On 09/04/15 16:04, Philip Ward wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
I have noticed the same bug today.
Our name servers crash when running "rndc reload" and the same text
seen by Tony Watt is seen in syslog.
Hi, That's "Tim Watts" :)
The only changes made recently to our DNS configuration is to add a
RPZ zone.
Interesting you should mention that.
I have an RPZ zone too and the problem has only just occurred for me too
after adding that (though I so heavily re-engineered my DNS last week
that it was not obvious this particular change might be the cause).
I am also wondering if my upping/downing of an interface is merely
causing bind9 to reload its config in response to an external change so
the bug might be exactly what you are seeing with an rndc reload?
My ignorance of the bind9 code means the above is just a speculative idea.
Removing the RPZ zone configuration enables named to reload.
I'm about to eat, but tomorrow I will disable my RPZ zone and see if
downing an interface still causes it to crash.
Best,
Tim
--
Tim Watts
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
