aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorVratko Polak <vrpolak@cisco.com>2019-02-25 13:00:43 +0100
committerPeter Mikus <pmikus@cisco.com>2019-02-26 10:33:02 +0000
commit3a066cb1f1b79c6fb14cb1e9c7871d0b19bb06a2 (patch)
treeaf1b7c75f3b9acb670377e03af007864922a37f6
parent53f44d52b513fae893e7774d77e3cff160ed9abe (diff)
De-duplicate unidirection support
- Latency is now reported in both directions. -- For unidirection that means additional -1/-1/-1. + Changed function argument wrapping to save vertical space. + Changed trex command synthesys to save vertical space. + Fixed handling of command line arguments to avoid superfluous spaces. + More care distinguishing (0, 1) from (port_0, port_1). - Not renaming the original port arguments. Change-Id: I7b1736b594f6edea52eb07e5bc76ef52d0af92f8 Signed-off-by: Vratko Polak <vrpolak@cisco.com>
-rw-r--r--resources/libraries/python/TrafficGenerator.py185
-rwxr-xr-xresources/tools/trex/trex_stateless_profile.py267
2 files changed, 106 insertions, 346 deletions
diff --git a/resources/libraries/python/TrafficGenerator.py b/resources/libraries/python/TrafficGenerator.py
index 6123b8753a..10fee9afbf 100644
--- a/resources/libraries/python/TrafficGenerator.py
+++ b/resources/libraries/python/TrafficGenerator.py
@@ -1,4 +1,4 @@
-# Copyright (c) 2018 Cisco and/or its affiliates.
+# Copyright (c) 2019 Cisco and/or its affiliates.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at:
@@ -168,13 +168,14 @@ class TrafficGenerator(AbstractMeasurer):
"""
return self._latency
- def initialize_traffic_generator(self, tg_node, tg_if1, tg_if2,
- tg_if1_adj_node, tg_if1_adj_if,
- tg_if2_adj_node, tg_if2_adj_if,
- test_type,
- tg_if1_dst_mac=None, tg_if2_dst_mac=None):
+ def initialize_traffic_generator(
+ self, tg_node, tg_if1, tg_if2, tg_if1_adj_node, tg_if1_adj_if,
+ tg_if2_adj_node, tg_if2_adj_if, test_type, tg_if1_dst_mac=None,
+ tg_if2_dst_mac=None):
"""TG initialization.
+ TODO: Document why do we need (and how do we use) _ifaces_reordered.
+
:param tg_node: Traffic generator node.
:param tg_if1: TG - name of first interface.
:param tg_if2: TG - name of second interface.
@@ -386,9 +387,10 @@ class TrafficGenerator(AbstractMeasurer):
if int(ret) != 0:
raise RuntimeError('TRex stateless runtime error')
- def trex_stl_start_remote_exec(self, duration, rate, framesize,
- traffic_type, async_call=False,
- latency=True, warmup_time=5.0):
+ def trex_stl_start_remote_exec(
+ self, duration, rate, framesize, traffic_type, async_call=False,
+ latency=True, warmup_time=5.0, unidirection=False, tx_port=0,
+ rx_port=1):
"""Execute script on remote node over ssh to start traffic.
:param duration: Time expresed in seconds for how long to send traffic.
@@ -399,6 +401,11 @@ class TrafficGenerator(AbstractMeasurer):
:param async_call: If enabled then don't wait for all incomming trafic.
:param latency: With latency measurement.
:param warmup_time: Warmup time period.
+ :param unidirection: Traffic is unidirectional. Default: False
+ :param tx_port: Traffic generator transmit port for first flow.
+ Default: 0
+ :param rx_port: Traffic generator receive port for first flow.
+ Default: 1
:type duration: float
:type rate: str
:type framesize: str
@@ -406,34 +413,33 @@ class TrafficGenerator(AbstractMeasurer):
:type async_call: bool
:type latency: bool
:type warmup_time: float
- :returns: Nothing
+ :type unidirection: bool
+ :type tx_port: int
+ :type rx_port: int
:raises RuntimeError: In case of TG driver issue.
"""
ssh = SSH()
ssh.connect(self._node)
+ reorder = self._ifaces_reordered # Just to make the next line fit.
