Bind 9.6.1: skipping zone transfer, but why ?

[bind9]

Hi list,

I'm currently developing a "management application" that will run on each of our nameservers 
and take care of domain creation, deletion and updates. Sort of "remote controlling" Bind via 
commands entered in a database. My setup is one master and three slaves, all running windows 
(2008, 3x2003 respectively) and all running Bind 9.6.1.

As an example: an update to a zone requires the master to re-generate the zonefile (with an 
increased serial) and then reload that particular zone. This will cause the master to issue 
notify-messages to the three slaves, which in turn will retrieve the updated zone file. All is 
well and good. 

This mostly works, but at times the slave servers seem to get into a "my master server is 
unreachable"-mode, which I can't get them out of unless I restart the ISC Bind service. Please 
take a look a the log below (from a slave). 

In line 1, the client receives a notify from the master. This is a create-scenario, so the 
master has just got a new _zones.conf and zonefile, and a "bind reconfig" command. However, the 
slave has not yet create its new _zones.conf file, so it isnt authorative yet. 
Then, a new _zones.conf is generated on the slave , and a "bind reconfig" is issued (line 2). 
Bind loads the new config file, but is skipping zone transfer as the master apparently is 
unreachable.

30-jun-2009 14:43:43.828 notify: notice: client 213.173.250.146#10182: received notify for 
zone '3yhta.dk': not authoritative
30-jun-2009 14:43:44.718 general: info: received control channel command 'reconfig'
30-jun-2009 14:43:44.718 general: info: loading configuration from 'C:\WINDOWS\system32
\dns\etc\named.conf'
30-jun-2009 14:43:44.718 general: info: using default UDP/IPv4 port range: [1024, 65535]
30-jun-2009 14:43:44.718 general: info: using default UDP/IPv6 port range: [1024, 65535]
30-jun-2009 14:43:44.718 network: info: no IPv6 interfaces found
30-jun-2009 14:43:44.781 general: info: reloading configuration succeeded
30-jun-2009 14:43:44.781 general: info: any newly configured zones are now loaded
30-jun-2009 14:43:44.781 general: info: zone 3yhta.dk/IN: refresh: skipping zone transfer as 
master 213.173.250.146#53 (source 0.0.0.0#0) is unreachable (cached)

Succeeding attempts to reload the zone, retransfer the zone yields the same " skipping zone 
transfer as master 213.173.250.146#53 (source 0.0.0.0#0) is unreachable (cached)" error, and a 
service restart caused this to show up:

30-jun-2009 14:48:22.828 xfer-in: error: transfer of '3yhta.dk/IN' from 213.173.250.146#53: 
failed to connect: connection refused
30-jun-2009 14:48:22.828 xfer-in: info: transfer of '3yhta.dk/IN' from 213.173.250.146#53: 
Transfer completed: 0 messages, 0 records, 0 bytes, 0.937 secs (0 bytes/sec)

I then restarted BOTH the service on the master and the slave, and the slave then successfully 
transferred the zone data, but my questions are:


1) "skipping zone transfer as master 213.173.250.146#53 (source 0.0.0.0#0) is unreachable 
(cached)" seem to indicate that the slave has cached a knowledge about the master being 
unreachable. It isn't. I can nslookup on the master from the slave just fine. What is wrong?

2) what causes "transfer of '3yhta.dk/IN' from 213.173.250.146#53: failed to connect: 
connection refused" ? There is no evidence of "connection refused" in the masters log, so where 
could this come from?


Thanks for your time in advance :o)

Best regards

Jan Hansen


s