.2,120.0.0.2]
388 2020-06-17 07:39:47,015 INFO Port 1, chain 2: IP dst range [110.0.0.33,110.0.0.49]
389 2020-06-17 07:39:47,015 INFO Port 1, chain 2: UDP src range [53,53]
390 2020-06-17 07:39:47,015 INFO Port 1, chain 2: UDP dst range [53,53]
393 2. Source IP is static, IP step is random and one UDP port used
395 NFVbench configuration options:
399 ip_addrs: ['110.0.0.0/8', '120.0.0.0/8']
400 ip_addrs_step: 'random'
401 tg_gateway_ip_addrs: ['1.1.0.100', '2.2.0.100']
402 tg_gateway_ip_addrs_step: 0.0.0.1
403 gateway_ip_addrs: ['1.1.0.2', '2.2.0.2']
404 gateway_ip_addrs_step: 0.0.0.1
406 To run NFVbench with 3 chains and 100 flows, use the following command:
410 nfvbench -c nfvbench.cfg --rate 10000pps -scc 3 -fc 100
412 The least common multiple for this configuration is lcm(16 777 216, 16 777 216, 1, 1) = 16 777 216.
413 .. note:: LCM method used IP pools sizes and UDP source and destination range sizes
415 Requested flow count is lower than configuration capacity. So, NFVbench will limit IP range to generate accurate flows:
419 2020-06-17 07:37:47,012 INFO Port 0, chain 0: IP src range [110.0.0.0,110.0.0.0]
420 2020-06-17 07:37:47,012 INFO Port 0, chain 0: IP dst range [120.0.0.0,120.0.0.15]
421 2020-06-17 07:37:47,012 INFO Port 0, chain 0: UDP src range [53,53]
422 2020-06-17 07:37:47,012 INFO Port 0, chain 0: UDP dst range [53,53]
423 2020-06-17 07:37:47,015 INFO Port 1, chain 0: IP src range [120.0.0.0,120.0.0.0]
424 2020-06-17 07:37:47,015 INFO Port 1, chain 0: IP dst range [110.0.0.0,110.0.0.15]
425 2020-06-17 07:37:47,015 INFO Port 1, chain 0: UDP src range [53,53]
426 2020-06-17 07:37:47,015 INFO Port 1, chain 0: UDP dst range [53,53]
428 2020-06-17 07:38:47,012 INFO Port 0, chain 1: IP src range [110.0.0.1,110.0.0.1]
429 2020-06-17 07:38:47,012 INFO Port 0, chain 1: IP dst range [120.0.0.16,120.0.0.32]
430 2020-06-17 07:38:47,012 INFO Port 0, chain 1: UDP src range [53,53]
431 2020-06-17 07:38:47,012 INFO Port 0, chain 1: UDP dst range [53,53]
432 2020-06-17 07:38:47,015 INFO Port 1, chain 1: IP src range [120.0.0.1,120.0.0.1]
433 2020-06-17 07:38:47,015 INFO Port 1, chain 1: IP dst range [110.0.0.16,110.0.0.32]
434 2020-06-17 07:38:47,015 INFO Port 1, chain 1: UDP src range [53,53]
435 2020-06-17 07:38:47,015 INFO Port 1, chain 1: UDP dst range [53,53]
437 2020-06-17 07:39:47,012 INFO Port 0, chain 2: IP src range [110.0.0.2,110.0.0.2]
438 2020-06-17 07:39:47,012 INFO Port 0, chain 2: IP dst range [120.0.0.33,120.0.0.49]
439 2020-06-17 07:39:47,012 INFO Port 0, chain 2: UDP src range [53,53]
440 2020-06-17 07:39:47,012 INFO Port 0, chain 2: UDP dst range [53,53]
441 2020-06-17 07:39:47,015 INFO Port 1, chain 2: IP src range [120.0.0.2,120.0.0.2]
442 2020-06-17 07:39:47,015 INFO Port 1, chain 2: IP dst range [110.0.0.33,110.0.0.49]
443 2020-06-17 07:39:47,015 INFO Port 1, chain 2: UDP src range [53,53]
444 2020-06-17 07:39:47,015 INFO Port 1, chain 2: UDP dst range [53,53]
445 2020-06-17 07:39:47,015 WARNING Using random step, the number of flows can be less than the requested number of flows due to repeatable multivariate random generation which can reproduce the same pattern of values
447 By using a random step the number of generated flows may be less than the number of requested flows. This is due to the probability of drawing the same value several times (Bernouillian drawing) from the IP range used and thus generating the same flow sequence.