+ p_0, p_1 = (rx_port, tx_port) if reorder else (tx_port, rx_port)
+ command = (
+ "sh -c '{tool}/resources/tools/trex/trex_stateless_profile.py"
+ " --profile {prof}/resources/traffic_profiles/trex/{traffic}.py"
+ " --duration {duration} --frame_size {framesize} --rate {rate}"
+ " --warmup_time {warmup} --port_0 {p_0} --port_1 {p_1}").format(
+ tool=Constants.REMOTE_FW_DIR, prof=Constants.REMOTE_FW_DIR,
+ traffic=traffic_type, duration=duration, framesize=framesize,
+ rate=rate, warmup=warmup_time, p_0=p_0, p_1=p_1)
+ if async_call:
+ command += " --async"
+ if latency:
+ command += " --latency"
+ if unidirection:
+ command += " --unidirection"
+ command += "'"
- _async = "--async" if async_call else ""
- _latency = "--latency" if latency else ""
- _p0, _p1 = (2, 1) if self._ifaces_reordered else (1, 2)
-
- profile_path = ("{0}/resources/traffic_profiles/trex/"
- "{1}.py".format(Constants.REMOTE_FW_DIR,
- traffic_type))
(ret, stdout, _) = ssh.exec_command(
- "sh -c "
- "'{0}/resources/tools/trex/trex_stateless_profile.py "
- "--profile {1} "
- "--duration {2} "
- "--frame_size {3} "
- "--rate {4} "
- "--warmup_time {5} "
- "--port_0 {6} "
- "--port_1 {7} "
- "{8} " # --async
- "{9}'". # --latency
- format(Constants.REMOTE_FW_DIR, profile_path, duration, framesize,
- rate, warmup_time, _p0 - 1, _p1 - 1, _async, _latency),
- timeout=float(duration) + 60)
+ command, timeout=float(duration) + 60)
if int(ret) != 0:
raise RuntimeError('TRex stateless runtime error')
@@ -446,95 +452,15 @@ class TrafficGenerator(AbstractMeasurer):
else:
# last line from console output
line = stdout.splitlines()[-1]
-
self._result = line
logger.info('TrafficGen result: {0}'.format(self._result))
-
self._received = self._result.split(', ')[1].split('=')[1]
self._sent = self._result.split(', ')[2].split('=')[1]
self._loss = self._result.split(', ')[3].split('=')[1]
-
self._latency = []
self._latency.append(self._result.split(', ')[4].split('=')[1])
self._latency.append(self._result.split(', ')[5].split('=')[1])
- def trex_stl_start_unidirection(
- self, duration, rate, framesize, traffic_type, async_call=False,
- latency=False, warmup_time=5.0, tx_port=0, rx_port=1):
- """Execute script on remote node over ssh to start unidirection traffic.
- The purpose of this function is to support performance test that need to
- measure unidirectional traffic, e.g. Load balancer maglev mode and l3dsr
- mode test.
-
- :param duration: Time expresed in seconds for how long to send traffic.
- :param rate: Traffic rate expressed with units (pps, %)
- :param framesize: L2 frame size to send (without padding and IPG).
- :param traffic_type: Module name as a traffic type identifier.
- See resources/traffic_profiles/trex for implemented modules.
- :param latency: With latency measurement.
- :param async_call: If enabled then don't wait for all incomming trafic.
- :param warmup_time: Warmup time period.
- :param tx_port: Traffic generator transmit port.
- :param rx_port: Traffic generator receive port.
- :type duration: float
- :type rate: str
- :type framesize: str
- :type traffic_type: str
- :type latency: bool
- :type async_call: bool
- :type warmup_time: float
- :type tx_port: integer
- :type rx_port: integer
- :raises RuntimeError: In case of TG driver issue.
