<a name="relnotes_download"></a>Download</h3></div></div></div>
<p>
The latest versions of BIND 9 software can always be found at
- <a class="link" href="http://www.isc.org/downloads/" target="_top">http://www.isc.org/downloads/</a>.
+ <a class="link" href="https://www.isc.org/download/" target="_top">https://www.isc.org/download/</a>.
There you will find additional information about each release,
source code, and pre-compiled versions for Microsoft Windows
operating systems.
license may contact ISC at <a class="link" href="https://www.isc.org/mission/contact/" target="_top">
https://www.isc.org/mission/contact/</a>.
</p>
+</div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.13"></a>Notes for BIND 9.11.13</h3></div></div></div>
+
+ <p>
+ None.
+ </p>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.12"></a>Notes for BIND 9.11.12</h3></div></div></div>
+
+ <p>
+ None.
+ </p>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.11"></a>Notes for BIND 9.11.11</h3></div></div></div>
+
+ <p>
+ None.
+ </p>
+
</div>
<div class="section">
<div class="titlepage"><div><div><h3 class="title">
-<a name="relnotes_security"></a>Security Fixes</h3></div></div></div>
- <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
- <p>
- A race condition could trigger an assertion failure when
- a large number of incoming packets were being rejected.
- This flaw is disclosed in CVE-2019-6471. [GL #942]
- </p>
- </li></ul></div>
-</div>
- <div class="section">
-<div class="titlepage"><div><div><h3 class="title">
-<a name="relnotes_features"></a>New Features</h3></div></div></div>
- <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
-<li class="listitem">
- <p>
- The new GeoIP2 API from MaxMind is now supported when BIND
- is compiled using <span class="command"><strong>configure --with-geoip2</strong></span>.
- The legacy GeoIP API can be used by compiling with
- <span class="command"><strong>configure --with-geoip</strong></span> instead. (Note that
- the databases for the legacy API are no longer maintained by
- MaxMind.)
- </p>
- <p>
- The default path to the GeoIP2 databases will be set based
- on the location of the <span class="command"><strong>libmaxminddb</strong></span> library;
- for example, if it is in <code class="filename">/usr/local/lib</code>,
- then the default path will be
- <code class="filename">/usr/local/share/GeoIP</code>.
- This value can be overridden in <code class="filename">named.conf</code>
- using the <span class="command"><strong>geoip-directory</strong></span> option.
- </p>
- <p>
- Some <span class="command"><strong>geoip</strong></span> ACL settings that were available with
- legacy GeoIP, including searches for <span class="command"><strong>netspeed</strong></span>,
- <span class="command"><strong>org</strong></span>, and three-letter ISO country codes, will
- no longer work when using GeoIP2. Supported GeoIP2 database
- types are <span class="command"><strong>country</strong></span>, <span class="command"><strong>city</strong></span>,
- <span class="command"><strong>domain</strong></span>, <span class="command"><strong>isp</strong></span>, and
- <span class="command"><strong>as</strong></span>. All of the databases support both IPv4
- and IPv6 lookups. [GL #182]
- </p>
- </li>
-<li class="listitem">
- <p>
- A SipHash 2-4 based DNS Cookie (RFC 7873) algorithm has been added.
- [GL #605]
- </p>
- <p>
- If you are running multiple DNS Servers (different versions of BIND 9
- or DNS server from multiple vendors) responding from the same IP
- address (anycast or load-balancing scenarios), you'll have to make
- sure that all the servers are configured with the same DNS Cookie
- algorithm and same Server Secret for the best performance.
- </p>
- </li>
-<li class="listitem">
- <p>
- DS records included in DNS referral messages can now be validated
- and cached immediately, reducing the number of queries needed for
- a DNSSEC validation. [GL #964]
- </p>
- </li>
-</ul></div>
-</div>
- <div class="section">
-<div class="titlepage"><div><div><h3 class="title">
-<a name="relnotes_bugs"></a>Bug Fixes</h3></div></div></div>
- <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
-<li class="listitem">
- <p>
- Glue address records were not being returned in responses
- to root priming queries; this has been corrected. [GL #1092]
- </p>
- </li>
-<li class="listitem">
- <p>
- Interaction between DNS64 and RPZ No Data rule (CNAME *.) could
- cause unexpected results; this has been fixed. [GL #1106]
- </p>
- </li>
-<li class="listitem">
- <p>
- <span class="command"><strong>named-checkconf</strong></span> now checks DNS64 prefixes
- to ensure bits 64-71 are zero. [GL #1159]
- </p>
- </li>
-<li class="listitem">
- <p>
- <span class="command"><strong>named-checkconf</strong></span> could crash during
- configuration if configured to use "geoip continent" ACLs with
- legacy GeoIP. [GL #1163]
- </p>
- </li>
-<li class="listitem">
- <p>
- <span class="command"><strong>named-checkconf</strong></span> now correctly reports a missing
- <span class="command"><strong>dnstap-output</strong></span> option when
- <span class="command"><strong>dnstap</strong></span> is set. [GL #1136]
- </p>
- </li>
-<li class="listitem">
- <p>
- Handle ETIMEDOUT error on connect() with a non-blocking
- socket. [GL #1133]
- </p>
- </li>
+<a name="relnotes-9.11.10"></a>Notes for BIND 9.11.10</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.10-features"></a>New Features</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ A SipHash 2-4 based DNS Cookie (RFC 7873) algorithm has been added.
+ [GL #605]
+ </p>
+ <p>
+ If you are running multiple DNS Servers (different versions of BIND 9
+ or DNS server from multiple vendors) responding from the same IP
+ address (anycast or load-balancing scenarios), you'll have to make
+ sure that all the servers are configured with the same DNS Cookie
+ algorithm and same Server Secret for the best performance.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ DS records included in DNS referral messages can now be validated
+ and cached immediately, reducing the number of queries needed for
+ a DNSSEC validation. [GL #964]
+ </p>
+ </li>
</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.10-bugs"></a>Bug Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ Interaction between DNS64 and RPZ No Data rule (CNAME *.) could
+ cause unexpected results; this has been fixed. [GL #1106]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named-checkconf</strong></span> now checks DNS64 prefixes
+ to ensure bits 64-71 are zero. [GL #1159]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named-checkconf</strong></span> could crash during
+ configuration if configured to use "geoip continent" ACLs with
+ legacy GeoIP. [GL #1163]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named-checkconf</strong></span> now correctly reports a missing
+ <span class="command"><strong>dnstap-output</strong></span> option when
+ <span class="command"><strong>dnstap</strong></span> is set. [GL #1136]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Handle ETIMEDOUT error on connect() with a non-blocking
+ socket. [GL #1133]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
</div>
<div class="section">
<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.9"></a>Notes for BIND 9.11.9</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.9-features"></a>New Features</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ The new GeoIP2 API from MaxMind is now supported when BIND
+ is compiled using <span class="command"><strong>configure --with-geoip2</strong></span>.
+ The legacy GeoIP API can be used by compiling with
+ <span class="command"><strong>configure --with-geoip</strong></span> instead. (Note that
+ the databases for the legacy API are no longer maintained by
+ MaxMind.)
+ </p>
+ <p>
+ The default path to the GeoIP2 databases will be set based
+ on the location of the <span class="command"><strong>libmaxminddb</strong></span> library;
+ for example, if it is in <code class="filename">/usr/local/lib</code>,
+ then the default path will be
+ <code class="filename">/usr/local/share/GeoIP</code>.
+ This value can be overridden in <code class="filename">named.conf</code>
+ using the <span class="command"><strong>geoip-directory</strong></span> option.
+ </p>
+ <p>
+ Some <span class="command"><strong>geoip</strong></span> ACL settings that were available with
+ legacy GeoIP, including searches for <span class="command"><strong>netspeed</strong></span>,
+ <span class="command"><strong>org</strong></span>, and three-letter ISO country codes, will
+ no longer work when using GeoIP2. Supported GeoIP2 database
+ types are <span class="command"><strong>country</strong></span>, <span class="command"><strong>city</strong></span>,
+ <span class="command"><strong>domain</strong></span>, <span class="command"><strong>isp</strong></span>, and
+ <span class="command"><strong>as</strong></span>. All of the databases support both IPv4
+ and IPv6 lookups. [GL #182]
+ </p>
+ </li></ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.9-bugs"></a>Bug Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ Glue address records were not being returned in responses
+ to root priming queries; this has been corrected. [GL #1092]
+ </p>
+ </li></ul></div>
+ </div>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.8"></a>Notes for BIND 9.11.8</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.8-security"></a>Security Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ A race condition could trigger an assertion failure when
+ a large number of incoming packets were being rejected.
+ This flaw is disclosed in CVE-2019-6471. [GL #942]
+ </p>
+ </li></ul></div>
+ </div>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.7"></a>Notes for BIND 9.11.7</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.7-security"></a>Security Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ The TCP client quota set using the <span class="command"><strong>tcp-clients</strong></span>
+ option could be exceeded in some cases. This could lead to
+ exhaustion of file descriptors. This flaw is disclosed in
+ CVE-2018-5743. [GL #615]
+ </p>
+ </li></ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.7-changes"></a>Feature Changes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ When <span class="command"><strong>trusted-keys</strong></span> and
+ <span class="command"><strong>managed-keys</strong></span> are both configured for the
+ same name, or when <span class="command"><strong>trusted-keys</strong></span> is used to
+ configure a trust anchor for the root zone and
+ <span class="command"><strong>dnssec-validation</strong></span> is set to
+ <code class="literal">auto</code>, automatic RFC 5011 key
+ rollovers will fail.
+ </p>
+ <p>
+ This combination of settings was never intended to work,
+ but there was no check for it in the parser. This has been
+ corrected; a warning is now logged. (In BIND 9.15 and
+ higher this error will be fatal.) [GL #868]
+ </p>
+ </li></ul></div>
+ </div>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.6"></a>Notes for BIND 9.11.6</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.6-security"></a>Security Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ Code change #4964, intended to prevent double signatures
+ when deleting an inactive zone DNSKEY in some situations,
+ introduced a new problem during zone processing in which
+ some delegation glue RRsets are incorrectly identified
+ as needing RRSIGs, which are then created for them using
+ the current active ZSK for the zone. In some, but not all
+ cases, the newly-signed RRsets are added to the zone's
+ NSEC/NSEC3 chain, but incompletely -- this can result in
+ a broken chain, affecting validation of proof of nonexistence
+ for records in the zone. [GL #771]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> could crash if it managed a DNSSEC
+ security root with <span class="command"><strong>managed-keys</strong></span> and the
+ authoritative zone rolled the key to an algorithm not supported
+ by BIND 9. This flaw is disclosed in CVE-2018-5745. [GL #780]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> leaked memory when processing a
+ request with multiple Key Tag EDNS options present. ISC
+ would like to thank Toshifumi Sakaguchi for bringing this
+ to our attention. This flaw is disclosed in CVE-2018-5744.
+ [GL #772]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Zone transfer controls for writable DLZ zones were not
+ effective as the <span class="command"><strong>allowzonexfr</strong></span> method was
+ not being called for such zones. This flaw is disclosed in
+ CVE-2019-6465. [GL #790]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.6-changes"></a>Feature Changes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ When compiled with IDN support, the <span class="command"><strong>dig</strong></span> and the
+ <span class="command"><strong>nslookup</strong></span> commands now disable IDN processing when
+ the standard output is not a tty (e.g. not used by human). The command
+ line options +idnin and +idnout need to be used to enable IDN
+ processing when <span class="command"><strong>dig</strong></span> or <span class="command"><strong>nslookup</strong></span>
+ is used from the shell scripts.
