bind9: failed to get request's destination: failure

[Debian Bug Importer]

Automatically imported from Debian bug report #257558 http://bugs.debian.org/257558

[Benoit Panizzon (panizzon)]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi LaMont

I figured out the error.
Apparently something has changed in the IPv6 handeling.

In previous Versions you had to disable listening to IPv4 because IPv6
included all IPv4 Address inthe ::ffff:ipv4 notation.

This seams to have changed now. After re-enabling listening to IPv4
everything works fine again and the client.c error is gone.

Regards
- -Benoit-
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFA597RCVq2G/yL7/ARApXaAJ4yn7BEwoOEqiDOt2xYvDA0sb0s9gCguNDL
HevZapj9zvboMqlea0M++hM=
=VTW/
-----END PGP SIGNATURE-----

[Debian Bug Importer (debzilla)]

Automatically imported from Debian bug report #257558 http://bugs.debian.org/257558

[Debian Bug Importer (debzilla)]

Message-Id: <email address hidden>
Date: Sun, 04 Jul 2004 12:01:07 +0200
From: Benoit Panizzon <email address hidden>
To: Debian Bug Tracking System <email address hidden>
Subject: bind9: failed to get request's destination: failure

Package: bind9
Version: 1:9.2.3+9.2.4-rc5-1
Severity: critical
Justification: breaks the whole system

Hi Bind9 Maintainer

Something must have changed since the last 'sarge' bind9 update.

I'm runing a IPv6 enabled system and had no troubles util 2-3 days ago when bind9 suddenly started to behave strangely.

Everything starts OK.
I can do dig AXFR @localhost zone.dom that works fine.

nmap shows that bind9 is listening to UDP on port 53...

But:

magma:~# host -a woody.ch localhost
Trying "woody.ch"
;; connection timed out; no servers could be reached

The only error I could figure out is shortly after starting bind9:

Jul 4 11:54:31 magma named[5415]: client.c:1317: unexpected error:
Jul 4 11:54:31 magma named[5415]: failed to get request's destination: failure

Other hostmasters also notifyed me, that their secondary zone on my ns don't work anymore.

Any idea of where the problem lies?

-Benoit-

-- System Information:
Debian Release: testing/unstable
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: i386 (i686)
Kernel: Linux 2.6.7
Locale: LANG=C, LC_CTYPE=C

Versions of packages bind9 depends on:
ii adduser 3.57 Add and remove users and groups
ii libc6 2.3.2.ds1-13 GNU C Library: Shared libraries an
ii libdns11 1:9.2.3+9.2.4-rc5-1 DNS Shared Library used by BIND
ii libisc7 1:9.2.3+9.2.4-rc5-1 ISC Shared Library used by BIND
ii libisccc0 1:9.2.3+9.2.4-rc5-1 Command Channel Library used by BI
ii libisccfg0 1:9.2.3+9.2.4-rc5-1 Config File Handling Library used
ii liblwres1 1:9.2.3+9.2.4-rc5-1 Lightweight Resolver Library used
ii libssl0.9.7 0.9.7d-3 SSL shared libraries
ii netbase 4.17 Basic TCP/IP networking system

-- no debconf information

[Debian Bug Importer (debzilla)]

Message-Id: <email address hidden>
Date: Sun, 4 Jul 2004 12:41:17 +0200
From: Benoit Panizzon <email address hidden>
To: <email address hidden>
Subject: Re: Bug#257558: Acknowledgement (bind9: failed to get request's destination: failure)

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi LaMont

I figured out the error.
Apparently something has changed in the IPv6 handeling.

In previous Versions you had to disable listening to IPv4 because IPv6=20
included all IPv4 Address inthe ::ffff:ipv4 notation.

This seams to have changed now. After re-enabling listening to IPv4=20
everything works fine again and the client.c error is gone.

Regards
=2D -Benoit-
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFA597RCVq2G/yL7/ARApXaAJ4yn7BEwoOEqiDOt2xYvDA0sb0s9gCguNDL
HevZapj9zvboMqlea0M++hM=3D
=3DVTW/
=2D----END PGP SIGNATURE-----

[Thomas Hood (jdthood-yahoo)]

severity 257558 important

[Debian Bug Importer (debzilla)]

Message-Id: <20040706123743.4854410D6D0@localhost>
Date: Tue, 6 Jul 2004 14:37:43 +0200 (CEST)
From: <email address hidden> (Thomas Hood)
To: <email address hidden>
Subject: severity of 257558 is important

severity 257558 important

[Matt Zimmerman (mdz)]

Remove myself from all these CCs now that we have the warty-bugs mailing list

[Matt Zimmerman (mdz)]

Increase severity of RC bugs to major, now that we have other, non-RC bugs in
the list

[Matt Zimmerman (mdz)]

LaMont, what are we going to do about this bug? Is it RC or not?

[Thom May (thombot)]

I don't think it's a bug for us - the reporter had changed the default config to
work around an upstream problem which was subsequently fixed.

