At 09:57 AM 6/22/2005 -0500, wayne wrote:
In <x4u0jr2cob(_dot_)fsf(_at_)footbone(_dot_)schlitt(_dot_)net> wayne
<wayne(_at_)schlitt(_dot_)net> writes:
Anyway, the answer John received on namedropers was to see "RFC 2181
section 5". I asked about the problems with djbdns on this subject.
There hasn't been time to get a response yet, but maybe tomorrow we
shall learn more.
I've gotten two responses back from people, and both indicate that
djbdns is broken with respect to sending complete RRsets. So, in
theory, you should only get complete RRsets, but in practice, you
can't depend on it.
I assume by "broken" you mean it can't handle more than 8 records in a set.
But then, domain owners that use braindead DNS software get what they
deserve and need to make sure they don't publish policies that trigger
bugs. Similarly email receivers that use braindead DNS software
shouldn't be doing LMAP-type checking.
I wonder if the current practice we are seeing of keeping MX sets less than
8 is a result of discovering these problems and making adjustments. I did
a little more digging, and here are the results for the top 20 domains in
senderbase.org:
domain MX A/MX
rr.com 6 6-7
yahoo.com 4 3-4
aol.com 4 4-5
hotmail.com 4 4
comcast.net 2 2
verizon.net 1 1
charter.com 1 8
proxad.net 2 1
hinet.net 1 1
auna.net 1 1
wanadoo.fr 1 13
interbusiness.it 1 1
rima-tde.net 1 1
blueyonder.co.uk 3 1
telesp.net.br 2 1
shawcable.net 2 1
t-dialin.net 1 1
prod-infinitum.com.mx. 1 1
pacbell.net 14 1
ameritech.net 6 1
I see only one with more than 8 A records in a set. Maybe they are using a
server that doesn't have a limit of 8. Seems like there might still be
problems depending on what software is used by the client.
I suggest encouraging multiple A records in a set (to reduce the DNS
traffic), but limiting the number to 8 per SRV name. This would cut DNS
queries that miss the cache by a factor of 8, and that should be good enough.
--
Dave
************************************************************ *
* David MacQuigg, PhD email: david_macquigg at yahoo.com * *
* IC Design Engineer phone: USA 520-721-4583 * * *
* Analog Design Methodologies * * *
* 9320 East Mikelyn Lane * * *
* VRS Consulting, P.C. Tucson, Arizona 85710 *
************************************************************ *