- port_limit : 2
version : 2
low_end : true
interfaces : ["eth0", "eth0"] # list of the interfaces to bind run ./dpdk_nic_bind.py --status
port_info : # set eh mac addr
- ip : 172.17.0.2
default_gw : 172.17.0.1
- ip : 172.17.0.2
default_gw : 172.17.0.1
c) Docker-B is acting like Server and its /etc/trex_cfg.yaml
- port_limit : 2
version : 2
low_end : true
interfaces : ["eth0", "veth1"] # list of the interfaces to bind run ./dpdk_nic_bind.py --status
port_info : # set eh mac addr
- ip : 1.1.1.1
default_gw : 2.2.2.2
- ip : 2.2.2.2
default_gw : 1.1.1.1
d) Then started the trex on client in interactive mode and ran "start -f stl/udp_1pkt_simple.py -m 10mbps -a –force".
stl/udp_1pkt_simple.py file is modified to have SrcIP= 172.17.0.2 , DstIP=172.17.0.3
I could see that Docker-A/Client is sending the packets and Docker-B/Server is receiving the packets.
My intention of this test is to measure throughput , bandwidth of my docker network.
Couple of questions:
1) In Stateless mode run (with above procedure) , is it possible to measure the latency/RTT for UDP traffic ?
2) Want to measure the PPM drop. Is it possible with current mode of run ?
3) From the results displayed on client side (attached below), i could see drop-rate as 16.95 Mbps.
If client and server are running in different docker instances , how could it figure out the drop-rate or drop packets ?