Re: Problem with DLZ and AXFR requests?

On Tue, Sep 15, 2009 at 8:40 AM,
<[email protected]> wrote:
> Message: 6
> Date: Tue, 15 Sep 2009 08:13:42 +1000
> From: Jon Austin <[email protected]>
> Subject: [Bind-dlz-testers] Problem with DLZ and AXFR requests?
> To: [email protected]
> Message-ID:
>        <[email protected]>
> Content-Type: text/plain; charset=ISO-8859-1
>
> Hi there
>
> I'm having a few problems with DLZ and BIND crashing when a slave
> server requests a zone transfer (my secondary DNS host does not
> support DLZ).

Forgot digest mode makes it difficult to reply in a thread. Scott Hendra write:

> Are you certain that you have -enable-threads=no when you built BIND,
> assuming you are only using MySql just for this case, or, since I was
> using shared MySql, I just sent in "-n 1" to named when it started,
> and the core dumping stopped.

I used the default build flags/config (apart from enabling dlz-mysql)
in debian/rules which has

        ./configure --prefix=/usr \
                --mandir=\$${prefix}/share/man \
                --infodir=\$${prefix}/share/info \
                --sysconfdir=/etc/bind \
                --localstatedir=/var/run/bind \
                --enable-threads \
                --with-libtool \
                --enable-shared \
                --enable-static \
                --with-openssl=/usr \
                --with-gssapi=/usr \
                --with-gnu-ld \
                --with-dlz-postgres=no \
                --with-dlz-mysql=yes \
                --with-dlz-bdb=yes \
                --with-dlz-filesystem=yes \
                --with-dlz-ldap=yes \
                --with-dlz-stub=yes \
                --enable-ipv6

So I guess it is built multithreaded. When I start up named with -n 1 I get

15-Sep-2009 07:31:13.764 starting BIND 9.5.0-P2 -g -d 1 -n 1 -p 53
15-Sep-2009 07:31:13.764 found 1 CPU, using 1 worker thread
15-Sep-2009 07:31:13.766 loading configuration from '/etc/bind/named.conf'
15-Sep-2009 07:31:13.789 listening on IPv6 interfaces, port 53
15-Sep-2009 07:31:13.790 listening on IPv4 interface lo, 127.0.0.1#53
15-Sep-2009 07:31:13.791 listening on IPv4 interface eth0, 10.208.127.145#53
15-Sep-2009 07:31:13.793 Loading 'Mysql zone' using driver mysql
15-Sep-2009 07:31:13.795 default max-cache-size (33554432) applies

the number of worker threads does not change when removing -n 1 arg

15-Sep-2009 07:39:01.236 starting BIND 9.5.0-P2 -g -d 1 -p 53
15-Sep-2009 07:39:01.236 found 1 CPU, using 1 worker thread
15-Sep-2009 07:39:01.261 loading configuration from '/etc/bind/named.conf'
15-Sep-2009 07:39:01.262 listening on IPv6 interfaces, port 53
15-Sep-2009 07:39:01.263 listening on IPv4 interface lo, 127.0.0.1#53
15-Sep-2009 07:39:01.264 listening on IPv4 interface eth0, 10.208.127.145#53
15-Sep-2009 07:39:01.265 Loading 'Mysql zone' using driver mysql
15-Sep-2009 07:39:01.289 default max-cache-size (33554432) applies

Does this mean it's running as single thread?

In any case, the server is getting 0 queries except for the test ones
I'm putting to it - so I would assume it's not a deadlock situation.
The server will happly respond for all RR requests (A, CNAME, NS, MX,
SOA) but core dumps the instant I send an AFXR request from another
bind server.

------------------------------------------------------------------------------
Come build with us! The BlackBerry&reg; Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay 
ahead of the curve. Join us from November 9&#45;12, 2009. Register now&#33;
http://p.sf.net/sfu/devconf