- """
- ssh = SSH()
- ssh.connect(self._node)
-
- _latency = "--latency" if latency else ""
- _async = "--async" if async_call else ""
-
- profile_path = ("{0}/resources/traffic_profiles/trex/"
- "{1}.py".format(Constants.REMOTE_FW_DIR,
- traffic_type))
- (ret, stdout, _) = ssh.exec_command(
- "sh -c "
- "'{0}/resources/tools/trex/trex_stateless_profile.py "
- "--profile {1} "
- "--duration {2} "
- "--frame_size {3} "
- "--rate {4} "
- "--warmup_time {5} "
- "--port_0 {6} "
- "--port_1 {7} "
- "{8} " # --async
- "{9} " # --latency
- "{10}'". # --unidirection
- format(Constants.REMOTE_FW_DIR, profile_path, duration, framesize,
- rate, warmup_time, tx_port, rx_port, _async, _latency,
- "--unidirection"),
- timeout=float(duration) + 60)
-
- if int(ret) != 0:
- raise RuntimeError('TRex unidirection runtime error')
- elif async_call:
- #no result
- self._received = None
- self._sent = None
- self._loss = None
- self._latency = None
- else:
- # last line from console output
- line = stdout.splitlines()[-1]
-
- self._result = line
- logger.info('TrafficGen result: {0}'.format(self._result))
-
- self._received = self._result.split(', ')[1].split('=')[1]
- self._sent = self._result.split(', ')[2].split('=')[1]
- self._loss = self._result.split(', ')[3].split('=')[1]
- self._latency = []
- self._latency.append(self._result.split(', ')[4].split('=')[1])
-
def stop_traffic_on_tg(self):
"""Stop all traffic on TG.
@@ -552,6 +478,19 @@ class TrafficGenerator(AbstractMeasurer):
rx_port=1):
"""Send traffic from all configured interfaces on TG.
+ Note that bidirectional traffic also contains flows
+ transmitted from rx_port and received in tx_port.
+ But some tests use asymmetric traffic, so those arguments are relevant.
+
+ Also note that traffic generator uses DPDK driver which might
+ reorder port numbers based on wiring and PCI numbering.
+ This method handles that, so argument values are invariant,
+ but you can see swapped valued in debug logs.
+
+ TODO: Is it better to have less descriptive argument names
+ just to make them less probable to be viewed as misleading or confusing?
+ See https://gerrit.fd.io/r/#/c/17625/11/resources/libraries/python/TrafficGenerator.py@406
+
:param duration: Duration of test traffic generation in seconds.
:param rate: Offered load per interface (e.g. 1%, 3gbps, 4mpps, ...).
:param framesize: Frame size (L2) in Bytes.
@@ -560,9 +499,11 @@ class TrafficGenerator(AbstractMeasurer):
:param warmup_time: Warmup phase in seconds.
:param async_call: Async mode.
:param latency: With latency measurement.
- :param unidirection: Traffic is unidirectional.
- :param tx_port: Traffic generator transmit port.
- :param rx_port: Traffic generator receive port.
+ :param unidirection: Traffic is unidirectional. Default: False
+ :param tx_port: Traffic generator transmit port for first flow.
+ Default: 0
+ :param rx_port: Traffic generator receive port for first flow.
+ Default: 1
:type duration: str
:type rate: str
:type framesize: str
@@ -571,8 +512,8 @@ class TrafficGenerator(AbstractMeasurer):
:type async_call: bool
:type latency: bool
:type unidirection: bool
- :type tx_port: integer
- :type rx_port: integer
+ :type tx_port: int
+ :type rx_port: int
:returns: TG output.
:rtype: str
:raises RuntimeError: If TG is not set, or if node is not TG,
@@ -590,15 +531,9 @@ class TrafficGenerator(AbstractMeasurer):
if node['subtype'] is None:
raise RuntimeError('TG subtype not defined')
elif node['subtype'] == NodeSubTypeTG.TREX:
- if unidirection:
- self.trex_stl_start_unidirection(duration, rate, framesize,
- traffic_type, tx_port,
- rx_port, async_call, latency,
- warmup_time)
- else:
- self.trex_stl_start_remote_exec(duration, rate, framesize,
- traffic_type, async_call,
- latency, warmup_time)
+ self.trex_stl_start_remote_exec(
+ duration, rate, framesize, traffic_type, async_call, latency,
+ warmup_time, unidirection, tx_port, rx_port)
else:
raise NotImplementedError("TG subtype not supported")
diff --git a/resources/tools/trex/trex_stateless_profile.py b/resources/tools/trex/trex_stateless_profile.py
index de29ff505a..61b244e21a 100755
--- a/resources/tools/trex/trex_stateless_profile.py
+++ b/resources/tools/trex/trex_stateless_profile.py
@@ -1,6 +1,6 @@
#!/usr/bin/python
-# Copyright (c) 2017 Cisco and/or its affiliates.
