failed while receiving responses: end of file
Kevin Darcy
kcd at daimlerchrysler.com
Fri Mar 26 01:13:04 UTC 2004
Carl Byington wrote:
>-----BEGIN PGP SIGNED MESSAGE-----
>Hash: SHA1
>
>Kevin Darcy <kcd at daimlerchrysler.com> wrote in message
>news:<c3qaf0$1gig$1 at sf1.isc.org>...
>
>[snip]
>
>
>
>>>Any clues?
>>>
>>>
>>>
>>Look at the logs on the master server. Seems like it is having
>>trouble
>>sending the zone transfer.
>>
>>
>
>The master server shows nothing in the syslog. Doing a 'dig @ns' from
>ns1 the zone transfer completes normally, and the syslog on ns shows
>the zone transfer. Doing 'rndc reload' on ns1 to force the same
>thing, the syslog on ns1 shows the errors, and the syslog on ns shows
>nothing (no syslog items from named).
>
>Does 'rndc trace 99' increase the amount of stuff that will be logged
>to the syslog, or do I need to add some stuff to named.conf to cause
>that to happen? The above tests were run with trace 99 on both
>servers.
>
>
Debugging won't be sent to syslog unless you have specifically
configured it to do so. By default, it'll get written to the file
named.run in the working directory. Frankly, though, debugging in BIND 9
still leaves a lot to be desired...
>I futzed around with the slave server, and it now loads k12ebooks
>properly, but another (new) zone (jchorton.com) is having the same
>problem. I have no idea what caused it to start working for k12ebooks
>- - possibly 'rndc reconfig', although I had stopped/restarted named
>before. I am sure that I did not change the zone file on the master
>server.
>
>For both problem zones, named-checkzone (on both servers) reports no
>errors.
>
I'm at a bit of a loss too. If it were me, at this point I'd probably
drop down to doing packet traces to try and find out at the very least
which side seems to be terminating the connection.
You're not going through a firewall or anything like that, are you?
- Kevin
More information about the bind-users
mailing list