[Pdns-users] problems with recursor

[Marko Kobal]

Hi,

My system:
FreeBSD 6.2-RELEASE-p7 i386
powerdns-2.9.21
powerdns-recursor-3.1.4_6


I have problems with recursor, but don't know exactly  what could be the issue. In general, the recursor works just fine, resolves any kind of domains, however, in some special cases, the resolving returns NXDOMAIN, althrough it should resolve the domain. The case is this:


1) Let's say I restart the recursor:

# /usr/local/etc/rc.d/pdns-recursor restart

2) Try to resolve the troubled domain:

# dig @127.0.0.1 syssvarun.infrax.si

; <<>> DiG 9.3.3 <<>> syssvarun.infrax.si
;; global options:  printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 64525
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 0

;; QUESTION SECTION:
;syssvarun.infrax.si.           IN      A

;; ANSWER SECTION:
syssvarun.infrax.si.    3461    IN      A       89.212.81.4

;; Query time: 1 msec
;; SERVER: 127.0.0.1#53(127.0.0.1)
;; WHEN: Sat Oct  6 10:23:01 2007
;; MSG SIZE  rcvd: 53



OK, everything is cool, domain has been properly resolved.

3) After some time, let's say, after 15 minutes I try to reslove it again:

# dig @127.0.0.1 syssvarun.infrax.si

; <<>> DiG 9.3.3 <<>> syssvarun.infrax.si
;; global options:  printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NXDOMAIN, id: 57947
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 0

;; QUESTION SECTION:
;syssvarun.infrax.si.           IN      A

;; AUTHORITY SECTION:
infrax.si.              3504    IN      SOA     svarun.infrax.si. hostmaster.infrax.si. 1189488858 3600 1200 604800 3600

;; Query time: 42 msec
;; SERVER: 127.0.0.1#53(127.0.0.1)
;; WHEN: Sat Oct  6 10:24:41 2007
;; MSG SIZE  rcvd: 91


??? Why NXDOMAIN? If I try to reslove that same domain via some other public DNS, it resloves it with NOERROR - *always*! Furthermore, it I thy to resolve some other domain within the same root domain, like "dig @127.0.0.1 hostsvarun.infrax.si", I *always* get NOERROR.

If I try and get back to step 1) - restart recursor, etc I always get the same scenario ... are there any problems with recursors caching?




-- 
Kind regards, Marko Kobal.