+ </p>
+ </li></ul></div>
+ </div>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.5"></a>Notes for BIND 9.11.5</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.5-security"></a>Security Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> could crash during recursive processing
+ of DNAME records when <span class="command"><strong>deny-answer-aliases</strong></span> was
+ in use. This flaw is disclosed in CVE-2018-5740. [GL #387]
+ </p>
+ </li></ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.5-features"></a>New Features</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ Two new update policy rule types have been added
+ <span class="command"><strong>krb5-selfsub</strong></span> and <span class="command"><strong>ms-selfsub</strong></span>
+ which allow machines with Kerberos principals to update
+ the name space at or below the machine names identified
+ in the respective principals.
+ </p>
+ </li></ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.5-changes"></a>Feature Changes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ The <span class="command"><strong>rndc nta</strong></span> command could not differentiate
+ between views of the same name but different class; this
+ has been corrected with the addition of a <span class="command"><strong>-class</strong></span>
+ option. [GL #105]
+ </p>
+ </li></ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.5-bugs"></a>Bug Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ When a negative trust anchor was added to multiple views
+ using <span class="command"><strong>rndc nta</strong></span>, the text returned via
+ <span class="command"><strong>rndc</strong></span> was incorrectly truncated after the
+ first line, making it appear that only one NTA had been
+ added. This has been fixed. [GL #105]
+ </p>
+ </li></ul></div>
+ </div>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.4"></a>Notes for BIND 9.11.4</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.4-security"></a>Security Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ When recursion is enabled but the <span class="command"><strong>allow-recursion</strong></span>
+ and <span class="command"><strong>allow-query-cache</strong></span> ACLs are not specified, they
+ should be limited to local networks, but they were inadvertently set
+ to match the default <span class="command"><strong>allow-query</strong></span>, thus allowing
+ remote queries. This flaw is disclosed in CVE-2018-5738. [GL #309]
+ </p>
+ </li></ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.4-features"></a>New Features</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> now supports the "root key sentinel"
+ mechanism. This enables validating resolvers to indicate
+ which trust anchors are configured for the root, so that
+ information about root key rollover status can be gathered.
+ To disable this feature, add
+ <span class="command"><strong>root-key-sentinel no;</strong></span> to
+ <code class="filename">named.conf</code>.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Added the ability not to return a DNS COOKIE option when one
+ is present in the request. To prevent a cookie being returned,
+ add <span class="command"><strong>answer-cookie no;</strong></span> to
+ <code class="filename">named.conf</code>. [GL #173]
+ </p>
+ <p>
+ <span class="command"><strong>answer-cookie no</strong></span> is only intended as a
+ temporary measure, for use when <span class="command"><strong>named</strong></span>
+ shares an IP address with other servers that do not yet
+ support DNS COOKIE. A mismatch between servers on the
+ same address is not expected to cause operational problems,
+ but the option to disable COOKIE responses so that all
+ servers have the same behavior is provided out of an
+ abundance of caution. DNS COOKIE is an important security
+ mechanism, and should not be disabled unless absolutely
+ necessary.
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.4-removed"></a>Removed Features</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> will now log a warning if the old
+ BIND now can be compiled against libidn2 library to add
+ IDNA2008 support. Previously BIND only supported IDNA2003
+ using (now obsolete) idnkit-1 library.
+ </p>
+ </li></ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.4-changes"></a>Feature Changes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dig +noidnin</strong></span> can be used to disable IDN
+ processing on the input domain name, when BIND is compiled
+ with IDN support.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Multiple <span class="command"><strong>cookie-secret</strong></span> clause are now
+ supported. The first <span class="command"><strong>cookie-secret</strong></span> in
+ <code class="filename">named.conf</code> is used to generate new
+ server cookies. Any others are used to accept old server
+ cookies or those generated by other servers using the
+ matching <span class="command"><strong>cookie-secret</strong></span>.
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.4-bugs"></a>Bug Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> now rejects excessively large
+ incremental (IXFR) zone transfers in order to prevent
+ possible corruption of journal files which could cause
+ <span class="command"><strong>named</strong></span> to abort when loading zones. [GL #339]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>rndc reload</strong></span> could cause <span class="command"><strong>named</strong></span>
+ to leak memory if it was invoked before the zone loading actions
+ from a previous <span class="command"><strong>rndc reload</strong></span> command were
+ completed. [RT #47076]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.3"></a>Notes for BIND 9.11.3</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.3-security"></a>Security Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ Addresses could be referenced after being freed during resolver
+ processing, causing an assertion failure. The chances of this
+ happening were remote, but the introduction of a delay in
+ resolution increased them. This bug is disclosed in
+ CVE-2017-3145. [RT #46839]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ update-policy rules that otherwise ignore the name field now
+ require that it be set to "." to ensure that any type list
+ present is properly interpreted. If the name field was omitted
+ from the rule declaration and a type list was present it wouldn't
+ be interpreted as expected.
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.3-removed"></a>Removed Features</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ The ISC DNSSEC Lookaside Validation (DLV) service has
+ been shut down; all DLV records in the dlv.isc.org zone
+ have been removed. References to the service have been
+ removed from BIND documentation. Lookaside validation
+ is no longer used by default by <span class="command"><strong>delv</strong></span>.
+ The DLV key has been removed from <code class="filename">bind.keys</code>.
+ Setting <span class="command"><strong>dnssec-lookaside</strong></span> to
+ <span class="command"><strong>auto</strong></span> or to use dlv.isc.org as a trust
+ anchor results in a warning being issued.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> will now log a warning if the old
+ root DNSSEC key is explicitly configured and has not been updated.
+ [RT #43670]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="proto_changes"></a>Protocol Changes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ BIND can now use the Ed25519 and Ed448 Edwards Curve DNSSEC
+ signing algorithms described in RFC 8080. Note, however, that
+ these algorithms must be supported in OpenSSL;
+ currently they are only available in the development branch
+ of OpenSSL at
+ <a class="link" href="https://github.com/openssl/openssl" target="_top">
+ https://github.com/openssl/openssl</a>.
+ [RT #44696]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ When parsing DNS messages, EDNS KEY TAG options are checked
+ for correctness. When printing messages (for example, in
+ <span class="command"><strong>dig</strong></span>), EDNS KEY TAG options are printed
+ in readable format.
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.3-changes"></a>Feature Changes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> will no longer start or accept
+ reconfiguration if <span class="command"><strong>managed-keys</strong></span> or
+ <span class="command"><strong>dnssec-validation auto</strong></span> are in use and
+ the managed-keys directory (specified by
+ <span class="command"><strong>managed-keys-directory</strong></span>, and defaulting
+ to the working directory if not specified),
+ is not writable by the effective user ID. [RT #46077]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Previously, <span class="command"><strong>update-policy local;</strong></span> accepted
+ updates from any source so long as they were signed by the
+ locally-generated session key. This has been further restricted;
+ updates are now only accepted from locally configured addresses.
+ [RT #45492]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.3-bugs"></a>Bug Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ Attempting to validate improperly unsigned CNAME responses
+ from secure zones could cause a validator loop. This caused
+ a delay in returning SERVFAIL and also increased the chances
+ of encountering the crash bug described in CVE-2017-3145.
+ [RT #46839]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ When <span class="command"><strong>named</strong></span> was reconfigured, failure of some
+ zones to load correctly could leave the system in an inconsistent
+ state; while generally harmless, this could lead to a crash later
+ when using <span class="command"><strong>rndc addzone</strong></span>. Reconfiguration changes
+ are now fully rolled back in the event of failure. [RT #45841]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Some header files included <isc/util.h> incorrectly as
+ it pollutes with namespace with non ISC_ macros and this should
+ only be done by explicitly including <isc/util.h>. This
+ has been corrected. Some code may depend on <isc/util.h>
+ being implicitly included via other header files. Such
+ code should explicitly include <isc/util.h>.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Zones created with <span class="command"><strong>rndc addzone</strong></span> could
+ temporarily fail to inherit the <span class="command"><strong>allow-transfer</strong></span>
+ ACL set in the <span class="command"><strong>options</strong></span> section of
+ <code class="filename">named.conf</code>. [RT #46603]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> failed to properly determine whether
+ there were active KSK and ZSK keys for an algorithm when
+ <span class="command"><strong>update-check-ksk</strong></span> was true (which is the
+ default setting). This could leave records unsigned
+ when rolling keys. [RT #46743] [RT #46754] [RT #46774]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.2"></a>Notes for BIND 9.11.2</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.2-security"></a>Security Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ An error in TSIG handling could permit unauthorized zone
+ transfers or zone updates. These flaws are disclosed in
+ CVE-2017-3142 and CVE-2017-3143. [RT #45383]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The BIND installer on Windows used an unquoted service path,
+ which can enable privilege escalation. This flaw is disclosed
+ in CVE-2017-3141. [RT #45229]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ With certain RPZ configurations, a response with TTL 0
+ could cause <span class="command"><strong>named</strong></span> to go into an infinite
+ query loop. This flaw is disclosed in CVE-2017-3140.
+ [RT #45181]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.2-changes"></a>Feature Changes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dig +ednsopt</strong></span> now accepts the names
+ for EDNS options in addition to numeric values. For example,
+ an EDNS Client-Subnet option could be sent using
+ <span class="command"><strong>dig +ednsopt=ecs:...</strong></span>. Thanks to
+ John Worley of Secure64 for the contribution. [RT #44461]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Threads in <span class="command"><strong>named</strong></span> are now set to human-readable
+ names to assist debugging on operating systems that support that.
+ Threads will have names such as "isc-timer", "isc-sockmgr",
+ "isc-worker0001", and so on. This will affect the reporting of
+ subsidiary thread names in <span class="command"><strong>ps</strong></span> and
+ <span class="command"><strong>top</strong></span>, but not the main thread. [RT #43234]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ DiG now warns about .local queries which are reserved for
+ Multicast DNS. [RT #44783]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.2-bugs"></a>Bug Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ Fixed a bug that was introduced in an earlier development
+ release which caused multi-packet AXFR and IXFR messages to fail
+ validation if not all packets contained TSIG records; this
+ caused interoperability problems with some other DNS
+ implementations. [RT #45509]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Reloading or reconfiguring <span class="command"><strong>named</strong></span> could
+ fail on some platforms when LMDB was in use. [RT #45203]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Due to some incorrectly deleted code, when BIND was
+ built with LMDB, zones that were deleted via
+ <span class="command"><strong>rndc delzone</strong></span> were removed from the
+ running server but were not removed from the new zone
+ database, so that deletion did not persist after a
+ server restart. This has been corrected. [RT #45185]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Semicolons are no longer escaped when printing CAA and
+ URI records. This may break applications that depend on the
+ presence of the backslash before the semicolon. [RT #45216]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ AD could be set on truncated answer with no records present
+ in the answer and authority sections. [RT #45140]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.1"></a>Notes for BIND 9.11.1</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.1-security"></a>Security Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ <span class="command"><strong>rndc ""</strong></span> could trigger an assertion failure
+ in <span class="command"><strong>named</strong></span>. This flaw is disclosed in
+ (CVE-2017-3138). [RT #44924]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Some chaining (i.e., type CNAME or DNAME) responses to upstream
+ queries could trigger assertion failures. This flaw is disclosed
+ in CVE-2017-3137. [RT #44734]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dns64</strong></span> with <span class="command"><strong>break-dnssec yes;</strong></span>
+ can result in an assertion failure. This flaw is disclosed in
+ CVE-2017-3136. [RT #44653]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ If a server is configured with a response policy zone (RPZ)
+ that rewrites an answer with local data, and is also configured
+ for DNS64 address mapping, a NULL pointer can be read
+ triggering a server crash. This flaw is disclosed in
+ CVE-2017-3135. [RT #44434]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ A coding error in the <code class="option">nxdomain-redirect</code>
+ feature could lead to an assertion failure if the redirection
+ namespace was served from a local authoritative data source
+ such as a local zone or a DLZ instead of via recursive
+ lookup. This flaw is disclosed in CVE-2016-9778. [RT #43837]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> could mishandle authority sections
+ with missing RRSIGs, triggering an assertion failure. This
+ flaw is disclosed in CVE-2016-9444. [RT #43632]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> mishandled some responses where
+ covering RRSIG records were returned without the requested
+ data, resulting in an assertion failure. This flaw is
+ disclosed in CVE-2016-9147. [RT #43548]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> incorrectly tried to cache TKEY
+ records which could trigger an assertion failure when there was
+ a class mismatch. This flaw is disclosed in CVE-2016-9131.
