path: root/contrib/bind9/FAQ.xml

<?xml-stylesheet href="common.css" type="text/css"?>
<!DOCTYPE article PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
       "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd" []>
<!--
 - Copyright (C) 2004-2007  Internet Systems Consortium, Inc. ("ISC")
 - Copyright (C) 2000-2003  Internet Software Consortium.
 -
 - Permission to use, copy, modify, and distribute this software for any
 - purpose with or without fee is hereby granted, provided that the above
 - copyright notice and this permission notice appear in all copies.
 -
 - THE SOFTWARE IS PROVIDED "AS IS" AND ISC DISCLAIMS ALL WARRANTIES WITH
 - REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY
 - AND FITNESS.  IN NO EVENT SHALL ISC BE LIABLE FOR ANY SPECIAL, DIRECT,
 - INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
 - LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE
 - OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
 - PERFORMANCE OF THIS SOFTWARE.
-->

<!-- $Id: FAQ.xml,v 1.4.4.8 2007/02/05 05:23:39 marka Exp $ -->

<article class="faq">
  <title>Frequently Asked Questions about BIND 9</title>
  <articleinfo>
    <copyright>
      <year>2004</year>
      <year>2005</year>
      <year>2006</year>
      <year>2007</year>
      <holder>Internet Systems Consortium, Inc. ("ISC")</holder>
    </copyright>
    <copyright>
      <year>2000</year>
      <year>2001</year>
      <year>2002</year>
      <year>2003</year>
      <holder>Internet Software Consortium.</holder>
    </copyright>
  </articleinfo>
  <qandaset defaultlabel='qanda'>
    <qandaentry>
      <question>
	<para>
	  Why doesn't -u work on Linux 2.2.x when I build with
	  --enable-threads?
	</para>
      </question>
      <answer>
	<para>
	  Linux threads do not fully implement the Posix threads
	  (pthreads) standard.  In particular, setuid() operates only
	  on the current thread, not the full process.  Because of
	  this limitation, BIND 9 cannot use setuid() on Linux as it
	  can on all other supported platforms.  setuid() cannot be
	  called before creating threads, since the server does not
	  start listening on reserved ports until after threads have
	  started.
	</para>
	<para>
	  In the 2.2.18 or 2.3.99-pre3 and newer kernels, the ability
	  to preserve capabilities across a setuid() call is present.
	  This allows BIND 9 to call setuid() early, while retaining
	  the ability to bind reserved ports.  This is a Linux-specific
	  hack.
	</para>
	<para>
	  On a 2.2 kernel, BIND 9 does drop many root privileges, so
	  it should be less of a security risk than a root process
	  that has not dropped privileges.
	</para>
	<para>
	  If Linux threads ever work correctly, this restriction will
	  go away.
	</para>
	<para>
	  Configuring BIND9 with the --disable-threads option (the
	  default) causes a non-threaded version to be built, which
	  will allow -u to be used.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Why do I get the following errors:
<programlisting>general: errno2result.c:109: unexpected error:
general: unable to convert errno to isc_result: 14: Bad address
client: UDP client handler shutting down due to fatal receive error: unexpected error</programlisting>
	</para>
      </question>
      <answer>
	<para>
	  This is the result of a Linux kernel bug.
	</para>
	<para>
	  See:
	  <ulink url="http://marc.theaimsgroup.com/?l=linux-netdev&amp;m=113081708031466&amp;w=2">http://marc.theaimsgroup.com/?l=linux-netdev&amp;m=113081708031466&amp;w=2</ulink>
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Why does named log the warning message <quote>no TTL specified -
	  using SOA MINTTL instead</quote>?
	</para>
      </question>
      <answer>
	<para>
	  Your zone file is illegal according to RFC1035.  It must either
	  have a line like:
	</para>
	<informalexample>
	  <programlisting>
$TTL 86400</programlisting>
	</informalexample>
	<para>
	  at the beginning, or the first record in it must have a TTL field,
	  like the "84600" in this example:
	</para>
	<informalexample>
	  <programlisting>
example.com. 86400 IN SOA ns hostmaster ( 1 3600 1800 1814400 3600 )</programlisting>
	</informalexample>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Why do I see 5 (or more) copies of named on Linux?
	</para>
      </question>
      <answer>
	<para>
	  Linux threads each show up as a process under ps.  The
	  approximate number of threads running is n+4, where n is
	  the number of CPUs.  Note that the amount of memory used
	  is not cumulative; if each process is using 10M of memory,
	  only a total of 10M is used.
	</para>
	<para>
	  Newer versions of Linux's ps command hide the individual threads
	  and require -L to display them.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Why does BIND 9 log <quote>permission denied</quote> errors accessing
	  its configuration files or zones on my Linux system even
	  though it is running as root?
	</para>
      </question>
