My IXFR/AXFR stopped suddenly
anandb at ripe.net
Sat Jul 7 12:48:27 UTC 2018
On 07/07/2018 13:31, Alan Clegg wrote:
>> Well, I just tried transferring zone using dig and it was successful
>> from slave
>> On slave
>> dig AXFR block.now @xx.xx.xx.xx
>> On master xfer-out.log
>> 07-Jul-2018 09:53:11.520 client xx.xx.xx.xx#16129 (immediate.block):
>> transfer of 'block.now/IN': AXFR started (serial 2018061016)
>> 07-Jul-2018 09:53:11.521 client xx.xx.xx.xx#16129 (immediate.block):
>> transfer of 'block.now/IN': AXFR ended
> This isn't using port 15455.
That's right Alan, but how does that matter? Everyone here seems
obssessed by port 15455. It seems everyone here needs a lesson in how
- Listens on port 53
- Serves zone block.now
- Sends UDP NOTIFY to slave at 184.108.40.206, from a RANDOM source port, and
destination port 15455.
- Listens in port 15455
- Loads zone block.now from the master at xx.xx.xx.xx
- AXFRs zone block.now from master xx.xx.xx.xx by initiating a TCP query
with a RANDOM source port, and destination port 53 (the default).
- Gets NOTIFY from master, from a RANDOM source port, and to destination
port 15455, when the zone block.now is updated
This is a perfectly legal setup. Blason says it wasn't working, but I
can't explain why. There may be something else wrong.
Of course, Blason is also being awkward with all of us, by showing us
only parts of his configuration, instead of showing all of it. Normally
I would try to help folk, but I have very little patience for folk who
deliberately hide their configs from us, and expect us to guess what the
problem may be.
Are you paying attention Blason? Help yourself by sharing your FULL
configs, and avoid all this second-guessing from people.
More information about the bind-users