+ [RT #43522]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ It was possible to trigger assertions when processing
+ responses containing answers of type DNAME. This flaw is
+ disclosed in CVE-2016-8864. [RT #43465]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Added the ability to specify the maximum number of records
+ permitted in a zone (<code class="option">max-records #;</code>).
+ This provides a mechanism to block overly large zone
+ transfers, which is a potential risk with slave zones from
+ other parties, as described in CVE-2016-6170.
+ [RT #42143]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.1-changes"></a>Feature Changes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dnstap</strong></span> now stores both the local and remote
+ addresses for all messages, instead of only the remote address.
+ The default output format for <span class="command"><strong>dnstap-read</strong></span> has
+ been updated to include these addresses, with the initiating
+ address first and the responding address second, separated by
+ "-%gt;" or "%lt;-" to indicate in which direction the message
+ was sent. [RT #43595]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Expanded and improved the YAML output from
+ <span class="command"><strong>dnstap-read -y</strong></span>: it now includes packet
+ size and a detailed breakdown of message contents.
+ [RT #43622] [RT #43642]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ If an ACL is specified with an address prefix in which the
+ prefix length is longer than the address portion (for example,
+ 192.0.2.1/8), <span class="command"><strong>named</strong></span> will now log a warning.
+ In future releases this will be a fatal configuration error.
+ [RT #43367]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.1-bugs"></a>Bug Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ A synthesized CNAME record appearing in a response before the
+ associated DNAME could be cached, when it should not have been.
+ This was a regression introduced while addressing CVE-2016-8864.
+ [RT #44318]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> could deadlock if multiple changes
+ to NSEC/NSEC3 parameters for the same zone were being processed
+ at the same time. [RT #42770]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> could trigger an assertion when
+ sending NOTIFY messages. [RT #44019]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Referencing a nonexistent zone in a <span class="command"><strong>response-policy</strong></span>
+ statement could cause an assertion failure during configuration.
+ [RT #43787]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>rndc addzone</strong></span> could cause a crash
+ when attempting to add a zone with a type other than
+ <span class="command"><strong>master</strong></span> or <span class="command"><strong>slave</strong></span>.
+ Such zones are now rejected. [RT #43665]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> could hang when encountering log
+ file names with large apparent gaps in version number (for
+ example, when files exist called "logfile.0", "logfile.1",
+ and "logfile.1482954169"). This is now handled correctly.
+ [RT #38688]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ If a zone was updated while <span class="command"><strong>named</strong></span> was
+ processing a query for nonexistent data, it could return
+ out-of-sync NSEC3 records causing potential DNSSEC validation
+ failure. [RT #43247]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.1-maint"></a>Maintenance</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ The built-in root hints have been updated to include an
+ IPv6 address (2001:500:12::d0d) for G.ROOT-SERVERS.NET.
+ </p>
+ </li></ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.1-misc"></a>Miscellaneous Notes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; "><li class="listitem">
+ <p>
+ Authoritative server support for the EDNS Client Subnet option
+ (ECS), introduced in BIND 9.11.0, was based on an early version
+ of the specification, and is now known to have incompatibilities
+ with other ECS implementations. It is also inefficient, requiring
+ a separate view for each answer, and is unable to correct for
+ overlapping subnets in the configuration. It is intended for
+ testing purposes but is not recommended for for production use.
+ This was not made sufficiently clear in the documentation at
+ the time of release.
+ </p>
+ </li></ul></div>
+ </div>
+
+</div>
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
+<a name="relnotes-9.11.0"></a>Notes for BIND 9.11.0</h3></div></div></div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.0-security"></a>Security Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ It was possible to trigger a assertion when rendering a
+ message using a specially crafted request. This flaw is
+ disclosed in CVE-2016-2776. [RT #43139]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ getrrsetbyname with a non absolute name could trigger an
+ infinite recursion bug in lwresd and named with lwres
+ configured if when combined with a search list entry the
+ resulting name is too long. This flaw is disclosed in
+ CVE-2016-2775. [RT #42694]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.0-features"></a>New Features</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ A new method of provisioning secondary servers called
+ "Catalog Zones" has been added. This is an implementation of
+ <a class="link" href="https://datatracker.ietf.org/doc/draft-muks-dnsop-dns-catalog-zones/" target="_top">
+ draft-muks-dnsop-dns-catalog-zones/
+ </a>.
+ </p>
+ <p>
+ A catalog zone is a regular DNS zone which contains a list
+ of "member zones", along with the configuration options for
+ each of those zones. When a server is configured to use a
+ catalog zone, all the zones listed in the catalog zone are
+ added to the local server as slave zones. When the catalog
+ zone is updated (e.g., by adding or removing zones, or
+ changing configuration options for existing zones) those
+ changes will be put into effect. Since the catalog zone is
+ itself a DNS zone, this means configuration changes can be
+ propagated to slaves using the standard AXFR/IXFR update
+ mechanism.
+ </p>
+ <p>
+ This feature should be considered experimental. It currently
+ supports only basic features; more advanced features such as
+ ACLs and TSIG keys are not yet supported. Example catalog
+ zone configurations can be found in the Chapter 9 of the
+ BIND Administrator Reference Manual.
+ </p>
+ <p>
+ Support for master entries with TSIG keys has been added to catalog
+ zones, as well as support for allow-query and allow-transfer.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Added an <span class="command"><strong>isc.rndc</strong></span> Python module, which allows
+ <span class="command"><strong>rndc</strong></span> commands to be sent from Python programs.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Added support for DynDB, a new interface for loading zone data
+ from an external database, developed by Red Hat for the FreeIPA
+ project. (Thanks in particular to Adam Tkac and Petr
+ Spacek of Red Hat for the contribution.)
+ </p>
+ <p>
+ Unlike the existing DLZ and SDB interfaces, which provide a
+ limited subset of database functionality within BIND -
+ translating DNS queries into real-time database lookups with
+ relatively poor performance and with no ability to handle
+ DNSSEC-signed data - DynDB is able to fully implement
+ and extend the database API used natively by BIND.
+ </p>
+ <p>
+ A DynDB module could pre-load data from an external data
+ source, then serve it with the same performance and
+ functionality as conventional BIND zones, and with the
+ ability to take advantage of database features not
+ available in BIND, such as multi-master replication.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Fetch quotas are now compiled in by default: they
+ no longer require BIND to be configured with
+ <span class="command"><strong>--enable-fetchlimit</strong></span>, as was the case
+ when the feature was introduced in BIND 9.10.3.
+ </p>
+ <p>
+ These quotas limit the queries that are sent by recursive
+ resolvers to authoritative servers experiencing denial-of-service
+ attacks. They can both reduce the harm done to authoritative
+ servers and also avoid the resource exhaustion that can be
+ experienced by recursive servers when they are being used as a
+ vehicle for such an attack.
+ </p>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: circle; ">
+<li class="listitem">
+ <p>
+ <code class="option">fetches-per-server</code> limits the number of
+ simultaneous queries that can be sent to any single
+ authoritative server. The configured value is a starting
+ point; it is automatically adjusted downward if the server is
+ partially or completely non-responsive. The algorithm used to
+ adjust the quota can be configured via the
+ <code class="option">fetch-quota-params</code> option.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <code class="option">fetches-per-zone</code> limits the number of
+ simultaneous queries that can be sent for names within a
+ single domain. (Note: Unlike "fetches-per-server", this
+ value is not self-tuning.)
+ </p>
+ </li>
+</ul></div>
+ <p>
+ Statistics counters have also been added to track the number
+ of queries affected by these quotas.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Added support for <span class="command"><strong>dnstap</strong></span>, a fast,
+ flexible method for capturing and logging DNS traffic,
+ developed by Robert Edmonds at Farsight Security, Inc.,
+ whose assistance is gratefully acknowledged.
+ </p>
+ <p>
+ To enable <span class="command"><strong>dnstap</strong></span> at compile time,
+ the <span class="command"><strong>fstrm</strong></span> and <span class="command"><strong>protobuf-c</strong></span>
+ libraries must be available, and BIND must be configured with
+ <code class="option">--enable-dnstap</code>.
+ </p>
+ <p>
+ A new utility <span class="command"><strong>dnstap-read</strong></span> has been added
+ to allow <span class="command"><strong>dnstap</strong></span> data to be presented in
+ a human-readable format.
+ </p>
+ <p>
+ <span class="command"><strong>rndc dnstap -roll</strong></span> causes <span class="command"><strong>dnstap</strong></span>
+ output files to be rolled like log files -- the most recent output
+ file is renamed with a <code class="filename">.0</code> suffix, the next
+ most recent with <code class="filename">.1</code>, etc. (Note that this
+ only works when <span class="command"><strong>dnstap</strong></span> output is being written
+ to a file, not to a UNIX domain socket.) An optional numerical
+ argument specifies how many backup log files to retain; if not
+ specified or set to 0, there is no limit.
+ </p>
+ <p>
+ <span class="command"><strong>rndc dnstap -reopen</strong></span> simply closes and reopens
+ the <span class="command"><strong>dnstap</strong></span> output channel without renaming
+ the output file.
+ </p>
+ <p>
+ For more information on <span class="command"><strong>dnstap</strong></span>, see
+ <a class="link" href="https://dnstap.info" target="_top">https://dnstap.info</a>.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ New statistics counters have been added to track traffic
+ sizes, as specified in RSSAC002. Query and response
+ message sizes are broken up into ranges of histogram buckets:
+ TCP and UDP queries of size 0-15, 16-31, ..., 272-288, and 288+,
+ and TCP and UDP responses of size 0-15, 16-31, ..., 4080-4095,
+ and 4096+. These values can be accessed via the XML and JSON
+ statistics channels at, for example,
+ <a class="link" href="http://localhost:8888/xml/v3/traffic" target="_top">http://localhost:8888/xml/v3/traffic</a>
+ or
+ <a class="link" href="http://localhost:8888/json/v1/traffic" target="_top">http://localhost:8888/json/v1/traffic</a>.
+ </p>
+ <p>
+ Statistics for RSSAC02v3 traffic-volume, traffic-sizes and
+ rcode-volume reporting are now collected.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ A new DNSSEC key management utility,
+ <span class="command"><strong>dnssec-keymgr</strong></span>, has been added. This tool
+ is meant to run unattended (e.g., under <span class="command"><strong>cron</strong></span>).
+ It reads a policy definition file
+ (default <code class="filename">/etc/dnssec-policy.conf</code>)
+ and creates or updates DNSSEC keys as necessary to ensure that a
+ zone's keys match the defined policy for that zone. New keys are
+ created whenever necessary to ensure rollovers occur correctly.
+ Existing keys' timing metadata is adjusted as needed to set the
+ correct rollover period, prepublication interval, etc. If
+ the configured policy changes, keys are corrected automatically.
+ See the <span class="command"><strong>dnssec-keymgr</strong></span> man page for full details.
+ </p>
+ <p>
+ Note: <span class="command"><strong>dnssec-keymgr</strong></span> depends on Python and on
+ the Python lex/yacc module, PLY. The other Python-based tools,
+ <span class="command"><strong>dnssec-coverage</strong></span> and
+ <span class="command"><strong>dnssec-checkds</strong></span>, have been
+ refactored and updated as part of this work.
+ </p>
+ <p>
+ <span class="command"><strong>dnssec-keymgr</strong></span> now takes a -r
+ <em class="replaceable"><code>randomfile</code></em> option.
+ </p>
+ <p>
+ (Many thanks to Sebastián
+ Castro for his assistance in developing this tool at the IETF
+ 95 Hackathon in Buenos Aires, April 2016.)