448 By using a high range of UDP ports couple with ``udp_port_step='random'`` the probability to reach the requested flow counts is greater.
449 As latency stream is a separate stream than data one and have his own random draw, NFVbench will use only one packet signature (same IP and ports used for all latency packets) to avoid flow count overflow.
450 So in some cases, generated flow count can be equal to the requested flow count + 1 (latency stream).
452 **For deterministic flow count we recommend to use a step different from random.**
455 3. Source IP is static, IP step is 5 and one UDP port used
457 NFVbench configuration options:
461 ip_addrs: ['110.0.0.0/8', '120.0.0.0/8']
462 ip_addrs_step: '0.0.0.5'
463 tg_gateway_ip_addrs: ['1.1.0.100', '2.2.0.100']
464 tg_gateway_ip_addrs_step: 0.0.0.1
465 gateway_ip_addrs: ['1.1.0.2', '2.2.0.2']
466 gateway_ip_addrs_step: 0.0.0.1
468 To run NFVbench with 3 chains and 100 flows, use the following command:
472 nfvbench -c nfvbench.cfg --rate 10000pps -scc 3 -fc 100
474 The least common multiple for this configuration is lcm(16 777 216, 16 777 216, 1, 1) = 16 777 216.
475 .. note:: LCM method used IP pools sizes and UDP source and destination range sizes
477 Requested flow count is lower than configuration capacity. So, NFVbench will limit IP range to generate accurate flows:
481 2020-06-17 07:37:47,012 INFO Port 0, chain 0: IP src range [110.0.0.0,110.0.0.0]
482 2020-06-17 07:37:47,012 INFO Port 0, chain 0: IP dst range [120.0.0.0,120.0.0.75]
483 2020-06-17 07:37:47,012 INFO Port 0, chain 0: UDP src range [53,53]
484 2020-06-17 07:37:47,012 INFO Port 0, chain 0: UDP dst range [53,53]
485 2020-06-17 07:37:47,015 INFO Port 1, chain 0: IP src range [120.0.0.0,120.0.0.0]
486 2020-06-17 07:37:47,015 INFO Port 1, chain 0: IP dst range [110.0.0.0,110.0.0.75]
487 2020-06-17 07:37:47,015 INFO Port 1, chain 0: UDP src range [53,53]
488 2020-06-17 07:37:47,015 INFO Port 1, chain 0: UDP dst range [53,53]
490 2020-06-17 07:38:47,012 INFO Port 0, chain 1: IP src range [110.0.0.5,110.0.0.5]
491 2020-06-17 07:38:47,012 INFO Port 0, chain 1: IP dst range [120.0.0.80,120.0.0.160]
492 2020-06-17 07:38:47,012 INFO Port 0, chain 1: UDP src range [53,53]
493 2020-06-17 07:38:47,012 INFO Port 0, chain 1: UDP dst range [53,53]
494 2020-06-17 07:38:47,015 INFO Port 1, chain 1: IP src range [120.0.0.5,120.0.0.5]
495 2020-06-17 07:38:47,015 INFO Port 1, chain 1: IP dst range [110.0.0.80,110.0.0.160]
496 2020-06-17 07:38:47,015 INFO Port 1, chain 1: UDP src range [53,53]
497 2020-06-17 07:38:47,015 INFO Port 1, chain 1: UDP dst range [53,53]
499 2020-06-17 07:39:47,012 INFO Port 0, chain 2: IP src range [110.0.0.10,110.0.0.10]
500 2020-06-17 07:39:47,012 INFO Port 0, chain 2: IP dst range [120.0.0.165,120.0.0.245]
501 2020-06-17 07:39:47,012 INFO Port 0, chain 2: UDP src range [53,53]
502 2020-06-17 07:39:47,012 INFO Port 0, chain 2: UDP dst range [53,53]
503 2020-06-17 07:39:47,015 INFO Port 1, chain 2: IP src range [120.0.0.10,120.0.0.10]
504 2020-06-17 07:39:47,015 INFO Port 1, chain 2: IP dst range [110.0.0.165,110.0.0.245]
505 2020-06-17 07:39:47,015 INFO Port 1, chain 2: UDP src range [53,53]
506 2020-06-17 07:39:47,015 INFO Port 1, chain 2: UDP dst range [53,53]
