[Pdns-users] PowerDNS hangs when used with recursor=

[Harald Kapper]

hi list,
from my experience I can confirm the described behaviour when using the supplied
recursor of the powerdns-rpm.

my solution was simply to use another recursor - as suggested by the
powerdns-docs.

in my case I simply used the previously running bind as the recursor using the
following named.conf:

[root at test01 root]# cat /etc/named.conf
// generated by named-bootconf.pl

options {
        directory "/var/named";

        listen-on port 5300 { 127.0.0.1; };

};

// 
// a caching only nameserver config
// 
controls {
        inet 127.0.0.1 allow { localhost; } keys { rndckey; };
};
zone "." IN {
        type hint;
        file "named.ca";
};

zone "localhost" IN {
        type master;
        file "localhost.zone";
        allow-update { none; };
};

zone "0.0.127.in-addr.arpa" IN {
        type master;
        file "named.local";
        allow-update { none; };
};

include "/etc/rndc.key";
---------------------------------------------------------------

this is based on a default-redhat-setup so your localhost-zones might vary, of
course you can also skip everything except the options for localhost and port
5300 as well as the "." zone for the root-nameservers.

for other recursor-backends refer to the powerdns-docs.
my experience with the above config: running for weeks without trouble.

best
Harald Kapper, icq# 36178328         kapper.net, inc.
managing director                    loeblichgasse  6
chief software development           1090 vienna, .at
tel +43 1 3195500-0, fax +43 1 3195502, hk at kapper.net
--------------quote.of.the.day--------------
 All programmers are optimists
 -- Frederick P. Brooks, Jr.

On Sat, 15 May 2004 14:59:07 +0200, Michel Stol <michel at powerdns.com> wrote:

>
>Lukas,
>
>Are you running both the recursor and PowerDNS on the same port?
>That would cause strange behaviour. You could specify port 5300
>to the recursor in recursor.conf and then:
>recursor=127.0.0.1:5300
>
>in your pdns.conf
>
>Greetings,
>
>- Michel Stol
>
>> I've got an installation of PowerDNS where it runs with
>> recursor=127.0.0.1, and a recursor is listening on said address.
>> After a while (2-7h), pdns_server stops responding to any query
>> it receives. No error message in the logfile. Nothing solid to
>> see when using strace.
>>
>> How could i debug this kind of problem? Running it under gdb
>> seems quite difficult, since its a threaded application, and i
>> have barely no clue on how to debug in gdb anyway.
>>
>> Could anyone tell me what information i should provide so that someone 
>> with more knowledge than myself could debug this?
>>
>> Below is the configuration file iam using:
>>
>> daemon=no
>> setuid=pdns
>> setgid=pdns
>> launch=gmysql
>> webserver=no
>> local-address=195.134.143.80
>> gmysql-socket=/var/run/symysql/symysql.sock
>> gmysql-user=pdns-master
>> gmysql-dbname=pdns
>> gmysql-password=[cut]
>> recursor=127.0.0.1
>> allow-recursion-override=yes
>> lazy-recursion=yes
>> master=on
>> slave=on
>>
>> _______________________________________________
>> Pdns-users mailing list
>> Pdns-users at mailman.powerdns.com
>> http://mailman.powerdns.com/mailman/listinfo/pdns-users
>>