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The serial number of a dynamically updatable zone can
+ now be set using
+ <span class="command"><strong>rndc signing -serial <em class="replaceable"><code>number</code></em> <em class="replaceable"><code>zonename</code></em></strong></span>.
+ This is particularly useful with <code class="option">inline-signing</code>
+ zones that have been reset. Setting the serial number to a value
+ larger than that on the slaves will trigger an AXFR-style
+ transfer.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ When answering recursive queries, SERVFAIL responses can now be
+ cached by the server for a limited time; subsequent queries for
+ the same query name and type will return another SERVFAIL until
+ the cache times out. This reduces the frequency of retries
+ when a query is persistently failing, which can be a burden
+ on recursive servers. The SERVFAIL cache timeout is controlled
+ by <code class="option">servfail-ttl</code>, which defaults to 1 second
+ and has an upper limit of 30.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The new <span class="command"><strong>rndc nta</strong></span> command can now be used to
+ set a "negative trust anchor" (NTA), disabling DNSSEC validation for
+ a specific domain; this can be used when responses from a domain
+ are known to be failing validation due to administrative error
+ rather than because of a spoofing attack. NTAs are strictly
+ temporary; by default they expire after one hour, but can be
+ configured to last up to one week. The default NTA lifetime
+ can be changed by setting the <code class="option">nta-lifetime</code> in
+ <code class="filename">named.conf</code>. When added, NTAs are stored in a
+ file (<code class="filename"><em class="replaceable"><code>viewname</code></em>.nta</code>)
+ in order to persist across restarts of the <span class="command"><strong>named</strong></span> server.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The EDNS Client Subnet (ECS) option is now supported for
+ authoritative servers; if a query contains an ECS option then
+ ACLs containing <code class="option">geoip</code> or <code class="option">ecs</code>
+ elements can match against the address encoded in the option.
+ This can be used to select a view for a query, so that different
+ answers can be provided depending on the client network.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The EDNS EXPIRE option has been implemented on the client
+ side, allowing a slave server to set the expiration timer
+ correctly when transferring zone data from another slave
+ server.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ A new <code class="option">masterfile-style</code> zone option controls
+ the formatting of text zone files: When set to
+ <code class="literal">full</code>, the zone file will dumped in
+ single-line-per-record format.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dig +ednsopt</strong></span> can now be used to set
+ arbitrary EDNS options in DNS requests.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dig +ednsflags</strong></span> can now be used to set
+ yet-to-be-defined EDNS flags in DNS requests.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dig +[no]ednsnegotiation</strong></span> can now be used enable /
+ disable EDNS version negotiation.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dig +header-only</strong></span> can now be used to send
+ queries without a question section.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dig +ttlunits</strong></span> causes <span class="command"><strong>dig</strong></span>
+ to print TTL values with time-unit suffixes: w, d, h, m, s for
+ weeks, days, hours, minutes, and seconds.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dig +zflag</strong></span> can be used to set the last
+ unassigned DNS header flag bit. This bit is normally zero.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dig +dscp=<em class="replaceable"><code>value</code></em></strong></span>
+ can now be used to set the DSCP code point in outgoing query
+ packets.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dig +mapped</strong></span> can now be used to determine
+ if mapped IPv4 addresses can be used.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>nslookup</strong></span> will now look up IPv6 as well
+ as IPv4 addresses by default. [RT #40420]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <code class="option">serial-update-method</code> can now be set to
+ <code class="literal">date</code>. On update, the serial number will
+ be set to the current date in YYYYMMDDNN format.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>dnssec-signzone -N date</strong></span> also sets the serial
+ number to YYYYMMDDNN.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named -L <em class="replaceable"><code>filename</code></em></strong></span>
+ causes <span class="command"><strong>named</strong></span> to send log messages to the
+ specified file by default instead of to the system log.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The rate limiter configured by the
+ <code class="option">serial-query-rate</code> option no longer covers
+ NOTIFY messages; those are now separately controlled by
+ <code class="option">notify-rate</code> and
+ <code class="option">startup-notify-rate</code> (the latter of which
+ controls the rate of NOTIFY messages sent when the server
+ is first started up or reconfigured).
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The default number of tasks and client objects available
+ for serving lightweight resolver queries have been increased,
+ and are now configurable via the new <code class="option">lwres-tasks</code>
+ and <code class="option">lwres-clients</code> options in
+ <code class="filename">named.conf</code>. [RT #35857]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Log output to files can now be buffered by specifying
+ <span class="command"><strong>buffered yes;</strong></span> when creating a channel.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>delv +tcp</strong></span> will exclusively use TCP when
+ sending queries.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> will now check to see whether
+ other name server processes are running before starting up.
+ This is implemented in two ways: 1) by refusing to start
+ if the configured network interfaces all return "address
+ in use", and 2) by attempting to acquire a lock on a file
+ specified by the <code class="option">lock-file</code> option or
+ the <span class="command"><strong>-X</strong></span> command line option. The
+ default lock file is
+ <code class="filename">/var/run/named/named.lock</code>.
+ Specifying <code class="literal">none</code> will disable the lock
+ file check.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>rndc delzone</strong></span> can now be applied to zones
+ which were configured in <code class="filename">named.conf</code>;
+ it is no longer restricted to zones which were added by
+ <span class="command"><strong>rndc addzone</strong></span>. (Note, however, that
+ this does not edit <code class="filename">named.conf</code>; the zone
+ must be removed from the configuration or it will return
+ when <span class="command"><strong>named</strong></span> is restarted or reloaded.)
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>rndc modzone</strong></span> can be used to reconfigure
+ a zone, using similar syntax to <span class="command"><strong>rndc addzone</strong></span>.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>rndc showzone</strong></span> displays the current
+ configuration for a specified zone.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ When BIND is built with the <span class="command"><strong>lmdb</strong></span> library
+ (Lightning Memory-Mapped Database), <span class="command"><strong>named</strong></span>
+ will store the configuration information for zones
+ that are added via <span class="command"><strong>rndc addzone</strong></span>
+ in a database, rather than in a flat "NZF" file. This
+ dramatically improves performance for
+ <span class="command"><strong>rndc delzone</strong></span> and
+ <span class="command"><strong>rndc modzone</strong></span>: deleting or changing
+ the contents of a database is much faster than rewriting
+ a text file.
+ </p>
+ <p>
+ On startup, if <span class="command"><strong>named</strong></span> finds an existing
+ NZF file, it will automatically convert it to the new NZD
+ database format.
+ </p>
+ <p>
+ To view the contents of an NZD, or to convert an
+ NZD back to an NZF file (for example, to revert back
+ to an earlier version of BIND which did not support the
+ NZD format), use the new command <span class="command"><strong>named-nzd2nzf</strong></span>
+ [RT #39837]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Added server-side support for pipelined TCP queries. Clients
+ may continue sending queries via TCP while previous queries are
+ processed in parallel. Responses are sent when they are
+ ready, not necessarily in the order in which the queries were
+ received.
+ </p>
+ <p>
+ To revert to the former behavior for a particular
+ client address or range of addresses, specify the address prefix
+ in the "keep-response-order" option. To revert to the former
+ behavior for all clients, use "keep-response-order { any; };".
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The new <span class="command"><strong>mdig</strong></span> command is a version of
+ <span class="command"><strong>dig</strong></span> that sends multiple pipelined
+ queries and then waits for responses, instead of sending one
+ query and waiting the response before sending the next. [RT #38261]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ To enable better monitoring and troubleshooting of RFC 5011
+ trust anchor management, the new <span class="command"><strong>rndc managed-keys</strong></span>
+ can be used to check status of trust anchors or to force keys
+ to be refreshed. Also, the managed-keys data file now has
+ easier-to-read comments. [RT #38458]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ An <span class="command"><strong>--enable-querytrace</strong></span> configure switch is
+ now available to enable very verbose query trace logging. This
+ option can only be set at compile time. This option has a
+ negative performance impact and should be used only for
+ debugging. [RT #37520]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ A new <span class="command"><strong>tcp-only</strong></span> option can be specified
+ in <span class="command"><strong>server</strong></span> statements to force
+ <span class="command"><strong>named</strong></span> to connect to the specified
+ server via TCP. [RT #37800]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The <span class="command"><strong>nxdomain-redirect</strong></span> option specifies
+ a DNS namespace to use for NXDOMAIN redirection. When a
+ recursive lookup returns NXDOMAIN, a second lookup is
+ initiated with the specified name appended to the query
+ name. This allows NXDOMAIN redirection data to be supplied
+ by multiple zones configured on the server, or by recursive
+ queries to other servers. (The older method, using
+ a single <span class="command"><strong>type redirect</strong></span> zone, has
+ better average performance but is less flexible.) [RT #37989]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The following types have been implemented: CSYNC, NINFO, RKEY,
+ SINK, TA, TALINK.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ A new <span class="command"><strong>message-compression</strong></span> option can be
+ used to specify whether or not to use name compression when
+ answering queries. Setting this to <strong class="userinput"><code>no</code></strong>
+ results in larger responses, but reduces CPU consumption and
+ may improve throughput. The default is <strong class="userinput"><code>yes</code></strong>.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ A <span class="command"><strong>read-only</strong></span> option is now available in the
+ <span class="command"><strong>controls</strong></span> statement to grant non-destructive
+ control channel access. In such cases, a restricted set of
+ <span class="command"><strong>rndc</strong></span> commands are allowed, which can
+ report information from <span class="command"><strong>named</strong></span>, but cannot
+ reconfigure or stop the server. By default, the control channel
+ access is <span class="emphasis"><em>not</em></span> restricted to these
+ read-only operations. [RT #40498]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ When loading a signed zone, <span class="command"><strong>named</strong></span> will
+ now check whether an RRSIG's inception time is in the future,
+ and if so, it will regenerate the RRSIG immediately. This helps
+ when a system's clock needs to be reset backwards.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The new <span class="command"><strong>minimal-any</strong></span> option reduces the size
+ of answers to UDP queries for type ANY by implementing one of
+ the strategies in "draft-ietf-dnsop-refuse-any": returning
+ a single arbitrarily-selected RRset that matches the query
+ name rather than returning all of the matching RRsets.
+ Thanks to Tony Finch for the contribution. [RT #41615]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> now provides feedback to the
+ owners of zones which have trust anchors configured
+ (<span class="command"><strong>trusted-keys</strong></span>,
+ <span class="command"><strong>managed-keys</strong></span>, <span class="command"><strong>dnssec-validation
+ auto;</strong></span> and <span class="command"><strong>dnssec-lookaside auto;</strong></span>)
+ by sending a daily query which encodes the keyids of the
+ configured trust anchors for the zone. This is controlled
+ by <span class="command"><strong>trust-anchor-telemetry</strong></span> and defaults
+ to yes.
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.0-changes"></a>Feature Changes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ The logging format used for <span class="command"><strong>querylog</strong></span> has been
+ altered. It now includes an additional field indicating the
+ address in memory of the client object processing the query.
+ </p>
+ <p>
+ The ISC DNSSEC Lookaside Validation (DLV) service is scheduled
+ to be disabled in 2017. A warning is now logged when
+ <span class="command"><strong>named</strong></span> is configured to use this service,
+ either explicitly or via <code class="option">dnssec-lookaside auto;</code>.
+ [RT #42207]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The timers returned by the statistics channel (indicating current
+ time, server boot time, and most recent reconfiguration time) are
+ now reported with millisecond accuracy. [RT #40082]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Updated the compiled-in addresses for H.ROOT-SERVERS.NET
+ and L.ROOT-SERVERS.NET.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ ACLs containing <span class="command"><strong>geoip asnum</strong></span> elements were
+ not correctly matched unless the full organization name was
+ specified in the ACL (as in
+ <span class="command"><strong>geoip asnum "AS1234 Example, Inc.";</strong></span>).
