Slave zone transfer issues

[JINMEI Tatuya / 神明達哉]

At Thu, 8 Dec 2011 03:16:19 +0000,
"Spain, Dr. Jeffry A." <spainj at countryday.net> wrote:

> > Hmm, assuming you still have this in your configuration:
> > "Xfrin": {"zones": [{"name": "jaspain.net", "master_addr": "2001:4870:20ca:158:4423:f19d:4ead:5c20"}]}
> 
> > and your BIND 9 master server really sent the notify from 2001:4870:20ca:158:4423:f19d:4ead:5c20 (maybe you want to check if the master server has multiple IPv6 addresses and might use a different one to send the notify), I believe it should immediately trigger a retransfer.  Perhaps you have a log starting with XFRIN_NOTIFY_UNKNOWN_MASTER?
> 
> Checking the bind9 master's IPv6 configuration, I found that I had failed to suppress its autoconfigured EUI-64 address 2001:4870:20ca:158:250:56ff:fe3f:11a1 after configuring the static address 2001:4870:20ca:158:4423:f19d:4ead:5c20. I fixed this, configured bind10 logging (DEBUG level 40), and did another zone update on the master. Using tcpdump, I saw that the master immediately sent a notify query to the bind10 server, and it responded. The bind10 log and the master log both showed an immediate zone transfer to the bind10 server via AXFR -- IXFR is not configured. Dig showed the correct serial number on the bind10 server. Thanks again. Jeff.

Okay, good to know that.  FYI, to diagnose things like these troubles,
the INFO log level should be normally sufficient.  Of course, if you
are okay with the verbosity, there's nothing wrong with setting it to
a higher level, though.

---
JINMEI, Tatuya
Internet Systems Consortium, Inc.