On Wed, Dec 12, 2012 at 12:08 AM, Ben Pfaff <b...@nicira.com> wrote: > I have always interpreted this statement that limits updating the cookie > to OFPC_MODIFY and OFPC_MODIFY_STRICT as meaning that: > > I don't see anything in the OF1.0 spec that mentions matching on a > cookie. I think that interpreting a sentence that says that delete > operations don't *update* a cookie to mean that modify operations > *match* on the cookie is overreaching.
Grr... you're definitely right (read: the sentence talks about the action you take when updating a flow entry, not the condition you use to select one). Thanks for the clarification. > This would change long-standing behavior in OVS, that has existed for > years, since before the OVS 1.0 release. The same behavior also exists > in the OpenFlow 1.0 reference implementation. This will break real > software. > > With this change, it becomes impossible to delete an individual flow in > OF1.0 without knowing its cookie. One effect is that it becomes > impossible to delete all the flows in a switch in a single operation > (which is useful and which has always been possible before) in OF1.0 if > there is more than one unique cookie in the flow table. I am, also, > almost certain that it will break NVP, which I have been told prefers to > assign random cookie values to some flows, which it then throws away, so > that it doesn't even keep track of what the cookies are. I definitely see the point about this breaking existing software on OVS. My concern is that other switches may have implemented this a different way -- Ali's original question had to do with code that I wrote in flowvisor that favored my interpretation. It's certainly possible that the brain damage was contained to just implementations written by me, but it's worth verifying. Particularly since spec is undefined for 1.0 and there is no ambiguity that the cookie *should* be used in a match in OF1.1+. Not to beat a dead horse, but just to provide some evidence that I'm not just insane, it was always my understanding that a cookie of -1 (NONE) was effectively a cookie wildcard, so if you wanted to do a MODIFY or DELETE and didn't know the cookie, you could set it to -1. But, that seems moot now, so probably not worth dwelling on. Thanks everyone for entertaining my ramblings. - Rob . _______________________________________________ openflow-discuss mailing list openflow-discuss@lists.stanford.edu https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