+ They can now match against the AS number alone (as in
+ <span class="command"><strong>geoip asnum "AS1234";</strong></span>).
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ When using native PKCS#11 cryptography (i.e.,
+ <span class="command"><strong>configure --enable-native-pkcs11</strong></span>) HSM PINs
+ of up to 256 characters can now be used.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ NXDOMAIN responses to queries of type DS are now cached separately
+ from those for other types. This helps when using "grafted" zones
+ of type forward, for which the parent zone does not contain a
+ delegation, such as local top-level domains. Previously a query
+ of type DS for such a zone could cause the zone apex to be cached
+ as NXDOMAIN, blocking all subsequent queries. (Note: This
+ change is only helpful when DNSSEC validation is not enabled.
+ "Grafted" zones without a delegation in the parent are not a
+ recommended configuration.)
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Update forwarding performance has been improved by allowing
+ a single TCP connection to be shared between multiple updates.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ By default, <span class="command"><strong>nsupdate</strong></span> will now check
+ the correctness of hostnames when adding records of type
+ A, AAAA, MX, SOA, NS, SRV or PTR. This behavior can be
+ disabled with <span class="command"><strong>check-names no</strong></span>.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Added support for OPENPGPKEY type.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The names of the files used to store managed keys and added
+ zones for each view are no longer based on the SHA256 hash
+ of the view name, except when this is necessary because the
+ view name contains characters that would be incompatible with use
+ as a file name. For views whose names do not contain forward
+ slashes ('/'), backslashes ('\'), or capital letters - which
+ could potentially cause namespace collision problems on
+ case-insensitive filesystems - files will now be named
+ after the view (for example, <code class="filename">internal.mkeys</code>
+ or <code class="filename">external.nzf</code>). However, to ensure
+ consistent behavior when upgrading, if a file using the old
+ name format is found to exist, it will continue to be used.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ "rndc" can now return text output of arbitrary size to
+ the caller. (Prior to this, certain commands such as
+ "rndc tsig-list" and "rndc zonestatus" could return
+ truncated output.)
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Errors reported when running <span class="command"><strong>rndc addzone</strong></span>
+ (e.g., when a zone file cannot be loaded) have been clarified
+ to make it easier to diagnose problems.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ When encountering an authoritative name server whose name is
+ an alias pointing to another name, the resolver treats
+ this as an error and skips to the next server. Previously
+ this happened silently; now the error will be logged to
+ the newly-created "cname" log category.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ If <span class="command"><strong>named</strong></span> is not configured to validate
+ answers, then allow fallback to plain DNS on timeout even when
+ we know the server supports EDNS. This will allow the server to
+ potentially resolve signed queries when TCP is being
+ blocked.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Large inline-signing changes should be less disruptive.
+ Signature generation is now done incrementally; the number
+ of signatures to be generated in each quantum is controlled
+ by "sig-signing-signatures <em class="replaceable"><code>number</code></em>;".
+ [RT #37927]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The experimental SIT option (code point 65001) of BIND
+ 9.10.0 through BIND 9.10.2 has been replaced with the COOKIE
+ option (code point 10). It is no longer experimental, and
+ is sent by default, by both <span class="command"><strong>named</strong></span> and
+ <span class="command"><strong>dig</strong></span>.
+ </p>
+ <p>
+ The SIT-related named.conf options have been marked as
+ obsolete, and are otherwise ignored.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ When <span class="command"><strong>dig</strong></span> receives a truncated (TC=1)
+ response or a BADCOOKIE response code from a server, it
+ will automatically retry the query using the server COOKIE
+ that was returned by the server in its initial response.
+ [RT #39047]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Retrieving the local port range from net.ipv4.ip_local_port_range
+ on Linux is now supported.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ A new <code class="option">nsip-wait-recurse</code> directive has been
+ added to RPZ, specifying whether to look up unknown name server
+ IP addresses and wait for a response before applying RPZ-NSIP rules.
+ The default is <strong class="userinput"><code>yes</code></strong>. If set to
+ <strong class="userinput"><code>no</code></strong>, <span class="command"><strong>named</strong></span> will only
+ apply RPZ-NSIP rules to servers whose addresses are already cached.
+ The addresses will be looked up in the background so the rule can
+ be applied on subsequent queries. This improves performance when
+ the cache is cold, at the cost of temporary imprecision in applying
+ policy directives. [RT #35009]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Within the <code class="option">response-policy</code> option, it is now
+ possible to configure RPZ rewrite logging on a per-zone basis
+ using the <code class="option">log</code> clause.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ The default preferred glue is now the address type of the
+ transport the query was received over.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ On machines with 2 or more processors (CPU), the default value
+ for the number of UDP listeners has been changed to the number
+ of detected processors minus one.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Zone transfers now use smaller message sizes to improve
+ message compression. This results in reduced network usage.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Added support for the AVC resource record type (Application
+ Visibility and Control).
+ </p>
+ <p>
+ Changed <span class="command"><strong>rndc reconfig</strong></span> behavior so that newly
+ added zones are loaded asynchronously and the loading does not
+ block the server.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>minimal-responses</strong></span> now takes two new
+ arguments: <code class="option">no-auth</code> suppresses
+ populating the authority section but not the additional
+ section; <code class="option">no-auth-recursive</code>
+ does the same but only when answering recursive queries.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ At server startup time, the queues for processing
+ notify and zone refresh queries are now processed in
+ LIFO rather than FIFO order, to speed up
+ loading of newly added zones. [RT #42825]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ When answering queries of type MX or SRV, TLSA records for
+ the target name are now included in the additional section
+ to speed up DANE processing. [RT #42894]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ <span class="command"><strong>named</strong></span> can now use the TCP Fast Open
+ mechanism on the server side, if supported by the
+ local operating system. [RT #42866]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h4 class="title">
+<a name="relnotes-9.11.0-bugs"></a>Bug Fixes</h4></div></div></div>
+ <div class="itemizedlist"><ul class="itemizedlist" style="list-style-type: disc; ">
+<li class="listitem">
+ <p>
+ Fixed a crash when calling <span class="command"><strong>rndc stats</strong></span> on some
+ Windows builds: some Visual Studio compilers generate code that
+ crashes when the "%z" printf() format specifier is used. [RT #42380]
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ Windows installs were failing due to triggering UAC without
+ the installation binary being signed.
+ </p>
+ </li>
+<li class="listitem">
+ <p>
+ A change in the internal binary representation of the RBT database
+ node structure enabled a race condition to occur (especially when
+ BIND was built with certain compilers or optimizer settings),
+ leading to inconsistent database state which caused random
+ assertion failures. [RT #42380]
+ </p>
+ </li>
+</ul></div>
+ </div>
+
+</div>
+
+ <div class="section">
+<div class="titlepage"><div><div><h3 class="title">
<a name="end_of_life"></a>End of Life</h3></div></div></div>
<p>
BIND 9.11 (Extended Support Version) will be supported until at
least December, 2021.
</p>
<p>
- See <a class="link" href="https://www.isc.org/downloads/software-support-policy/" target="_top">https://www.isc.org/downloads/software-support-policy/</a>
+ See <a class="link" href="https://kb.isc.org/docs/aa-00896" target="_top">https://kb.isc.org/docs/aa-00896</a>
for details of ISC's software support policy.
</p>
</div>
Thank you to everyone who assisted us in making this release possible.
If you would like to contribute to ISC to assist us in continuing to
make quality open source software, please visit our donations page at
- <a class="link" href="http://www.isc.org/donate/" target="_top">http://www.isc.org/donate/</a>.
+ <a class="link" href="https://www.isc.org/donate/" target="_top">https://www.isc.org/donate/</a>.
</p>
</div>
</div>
Download
-The latest versions of BIND 9 software can always be found at http://
-www.isc.org/downloads/. There you will find additional information about
+The latest versions of BIND 9 software can always be found at https://
+www.isc.org/download/. There you will find additional information about
each release, source code, and pre-compiled versions for Microsoft Windows
operating systems.
or who wish to discuss how to comply with the license may contact ISC at
https://www.isc.org/mission/contact/.
-Security Fixes
+Notes for BIND 9.11.13
- * A race condition could trigger an assertion failure when a large
- number of incoming packets were being rejected. This flaw is disclosed
- in CVE-2019-6471. [GL #942]
+None.
-New Features
+Notes for BIND 9.11.12
- * The new GeoIP2 API from MaxMind is now supported when BIND is compiled
- using configure --with-geoip2. The legacy GeoIP API can be used by
- compiling with configure --with-geoip instead. (Note that the
- databases for the legacy API are no longer maintained by MaxMind.)
+None.
- The default path to the GeoIP2 databases will be set based on the
- location of the libmaxminddb library; for example, if it is in /usr/
- local/lib, then the default path will be /usr/local/share/GeoIP. This
- value can be overridden in named.conf using the geoip-directory
- option.
+Notes for BIND 9.11.11
- Some geoip ACL settings that were available with legacy GeoIP,
- including searches for netspeed, org, and three-letter ISO country
- codes, will no longer work when using GeoIP2. Supported GeoIP2
- database types are country, city, domain, isp, and as. All of the
- databases support both IPv4 and IPv6 lookups. [GL #182]
+None.
+
+Notes for BIND 9.11.10
+
+New Features
* A SipHash 2-4 based DNS Cookie (RFC 7873) algorithm has been added.
[GL #605]
Bug Fixes
- * Glue address records were not being returned in responses to root
- priming queries; this has been corrected. [GL #1092]
-
* Interaction between DNS64 and RPZ No Data rule (CNAME *.) could cause
unexpected results; this has been fixed. [GL #1106]
* Handle ETIMEDOUT error on connect() with a non-blocking socket. [GL #
1133]
+Notes for BIND 9.11.9
+
+New Features
+
+ * The new GeoIP2 API from MaxMind is now supported when BIND is compiled
+ using configure --with-geoip2. The legacy GeoIP API can be used by
+ compiling with configure --with-geoip instead. (Note that the
+ databases for the legacy API are no longer maintained by MaxMind.)
+
+ The default path to the GeoIP2 databases will be set based on the
+ location of the libmaxminddb library; for example, if it is in /usr/
+ local/lib, then the default path will be /usr/local/share/GeoIP. This
+ value can be overridden in named.conf using the geoip-directory
+ option.
+
+ Some geoip ACL settings that were available with legacy GeoIP,
+ including searches for netspeed, org, and three-letter ISO country
+ codes, will no longer work when using GeoIP2. Supported GeoIP2
+ database types are country, city, domain, isp, and as. All of the
+ databases support both IPv4 and IPv6 lookups. [GL #182]
+
+Bug Fixes
+
+ * Glue address records were not being returned in responses to root
+ priming queries; this has been corrected. [GL #1092]
+
+Notes for BIND 9.11.8
+
+Security Fixes
+
+ * A race condition could trigger an assertion failure when a large
+ number of incoming packets were being rejected. This flaw is disclosed
+ in CVE-2019-6471. [GL #942]
+
+Notes for BIND 9.11.7
+
+Security Fixes
+
+ * The TCP client quota set using the tcp-clients option could be
+ exceeded in some cases. This could lead to exhaustion of file
+ descriptors. This flaw is disclosed in CVE-2018-5743. [GL #615]
+
+Feature Changes
+
+ * When trusted-keys and managed-keys are both configured for the same
+ name, or when trusted-keys is used to configure a trust anchor for the
+ root zone and dnssec-validation is set to auto, automatic RFC 5011 key
+ rollovers will fail.
+
+ This combination of settings was never intended to work, but there was
+ no check for it in the parser. This has been corrected; a warning is
+ now logged. (In BIND 9.15 and higher this error will be fatal.) [GL #
+ 868]
+
+Notes for BIND 9.11.6
+
+Security Fixes
+
+ * Code change #4964, intended to prevent double signatures when deleting
+ an inactive zone DNSKEY in some situations, introduced a new problem
+ during zone processing in which some delegation glue RRsets are
+ incorrectly identified as needing RRSIGs, which are then created for
+ them using the current active ZSK for the zone. In some, but not all
+ cases, the newly-signed RRsets are added to the zone's NSEC/NSEC3
+ chain, but incompletely -- this can result in a broken chain,
+ affecting validation of proof of nonexistence for records in the zone.