+# Copyright (c) 2019 Cisco and/or its affiliates.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at:
@@ -98,43 +98,6 @@ def simple_burst(profile_file, duration, framesize, rate, warmup_time, port_0,
:type unidirection: bool
"""
- #unidirection traffic
- if unidirection:
- send_traffic_unidirection(profile_file, duration, framesize, rate,
- warmup_time, port_0, port_1, latency,
- async_start)
- #bidirection traffic
- else:
- send_traffic_bidirection(profile_file, duration, framesize, rate,
- warmup_time, port_0, port_1, latency,
- async_start)
-
-
-def send_traffic_bidirection(profile_file, duration, framesize, rate,
- warmup_time, port_0, port_1, latency,
- async_start=False):
- """Send traffic bidirection and measure packet loss and latency.
-
- :param profile_file: A python module with T-rex traffic profile.
- :param framesize: Frame size.
- :param duration: Duration of traffic run in seconds (-1=infinite).
- :param rate: Traffic rate [percentage, pps, bps].
- :param warmup_time: Traffic warm-up time in seconds, 0 = disable.
- :param port_0: Port 0 on the traffic generator.
- :param port_1: Port 1 on the traffic generator.
- :param latency: With latency stats.
- :param async_start: Start the traffic and exit.
- :type profile_file: str
- :type framesize: int or str
- :type duration: float
- :type rate: str
- :type warmup_time: float
- :type port_0: int
- :type port_1: int
- :type latency: bool
- :type async_start: bool
- """
-
client = None
total_rcvd = 0
total_sent = 0
@@ -167,33 +130,37 @@ def send_traffic_bidirection(profile_file, duration, framesize, rate,
resolve=False)
if isinstance(framesize, int):
client.add_streams(streams[0], ports=[port_0])
- client.add_streams(streams[1], ports=[port_1])
+ if not unidirection:
+ client.add_streams(streams[1], ports=[port_1])
elif isinstance(framesize, str):
client.add_streams(streams[0:3], ports=[port_0])
- client.add_streams(streams[3:6], ports=[port_1])
+ if not unidirection:
+ client.add_streams(streams[3:6], ports=[port_1])
if latency:
try:
if isinstance(framesize, int):
client.add_streams(streams[2], ports=[port_0])
- client.add_streams(streams[3], ports=[port_1])
+ if not unidirection:
+ client.add_streams(streams[3], ports=[port_1])
elif isinstance(framesize, str):
latency = False
except STLError:
# Disable latency if NIC does not support requested stream type
print("##### FAILED to add latency streams #####")
latency = False
+ ports = [port_0]
+ if not unidirection:
+ ports.append(port_1)
# Warm-up phase:
if warmup_time > 0:
# Clear the stats before injecting:
client.clear_stats()
# Choose rate and start traffic:
- client.start(ports=[port_0, port_1], mult=rate,
- duration=warmup_time)
+ client.start(ports=ports, mult=rate, duration=warmup_time)
# Block until done:
- client.wait_on_traffic(ports=[port_0, port_1],
- timeout=warmup_time+30)
+ client.wait_on_traffic(ports=ports, timeout=warmup_time+30)
if client.get_warnings():
for warning in client.get_warnings():
@@ -206,11 +173,15 @@ def send_traffic_bidirection(profile_file, duration, framesize, rate,
print(json.dumps(stats, indent=4, separators=(',', ': '),
sort_keys=True))
- lost_a = stats[0]["opackets"] - stats[1]["ipackets"]
- lost_b = stats[1]["opackets"] - stats[0]["ipackets"]
+ lost_a = stats[port_0]["opackets"] - stats[port_1]["ipackets"]
+ if not unidirection:
+ lost_b = stats[port_1]["opackets"] - stats[port_0]["ipackets"]
- print("\npackets lost from 0 --> 1: {0} pkts".format(lost_a))
- print("packets lost from 1 --> 0: {0} pkts".format(lost_b))
