Goal: First quickly setting up LINC-OE simple optical topology controlled by iControl and a packet simple topology controlled by POX. Then couple of experiments are performed.
Requirements: A basic knowlege of LINC-OE, TAP interfaces, POX, Erlang language , Lambda Switchin and linux CLI is required. Doing this tutorial is a must.
Environment: I have used the VM from sdn hub, I recommond you do the same. Link for installation is provided below: http://sdnhub.org/tutorials/sdn-tutorial-vm/
Road Map: This document has two sections for setup:
- setting up the optical and packet network
- runs some experiments
Notations:
>means the linuc command lineiControl>Means the iControll command line
a. Run iControl:
> cd loom/iControl
> rel/icontrol/bin/icontrol consoleThe iControl starts and listens on 0.0.0.0:6653
b. Clearting the tap interfaces: <\br>
For this section I have created a bash script called Mininet_LINC_script/TapSetup.bash that takes care of tap interfaces.
> sudo bash TapSetup.bash 7 upThe first argument is the number of tap interfaces and the second one is up which also brings the interfaces up.
c. Set up the sys.config file:
rel/files/sys.config file for the network shown above should looks as following:
[{linc,
[{of_config,disabled},
{capable_switch_ports,
[{port,1,[{interface,"tap1"}]},
{port,2,[{interface,"tap2"}]},
{port,3,[{interface,"tap3"}]},
{port,4,[{interface,"dummy"}, {type, optical}]},
{port,5,[{interface,"dummy"}, {type, optical}]},
{port,6,[{interface,"tap4"}]},
{port,7,[{interface,"dummy"}, {type, optical}]},
{port,8,[{interface,"tap5"}]},
{port,9,[{interface,"dummy"}, {type, optical}]},
{port,10,[{interface,"tap6"}]},
{port,11,[{interface,"tap7"}]}
]},
{capable_switch_queues, []},
{optical_links, [{{1,4}, {2,1}}, {{2,3},{3,1}}]},
{logical_switches,
[{switch,1,
[{backend,linc_us4_oe},
{controllers,[{"Switch0-Controller","localhost",6653,tcp}]},
{controllers_listener,disabled},
{queues_status,disabled},
{datapath_id, "00:00:00:00:00:01:00:01"},
{ports,[{port,1,[{queues,[]}, {port_no, 1}]},
{port,2,[{queues,[]}, {port_no, 2}]},
{port,3,[{queues,[]}, {port_no, 3}]},
{port,4,[{queues,[]}, {port_no, 4}]}
]}]},
{switch,2,
[{backend,linc_us4_oe},
{controllers,[{"Switch0-Controller","localhost",6653,tcp}]},
{controllers_listener,disabled},
{queues_status,disabled},
{datapath_id, "00:00:00:00:00:01:00:02"},
{ports,[{port,5,[{queues,[]}, {port_no, 1}]},
{port,6,[{queues,[]}, {port_no, 2}]},
{port,7,[{queues,[]}, {port_no, 3}]},
{port,8,[{queues,[]}, {port_no, 4}]}
]}]},
{switch,3,
[{backend,linc_us4_oe},
{controllers,[{"Switch0-Controller","localhost",6653,tcp}]},
{controllers_listener,disabled},
{queues_status,disabled},
{datapath_id, "00:00:00:00:00:01:00:03"},
{ports,[{port,9,[{queues,[]}, {port_no, 1}]},
{port,10,[{queues,[]}, {port_no, 2}]},
{port,11,[{queues,[]}, {port_no, 3}]}
]}]}
]}]},
{of_protocol, [{no_multipart, false}]},
{enetconf,
[{capabilities,[{base,{1,1}},{startup,{1,0}},{'writable-running',{1,0}}]},
{callback_module,linc_ofconfig},
{sshd_ip,any},
{sshd_port,1830},
{sshd_user_passwords,[{"linc","linc"}]}]},
{epcap,
[{verbose, false},
{stats_interval, 10},
{buffer_size, 73400320}]},
{lager,
[{handlers,
[{lager_console_backend,debug},
{lager_file_backend,
[{"log/error.log",error,10485760,"$D0",5},
{"log/debug.log",debug,10485760,"$D0",5},
{"log/console.log",info,10485760,"$D0",5}]}]}]},
{sasl,
[{sasl_error_logger,{file,"log/sasl-error.log"}},
{errlog_type,error},
{error_logger_mf_dir,"log/sasl"},
{error_logger_mf_maxbytes,1048576000000},
{error_logger_mf_maxfiles,5}]},
{sync,
[{excluded_modules, [procket]}]}].d. start LINC-OE:
> make rel && sudo rel/linc/bin/linc consolee. Add some flows to the optical switches. Type the following commands in the iControl window:
iControl> iof:oe_flow_tw(2,100,1,2,20).
iControl> iof:oe_flow_ww(1,100,1,20,2,20).
iControl> iof:oe_flow_wt(3,100,1,20,2).
iControl> iof:oe_flow_wt(2,100,2,20,1).
iControl> iof:oe_flow_ww(1,100,2,20,1,20).
iControl> iof:oe_flow_tw(3,100,2,1,20).Note that you need to add flows for both directions.
