[Pdns-users] the updating of SLAVE-zones

[Wouter de Jong]

Hello,

Anyone who can tell me if I forget something to do with the following in
mind:

If I update the SOA-record of a NATIVE zone, it work immediately
correct.
However, our slave nameservers are updating very slow, despite the
refresh of 60 seconds in the SOA-record and the slave-cycle-interval=60
setting on our slave nameservers. Also, it sets the 'prio' of the slave-record on a
non-zero value, for example '53248' or '32768'.

In the logs on the slave I read this:
Mar 17 14:53:15 blackbox pdns[46940]: Domain db.widexs.nl is stale,
master serial 2003020703, our serial 2003020702

And then:
Mar 17 15:04:23 blackbox pdns[46940]: AXFR started for 'db.widexs.nl',
transaction started
Mar 17 15:04:23 blackbox pdns[46940]: AXFR done for 'db.widexs.nl', zone
committed


That's not a refresh of 60 (slave-cycle-interval) + 60 (refresh in SOA-record) = 120 seconds, but of 11 minutes.

The (relevant) config of our primary server:
# the ip-addresses of our slave nameservers (a few other ip's i omitted)
allow-axfr-ips=212.204.207.192,62.250.7.3
disable-axfr=yes
disable-tcp=no
master=yes
strict-rfc-axfrs=no

MySQL-records:
+------+-----------+-------------------------+------+-------------------------------------------------------------------+------+------+-------------+
| id   | domain_id | name                    | type | content
| ttl  | prio | change_date |
+------+-----------+-------------------------+------+-------------------------------------------------------------------+------+------+-------------+
| 4857 |       102 | db.widexs.nl            | SOA  | ns1.widexs.nl.
hostmaster.widexs.nl. 2003020703 60 3600 604800 60 |   60 |    0 |
NULL |
+-----+--------------+--------+------------+--------+-----------------+---------+
| id  | name         | master | last_check | type   | notified_serial |
account |
+-----+--------------+--------+------------+--------+-----------------+---------+
| 102 | db.widexs.nl | NULL   |       NULL | NATIVE |            NULL |
NULL    |



The (relevant) config of our secondary server:
disable-axfr=yes
disable-tcp=no
slave=yes
slave-cycle-interval=60
strict-rfc-axfrs=no

MySQL-records:
+--------+-----------+-------------------------+------+-------------------------------------------------------------------+------+-------+-------------+
| id     | domain_id | name                    | type | content
| ttl  | prio  | change_date |
+--------+-----------+-------------------------+------+-------------------------------------------------------------------+------+-------+-------------+
| 748503 |       102 | db.widexs.nl            | SOA  | ns1.widexs.nl
hostmaster.widexs.nl 2003020703 60 3600 604800 60 |   60 | 32768 |
NULL |
+-----+--------------+-----------------+------------+-------+-----------------+---------+
| id  | name         | master          | last_check | type  |
notified_serial | account |
+-----+--------------+-----------------+------------+-------+-----------------+---------+
| 102 | db.widexs.nl | 212.204.192.252 | 1047909929 | SLAVE |
NULL | NULL    |



We're using the gmysql backend (MySQL 3.23) with InnoDB-tables.
Also issuing a 'pdns_control retrieve db.widexs.nl' doesn't help,
despite the docs say:
'Retrieve a slave domain from its master. Done nearly immediatly.'


Thanks in advance for a useful reply :P

-- 
Met vriendelijke groet/With kind regards,

Wouter de Jong
System-Administrator

CABLE & WIRELESS
Delivering the Internet promise(tm)
URL:    http://www.widexs.nl

Email:  Wouter.deJong at cw.com
Tel:    +31 (0) 23 5698070
Fax:    +31 (0) 23 5698099


***************************************************************************************
This message may contain information which is confidential or privileged.
If you are not the intended recipient, please advise the sender immediately
by reply e-mail and delete this message and any attachments without retaining a copy.
***************************************************************************************