508 4. Source IP is static, IP and UDP ranges sizes greater than requested flow count, UDP step is random
510 NFVbench configuration options:
514 ip_addrs: ['110.0.0.0/29', '120.0.0.0/30']
515 tg_gateway_ip_addrs: ['1.1.0.100', '2.2.0.100']
516 tg_gateway_ip_addrs_step: 0.0.0.1
517 gateway_ip_addrs: ['1.1.0.2', '2.2.0.2']
518 gateway_ip_addrs_step: 0.0.0.1
519 udp_src_port: ['10', '14']
520 udp_dst_port: ['20', '25']
521 udp_port_step: 'random'
523 To run NFVbench with 3 chains and 100 flows, use the following command:
527 nfvbench -c nfvbench.cfg --rate 10000pps -scc 3 -fc 100
529 The least common multiple for this configuration is lcm(8, 4, 5, 6) = 120.
530 .. note:: LCM method used IP pools sizes and UDP source and destination range sizes
532 Requested flow count is higher than IP range (8 and 4 IP addresses available) and UDP (5 and 6 ports available) configuration capacity.
533 As the combination of ranges does not permit to obtain an accurate flow count, NFVbench will override the `udp_port_step` property to '1' (was 'random') to allow flows creation.
534 A warning log will appear to inform NFVbench user that step properties will be overriden
535 So, NFVbench will determine each pool size to generate accurate flows:
539 2020-06-17 07:37:47,010 WARNING Current values of ip_addrs_step and/or udp_port_step properties do not allow to control an accurate flow count. Values will be overridden as follows:
540 2020-06-17 07:37:47,011 INFO udp_port_step='1' (previous value: udp_port_step='random')
541 2020-06-17 07:37:47,012 INFO Port 0, chain 0: IP src range [110.0.0.0,110.0.0.0]
542 2020-06-17 07:37:47,012 INFO Port 0, chain 0: IP dst range [120.0.0.0,120.0.0.0]
543 2020-06-17 07:37:47,012 INFO Port 0, chain 0: UDP src range [10,14]
544 2020-06-17 07:37:47,012 INFO Port 0, chain 0: UDP dst range [20,25]
545 2020-06-17 07:37:47,013 WARNING Current values of ip_addrs_step and/or udp_port_step properties do not allow to control an accurate flow count. Values will be overridden as follows:
546 2020-06-17 07:37:47,013 INFO udp_port_step='1' (previous value: udp_port_step='random'
547 2020-06-17 07:37:47,015 INFO Port 1, chain 0: IP src range [120.0.0.0,120.0.0.0]
548 2020-06-17 07:37:47,015 INFO Port 1, chain 0: IP dst range [110.0.0.0,110.0.0.0]
549 2020-06-17 07:37:47,015 INFO Port 1, chain 0: UDP src range [10,14]
550 2020-06-17 07:37:47,015 INFO Port 1, chain 0: UDP dst range [20,25]
552 2020-06-17 07:38:47,010 WARNING Current values of ip_addrs_step and/or udp_port_step properties do not allow to control an accurate flow count. Values will be overridden as follows:
553 2020-06-17 07:38:47,011 INFO udp_port_step='1' (previous value: udp_port_step='random'
554 2020-06-17 07:38:47,012 INFO Port 0, chain 1: IP src range [110.0.0.1,110.0.0.1]
555 2020-06-17 07:38:47,012 INFO Port 0, chain 1: IP dst range [120.0.0.1,120.0.0.1]
556 2020-06-17 07:38:47,012 INFO Port 0, chain 1: UDP src range [10,14]
557 2020-06-17 07:38:47,012 INFO Port 0, chain 1: UDP dst range [20,25]
558 2020-06-17 07:38:47,013 WARNING Current values of ip_addrs_step and/or udp_port_step properties do not allow to control an accurate flow count. Values will be overridden as follows:
559 2020-06-17 07:38:47,013 INFO udp_port_step='1' (previous value: udp_port_step='random'