So now if put stuff (packets) in one of the tap interfaces (using tcpreply) it will apear on the tap.
Follow the LINCoe_and_iControl.md tutorial to check if you have done all the steps right.
a. Run POX
> ./pox.py log.level --debug forwarding.tutorial_l2_hubThis will instanciate a hub controller. So here is what is happening nder the hood: a packet comes on one port of the packet switch, the switch doesn't know what to do so it asks the pox controller. The pox will tell the switch to broadcast the packet (meaning send it to all ports; acts like a hub). This would send the packet on to the port which is connected to the tap interface.
The pox will start up and listens on 0.0.0.0:6633
b. Move to the directory where this repository is cloned into, and tun python topo using python API.
> sudo -E python ComplexMultiTopo.pyThe ComplexMultiTopo.py file will create a packet topolgy which includes two switches and two host.
The switches are connected to hosts on one end and on the other end they are connected to tap interfaces.
Note that tap interfaces are connected linc-oe switches.
Here is the topolgu of what we have right now:
Now the optical and packet network are ready and connected.
If you run pingall on mininet CLI, all of the packets should be droped.
mininet> pingall
*** Ping: testing ping reachability
h1 -> X X X X X X
h2 -> X X X X X X
h3 -> X X X X X X
h4 -> X X X X X X
h5 -> X X X X X X
h6 -> X X X X X X
h7 -> X X X X X X
*** Results: 100% dropped (0/42 received)
Of course you can have wireshark probing any of the interfaces.
Before we add flows we need to check what the switch key is for each optical switch. That can be done using
iControl> iof:switches(). The bove with return something like:
([email protected])1> iof:switches().
Switch-Key DatapathId IpAddr Version
---------- -------------------------------- ----------------- -------
*1 00:00:00:00:00:01:00:01 {127,0,0,1} 4
2 00:00:00:00:00:01:00:03 {127,0,0,1} 4
3 00:00:00:00:00:01:00:02 {127,0,0,1} 4
ok
The first line means: the switch with Datapath ID (DPID) of 00:00:00:00:00:01:00:02 has the switch key value of 1. The second and third can be observed similarly.
Note that the follwoing code may need to be changed based on switch keys.
Now, let's add some flows such that h1 can ping h7.
iControl> iof:oe_flow_tw(1,100,1,4,20).
iControl> iof:oe_flow_ww(3,100,1,20,3,20).
iControl> iof:oe_flow_wt(2,100,1,20,3).The above would create one side of the path from h1 to h7.
iControl> iof:oe_flow_wt(1,100,4,20,1).
iControl> iof:oe_flow_ww(3,100,3,20,1,20).
iControl> iof:oe_flow_tw(2,100,3,1,20). Now, if you try pingall only pings between h1 and h7 succeed.
mininet> pingall
*** Ping: testing ping reachability
h1 -> X X X X X h7
h2 -> X X X X X X
h3 -> X X X X X X
h4 -> X X X X X X
h5 -> X X X X X X
h6 -> X X X X X X
h7 -> h1 X X X X X
*** Results: 95% dropped (2/42 received)
We already have the flows for purpple channel. Let's add more flows for brown channel:
iControl> iof:oe_flow_tw(1,100,2,4,40).
iControl> iof:oe_flow_wt(3,100,1,40,2).
iControl> iof:oe_flow_wt(1,100,4,40,2).
iControl> iof:oe_flow_tw(3,100,2,1,40). So far we have two circuit and therefor a ping from h1 to h7 and from h2 to h4 must work. Bellow is the outpur of pingall which makes sense:
mininet> pingall
*** Ping: testing ping reachability
h1 -> X X X X X h7
h2 -> X X h4 X X X
h3 -> X X X X X X
h4 -> X h2 X X X X
h5 -> X X X X X X
h6 -> X X X X X X
h7 -> h1 X X X X X
*** Results: 90% dropped (4/42 received)
Now for the dark yellow channel:
iControl> iof:oe_flow_tw(1,100,3,4,30).
iControl> iof:oe_flow_ww(3,100,1,30,3,30).
iControl> iof:oe_flow_wt(2,100,1,30,2).
iControl> iof:oe_flow_wt(1,100,4,30,3).
iControl> iof:oe_flow_ww(3,100,3,30,1,30).
iControl> iof:oe_flow_tw(2,100,2,1,30). With the above set of flows, pingall should return:
mininet> pingall
*** Ping: testing ping reachability
h1 -> X X X X X h7
h2 -> X X h4 X X X
h3 -> X X X X h6 X
h4 -> X h2 X X X X
h5 -> X X X X X X
h6 -> X X h3 X X X
h7 -> h1 X X X X X
*** Results: 85% dropped (6/42 received)