+ [GL #771]
+
+ * named could crash if it managed a DNSSEC security root with
+ managed-keys and the authoritative zone rolled the key to an algorithm
+ not supported by BIND 9. This flaw is disclosed in CVE-2018-5745. [GL
+ #780]
+
+ * named leaked memory when processing a request with multiple Key Tag
+ EDNS options present. ISC would like to thank Toshifumi Sakaguchi for
+ bringing this to our attention. This flaw is disclosed in
+ CVE-2018-5744. [GL #772]
+
+ * Zone transfer controls for writable DLZ zones were not effective as
+ the allowzonexfr method was not being called for such zones. This flaw
+ is disclosed in CVE-2019-6465. [GL #790]
+
+Feature Changes
+
+ * When compiled with IDN support, the dig and the nslookup commands now
+ disable IDN processing when the standard output is not a tty (e.g. not
+ used by human). The command line options +idnin and +idnout need to be
+ used to enable IDN processing when dig or nslookup is used from the
+ shell scripts.
+
+Notes for BIND 9.11.5
+
+Security Fixes
+
+ * named could crash during recursive processing of DNAME records when
+ deny-answer-aliases was in use. This flaw is disclosed in
+ CVE-2018-5740. [GL #387]
+
+New Features
+
+ * Two new update policy rule types have been added krb5-selfsub and
+ ms-selfsub which allow machines with Kerberos principals to update the
+ name space at or below the machine names identified in the respective
+ principals.
+
+Feature Changes
+
+ * The rndc nta command could not differentiate between views of the same
+ name but different class; this has been corrected with the addition of
+ a -class option. [GL #105]
+
+Bug Fixes
+
+ * When a negative trust anchor was added to multiple views using rndc
+ nta, the text returned via rndc was incorrectly truncated after the
+ first line, making it appear that only one NTA had been added. This
+ has been fixed. [GL #105]
+
+Notes for BIND 9.11.4
+
+Security Fixes
+
+ * When recursion is enabled but the allow-recursion and
+ allow-query-cache ACLs are not specified, they should be limited to
+ local networks, but they were inadvertently set to match the default
+ allow-query, thus allowing remote queries. This flaw is disclosed in
+ CVE-2018-5738. [GL #309]
+
+New Features
+
+ * named now supports the "root key sentinel" mechanism. This enables
+ validating resolvers to indicate which trust anchors are configured
+ for the root, so that information about root key rollover status can
+ be gathered. To disable this feature, add root-key-sentinel no; to
+ named.conf.
+
+ * Added the ability not to return a DNS COOKIE option when one is
+ present in the request. To prevent a cookie being returned, add
+ answer-cookie no; to named.conf. [GL #173]
+
+ answer-cookie no is only intended as a temporary measure, for use when
+ named shares an IP address with other servers that do not yet support
+ DNS COOKIE. A mismatch between servers on the same address is not
+ expected to cause operational problems, but the option to disable
+ COOKIE responses so that all servers have the same behavior is
+ provided out of an abundance of caution. DNS COOKIE is an important
+ security mechanism, and should not be disabled unless absolutely
+ necessary.
+
+Removed Features
+
+ * named will now log a warning if the old BIND now can be compiled
+ against libidn2 library to add IDNA2008 support. Previously BIND only
+ supported IDNA2003 using (now obsolete) idnkit-1 library.
+
+Feature Changes
+
+ * dig +noidnin can be used to disable IDN processing on the input domain
+ name, when BIND is compiled with IDN support.
+
+ * Multiple cookie-secret clause are now supported. The first
+ cookie-secret in named.conf is used to generate new server cookies.
+ Any others are used to accept old server cookies or those generated by
+ other servers using the matching cookie-secret.
+
+Bug Fixes
+
+ * named now rejects excessively large incremental (IXFR) zone transfers
+ in order to prevent possible corruption of journal files which could
+ cause named to abort when loading zones. [GL #339]
+
+ * rndc reload could cause named to leak memory if it was invoked before
+ the zone loading actions from a previous rndc reload command were
+ completed. [RT #47076]
+
+Notes for BIND 9.11.3
+
+Security Fixes
+
+ * Addresses could be referenced after being freed during resolver
+ processing, causing an assertion failure. The chances of this
+ happening were remote, but the introduction of a delay in resolution
+ increased them. This bug is disclosed in CVE-2017-3145. [RT #46839]
+
+ * update-policy rules that otherwise ignore the name field now require
+ that it be set to "." to ensure that any type list present is properly
+ interpreted. If the name field was omitted from the rule declaration
+ and a type list was present it wouldn't be interpreted as expected.
+
+Removed Features
+
+ * The ISC DNSSEC Lookaside Validation (DLV) service has been shut down;
+ all DLV records in the dlv.isc.org zone have been removed. References
+ to the service have been removed from BIND documentation. Lookaside
+ validation is no longer used by default by delv. The DLV key has been
+ removed from bind.keys. Setting dnssec-lookaside to auto or to use
+ dlv.isc.org as a trust anchor results in a warning being issued.
+
+ * named will now log a warning if the old root DNSSEC key is explicitly
+ configured and has not been updated. [RT #43670]
+
+Protocol Changes
+
+ * BIND can now use the Ed25519 and Ed448 Edwards Curve DNSSEC signing
+ algorithms described in RFC 8080. Note, however, that these algorithms
+ must be supported in OpenSSL; currently they are only available in the
+ development branch of OpenSSL at https://github.com/openssl/openssl.
+ [RT #44696]
+
+ * When parsing DNS messages, EDNS KEY TAG options are checked for
+ correctness. When printing messages (for example, in dig), EDNS KEY
+ TAG options are printed in readable format.
+
+Feature Changes
+
+ * named will no longer start or accept reconfiguration if managed-keys
+ or dnssec-validation auto are in use and the managed-keys directory
+ (specified by managed-keys-directory, and defaulting to the working
+ directory if not specified), is not writable by the effective user ID.
+ [RT #46077]
+
+ * Previously, update-policy local; accepted updates from any source so
+ long as they were signed by the locally-generated session key. This
+ has been further restricted; updates are now only accepted from
+ locally configured addresses. [RT #45492]
+
+Bug Fixes
+
+ * Attempting to validate improperly unsigned CNAME responses from secure
+ zones could cause a validator loop. This caused a delay in returning
+ SERVFAIL and also increased the chances of encountering the crash bug
+ described in CVE-2017-3145. [RT #46839]
+
+ * When named was reconfigured, failure of some zones to load correctly
+ could leave the system in an inconsistent state; while generally
+ harmless, this could lead to a crash later when using rndc addzone.
+ Reconfiguration changes are now fully rolled back in the event of
+ failure. [RT #45841]
+
+ * Some header files included <isc/util.h> incorrectly as it pollutes
+ with namespace with non ISC_ macros and this should only be done by
+ explicitly including <isc/util.h>. This has been corrected. Some code
+ may depend on <isc/util.h> being implicitly included via other header
+ files. Such code should explicitly include <isc/util.h>.
+
+ * Zones created with rndc addzone could temporarily fail to inherit the
+ allow-transfer ACL set in the options section of named.conf. [RT #
+ 46603]
+
+ * named failed to properly determine whether there were active KSK and
+ ZSK keys for an algorithm when update-check-ksk was true (which is the
+ default setting). This could leave records unsigned when rolling keys.
+ [RT #46743] [RT #46754] [RT #46774]
+
+Notes for BIND 9.11.2
+
+Security Fixes
+
+ * An error in TSIG handling could permit unauthorized zone transfers or
+ zone updates. These flaws are disclosed in CVE-2017-3142 and
+ CVE-2017-3143. [RT #45383]
+
+ * The BIND installer on Windows used an unquoted service path, which can
+ enable privilege escalation. This flaw is disclosed in CVE-2017-3141.
+ [RT #45229]
+
+ * With certain RPZ configurations, a response with TTL 0 could cause
+ named to go into an infinite query loop. This flaw is disclosed in
+ CVE-2017-3140. [RT #45181]
+
+Feature Changes
+
+ * dig +ednsopt now accepts the names for EDNS options in addition to
+ numeric values. For example, an EDNS Client-Subnet option could be
+ sent using dig +ednsopt=ecs:.... Thanks to John Worley of Secure64 for
+ the contribution. [RT #44461]
+
+ * Threads in named are now set to human-readable names to assist
+ debugging on operating systems that support that. Threads will have
+ names such as "isc-timer", "isc-sockmgr", "isc-worker0001", and so on.
+ This will affect the reporting of subsidiary thread names in ps and
+ top, but not the main thread. [RT #43234]
+
+ * DiG now warns about .local queries which are reserved for Multicast
+ DNS. [RT #44783]
+
+Bug Fixes
+
+ * Fixed a bug that was introduced in an earlier development release
+ which caused multi-packet AXFR and IXFR messages to fail validation if
+ not all packets contained TSIG records; this caused interoperability
+ problems with some other DNS implementations. [RT #45509]
+
+ * Reloading or reconfiguring named could fail on some platforms when
+ LMDB was in use. [RT #45203]
+
+ * Due to some incorrectly deleted code, when BIND was built with LMDB,
+ zones that were deleted via rndc delzone were removed from the running
+ server but were not removed from the new zone database, so that
+ deletion did not persist after a server restart. This has been
+ corrected. [RT #45185]
+
+ * Semicolons are no longer escaped when printing CAA and URI records.
+ This may break applications that depend on the presence of the
+ backslash before the semicolon. [RT #45216]
+
+ * AD could be set on truncated answer with no records present in the
+ answer and authority sections. [RT #45140]
+
+Notes for BIND 9.11.1
+
+Security Fixes
+
+ * rndc "" could trigger an assertion failure in named. This flaw is
+ disclosed in (CVE-2017-3138). [RT #44924]
+
+ * Some chaining (i.e., type CNAME or DNAME) responses to upstream
+ queries could trigger assertion failures. This flaw is disclosed in
+ CVE-2017-3137. [RT #44734]
+
+ * dns64 with break-dnssec yes; can result in an assertion failure. This
+ flaw is disclosed in CVE-2017-3136. [RT #44653]
+
+ * If a server is configured with a response policy zone (RPZ) that
+ rewrites an answer with local data, and is also configured for DNS64
+ address mapping, a NULL pointer can be read triggering a server crash.