[Michael Gernoth (debian-gernoth)]

On Sun, Jul 04, 2004 at 12:01:07PM +0200, Benoit Panizzon wrote:
> Package: bind9
> Version: 1:9.2.3+9.2.4-rc5-1
>
> magma:~# host -a woody.ch localhost
> Trying "woody.ch"
> ;; connection timed out; no servers could be reached
>
> The only error I could figure out is shortly after starting bind9:
>
> Jul 4 11:54:31 magma named[5415]: client.c:1317: unexpected error:
> Jul 4 11:54:31 magma named[5415]: failed to get request's destination: failure

I have just hit the same problem on my IPv6 enabled Nameserver.
It is listening on both IPv4 and IPv6 and the error occurs after an
undefined number of successful external DNS-lookups via IPv6.
I'm then unable to query the server via IPv6 any further, but restarting
it solves the problem.

I have set listen-on { any; } and listen-on-v6 { any; } in my config.

Regards,
  Michael

[Debian Bug Importer (debzilla)]

Message-ID: <email address hidden>
Date: Fri, 24 Sep 2004 19:45:36 +0200
From: Michael Gernoth <email address hidden>
To: <email address hidden>
Subject: Re: bind9: failed to get request's destination: failure

On Sun, Jul 04, 2004 at 12:01:07PM +0200, Benoit Panizzon wrote:
> Package: bind9
> Version: 1:9.2.3+9.2.4-rc5-1
>
> magma:~# host -a woody.ch localhost
> Trying "woody.ch"
> ;; connection timed out; no servers could be reached
>
> The only error I could figure out is shortly after starting bind9:
>
> Jul 4 11:54:31 magma named[5415]: client.c:1317: unexpected error:
> Jul 4 11:54:31 magma named[5415]: failed to get request's destination: failure

I have just hit the same problem on my IPv6 enabled Nameserver.
It is listening on both IPv4 and IPv6 and the error occurs after an
undefined number of successful external DNS-lookups via IPv6.
I'm then unable to query the server via IPv6 any further, but restarting
it solves the problem.

I have set listen-on { any; } and listen-on-v6 { any; } in my config.

Regards,
  Michael

[Michael Gernoth (debian-gernoth)]

On Fri, Sep 24, 2004 at 07:45:36PM +0200, Michael Gernoth wrote:
> I have just hit the same problem on my IPv6 enabled Nameserver.
> It is listening on both IPv4 and IPv6 and the error occurs after an
> undefined number of successful external DNS-lookups via IPv6.
> I'm then unable to query the server via IPv6 any further, but restarting
> it solves the problem.

I updated to bind9 1:9.2.4-1 a few hours after my mail, and this version
is stable with an unchanged config since then.
So I think the package currently in testing is broken.

Regards,
  Michael

[Debian Bug Importer (debzilla)]

Message-ID: <email address hidden>
Date: Fri, 1 Oct 2004 17:53:55 +0200
From: Michael Gernoth <email address hidden>
To: <email address hidden>
Subject: Re: bind9: failed to get request's destination: failure

On Fri, Sep 24, 2004 at 07:45:36PM +0200, Michael Gernoth wrote:
> I have just hit the same problem on my IPv6 enabled Nameserver.
> It is listening on both IPv4 and IPv6 and the error occurs after an
> undefined number of successful external DNS-lookups via IPv6.
> I'm then unable to query the server via IPv6 any further, but restarting
> it solves the problem.

I updated to bind9 1:9.2.4-1 a few hours after my mail, and this version
is stable with an unchanged config since then.
So I think the package currently in testing is broken.

Regards,
  Michael

[Benoit Panizzon (bugreport-spam-woody)]

Package: bind9
Version: 1:9.2.4-1
Followup-For: Bug #257558

Hi

It looks like the problem is still apearing from time to time.

As i only want to liste to one IPv4 address I have:

        listen-on-v6 { any; };
        listen-on { 157.161.57.1; };

Don't know if this creates the problem... but anyway bind stopps accepting dns updates via ipv6 after a while...

-Benoit-

-- System Information:
Debian Release: 3.1
Architecture: i386 (i686)
Kernel: Linux 2.6.15
Locale: LANG=de_CH, LC_CTYPE=de_CH (charmap=ISO-8859-1)

Versions of packages bind9 depends on:
ii adduser 3.63 Add and remove users and groups
ii libc6 2.3.2.ds1-22sarge3 GNU C Library: Shared libraries an
ii libdns16 1:9.2.4-1 DNS Shared Library used by BIND
ii libisc7 1:9.2.4-1 ISC Shared Library used by BIND
ii libisccc0 1:9.2.4-1 Command Channel Library used by BI
ii libisccfg0 1:9.2.4-1 Config File Handling Library used
ii liblwres1 1:9.2.4-1 Lightweight Resolver Library used
ii libssl0.9.7 0.9.7e-3sarge1 SSL shared libraries
ii netbase 4.21 Basic TCP/IP networking system

-- no debconf information

[rusivi2 (rusivi2-deactivatedaccount)]

Thank you for reporting this bug.

Does this occur in Lucid?