      <answer>
	<para>
	  On Linux, BIND 9 drops most of its root privileges on
	  startup.  This including the privilege to open files owned
	  by other users.  Therefore, if the server is running as
	  root, the configuration files and zone files should also
	  be owned by root.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Why do I get errors like <quote>dns_zone_load: zone foo/IN: loading
	  master file bar: ran out of space</quote>?
	</para>
      </question>
      <answer>
	<para>
	This is often caused by TXT records with missing close
	quotes.  Check that all TXT records containing quoted strings
	have both open and close quotes.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	How do I produce a usable core file from a multi-threaded
	named on Linux?
	</para>
      </question>
      <answer>
	<para>
	If the Linux kernel is 2.4.7 or newer, multi-threaded core
	dumps are usable (that is, the correct thread is dumped).
	Otherwise, if using a 2.2 kernel, apply the kernel patch
	found in contrib/linux/coredump-patch and rebuild the kernel.
	This patch will cause multi-threaded programs to dump the
	correct thread.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  How do I restrict people from looking up the server version?
	</para>
      </question>
      <answer>
	<para>
	  Put a "version" option containing something other than the
	  real version in the "options" section of named.conf.  Note
	  doing this will not prevent attacks and may impede people
	  trying to diagnose problems with your server.  Also it is
	  possible to "fingerprint" nameservers to determine their
	  version.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  How do I restrict only remote users from looking up the
	  server version?
	</para>
      </question>
      <answer>
	<para>
	  The following view statement will intercept lookups as the
	  internal view that holds the version information will be
	  matched last.  The caveats of the previous answer still
	  apply, of course.
	</para>
	<informalexample>
	  <programlisting>
view "chaos" chaos {
	match-clients { &lt;those to be refused&gt;; };
	allow-query { none; };
	zone "." {
		type hint;
		file "/dev/null";  // or any empty file
	};
};</programlisting>
	</informalexample>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  What do <quote>no source of entropy found</quote> or <quote>could not
	  open entropy source foo</quote> mean?
	</para>
      </question>
      <answer>
	<para>
	  The server requires a source of entropy to perform certain
	  operations, mostly DNSSEC related.  These messages indicate
	  that you have no source of entropy.  On systems with
	  /dev/random or an equivalent, it is used by default.  A
	  source of entropy can also be defined using the random-device
	  option in named.conf.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I installed BIND 9 and restarted named, but it's still BIND 8.  Why?
	</para>
      </question>
      <answer>
	<para>
	  BIND 9 is installed under /usr/local by default.  BIND 8
	  is often installed under /usr.  Check that the correct named
	  is running.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I'm trying to use TSIG to authenticate dynamic updates or
	  zone transfers.  I'm sure I have the keys set up correctly,
	  but the server is rejecting the TSIG.  Why?
	</para>
      </question>
      <answer>
	<para>
	  This may be a clock skew problem.  Check that the the clocks
	  on the client and server are properly synchronised (e.g.,
	  using ntp).
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I'm trying to compile BIND 9, and "make" is failing due to
	  files not being found.  Why?
	</para>
      </question>
      <answer>
	<para>
	  Using a parallel or distributed "make" to build BIND 9 is
	  not supported, and doesn't work.  If you are using one of
	  these, use normal make or gmake instead.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I have a BIND 9 master and a BIND 8.2.3 slave, and the
	  master is logging error messages like <quote>notify to 10.0.0.1#53
	  failed: unexpected end of input</quote>.  What's wrong?
	</para>
      </question>
      <answer>
	<para>
	  This error message is caused by a known bug in BIND 8.2.3
	  and is fixed in BIND 8.2.4.  It can be safely ignored - the
	  notify has been acted on by the slave despite the error
	  message.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I keep getting log messages like the following.  Why?
	</para>
	<para>
	  Dec  4 23:47:59 client 10.0.0.1#1355: updating zone
	  'example.com/IN': update failed: 'RRset exists (value
	  dependent)' prerequisite not satisfied (NXRRSET)
	</para>
      </question>
      <answer>
	<para>
	  DNS updates allow the update request to test to see if
	  certain conditions are met prior to proceeding with the
	  update.  The message above is saying that conditions were
	  not met and the update is not proceeding.  See doc/rfc/rfc2136.txt