+ print("\npackets lost from {p_0} --> {p_1}: {v} pkts".format(
+ p_0=port_0, p_1=port_1, v=lost_a))
+ if not unidirection:
+ print("packets lost from {p_1} --> {p_0}: {v} pkts".format(
+ p_0=port_0, p_1=port_1, v=lost_b))
# Clear the stats before injecting:
client.clear_stats()
@@ -218,11 +189,11 @@ def send_traffic_bidirection(profile_file, duration, framesize, rate,
lost_b = 0
# Choose rate and start traffic:
- client.start(ports=[port_0, port_1], mult=rate, duration=duration)
+ client.start(ports=ports, mult=rate, duration=duration)
if not async_start:
# Block until done:
- client.wait_on_traffic(ports=[port_0, port_1], timeout=duration+30)
+ client.wait_on_traffic(ports=ports, timeout=duration+30)
if client.get_warnings():
for warning in client.get_warnings():
@@ -235,24 +206,33 @@ def send_traffic_bidirection(profile_file, duration, framesize, rate,
print(json.dumps(stats, indent=4, separators=(',', ': '),
sort_keys=True))
- lost_a = stats[0]["opackets"] - stats[1]["ipackets"]
- lost_b = stats[1]["opackets"] - stats[0]["ipackets"]
+ lost_a = stats[port_0]["opackets"] - stats[port_1]["ipackets"]
+ if not unidirection:
+ lost_b = stats[port_1]["opackets"] - stats[port_0]["ipackets"]
if latency:
lat_a = fmt_latency(
- str(stats["latency"][0]["latency"]["total_min"]),
- str(stats["latency"][0]["latency"]["average"]),
- str(stats["latency"][0]["latency"]["total_max"]))
- lat_b = fmt_latency(
- str(stats["latency"][1]["latency"]["total_min"]),
- str(stats["latency"][1]["latency"]["average"]),
- str(stats["latency"][1]["latency"]["total_max"]))
-
- total_sent = stats[0]["opackets"] + stats[1]["opackets"]
- total_rcvd = stats[0]["ipackets"] + stats[1]["ipackets"]
-
- print("\npackets lost from 0 --> 1: {0} pkts".format(lost_a))
- print("packets lost from 1 --> 0: {0} pkts".format(lost_b))
+ str(stats["latency"][port_0]["latency"]["total_min"]),
+ str(stats["latency"][port_0]["latency"]["average"]),
+ str(stats["latency"][port_0]["latency"]["total_max"]))
+ if not unidirection:
+ lat_b = fmt_latency(
+ str(stats["latency"][port_1]["latency"]["total_min"]),
+ str(stats["latency"][port_1]["latency"]["average"]),
+ str(stats["latency"][port_1]["latency"]["total_max"]))
+
+ if not unidirection:
+ total_sent = stats[0]["opackets"] + stats[1]["opackets"]
+ total_rcvd = stats[0]["ipackets"] + stats[1]["ipackets"]
+ else:
+ total_sent = stats[port_0]["opackets"]
+ total_rcvd = stats[port_1]["ipackets"]
+
+ print("\npackets lost from {p_0} --> {p_1}: {v} pkts".format(
+ p_0=port_0, p_1=port_1, v=lost_a))
+ if not unidirection:
+ print("packets lost from {p_1} --> {p_0}: {v} pkts".format(
+ p_0=port_0, p_1=port_1, v=lost_b))
except STLError as err:
sys.stderr.write("{0}\n".format(err))
@@ -272,161 +252,6 @@ def send_traffic_bidirection(profile_file, duration, framesize, rate,
lat_a, lat_b))
-def send_traffic_unidirection(profile_file, duration, framesize, rate,
- warmup_time, port_0, port_1, latency,
- async_start=False):
- """Send traffic unidirection and measure packet loss and latency.
-
- :param profile_file: A python module with T-rex traffic profile.
- :param framesize: Frame size.
- :param duration: Duration of traffic run in seconds (-1=infinite).
- :param rate: Traffic rate [percentage, pps, bps].
- :param warmup_time: Traffic warm-up time in seconds, 0 = disable.
- :param port_0: Port 0 on the traffic generator.
- :param port_1: Port 1 on the traffic generator.
- :param latency: With latency stats.
- :param async_start: Start the traffic and exit.
