IPv6 site snapshot

[Jared Mauch]

On Apr 28, 2011, at 4:39 PM, Doug Barton wrote:

> On 04/27/2011 13:25, Jared Mauch wrote:
>> 	1) Appears there is a FreeBSD bug with getaddrinfo() when it queries for AAAA when it points to a cname.
>> 
>> http://www.freebsd.org/cgi/query-pr.cgi?pr=156684&cat=
> 
> Ok, according to the discussion on bind-users it seems the deciding factor should be whether or not a name server error is returned. If not, then EAI_NONAME is the right answer (for example, with a chain of CNAMEs that do not end in an AAAA record). If the name server does return an error (like SERVFAIL or NXDOMAIN) then EAI_FAIL is the right answer. You can see the reasoning for this in the thread at https://lists.isc.org/pipermail/bind-users/2011-April/083542.html if you're interested.
> 
> In the meantime, I'm going to see who I have to poke to get the bug in FreeBSD fixed.

Thanks for helping carry the water over to bind-users.

hope to see this appear in freebsd-update soon enough :)  i really want to try to abuse our resolvers some more :)

Also, for others, if I understand Doug correctly, if you are using the BIND/ISC resolver you may see it return EAI_FAIL vs EAI_NONAME.  So my warning about the code is likely a bit broader depending on your environment.

- jared

PS: Detailed Test info
MacOS 10.6.7: OK
Linux (various glibcs): OK
FreeBSD 8.1/8.2 (tested): EAI_FAIL
Solaris 10: OK

I can dig up the exact glibcs if people are curious, and I have some other testing results that returns OK as well for some unreleased OSes.