	  for more details on prerequisites.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I keep getting log messages like the following.  Why?
	</para>
	<para>
	  Jun 21 12:00:00.000 client 10.0.0.1#1234: update denied
	</para>
      </question>
      <answer>
	<para>
	  Someone is trying to update your DNS data using the RFC2136
	  Dynamic Update protocol.  Windows 2000 machines have a habit
	  of sending dynamic update requests to DNS servers without
	  being specifically configured to do so.  If the update
	  requests are coming from a Windows 2000 machine, see
	  <ulink
	   url="http://support.microsoft.com/support/kb/articles/q246/8/04.asp">
	    http://support.microsoft.com/support/kb/articles/q246/8/04.asp
	  </ulink>
	  for information about how to turn them off.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I see a log message like the following.  Why?
	</para>
	<para>
	  couldn't open pid file '/var/run/named.pid': Permission denied
	</para>
      </question>
      <answer>
	<para>
	  You are most likely running named as a non-root user, and
	  that user does not have permission to write in /var/run.
	  The common ways of fixing this are to create a /var/run/named
	  directory owned by the named user and set pid-file to
	  "/var/run/named/named.pid", or set pid-file to "named.pid",
	  which will put the file in the directory specified by the
	  directory option (which, in this case, must be writable by
	  the named user).
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  When I do a "dig . ns", many of the A records for the root
	  servers are missing.  Why?
	</para>
      </question>
      <answer>
	<para>
	  This is normal and harmless.  It is a somewhat confusing
	  side effect of the way BIND 9 does RFC2181 trust ranking
	  and of the efforts BIND 9 makes to avoid promoting glue
	  into answers.
	</para>
	<para>
	 When BIND 9 first starts up and primes its cache, it receives
	 the root server addresses as additional data in an authoritative
	 response from a root server, and these records are eligible
	 for inclusion as additional data in responses.  Subsequently
	 it receives a subset of the root server addresses as
	 additional data in a non-authoritative (referral) response
	 from a root server.  This causes the addresses to now be
	 considered non-authoritative (glue) data, which is not
	 eligible for inclusion in responses.
	</para>
	<para>
	 The server does have a complete set of root server addresses
	 cached at all times, it just may not include all of them
	 as additional data, depending on whether they were last
	 received as answers or as glue.  You can always look up the
	 addresses with explicit queries like "dig a.root-servers.net A".
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Zone transfers from my BIND 9 master to my Windows 2000
	  slave fail.  Why?
	</para>
      </question>
      <answer>
	<para>
	  This may be caused by a bug in the Windows 2000 DNS server
	  where DNS messages larger than 16K are not handled properly.
	  This can be worked around by setting the option "transfer-format
	  one-answer;".  Also check whether your zone contains domain
	  names with embedded spaces or other special characters,
	  like "John\032Doe\213s\032Computer", since such names have
	  been known to cause Windows 2000 slaves to incorrectly
	  reject the zone.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Why don't my zones reload when I do an "rndc reload" or SIGHUP?
	</para>
      </question>
      <answer>
	<para>
	  A zone can be updated either by editing zone files and
	  reloading the server or by dynamic update, but not both.
	  If you have enabled dynamic update for a zone using the
	  "allow-update" option, you are not supposed to edit the
	  zone file by hand, and the server will not attempt to reload
	  it.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I can query the nameserver from the nameserver but not from other
	  machines.  Why?
	</para>
      </question>
      <answer>
	<para>
	  This is usually the result of the firewall configuration stopping
	  the queries and / or the replies.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  How can I make a server a slave for both an internal and
	  an external view at the same time?  When I tried, both views
	  on the slave were transferred from the same view on the master.
	</para>
      </question>
      <answer>
	<para>
	  You will need to give the master and slave multiple IP
	  addresses and use those to make sure you reach the correct
	  view on the other machine.
	</para>
	<informalexample>
	  <programlisting>
Master: 10.0.1.1 (internal), 10.0.1.2 (external, IP alias)
    internal:
	match-clients { !10.0.1.2; !10.0.1.4; 10.0.1/24; };
		notify-source 10.0.1.1;
		transfer-source 10.0.1.1;
		query-source address 10.0.1.1;
    external:
	match-clients { any; };
	recursion no;	// don't offer recursion to the world
	notify-source 10.0.1.2;
	transfer-source 10.0.1.2;
	query-source address 10.0.1.2;

