User login

Weekly report, 4/06/2014




Following the authoring of my last weekly report, I successfully discovered and resolved the packet truncation issue I was having. It appears that there is a (possibly reintroduced) bug in the Open vSwitch implementation where the OFPCML_NO_BUFFER option (which instructs the switch not to create a buffer for an incoming packet) was being ignored where the incoming packets came in on a flow set with a priority of 0. Changing the priority to a different number allowed us to successfully read the full contents of a DHCP packet, options and all - no truncation. I was able to discover the problem by using scapy to throw some fixed-length packets to my controller and observe how the controller interpreted them as they were intercepted by different flows. A simple fix but exasperating to find.

I've spent the rest of this week assembling my presentation for my in-class honours practice talk, which I gave it today. It went ok, but I ran over time and found I had too many slides with too much content - but then again I found it difficult to find a happy medium in accommodating my audiences understanding of some of the concepts I was to speak about. Hat-tip to Brad and other members of the WAND group for their constructive feedback about my draft presentation.

Other than that, I have spent the rest of my project time this week working on my interim project report. I haven't found it too much of a struggle to grasp Latex, and now I'm finding myself wondering how I managed report writing before now without it. I've spent a lot more time than I intended writing longer introduction and background chapters, with the intention of borrowing some of it for my final report where applicable.