[Pdns-users] External CNAME w/ auth and recursor

[Sebastien Luttringer]

Augie Schwer wrote:
> We run the latest auth. code out of SVN and the official recursor
> release and I don't have any problems like you have described; you
> should at least be running the latest official release of the auth.
> code: 2.9.21 .  --Augie
>
> On Tue, Mar 4, 2008 at 2:51 AM, Sebastien Luttringer
> <sebastien.luttringer at smartjog.com> wrote:
>   
>> Sebastien Luttringer wrote:
>>  > Hello,
>>  >
>>  > With a pdns 2.9.20-8 (from debian package), i can't resolve external
>>  > cname. pdns is authoritative for my domains and recursive for others. I
>>  > know it's bad, but it's not my choise.
>>  >
>>  > After some web browsing, i don't find a conf option to allow powerdns to
>>  > use its resolver to answer to external cname.
>>  >
>>  > I've read this :
>>  > http://www.nabble.com/CNAME-answer-problem--td12085558.html which
>>  > explain that everything it's ok. It's right for a authoritative server,
>>  > but if we want to use pdns as a recursive and authoritative server
>>  > external CNAME should be resolver in first part by pdns-server (for
>>  > cname) and pdns-recursor (for cname answer). Isn't it ?
>>  >
>>  > This is really impossible or i do something bad ?
>>  >
>>  Somebody have the same behaviour with powerdns or it's only for me ?
>>
>>
>>
>>  --
>>  Sebastien "Seblu" Luttringer      sebastien.luttringer at smartjog.com
>>  Smartjog SA                       http://www.smartjog.com/
>>
>>  _______________________________________________
>>  Pdns-users mailing list
>>  Pdns-users at mailman.powerdns.com
>>  http://mailman.powerdns.com/mailman/listinfo/pdns-user
I've try with home compiled sources (2.9.21) fetch on pdns website and i
have the same problem.

Pdns start correctly

# pdns_server 
Mar 05 11:48:28 This is a standalone pdns
Mar 05 11:48:28 Listening on controlsocket in '/var/run/pdns.controlsocket'
Mar 05 11:48:28 It is advised to bind to explicit addresses with the --local-address option
Mar 05 11:48:28 UDP server bound to 0.0.0.0:53
Mar 05 11:48:28 TCP server bound to 0.0.0.0:53
Mar 05 11:48:28 PowerDNS 2.9.21 (C) 2001-2006 PowerDNS.COM BV (Mar  5 2008, 11:34:56, gcc 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)) starting up
Mar 05 11:48:28 PowerDNS comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to redistribute it according to the terms of the GPL version 2.
Mar 05 11:48:28 DNS Proxy launched, local port 19874, remote 10.0.2.3:53
Mar 05 11:48:28 Creating backend connection for TCP
Mar 05 11:48:28 gmysql Connection succesful
Mar 05 11:48:28 About to create 3 backend threads for UDP
Mar 05 11:48:28 gmysql Connection succesful

When i try to resolv an internal cname

#host ns.tdf-pmm.wan 127.0.0.1
Using domain server:
Name: 127.0.0.1
Address: 127.0.0.1#53
Aliases: 

ns.tdf-pmm.wan is an alias for ns-1.tdf-pmm.wan.
ns.tdf-pmm.wan is an alias for ns-2.tdf-pmm.wan.
ns-2.tdf-pmm.wan has address 10.20.0.2

When i try to resolv an external cname

# host test.tdf-pmm.wan 127.0.0.1
Using domain server:
Name: 127.0.0.1
Address: 127.0.0.1#53
Aliases: 

Host test.tdf-pmm.wan not found: 3(NXDOMAIN)

But if i request a cname resolution no problem !

# host -t cname test.tdf-pmm.wan 127.0.0.1
Using domain server:
Name: 127.0.0.1
Address: 127.0.0.1#53
Aliases: 

test.tdf-pmm.wan is an alias for hades.seblu.net.

The resolution of this address is ok !

# host hades.seblu.net 127.0.0.1
Using domain server:
Name: 127.0.0.1
Address: 127.0.0.1#53
Aliases: 

hades.seblu.net is an alias for seblu.net.
seblu.net has address 88.191.33.22

And the db table seems to be right.

mysql> select name,type,content from records where type='cname';
+---------------------+-------+----------------------+
| name                | type  | content              |
+---------------------+-------+----------------------+
| ns.tdf-pmm.wan      | CNAME | ns-1.tdf-pmm.wan     |
| ns.tdf-pmm.wan      | CNAME | ns-2.tdf-pmm.wan     |
| test.tdf-pmm.wan    | CNAME | hades.seblu.net      |
+---------------------+-------+----------------------+
3 rows in set (0.00 sec)

This is really strange ! This can be a bug ? I should submit a bugrequest ?

-- 
Sebastien "Seblu" Luttringer      sebastien.luttringer at smartjog.com
Smartjog SA                       http://www.smartjog.com/