Slave: 10.0.1.3 (internal), 10.0.1.4 (external, IP alias)
    internal:
	match-clients { !10.0.1.2; !10.0.1.4; 10.0.1/24; };
	notify-source 10.0.1.3;
	transfer-source 10.0.1.3;
	query-source address 10.0.1.3;
   external:
	match-clients { any; };
	recursion no;	// don't offer recursion to the world
	notify-source 10.0.1.4;
	transfer-source 10.0.1.4;
	query-source address 10.0.1.4;</programlisting>
	</informalexample>
	<para>
	  You put the external address on the alias so that all the other
	  dns clients on these boxes see the internal view by default.
	</para>
      </answer>
      <answer>
	<para>
	  BIND 9.3 and later: Use TSIG to select the appropriate view.
	</para>
	<informalexample>
	  <programlisting>
Master 10.0.1.1:
	key "external" {
		algorithm hmac-md5;
		secret "xxxxxxxx";
	};
	view "internal" {
		match-clients { !key external; 10.0.1/24; };
		...
	};
	view "external" {
		match-clients { key external; any; };
		server 10.0.1.2 { keys external; };
		recursion no;
		...
	};

Slave 10.0.1.2:
	key "external" {
		algorithm hmac-md5;
		secret "xxxxxxxx";
	};
	view "internal" {
		match-clients { !key external; 10.0.1/24; };
		...
	};
	view "external" {
		match-clients { key external; any; };
		server 10.0.1.1 { keys external; };
		recursion no;
		...
	};</programlisting>
	</informalexample>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I have FreeBSD 4.x and "rndc-confgen -a" just sits there.
	</para>
      </question>
      <answer>
	<para>
	  /dev/random is not configured.  Use rndcontrol(8) to tell
	  the kernel to use certain interrupts as a source of random
	  events.  You can make this permanent by setting rand_irqs
	  in /etc/rc.conf.
	</para>
	<informalexample>
	  <programlisting>
/etc/rc.conf
rand_irqs="3 14 15"</programlisting>
	</informalexample>
	<para>
	  See also
	  <ulink url="http://people.freebsd.org/~dougb/randomness.html">
	    http://people.freebsd.org/~dougb/randomness.html
	  </ulink>
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Why is named listening on UDP port other than 53?
	</para>
      </question>
      <answer>
	<para>
	  Named uses a system selected port to make queries of other
	  nameservers.  This behaviour can be overridden by using
	  query-source to lock down the port and/or address.  See
	  also notify-source and transfer-source.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I get error messages like <quote>multiple RRs of singleton type</quote>
	  and <quote>CNAME and other data</quote> when transferring a zone.  What
	  does this mean?
	</para>
      </question>
      <answer>
	<para>
	  These indicate a malformed master zone.  You can identify
	  the exact records involved by transferring the zone using
	  dig then running named-checkzone on it.
	</para>
	<informalexample>
	  <programlisting>
dig axfr example.com @master-server &gt; tmp
named-checkzone example.com tmp</programlisting>
	</informalexample>
	<para>
	  A CNAME record cannot exist with the same name as another record
	  except for the DNSSEC records which prove its existence (NSEC).
	</para>
	<para>
	  RFC 1034, Section 3.6.2: <quote>If a CNAME RR is present at a node,
	  no other data should be present; this ensures that the data for a
	  canonical name and its aliases cannot be different.  This rule also
	  insures that a cached CNAME can be used without checking with an
	  authoritative server for other RR types.</quote>
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I get error messages like <quote>named.conf:99: unexpected end
	  of input</quote> where 99 is the last line of named.conf.
	</para>
      </question>
      <answer>
	<para>
	  Some text editors (notepad and wordpad) fail to put a line
	  title indication (e.g. CR/LF) on the last line of a
	  text file.  This can be fixed by "adding" a blank line to
	  the end of the file.  Named expects to see EOF immediately
	  after EOL and treats text files where this is not met as
	  truncated.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I get warning messages like <quote>zone example.com/IN: refresh:
	  failure trying master 1.2.3.4#53: timed out</quote>.
	</para>
      </question>
      <answer>
	<para>
	Check that you can make UDP queries from the slave to the master
	</para>
	<informalexample>
	  <programlisting>
dig +norec example.com soa @1.2.3.4</programlisting>
	</informalexample>
	<para>
	  You could be generating queries faster than the slave can
	  cope with.  Lower the serial query rate.
	</para>
	<informalexample>
	  <programlisting>
serial-query-rate 5; // default 20</programlisting>
	</informalexample>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  How do I share a dynamic zone between multiple views?
	</para>
      </question>
      <answer>
	<para>
	  You choose one view to be master and the second a slave and
	  transfer the zone between views.
	</para>
	<informalexample>
	  <programlisting>
Master 10.0.1.1:
	key "external" {
		algorithm hmac-md5;
		secret "xxxxxxxx";
	};