560 2020-06-17 07:38:47,015 INFO Port 1, chain 1: IP src range [120.0.0.1,120.0.0.1]
561 2020-06-17 07:38:47,015 INFO Port 1, chain 1: IP dst range [110.0.0.1,110.0.0.1]
562 2020-06-17 07:38:47,015 INFO Port 1, chain 1: UDP src range [10,14]
563 2020-06-17 07:38:47,015 INFO Port 1, chain 1: UDP dst range [20,25]
565 2020-06-17 07:39:47,010 WARNING Current values of ip_addrs_step and/or udp_port_step properties do not allow to control an accurate flow count. Values will be overridden as follows:
566 2020-06-17 07:39:47,011 INFO udp_port_step='1' (previous value: udp_port_step='random'
567 2020-06-17 07:39:47,012 INFO Port 0, chain 2: IP src range [110.0.0.2,110.0.0.2]
568 2020-06-17 07:39:47,012 INFO Port 0, chain 2: IP dst range [120.0.0.2,120.0.0.2]
569 2020-06-17 07:39:47,012 INFO Port 0, chain 2: UDP src range [10,14]
570 2020-06-17 07:39:47,012 INFO Port 0, chain 2: UDP dst range [20,25]
571 2020-06-17 07:39:47,013 WARNING Current values of ip_addrs_step and/or udp_port_step properties do not allow to control an accurate flow count. Values will be overridden as follows:
572 2020-06-17 07:39:47,013 INFO udp_port_step='1' (previous value: udp_port_step='random'
573 2020-06-17 07:39:47,015 INFO Port 1, chain 2: IP src range [120.0.0.2,120.0.0.2]
574 2020-06-17 07:39:47,015 INFO Port 1, chain 2: IP dst range [110.0.0.2,110.0.0.2]
575 2020-06-17 07:39:47,015 INFO Port 1, chain 2: UDP src range [10,14]
576 2020-06-17 07:39:47,015 INFO Port 1, chain 2: UDP dst range [20,25]
579 Traffic Configuration via CLI
580 -----------------------------
582 While traffic configuration can be modified using the configuration file, it can be inconvenient to have to change the configuration file everytime
583 you need to change a traffic configuration option. Traffic configuration options can be overridden with a few CLI options.
585 Here is an example of configuring traffic via CLI:
589 nfvbench --rate 10kpps --service-chain-count 2 -fs 64 -fs IMIX -fs 1518 --unidir
591 This command will run NFVbench with a unidirectional flow for three packet sizes 64B, IMIX, and 1518B.
595 * ``--rate 10kpps`` : defines rate of packets sent by traffic generator (total TX rate)
596 * ``-scc 2`` or ``--service-chain-count 2`` : specifies number of parallel chains of given flow to run (default to 1)
597 * ``-fs 64`` or ``--frame-size 64``: add the specified frame size to the list of frame sizes to run
598 * ``--unidir`` : run traffic with unidirectional flow (default to bidirectional flow)
604 NFVbench will dicover the MAC addresses to use for generated frames using:
605 - either OpenStack discovery (find the MAC of an existing VM) in the case of PVP and PVVP service chains
606 - or using dynamic ARP discovery (find MAC from IP) in the case of external chains.
607 - In case of L3 chain with SDN-GW or router between traffic generator and loop VM ARP is needed to discover SDN-GW mac addresses, use ``--loop-vm-arp`` flag or ``loop_vm_arp: true`` in config file.
609 Status and Cleanup of NFVbench Resources
610 ----------------------------------------
612 The --status option will display the status of NFVbench and list any NFVbench resources. You need to pass the OpenStack RC
613 file in order to connect to OpenStack.