+ This flaw is disclosed in CVE-2017-3135. [RT #44434]
+
+ * A coding error in the nxdomain-redirect feature could lead to an
+ assertion failure if the redirection namespace was served from a local
+ authoritative data source such as a local zone or a DLZ instead of via
+ recursive lookup. This flaw is disclosed in CVE-2016-9778. [RT #43837]
+
+ * named could mishandle authority sections with missing RRSIGs,
+ triggering an assertion failure. This flaw is disclosed in
+ CVE-2016-9444. [RT #43632]
+
+ * named mishandled some responses where covering RRSIG records were
+ returned without the requested data, resulting in an assertion
+ failure. This flaw is disclosed in CVE-2016-9147. [RT #43548]
+
+ * named incorrectly tried to cache TKEY records which could trigger an
+ assertion failure when there was a class mismatch. This flaw is
+ disclosed in CVE-2016-9131. [RT #43522]
+
+ * It was possible to trigger assertions when processing responses
+ containing answers of type DNAME. This flaw is disclosed in
+ CVE-2016-8864. [RT #43465]
+
+ * Added the ability to specify the maximum number of records permitted
+ in a zone (max-records #;). This provides a mechanism to block overly
+ large zone transfers, which is a potential risk with slave zones from
+ other parties, as described in CVE-2016-6170. [RT #42143]
+
+Feature Changes
+
+ * dnstap now stores both the local and remote addresses for all
+ messages, instead of only the remote address. The default output
+ format for dnstap-read has been updated to include these addresses,
+ with the initiating address first and the responding address second,
+ separated by "-%gt;" or "%lt;-" to indicate in which direction the
+ message was sent. [RT #43595]
+
+ * Expanded and improved the YAML output from dnstap-read -y: it now
+ includes packet size and a detailed breakdown of message contents. [RT
+ #43622] [RT #43642]
+
+ * If an ACL is specified with an address prefix in which the prefix
+ length is longer than the address portion (for example, 192.0.2.1/8),
+ named will now log a warning. In future releases this will be a fatal
+ configuration error. [RT #43367]
+
+Bug Fixes
+
+ * A synthesized CNAME record appearing in a response before the
+ associated DNAME could be cached, when it should not have been. This
+ was a regression introduced while addressing CVE-2016-8864. [RT #
+ 44318]
+
+ * named could deadlock if multiple changes to NSEC/NSEC3 parameters for
+ the same zone were being processed at the same time. [RT #42770]
+
+ * named could trigger an assertion when sending NOTIFY messages. [RT #
+ 44019]
+
+ * Referencing a nonexistent zone in a response-policy statement could
+ cause an assertion failure during configuration. [RT #43787]
+
+ * rndc addzone could cause a crash when attempting to add a zone with a
+ type other than master or slave. Such zones are now rejected. [RT #
+ 43665]
+
+ * named could hang when encountering log file names with large apparent
+ gaps in version number (for example, when files exist called
+ "logfile.0", "logfile.1", and "logfile.1482954169"). This is now
+ handled correctly. [RT #38688]
+
+ * If a zone was updated while named was processing a query for
+ nonexistent data, it could return out-of-sync NSEC3 records causing
+ potential DNSSEC validation failure. [RT #43247]
+
+Maintenance
+
+ * The built-in root hints have been updated to include an IPv6 address
+ (2001:500:12::d0d) for G.ROOT-SERVERS.NET.
+
+Miscellaneous Notes
+
+ * Authoritative server support for the EDNS Client Subnet option (ECS),
+ introduced in BIND 9.11.0, was based on an early version of the
+ specification, and is now known to have incompatibilities with other
+ ECS implementations. It is also inefficient, requiring a separate view
+ for each answer, and is unable to correct for overlapping subnets in
+ the configuration. It is intended for testing purposes but is not
+ recommended for for production use. This was not made sufficiently
+ clear in the documentation at the time of release.
+
+Notes for BIND 9.11.0
+
+Security Fixes
+
+ * It was possible to trigger a assertion when rendering a message using
+ a specially crafted request. This flaw is disclosed in CVE-2016-2776.
+ [RT #43139]
+
+ * getrrsetbyname with a non absolute name could trigger an infinite
+ recursion bug in lwresd and named with lwres configured if when
+ combined with a search list entry the resulting name is too long. This
+ flaw is disclosed in CVE-2016-2775. [RT #42694]
+
+New Features
+
+ * A new method of provisioning secondary servers called "Catalog Zones"
+ has been added. This is an implementation of
+ draft-muks-dnsop-dns-catalog-zones/ .
+
+ A catalog zone is a regular DNS zone which contains a list of "member
+ zones", along with the configuration options for each of those zones.
+ When a server is configured to use a catalog zone, all the zones
+ listed in the catalog zone are added to the local server as slave
+ zones. When the catalog zone is updated (e.g., by adding or removing
+ zones, or changing configuration options for existing zones) those
+ changes will be put into effect. Since the catalog zone is itself a
+ DNS zone, this means configuration changes can be propagated to slaves
+ using the standard AXFR/IXFR update mechanism.
+
+ This feature should be considered experimental. It currently supports
+ only basic features; more advanced features such as ACLs and TSIG keys
+ are not yet supported. Example catalog zone configurations can be
+ found in the Chapter 9 of the BIND Administrator Reference Manual.
+
+ Support for master entries with TSIG keys has been added to catalog
+ zones, as well as support for allow-query and allow-transfer.
+
+ * Added an isc.rndc Python module, which allows rndc commands to be sent
+ from Python programs.
+
+ * Added support for DynDB, a new interface for loading zone data from an
+ external database, developed by Red Hat for the FreeIPA project.
+ (Thanks in particular to Adam Tkac and Petr Spacek of Red Hat for the
+ contribution.)
+
+ Unlike the existing DLZ and SDB interfaces, which provide a limited
+ subset of database functionality within BIND - translating DNS queries
+ into real-time database lookups with relatively poor performance and
+ with no ability to handle DNSSEC-signed data - DynDB is able to fully
+ implement and extend the database API used natively by BIND.
+
+ A DynDB module could pre-load data from an external data source, then
+ serve it with the same performance and functionality as conventional
+ BIND zones, and with the ability to take advantage of database
+ features not available in BIND, such as multi-master replication.
+
+ * Fetch quotas are now compiled in by default: they no longer require
+ BIND to be configured with --enable-fetchlimit, as was the case when
+ the feature was introduced in BIND 9.10.3.
+
+ These quotas limit the queries that are sent by recursive resolvers to
+ authoritative servers experiencing denial-of-service attacks. They can
+ both reduce the harm done to authoritative servers and also avoid the
+ resource exhaustion that can be experienced by recursive servers when
+ they are being used as a vehicle for such an attack.
+
+ + fetches-per-server limits the number of simultaneous queries that
+ can be sent to any single authoritative server. The configured
+ value is a starting point; it is automatically adjusted downward
+ if the server is partially or completely non-responsive. The
+ algorithm used to adjust the quota can be configured via the
+ fetch-quota-params option.
+
+ + fetches-per-zone limits the number of simultaneous queries that
+ can be sent for names within a single domain. (Note: Unlike
+ "fetches-per-server", this value is not self-tuning.)
+
+ Statistics counters have also been added to track the number of
+ queries affected by these quotas.
+
+ * Added support for dnstap, a fast, flexible method for capturing and
+ logging DNS traffic, developed by Robert Edmonds at Farsight Security,
+ Inc., whose assistance is gratefully acknowledged.
+
+ To enable dnstap at compile time, the fstrm and protobuf-c libraries
+ must be available, and BIND must be configured with --enable-dnstap.
+
+ A new utility dnstap-read has been added to allow dnstap data to be
+ presented in a human-readable format.
+
+ rndc dnstap -roll causes dnstap output files to be rolled like log
+ files -- the most recent output file is renamed with a .0 suffix, the
+ next most recent with .1, etc. (Note that this only works when dnstap
+ output is being written to a file, not to a UNIX domain socket.) An
+ optional numerical argument specifies how many backup log files to
+ retain; if not specified or set to 0, there is no limit.
+
+ rndc dnstap -reopen simply closes and reopens the dnstap output
+ channel without renaming the output file.
+
+ For more information on dnstap, see https://dnstap.info.
+
+ * New statistics counters have been added to track traffic sizes, as
+ specified in RSSAC002. Query and response message sizes are broken up
+ into ranges of histogram buckets: TCP and UDP queries of size 0-15,
+ 16-31, ..., 272-288, and 288+, and TCP and UDP responses of size 0-15,
+ 16-31, ..., 4080-4095, and 4096+. These values can be accessed via the
+ XML and JSON statistics channels at, for example, http://
+ localhost:8888/xml/v3/traffic or http://localhost:8888/json/v1/traffic
+ .
+
+ Statistics for RSSAC02v3 traffic-volume, traffic-sizes and
+ rcode-volume reporting are now collected.
+
+ * A new DNSSEC key management utility, dnssec-keymgr, has been added.
+ This tool is meant to run unattended (e.g., under cron). It reads a
+ policy definition file (default /etc/dnssec-policy.conf) and creates
+ or updates DNSSEC keys as necessary to ensure that a zone's keys match
+ the defined policy for that zone. New keys are created whenever
+ necessary to ensure rollovers occur correctly. Existing keys' timing
+ metadata is adjusted as needed to set the correct rollover period,
+ prepublication interval, etc. If the configured policy changes, keys
+ are corrected automatically. See the dnssec-keymgr man page for full
+ details.
+
+ Note: dnssec-keymgr depends on Python and on the Python lex/yacc
+ module, PLY. The other Python-based tools, dnssec-coverage and
+ dnssec-checkds, have been refactored and updated as part of this work.
+
+ dnssec-keymgr now takes a -r randomfile option.
+
+ (Many thanks to Sebasti?n Castro for his assistance in developing this
+ tool at the IETF 95 Hackathon in Buenos Aires, April 2016.)
+
+ * The serial number of a dynamically updatable zone can now be set using
+ rndc signing -serial number zonename. This is particularly useful with
+ inline-signing zones that have been reset. Setting the serial number
+ to a value larger than that on the slaves will trigger an AXFR-style
+ transfer.
+
+ * When answering recursive queries, SERVFAIL responses can now be cached
+ by the server for a limited time; subsequent queries for the same
+ query name and type will return another SERVFAIL until the cache times
+ out. This reduces the frequency of retries when a query is
+ persistently failing, which can be a burden on recursive servers. The
+ SERVFAIL cache timeout is controlled by servfail-ttl, which defaults
+ to 1 second and has an upper limit of 30.
+
+ * The new rndc nta command can now be used to set a "negative trust
+ anchor" (NTA), disabling DNSSEC validation for a specific domain; this
+ can be used when responses from a domain are known to be failing
+ validation due to administrative error rather than because of a
+ spoofing attack. NTAs are strictly temporary; by default they expire
+ after one hour, but can be configured to last up to one week. The
+ default NTA lifetime can be changed by setting the nta-lifetime in
+ named.conf. When added, NTAs are stored in a file (viewname.nta) in
+ order to persist across restarts of the named server.
+
+ * The EDNS Client Subnet (ECS) option is now supported for authoritative
+ servers; if a query contains an ECS option then ACLs containing geoip
+ or ecs elements can match against the address encoded in the option.
+ This can be used to select a view for a query, so that different
+ answers can be provided depending on the client network.
+
+ * The EDNS EXPIRE option has been implemented on the client side,
+ allowing a slave server to set the expiration timer correctly when
+ transferring zone data from another slave server.
+
+ * A new masterfile-style zone option controls the formatting of text
+ zone files: When set to full, the zone file will dumped in
+ single-line-per-record format.
+
+ * dig +ednsopt can now be used to set arbitrary EDNS options in DNS
+ requests.
+
+ * dig +ednsflags can now be used to set yet-to-be-defined EDNS flags in
+ DNS requests.
+
+ * dig +[no]ednsnegotiation can now be used enable / disable EDNS version
+ negotiation.
+
+ * dig +header-only can now be used to send queries without a question
+ section.
+
+ * dig +ttlunits causes dig to print TTL values with time-unit suffixes:
+ w, d, h, m, s for weeks, days, hours, minutes, and seconds.
+
+ * dig +zflag can be used to set the last unassigned DNS header flag bit.
+ This bit is normally zero.
+
+ * dig +dscp=value can now be used to set the DSCP code point in outgoing
+ query packets.
+
+ * dig +mapped can now be used to determine if mapped IPv4 addresses can
+ be used.
+
+ * nslookup will now look up IPv6 as well as IPv4 addresses by default.
+ [RT #40420]
+
+ * serial-update-method can now be set to date. On update, the serial
+ number will be set to the current date in YYYYMMDDNN format.
+
+ * dnssec-signzone -N date also sets the serial number to YYYYMMDDNN.
+
+ * named -L filename causes named to send log messages to the specified
+ file by default instead of to the system log.
+
+ * The rate limiter configured by the serial-query-rate option no longer
+ covers NOTIFY messages; those are now separately controlled by
+ notify-rate and startup-notify-rate (the latter of which controls the
+ rate of NOTIFY messages sent when the server is first started up or
+ reconfigured).