	key "mykey" {
		algorithm hmac-md5;
		secret "yyyyyyyy";
	};

	view "internal" {
		match-clients { !external; 10.0.1/24; };
		server 10.0.1.1 {
			/* Deliver notify messages to external view. */
			keys { external; };
		};
		zone "example.com" {
			type master;
			file "internal/example.db";
			allow-update { key mykey; };
			notify-also { 10.0.1.1; };
		};
	};

	view "external" {
		match-clients { external; any; };
		zone "example.com" {
			type slave;
			file "external/example.db";
			masters { 10.0.1.1; };
			transfer-source { 10.0.1.1; };
			// allow-update-forwarding { any; };
			// allow-notify { ... };
		};
	};</programlisting>
	</informalexample>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I get a error message like <quote>zone wireless.ietf56.ietf.org/IN:
	  loading master file primaries/wireless.ietf56.ietf.org: no
	  owner</quote>.
	</para>
      </question>
      <answer>
	<para>
	  This error is produced when a line in the master file
	  contains leading white space (tab/space) but the is no
	  current record owner name to inherit the name from.  Usually
	  this is the result of putting white space before a comment.
	  Forgetting the "@" for the SOA record or indenting the master
	  file.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Why are my logs in GMT (UTC).
	</para>
      </question>
      <answer>
	<para>
	  You are running chrooted (-t) and have not supplied local timezone
	  information in the chroot area.
	</para>
	<simplelist>
	  <member>FreeBSD: /etc/localtime</member>
	  <member>Solaris: /etc/TIMEZONE and /usr/share/lib/zoneinfo</member>
	  <member>OSF: /etc/zoneinfo/localtime</member>
	  </simplelist>
	<para>
	  See also tzset(3) and zic(8).
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I get the error message <quote>named: capset failed: Operation
	  not permitted</quote> when starting named.
	</para>
      </question>
      <answer>
	<para>
	  The capability module, part of "Linux Security Modules/LSM",
	  has not been loaded into the kernel.  See insmod(8).
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I get <quote>rndc: connect failed: connection refused</quote> when
	  I try to run rndc.
	</para>
      </question>
      <answer>
	<para>
	  This is usually a configuration error.
	</para>
	<para>
	  First ensure that named is running and no errors are being
	  reported at startup (/var/log/messages or equivalent).
	  Running "named -g &lt;usual arguments&gt;" from a title
	  can help at this point.
	</para>
	<para>
	  Secondly ensure that named is configured to use rndc either
	  by "rndc-confgen -a", rndc-confgen or manually.  The
	  Administrators Reference manual has details on how to do
	  this.
	</para>
	<para>
	  Old versions of rndc-confgen used localhost rather than
	  127.0.0.1 in /etc/rndc.conf for the default server.  Update
	  /etc/rndc.conf if necessary so that the default server
	  listed in /etc/rndc.conf matches the addresses used in
	  named.conf.  "localhost" has two address (127.0.0.1 and
	  ::1).
	</para>
	<para>
	  If you use "rndc-confgen -a" and named is running with -t or -u
	  ensure that /etc/rndc.conf has the correct ownership and that
	  a copy is in the chroot area.  You can do this by re-running
	  "rndc-confgen -a" with appropriate -t and -u arguments.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I don't get RRSIG's returned when I use "dig +dnssec".
	</para>
      </question>
      <answer>
	<para>
	  You need to ensure DNSSEC is enabled (dnssec-enable yes;).
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I get <quote>Error 1067</quote> when starting named under Windows.
	</para>
      </question>
      <answer>
	<para>
	  This is the service manager saying that named exited.   You
	  need to examine the Application log in the EventViewer to
	  find out why.
	</para>
	<para>
	  Common causes are that you failed to create "named.conf"
	  (usually "C:\windows\dns\etc\named.conf") or failed to
	  specify the directory in named.conf.
	</para>
	<informalexample>
	  <programlisting>
options {
	Directory "C:\windows\dns\etc";
};</programlisting>
	</informalexample>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I get <quote>transfer of 'example.net/IN' from 192.168.4.12#53:
	  failed while receiving responses: permission denied</quote> error
	  messages.
	</para>
      </question>
      <answer>
	<para>
	  These indicate a filesystem permission error preventing
	  named creating / renaming the temporary file.  These will
	  usually also have other associated error messages like
	</para>
	<informalexample>
	  <programlisting>
"dumping master file: sl/tmp-XXXX5il3sQ: open: permission denied"</programlisting>
	</informalexample>
	<para>
	  Named needs write permission on the directory containing
	  the file.  Named writes the new cache file to a temporary
	  file then renames it to the name specified in named.conf
	  to ensure that the contents are always complete.  This is
	  to prevent named loading a partial zone in the event of
	  power failure or similar interrupting the write of the
	  master file.
	</para>
	<para>
	  Note file names are relative to the directory specified in
	  options and any chroot directory  ([&lt;chroot
	  dir&gt;/][&lt;options dir&gt;]).
	</para>
	<informalexample>
	  <para>
	    If named is invoked as "named -t /chroot/DNS" with
	    the following named.conf then "/chroot/DNS/var/named/sl"
	    needs to be writable by the user named is running as.
	  </para>
	  <programlisting>
options {
	directory "/var/named";
};