617 # nfvbench --status -r /tmp/nfvbench/openrc
618 2018-04-09 17:05:48,682 INFO Version: 1.3.2.dev1
619 2018-04-09 17:05:48,683 INFO Status: idle
620 2018-04-09 17:05:48,757 INFO Discovering instances nfvbench-loop-vm...
621 2018-04-09 17:05:49,252 INFO Discovering flavor nfvbench.medium...
622 2018-04-09 17:05:49,281 INFO Discovering networks...
623 2018-04-09 17:05:49,365 INFO No matching NFVbench resources found
626 The Status can be either "idle" or "busy (run pending)".
628 The --cleanup option will first discover resources created by NFVbench and prompt if you want to proceed with cleaning them up.
633 # nfvbench --cleanup -r /tmp/nfvbench/openrc
634 2018-04-09 16:58:00,204 INFO Version: 1.3.2.dev1
635 2018-04-09 16:58:00,205 INFO Status: idle
636 2018-04-09 16:58:00,279 INFO Discovering instances nfvbench-loop-vm...
637 2018-04-09 16:58:00,829 INFO Discovering flavor nfvbench.medium...
638 2018-04-09 16:58:00,876 INFO Discovering networks...
639 2018-04-09 16:58:00,960 INFO Discovering ports...
640 2018-04-09 16:58:01,012 INFO Discovered 6 NFVbench resources:
641 +----------+-------------------+--------------------------------------+
642 | Type | Name | UUID |
643 |----------+-------------------+--------------------------------------|
644 | Instance | nfvbench-loop-vm0 | b039b858-777e-467e-99fb-362f856f4a94 |
645 | Flavor | nfvbench.medium | a027003c-ad86-4f24-b676-2b05bb06adc0 |
646 | Network | nfvbench-net0 | bca8d183-538e-4965-880e-fd92d48bfe0d |
647 | Network | nfvbench-net1 | c582a201-8279-4309-8084-7edd6511092c |
648 | Port | | 67740862-80ac-4371-b04e-58a0b0f05085 |
649 | Port | | b5db95b9-e419-4725-951a-9a8f7841e66a |
650 +----------+-------------------+--------------------------------------+
651 2018-04-09 16:58:01,013 INFO NFVbench will delete all resources shown...
652 Are you sure? (y/n) y
653 2018-04-09 16:58:01,865 INFO Deleting instance nfvbench-loop-vm0...
654 2018-04-09 16:58:02,058 INFO Waiting for 1 instances to be fully deleted...
655 2018-04-09 16:58:02,182 INFO 1 yet to be deleted by Nova, retries left=6...
656 2018-04-09 16:58:04,506 INFO 1 yet to be deleted by Nova, retries left=5...
657 2018-04-09 16:58:06,636 INFO 1 yet to be deleted by Nova, retries left=4...
658 2018-04-09 16:58:08,701 INFO Deleting flavor nfvbench.medium...
659 2018-04-09 16:58:08,729 INFO Deleting port 67740862-80ac-4371-b04e-58a0b0f05085...
660 2018-04-09 16:58:09,102 INFO Deleting port b5db95b9-e419-4725-951a-9a8f7841e66a...
661 2018-04-09 16:58:09,620 INFO Deleting network nfvbench-net0...
662 2018-04-09 16:58:10,357 INFO Deleting network nfvbench-net1...
665 The --force-cleanup option will do the same but without prompting for confirmation.
667 Service mode for TRex
668 ---------------------
670 The ``--service-mode`` option allows you to capture traffic on a TRex window during the NFVBench test. Thus, you will be
671 able to capture packets generated by TRex to observe many information on it.
677 nfvbench ``--service-mode``
679 .. note:: It is preferable to define the minimum rate (2002 pps) to have a better capture
681 In another bash window, you should connect to the TRex console doing:
685 cd /opt/trex/vX.XX/ # use completion here to find your corresponding TRex version
686 ./trex-console --python3 -r
687 # capture on port number 0
688 capture monitor start --rx 0 -v
693 Start this capture once you have started the NFVBench test, and you will observe packets on the TRex console:
697 #26342 Port: 0 ◀── RX
701 Type: UDP, Size: 66 B, TS: 26.30 [sec]
705 dst = a0:36:9f:7a:58:8e
706 src = fa:16:3e:57:8f:df
733 load = "xx\xab'\x01\x00?s\x00\x00\xbci\xf0_{U~\x00"
737 Check on the NFVBench window that the following log appears just before the testing phase:
741 2019-10-21 09:38:51,532 INFO Starting to generate traffic...