+
+ * The default number of tasks and client objects available for serving
+ lightweight resolver queries have been increased, and are now
+ configurable via the new lwres-tasks and lwres-clients options in
+ named.conf. [RT #35857]
+
+ * Log output to files can now be buffered by specifying buffered yes;
+ when creating a channel.
+
+ * delv +tcp will exclusively use TCP when sending queries.
+
+ * named will now check to see whether other name server processes are
+ running before starting up. This is implemented in two ways: 1) by
+ refusing to start if the configured network interfaces all return
+ "address in use", and 2) by attempting to acquire a lock on a file
+ specified by the lock-file option or the -X command line option. The
+ default lock file is /var/run/named/named.lock. Specifying none will
+ disable the lock file check.
+
+ * rndc delzone can now be applied to zones which were configured in
+ named.conf; it is no longer restricted to zones which were added by
+ rndc addzone. (Note, however, that this does not edit named.conf; the
+ zone must be removed from the configuration or it will return when
+ named is restarted or reloaded.)
+
+ * rndc modzone can be used to reconfigure a zone, using similar syntax
+ to rndc addzone.
+
+ * rndc showzone displays the current configuration for a specified zone.
+
+ * When BIND is built with the lmdb library (Lightning Memory-Mapped
+ Database), named will store the configuration information for zones
+ that are added via rndc addzone in a database, rather than in a flat
+ "NZF" file. This dramatically improves performance for rndc delzone
+ and rndc modzone: deleting or changing the contents of a database is
+ much faster than rewriting a text file.
+
+ On startup, if named finds an existing NZF file, it will automatically
+ convert it to the new NZD database format.
+
+ To view the contents of an NZD, or to convert an NZD back to an NZF
+ file (for example, to revert back to an earlier version of BIND which
+ did not support the NZD format), use the new command named-nzd2nzf [RT
+ #39837]
+
+ * Added server-side support for pipelined TCP queries. Clients may
+ continue sending queries via TCP while previous queries are processed
+ in parallel. Responses are sent when they are ready, not necessarily
+ in the order in which the queries were received.
+
+ To revert to the former behavior for a particular client address or
+ range of addresses, specify the address prefix in the
+ "keep-response-order" option. To revert to the former behavior for all
+ clients, use "keep-response-order { any; };".
+
+ * The new mdig command is a version of dig that sends multiple pipelined
+ queries and then waits for responses, instead of sending one query and
+ waiting the response before sending the next. [RT #38261]
+
+ * To enable better monitoring and troubleshooting of RFC 5011 trust
+ anchor management, the new rndc managed-keys can be used to check
+ status of trust anchors or to force keys to be refreshed. Also, the
+ managed-keys data file now has easier-to-read comments. [RT #38458]
+
+ * An --enable-querytrace configure switch is now available to enable
+ very verbose query trace logging. This option can only be set at
+ compile time. This option has a negative performance impact and should
+ be used only for debugging. [RT #37520]
+
+ * A new tcp-only option can be specified in server statements to force
+ named to connect to the specified server via TCP. [RT #37800]
+
+ * The nxdomain-redirect option specifies a DNS namespace to use for
+ NXDOMAIN redirection. When a recursive lookup returns NXDOMAIN, a
+ second lookup is initiated with the specified name appended to the
+ query name. This allows NXDOMAIN redirection data to be supplied by
+ multiple zones configured on the server, or by recursive queries to
+ other servers. (The older method, using a single type redirect zone,
+ has better average performance but is less flexible.) [RT #37989]
+
+ * The following types have been implemented: CSYNC, NINFO, RKEY, SINK,
+ TA, TALINK.
+
+ * A new message-compression option can be used to specify whether or not
+ to use name compression when answering queries. Setting this to no
+ results in larger responses, but reduces CPU consumption and may
+ improve throughput. The default is yes.
+
+ * A read-only option is now available in the controls statement to grant
+ non-destructive control channel access. In such cases, a restricted
+ set of rndc commands are allowed, which can report information from
+ named, but cannot reconfigure or stop the server. By default, the
+ control channel access is not restricted to these read-only
+ operations. [RT #40498]
+
+ * When loading a signed zone, named will now check whether an RRSIG's
+ inception time is in the future, and if so, it will regenerate the
+ RRSIG immediately. This helps when a system's clock needs to be reset
+ backwards.
+
+ * The new minimal-any option reduces the size of answers to UDP queries
+ for type ANY by implementing one of the strategies in
+ "draft-ietf-dnsop-refuse-any": returning a single arbitrarily-selected
+ RRset that matches the query name rather than returning all of the
+ matching RRsets. Thanks to Tony Finch for the contribution. [RT #
+ 41615]
+
+ * named now provides feedback to the owners of zones which have trust
+ anchors configured (trusted-keys, managed-keys, dnssec-validation
+ auto; and dnssec-lookaside auto;) by sending a daily query which
+ encodes the keyids of the configured trust anchors for the zone. This
+ is controlled by trust-anchor-telemetry and defaults to yes.
+
+Feature Changes
+
+ * The logging format used for querylog has been altered. It now includes
+ an additional field indicating the address in memory of the client
+ object processing the query.
+
+ The ISC DNSSEC Lookaside Validation (DLV) service is scheduled to be
+ disabled in 2017. A warning is now logged when named is configured to
+ use this service, either explicitly or via dnssec-lookaside auto;. [RT
+ #42207]
+
+ * The timers returned by the statistics channel (indicating current
+ time, server boot time, and most recent reconfiguration time) are now
+ reported with millisecond accuracy. [RT #40082]
+
+ * Updated the compiled-in addresses for H.ROOT-SERVERS.NET and
+ L.ROOT-SERVERS.NET.
+
+ * ACLs containing geoip asnum elements were not correctly matched unless
+ the full organization name was specified in the ACL (as in geoip asnum
+ "AS1234 Example, Inc.";). They can now match against the AS number
+ alone (as in geoip asnum "AS1234";).
+
+ * When using native PKCS#11 cryptography (i.e., configure
+ --enable-native-pkcs11) HSM PINs of up to 256 characters can now be
+ used.
+
+ * NXDOMAIN responses to queries of type DS are now cached separately
+ from those for other types. This helps when using "grafted" zones of
+ type forward, for which the parent zone does not contain a delegation,
+ such as local top-level domains. Previously a query of type DS for
+ such a zone could cause the zone apex to be cached as NXDOMAIN,
+ blocking all subsequent queries. (Note: This change is only helpful
+ when DNSSEC validation is not enabled. "Grafted" zones without a
+ delegation in the parent are not a recommended configuration.)
+
+ * Update forwarding performance has been improved by allowing a single
+ TCP connection to be shared between multiple updates.
+
+ * By default, nsupdate will now check the correctness of hostnames when
+ adding records of type A, AAAA, MX, SOA, NS, SRV or PTR. This behavior
+ can be disabled with check-names no.
+
+ * Added support for OPENPGPKEY type.
+
+ * The names of the files used to store managed keys and added zones for
+ each view are no longer based on the SHA256 hash of the view name,
+ except when this is necessary because the view name contains
+ characters that would be incompatible with use as a file name. For
+ views whose names do not contain forward slashes ('/'), backslashes ('
+ \'), or capital letters - which could potentially cause namespace
+ collision problems on case-insensitive filesystems - files will now be
+ named after the view (for example, internal.mkeys or external.nzf).
+ However, to ensure consistent behavior when upgrading, if a file using
+ the old name format is found to exist, it will continue to be used.
+
+ * "rndc" can now return text output of arbitrary size to the caller.
+ (Prior to this, certain commands such as "rndc tsig-list" and "rndc
+ zonestatus" could return truncated output.)
+
+ * Errors reported when running rndc addzone (e.g., when a zone file
+ cannot be loaded) have been clarified to make it easier to diagnose
+ problems.
+
+ * When encountering an authoritative name server whose name is an alias
+ pointing to another name, the resolver treats this as an error and
+ skips to the next server. Previously this happened silently; now the
+ error will be logged to the newly-created "cname" log category.
+
+ * If named is not configured to validate answers, then allow fallback to
+ plain DNS on timeout even when we know the server supports EDNS. This
+ will allow the server to potentially resolve signed queries when TCP
+ is being blocked.
+
+ * Large inline-signing changes should be less disruptive. Signature
+ generation is now done incrementally; the number of signatures to be
+ generated in each quantum is controlled by "sig-signing-signatures
+ number;". [RT #37927]
+
+ * The experimental SIT option (code point 65001) of BIND 9.10.0 through
+ BIND 9.10.2 has been replaced with the COOKIE option (code point 10).
+ It is no longer experimental, and is sent by default, by both named
+ and dig.
+
+ The SIT-related named.conf options have been marked as obsolete, and
+ are otherwise ignored.
+
+ * When dig receives a truncated (TC=1) response or a BADCOOKIE response
+ code from a server, it will automatically retry the query using the
+ server COOKIE that was returned by the server in its initial response.
+ [RT #39047]
+
+ * Retrieving the local port range from net.ipv4.ip_local_port_range on
+ Linux is now supported.
+
+ * A new nsip-wait-recurse directive has been added to RPZ, specifying
+ whether to look up unknown name server IP addresses and wait for a
+ response before applying RPZ-NSIP rules. The default is yes. If set to
+ no, named will only apply RPZ-NSIP rules to servers whose addresses
+ are already cached. The addresses will be looked up in the background
+ so the rule can be applied on subsequent queries. This improves
+ performance when the cache is cold, at the cost of temporary
+ imprecision in applying policy directives. [RT #35009]
+
+ * Within the response-policy option, it is now possible to configure RPZ
+ rewrite logging on a per-zone basis using the log clause.
+
+ * The default preferred glue is now the address type of the transport
+ the query was received over.
+
+ * On machines with 2 or more processors (CPU), the default value for the
+ number of UDP listeners has been changed to the number of detected
+ processors minus one.
+
+ * Zone transfers now use smaller message sizes to improve message
+ compression. This results in reduced network usage.
+
+ * Added support for the AVC resource record type (Application Visibility
+ and Control).
+
+ Changed rndc reconfig behavior so that newly added zones are loaded
+ asynchronously and the loading does not block the server.
+
+ * minimal-responses now takes two new arguments: no-auth suppresses
+ populating the authority section but not the additional section;
+ no-auth-recursive does the same but only when answering recursive
+ queries.
+
+ * At server startup time, the queues for processing notify and zone
+ refresh queries are now processed in LIFO rather than FIFO order, to
+ speed up loading of newly added zones. [RT #42825]
+
+ * When answering queries of type MX or SRV, TLSA records for the target
+ name are now included in the additional section to speed up DANE
+ processing. [RT #42894]
+
+ * named can now use the TCP Fast Open mechanism on the server side, if
+ supported by the local operating system. [RT #42866]
+
+Bug Fixes
+
+ * Fixed a crash when calling rndc stats on some Windows builds: some
+ Visual Studio compilers generate code that crashes when the "%z"
+ printf() format specifier is used. [RT #42380]
+
+ * Windows installs were failing due to triggering UAC without the
+ installation binary being signed.
+
+ * A change in the internal binary representation of the RBT database
+ node structure enabled a race condition to occur (especially when BIND
+ was built with certain compilers or optimizer settings), leading to
+ inconsistent database state which caused random assertion failures.
+ [RT #42380]
+
End of Life
BIND 9.11 (Extended Support Version) will be supported until at least
December, 2021.
-See https://www.isc.org/downloads/software-support-policy/ for details of
-ISC's software support policy.
+See https://kb.isc.org/docs/aa-00896 for details of ISC's software support
+policy.
Thank You
Thank you to everyone who assisted us in making this release possible. If
you would like to contribute to ISC to assist us in continuing to make
-quality open source software, please visit our donations page at http://
+quality open source software, please visit our donations page at https://
www.isc.org/donate/.
projects / thirdparty / bind9.git / commitdiff