zone "example.net" {
	type slave;
	file "sl/example.net";
	masters { 192.168.4.12; };
};</programlisting>
	</informalexample>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  How do I integrate BIND 9 and Solaris SMF
	</para>
      </question>
      <answer>
	<para>
	  Sun has a blog entry describing how to do this.
	</para>
	<para>
	  <ulink
	  url="http://blogs.sun.com/roller/page/anay/Weblog?catname=%2FSolaris">
	     http://blogs.sun.com/roller/page/anay/Weblog?catname=%2FSolaris
	  </ulink>
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Can a NS record refer to a CNAME.
	</para>
      </question>
      <answer>
	<para>
	  No.  The rules for glue (copies of the *address* records
	  in the parent zones) and additional section processing do
	  not allow it to work.
	</para>
	<para>
	  You would have to add both the CNAME and address records
	  (A/AAAA) as glue to the parent zone and have CNAMEs be
	  followed when doing additional section processing to make
	  it work.  No nameserver implementation supports either of
	  these requirements.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  What does <quote>RFC 1918 response from Internet for
	  0.0.0.10.IN-ADDR.ARPA</quote> mean?
	</para>
      </question>
      <answer>
	<para>
	  If the IN-ADDR.ARPA name covered refers to a internal address
	  space you are using then you have failed to follow RFC 1918
	  usage rules and are leaking queries to the Internet.  You
	  should establish your own zones for these addresses to prevent
	  you querying the Internet's name servers for these addresses.
	  Please see <ulink url="http://as112.net/">http://as112.net/</ulink>
	  for details of the problems you are causing and the counter
	  measures that have had to be deployed.
	</para>
	<para>
	  If you are not using these private addresses then a client
	  has queried for them.  You can just ignore the messages,
	  get the offending client to stop sending you these messages
	  as they are most probably leaking them or setup your own zones
	  empty zones to serve answers to these queries.
	</para>
	<informalexample>
	  <programlisting>
zone "10.IN-ADDR.ARPA" {
	type master;
	file "empty";
};

zone "16.172.IN-ADDR.ARPA" {
	type master;
	file "empty";
};

...

zone "31.172.IN-ADDR.ARPA" {
	type master;
	file "empty";
};

zone "168.192.IN-ADDR.ARPA" {
	type master;
	file "empty";
};

empty:
@ 10800 IN SOA &lt;name-of-server&gt;. &lt;contact-email&gt;. (
	       1 3600 1200 604800 10800 )
@ 10800 IN NS &lt;name-of-server&gt;.</programlisting>
	</informalexample>
	<para>
	<note>
	  Future versions of named are likely to do this automatically.
	</note>
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	   I'm running BIND on Red Hat Enterprise Linux or Fedora Core -
	</para>
	<para>
	  Why can't named update slave zone database files?
	</para>
	<para>
	  Why can't named create DDNS journal files or update
	  the master zones from journals?
	</para>
	<para>
	  Why can't named create custom log files?
	</para>
      </question>

