Sorry for the delay, I was in a badly connected place. On Wed 21/Aug/2013 11:35:17 +0200 Matus UHLAR - fantomas wrote: > On 20.08.13 19:13, Alessandro Vesely wrote: >> BTW, 4408bis just entered Last Call. There are not many differences, >> but it seems to be a good occasion to review SPF implementations. > > oh, hell. again that "people have not started using SPF, let's forbid it". > same like "people don't go forward, let's go back then"
The spec was buggy on what type servers should look up, and querying both --which is what Courier does-- is a waste of resources. Skipping an unnecessary query seems to be rather an acknowledgment that type 99 never flew, than a step backward. > are there any other important differences? Not much. DNS lookup limits are specified more precisely, and the notion of "void lookups" is formalized: Evaluation can stop with permerror after more than two queries return an empty answer. I'd be curious to see if the test suite will be upgraded. IIRC, Courier implementation is nearer to SPF "classic" than RFC 4408, for non-pass results. ------------------------------------------------------------------------------ Introducing Performance Central, a new site from SourceForge and AppDynamics. Performance Central is your source for news, insights, analysis and resources for efficient Application Performance Management. Visit us today! http://pubads.g.doubleclick.net/gampad/clk?id=48897511&iu=/4140/ostg.clktrk _______________________________________________ courier-users mailing list courier-users@lists.sourceforge.net Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users
