<?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" []>

<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>