      <answer>
	<para>
	  Red Hat Security Enhanced Linux (SELinux) policy security
	  protections :
	</para>

	<para>
	   Red Hat have adopted the National Security Agency's
	   SELinux security policy ( see http://www.nsa.gov/selinux
	   ) and recommendations for BIND security , which are more
	   secure than running named in a chroot and make use of
	   the bind-chroot environment unnecessary .
	</para>

	<para>
	  By default, named is not allowed by the SELinux policy
	  to write, create or delete any files EXCEPT in these
	  directories:
	  <informalexample>
	    <programlisting>
$ROOTDIR/var/named/slaves
$ROOTDIR/var/named/data
$ROOTDIR/var/tmp
	    </programlisting>
	  </informalexample>
	  where $ROOTDIR may be set in /etc/sysconfig/named if
	  bind-chroot is installed.
	</para>

	<para>
	  The SELinux policy particularly does NOT allow named to modify
	  the $ROOTDIR/var/named directory, the default location for master
	  zone database files.
	</para>

	<para>
	  SELinux policy overrules file access permissions - so
	  even if all the files under /var/named have ownership
	  named:named and mode rw-rw-r--, named will still not be
	  able to write or create files except in the directories
	  above, with SELinux in Enforcing mode.
	</para>
  
	<para>
	  So, to allow named to update slave or DDNS zone files,
	  it is best to locate them in $ROOTDIR/var/named/slaves,
	  with named.conf zone statements such as:
	  <informalexample>
	    <programlisting>
zone "slave.zone." IN {
	type slave;
	file "slaves/slave.zone.db";
	...
};   
zone "ddns.zone." IN  {
	type master;
	allow-updates {...};
	file "slaves/ddns.zone.db";
};
	    </programlisting>
	  </informalexample>
	</para>

	<para>
	  To allow named to create its cache dump and statistics
	  files, for example, you could use named.conf options
	  statements such as:
	  <informalexample>
	    <programlisting>
options {
	...
	dump-file "/var/named/data/cache_dump.db";
	statistics-file "/var/named/data/named_stats.txt";
	...
};
	    </programlisting>
	  </informalexample>
	</para>

	<para>
	  You can also tell SELinux to allow named to update any
	  zone database files, by setting the SELinux tunable boolean
	  parameter 'named_write_master_zones=1', using the
	  system-config-securitylevel GUI, using the 'setsebool'
	  command, or in /etc/selinux/targeted/booleans.
	</para>
  
	<para>
	  You can disable SELinux protection for named entirely by
	  setting the 'named_disable_trans=1' SELinux tunable boolean
	  parameter.
	</para>
    
	<para>
	  The SELinux named policy defines these SELinux contexts for named:
	  <informalexample>
	    <programlisting>
named_zone_t : for zone database files       - $ROOTDIR/var/named/*
named_conf_t : for named configuration files - $ROOTDIR/etc/{named,rndc}.*
named_cache_t: for files modifiable by named - $ROOTDIR/var/{tmp,named/{slaves,data}}
	    </programlisting>
	  </informalexample>
	</para>
   
	<para>
	  If you want to retain use of the SELinux policy for named,
	  and put named files in different locations, you can do
	  so by changing the context of the custom file locations
	  .
	</para>

	<para>
	  To create a custom configuration file location, e.g.
	  '/root/named.conf', to use with the 'named -c' option,
	  do:
	  <informalexample>
	    <programlisting>
# chcon system_u:object_r:named_conf_t /root/named.conf
	    </programlisting>
	  </informalexample>
	</para>
  
	<para>
	  To create a custom modifiable named data location, e.g.
	  '/var/log/named' for a log file, do:
	  <informalexample>
	    <programlisting>
# chcon system_u:object_r:named_cache_t /var/log/named
	    </programlisting>
	  </informalexample>
	</para>
   
	<para>
   To create a custom zone file location, e.g. /root/zones/, do:
	  <informalexample>
	    <programlisting>
# chcon system_u:object_r:named_zone_t /root/zones/{.,*}
	    </programlisting>
	  </informalexample>
	</para>
  
