Hello Ulrich,
I believe you are testing your code against broken implementations!
That's why my resolver logged the error "bad cookie". You need to look
elsewhere...
Daniel
On 24.08.16 13:07, Ulrich Wisser wrote:
Hi Daniel!
Thank you for the pointers to some cookie enabled servers!
I tried my code and I see a strange behavior. I even checked with wireshark.
I do send a request with the cookie option. The response contains a
cookie opt from the server, but I do not get back my client cookie.
My understanding of the RFC7873 is that the server copies the client
cookie from the client request and appends its own server cookie.
Here is what I see
Server 212.80.96.32 Port 53 Zone interconnect.ch <http://interconnect.ch>
Send Cookie 0x329dadee3437b0d1
Client Cookie 0xf34d81b343a10000 Server Cookie
0x40cec0ebbf00000040cec0ebbf000000
Test9 failure! Client cookie did not match
I tested with several doing something wrong?
/Ulrich
Daniel Stirnimann <daniel.stirnimann(a)switch.ch
<mailto:daniel.stirnimann@switch.ch>> schrieb am Mo., 22. Aug. 2016 um
14:49 Uhr:
As for other authoritatives, BIND should support
the cookies
AFAIK. But
I haven't tried.
BIND 9.11 does which is due to be released within the next weeks. I'm
testing BIND 9.11b3 internally but unfortunately not on a public
reachable IP address. I also have a BIND 9.11b3 host as a resolver
configured. "bad cookie" responses are logged in the "resolver"
log
category. So, I'm not sure the following auth. servers are good
candidates to test your code ;-)
85.195.97.250
2a01:7480:1:100
89.25.242.242
185.60.169.2
193.93.22.133
Daniel
_______________________________________________
knot-dns-users mailing list
knot-dns-users(a)lists.nic.cz <mailto:knot-dns-users@lists.nic.cz>
https://lists.nic.cz/cgi-bin/mailman/listinfo/knot-dns-users
--
Ulrich Wisser
ulrich(a)wisser.se <mailto:ulrich@wisser.se>