On Thursday 16 July 2009 17:19:23 Victor Duchovni wrote: > > People publishing multiple PTR records are IMHO misguided.
I fear the folks who wrote RFC1033 used the term "official name" for where a reverse PTR record should point. I'm sure they meant "canonical", which I'm assured is an outdated concept with regard to DNS resource records. But multiple PTR records are an annoying fact, I've seen several bits of software croak when the number of PTR records caused responses for reverse lookup to exceed the allowed size of response for their specific buffer. Mostly hosts that add a PTR record for every website on that IP address. But hey I've only done DNS for umpteen years, and I'm still learning about the vaguaries of interpretation people feel is okay to allow (I only just learnt why so many Linux boxes are now assigning hostname to 127.0.1.1 - don't do it on your email servers it is for DHCP clients that wander the world and want to keep their own internal name so they know who they are).