	<para>
	  See these man-pages for more information : selinux(8),
	  named_selinux(8), chcon(1), setsebool(8)
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  I want to forward all DNS queries from my caching nameserver to
	  another server. But there are some domains which have to be
	  served locally, via rbldnsd.
	</para>
	<para>
	  How do I achieve this ?
	</para>
      </question>
      <answer>
        <programlisting>
options {
	forward only;
	forwarders { &lt;ip.of.primary.nameserver&gt;; };
};

zone "sbl-xbl.spamhaus.org" {
	type forward; forward only;
	forwarders { &lt;ip.of.rbldns.server&gt; port 530; };
};

zone "list.dsbl.org" {
	type forward; forward only;
	forwarders { &lt;ip.of.rbldns.server&gt; port 530; };
};
        </programlisting>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Will named be affected by the 2007 changes to daylight savings
	  rules in the US.
	</para>
      </question>
      <answer>
	<para>
	  No, so long as the machines internal clock (as reported
	  by "date -u") remains at UTC.  The only visible change
	  if you fail to upgrade your OS, if you are in a affected
	  area, will be that log messages will be a hour out during
	  the period where the old rules do not match the new rules.
	</para>
	<para>
	  For most OS's this change just means that you need to
	  update the conversion rules from UTC to local time.
	  Normally this involves updating a file in /etc (which
	  sets the default timezone for the machine) and possibly
	  a directory which has all the conversion rules for the
	  world (e.g. /usr/share/zoneinfo).  When updating the OS
	  do not forget to update any chroot areas as well.
	  See your OS's documentation for more details.
	</para>
	<para>
	  The local timezone conversion rules can also be done on
	  a individual basis by setting the TZ environment variable
	  appropriately.  See your OS's documentation for more
	  details.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Why do we get the following warning at run time:
<programlisting>kernel: process `named' is using obsolete setsockopt SO_BSDCOMPAT</programlisting>
	</para>
      </question>
      <answer>
	<para>
          The early Linux kernels broke sendto() by having it return     
          that a ICMP unreachable had be received for non connected
          UDP sockets.  This made non connected UDP sockets work like
          connected UDP socket which is fine when you are only talking
          to one destination.  Named however talks to multiple
          destinations and it caused problems.
	</para>
	<para>
          Rather than fix sendto() to just have BSD behaviour they added
          SO_BSDCOMPAT to turn BSD behaviour on/off on a per socket basis.
	</para>
	<para>
          Later they decided to make BSD behaviour the default and
          to aggressively track down applications that used SO_BSDCOMPAT
          by issuing a warning.  This is the sort of things vendors
          do in alpha/beta stages of a release so that their code is
          clean.  They then turn the warning *off* for release code.
	</para>
	<para>
          We still have customers that have kernels that require
          SO_BSDCOMPAT to operate.  We therefore cannot remove the
          setsockopt(SO_BSDCOMPAT) call.
	</para>
	<para>
          Now most/all portable applications that use SO_BSDCOMPAT use it
          conditionally manner so just removing SO_BSDCOMPAT from the
          header file would be safe as long as the binary was not to
          be moved between systems.  BIND's use is conditional.
	</para>
	<para>
	  In short, the Linux developers should either, remove the #define for
	  SO_BSDCOMPAT, and/or remove the warning.
	</para>
      </answer>
    </qandaentry>

    <qandaentry>
      <question>
	<para>
	  Isn't "make install"  supposed to generate a default named.conf?
	</para>
      </question>
      <answer>
	<para>
	  Short Answer: No. 
	</para>
	<para>
	  Long Answer: There really isn't a default configuration which fits
	  any site perfectly.  There are lots of decisions that need to
	  be made and there is no consensus on what the defaults should be.
	  For example FreeBSD uses /etc/namedb as the location where the
	  configuration files for named are stored.  Others use /var/named.
	</para>
	<para>
	  What addresses to listen on?  For a laptop on the move a lot
	  you may only want to listen on the loop back interfaces.
	</para>
	<para>
	  Who do you offer recursive service to?  Is there are firewall
	  to consider?  If so is it stateless or stateful.  Are you
	  directly on the Internet?  Are you on a private network? Are
	  you on a NAT'd network? The answers
	  to all these questions change how you configure even a
	  caching name server.
	</para>
      </answer>
    </qandaentry>

  </qandaset>
</article>