Try: ./pox.py log --level DEBUG --file=pox.log,w openflow.discovery --eat-early-packets=True forwarding.l2_learning host_tracker
Or: ./pox.py log --level DEBUG --file=pox.log,w openflow.discovery forwarding.l2_learning --hold-down=10 host_tracker See if that improves things. You should probably have to wait a few moments before traffic will pass. -- Murphy > Hi, > > I'm having problem with host_tracker.py memory (correct me if I'm stating it > wrong). When I I start a topology in mininet, I'm getting wrong(or > unexpected) behaviour from host_tracker.py. > > Situation: > Rebooted ubuntu 64 bit OS and executed 'sudo mn -c' > 1. Started pox controller in carp branch with './pox.py log --file=pox.log,w > forwarding.l2_learning samples.httopo' > <Screenshot from 2013-09-14 22:13:44.png> > > 2. Started a topology in mininet 'sudo mn --topo linear,3 --controller remote' > <Screenshot from 2013-09-14 22:14:20.png> > > and when I observe the log... I saw many duplicate hosts found and it moves > from one switch to another switch. and I get wrong topology Information. In > this case I get information like where two hosts are attached to one host > (supposted to be one host per switch in this case). > > I have attached the pox log too. Plz let me know how to come out of this issue