742 2019-10-21 09:38:51,532 INFO Running traffic generator
743 2019-10-21 09:38:51,541 INFO ``Service mode is enabled``
744 2019-10-21 09:38:52,552 INFO TX: 2004; RX: 2003; Est. Dropped: 1; Est. Drop rate: 0.0499%
745 2019-10-21 09:38:53,559 INFO TX: 4013; RX: 4011; Est. Dropped: 2; Est. Drop rate: 0.0498%
747 Recording packet using service mode for TRex
748 --------------------------------------------
750 Check on the NFVBench window that the following log appears just before the testing phase:
754 2019-10-21 09:38:51,532 INFO Starting to generate traffic...
755 2019-10-21 09:38:51,532 INFO Running traffic generator
756 2019-10-21 09:38:51,541 INFO ``Service mode is enabled``
757 2019-10-21 09:38:52,552 INFO TX: 2004; RX: 2003; Est. Dropped: 1; Est. Drop rate: 0.0499%
759 In another bash window, you should connect to the TRex console doing :
763 cd /opt/trex/vX.XX/ #use completion here to find your corresponding TRex version
765 capture record start --rx [port number] --limit 10000
766 .. note::Start this capture once traffic generation is started (after ``Service mode is enabled`` log)
768 Check on the TRex window that the following log appears just after capture is started:
772 Starting packet capturing up to 10000 packets [SUCCESS]
773 *** Capturing ID is set to '8' ***
774 *** Please call 'capture record stop --id 8 -o <out.pcap>' when done ***
776 Then **before end of traffic generation**, stop capture and save it as a PCAP file:
780 capture record stop --id 8 -o /tmp/nfvb/record.pcap
781 .. note:: Provide a shared path with between NFVbench container and your host to retrieve pcap file
783 Check on the TRex window that the following log appears just after capture is started:
787 Stopping packet capture 8 [SUCCESS]
788 Writing up to 10000 packets to '/tmp/nfvb/record.pcap' [SUCCESS]
789 Removing PCAP capture 8 from server [SUCCESS]
794 The ``--user-info`` option allows you to pass custom information as a JSON string.
795 This information will be available through JSON output and also exported to ``fluentd`` and can be used in results post-processing.
801 nfvbench ``--user-info='{"status":"explore","description":{"target":"lab","ok":true,"version":2020}'``
803 .. note:: only JSON string is allowed
805 ``--user-info`` can be used for determining theoretical max rate. In some cases, an overhead encapsulation exists between NFVbench and SUT so NFVbench will not reach line rate inside SUT due to this extra encapsulation.
806 To calculate this theoretical line rate inside SUT, NFVbench will use a reserved key: ``extra_encapsulation_bytes`` in ``--user-info`` property.
810 nfvbench ``--user-info='{"extra_encapsulation_bytes": 28}'``
813 As a result, NFVbench will return two values ``theoretical_tx_rate_bps`` and ``theoretical_tx_rate_pps``:
819 "initial_rate_type": "rate_percent",
820 "l2frame_size": "64",
821 "load_percent_per_direction": 100.0,
822 "rate_bps": 20000000000.0,
823 "rate_percent": 200.0,
824 "rate_pps": 29761904,
827 "offered_tx_rate_bps": 15000000000.0,
829 "theoretical_tx_rate_bps": 15000000000.0,
830 "theoretical_tx_rate_pps": 22321428.57142857,
831 "total_tx_rate": 22321428
834 In the above example, line rate is 20Gbps but NFVbench is outside SUT and a SDN gateway add an extra encapsulation of 28 bytes.
835 Overall, theoretical line rate inside SUT is only 15 Gbps for 64 bytes packet size and it will be this max capacity treated by the target compute node.