- :type profile_file: str
- :type framesize: int or str
- :type duration: float
- :type rate: str
- :type warmup_time: float
- :type port_0: int
- :type port_1: int
- :type latency: bool
- :type async_start: bool
- """
-
- client = None
- total_rcvd = 0
- total_sent = 0
- lost_a = 0
- lat_a = "-1/-1/-1"
-
- # Read the profile:
- try:
- print("### Profile file:\n{}".format(profile_file))
- profile = STLProfile.load(profile_file, direction=0, port_id=0,
- framesize=framesize)
- streams = profile.get_streams()
- except STLError as err:
- print("Error while loading profile '{0}' {1}".format(profile_file, err))
- sys.exit(1)
-
- try:
- # Create the client:
- client = STLClient(verbose_level=LoggerApi.VERBOSE_QUIET)
- # Connect to server:
- client.connect()
- # Prepare our ports:
- if port_0 == port_1:
- client.reset(ports=[port_0])
- client.remove_all_streams(ports=[port_0])
-
- if "macsrc" in profile_file:
- client.set_port_attr(ports=[port_0], promiscuous=True,
- resolve=False)
- else:
- client.reset(ports=[port_0, port_1])
- client.remove_all_streams(ports=[port_0, port_1])
-
- if "macsrc" in profile_file:
- client.set_port_attr(ports=[port_0, port_1], promiscuous=True,
- resolve=False)
-
- if isinstance(framesize, int):
- client.add_streams(streams[0], ports=[port_0])
- elif isinstance(framesize, str):
- client.add_streams(streams[0:3], ports=[port_0])
- if latency:
- try:
- if isinstance(framesize, int):
- client.add_streams(streams[2], ports=[port_0])
- elif isinstance(framesize, str):
- latency = False
- except STLError:
- # Disable latency if NIC does not support requested stream type
- print("##### FAILED to add latency streams #####")
- latency = False
-
- # Warm-up phase:
- if warmup_time > 0:
- # Clear the stats before injecting:
- client.clear_stats()
-
- # Choose rate and start traffic:
- client.start(ports=[port_0], mult=rate,
- duration=warmup_time)
-
- # Block until done:
- client.wait_on_traffic(ports=[port_0],
- timeout=warmup_time+30)
-
- if client.get_warnings():
- for warning in client.get_warnings():
- print(warning)
-
- # Read the stats after the test:
- stats = client.get_stats()
-
- print("##### Warmup statistics #####")
- print(json.dumps(stats, indent=4, separators=(',', ': '),
- sort_keys=True))
-
- lost_a = stats[port_0]["opackets"] - stats[port_1]["ipackets"]
- print("\npackets lost : {0} pkts".format(lost_a))
-
- # Clear the stats before injecting:
- client.clear_stats()
- lost_a = 0
-
- # Choose rate and start traffic:
- client.start(ports=[port_0], mult=rate, duration=duration)
-
- if not async_start:
- # Block until done:
- client.wait_on_traffic(ports=[port_0], timeout=duration+30)
-
- if client.get_warnings():
- for warning in client.get_warnings():
- print(warning)
-
- # Read the stats after the test
- stats = client.get_stats()
-
- print("##### Statistics #####")
- print(json.dumps(stats, indent=4, separators=(',', ': '),
- sort_keys=True))
-
- lost_a = stats[port_0]["opackets"] - stats[port_1]["ipackets"]
-
- if latency:
- lat_a = fmt_latency(
- str(stats["latency"][0]["latency"]["total_min"]),
- str(stats["latency"][0]["latency"]["average"]),
- str(stats["latency"][0]["latency"]["total_max"]))
-
- total_sent = stats[port_0]["opackets"]
- total_rcvd = stats[port_1]["ipackets"]
-
- print("\npackets lost : {0} pkts".format(lost_a))
-
- except STLError as err:
- sys.stderr.write("{0}\n".format(err))
- sys.exit(1)
-
- finally:
- if async_start:
- if client:
- client.disconnect(stop_traffic=False, release_ports=True)
- else:
- if client:
- client.disconnect()
- print("rate={0}, totalReceived={1}, totalSent={2}, "
- "frameLoss={3}, latencyStream0(usec)={4}".
- format(rate, total_rcvd, total_sent, lost_a, lat_a))
-
-
def main():
"""Main function for the traffic generator using T-rex.