notes.xml revision bee8d5b202f6a8be7ca7165b98caf3b737d9f3d0
<!DOCTYPE book [
<!ENTITY Scaron "Š">
<!ENTITY ccaron "č">
<!ENTITY mdash "—">
<!ENTITY ouml "ö">]>
<!--
- Copyright (C) 2014-2016 Internet Systems Consortium, Inc. ("ISC")
-
- purpose with or without fee is hereby granted, provided that the above
- copyright notice and this permission notice appear in all copies.
-
- THE SOFTWARE IS PROVIDED "AS IS" AND ISC DISCLAIMS ALL WARRANTIES WITH
- REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY
- AND FITNESS. IN NO EVENT SHALL ISC BE LIABLE FOR ANY SPECIAL, DIRECT,
- INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
- LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE
- OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
- PERFORMANCE OF THIS SOFTWARE.
-->
<section xml:id="relnotes_intro"><info><title>Introduction</title></info>
<para>
BIND 9.11.0 is a new feature release of BIND, still under development.
This document summarizes new features and functional changes that
have been introduced on this branch. With each development
release leading up to the final BIND 9.11.0 release, this document
will be updated with additional features added and bugs fixed.
</para>
</section>
<section xml:id="relnotes_download"><info><title>Download</title></info>
<para>
The latest versions of BIND 9 software can always be found at
<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://www.isc.org/downloads/">http://www.isc.org/downloads/</link>.
There you will find additional information about each release,
source code, and pre-compiled versions for Microsoft Windows
operating systems.
</para>
</section>
<section xml:id="relnotes_security"><info><title>Security Fixes</title></info>
<itemizedlist>
<listitem>
<para>
Duplicate EDNS COOKIE options in a response could trigger
an assertion failure. This flaw is disclosed in CVE-2016-2088.
[RT #41809]
</para>
</listitem>
<listitem>
<para>
Insufficient testing when parsing a message allowed
records with an incorrect class to be be accepted,
triggering a REQUIRE failure when those records
were subsequently cached. This flaw is disclosed
in CVE-2015-8000. [RT #40987]
</para>
</listitem>
<listitem>
<para>
Incorrect reference counting could result in an INSIST
failure if a socket error occurred while performing a
lookup. This flaw is disclosed in CVE-2015-8461. [RT#40945]
</para>
</listitem>
<listitem>
<para>
An incorrect boundary check in the OPENPGPKEY rdatatype
could trigger an assertion failure. This flaw is disclosed
in CVE-2015-5986. [RT #40286]
</para>
</listitem>
<listitem>
<para>
A buffer accounting error could trigger an assertion failure
when parsing certain malformed DNSSEC keys.
</para>
<para>
This flaw was discovered by Hanno Böck of the Fuzzing
Project, and is disclosed in CVE-2015-5722. [RT #40212]
</para>
</listitem>
<listitem>
<para>
A specially crafted query could trigger an assertion failure
in message.c.
</para>
<para>
This flaw was discovered by Jonathan Foote, and is disclosed
in CVE-2015-5477. [RT #40046]
</para>
</listitem>
<listitem>
<para>
On servers configured to perform DNSSEC validation, an
assertion failure could be triggered on answers from
a specially configured server.
</para>
<para>
This flaw was discovered by Breno Silveira Soares, and is
disclosed in CVE-2015-4620. [RT #39795]
</para>
</listitem>
<listitem>
<para>
On servers configured to perform DNSSEC validation using
managed trust anchors (i.e., keys configured explicitly
via <command>managed-keys</command>, or implicitly
via <command>dnssec-validation auto;</command> or
<command>dnssec-lookaside auto;</command>), revoking
a trust anchor and sending a new untrusted replacement
could cause <command>named</command> to crash with an
assertion failure. This could occur in the event of a
botched key rollover, or potentially as a result of a
deliberate attack if the attacker was in position to
monitor the victim's DNS traffic.
</para>
<para>
This flaw was discovered by Jan-Piet Mens, and is
disclosed in CVE-2015-1349. [RT #38344]
</para>
</listitem>
<listitem>
<para>
A flaw in delegation handling could be exploited to put
<command>named</command> into an infinite loop, in which
each lookup of a name server triggered additional lookups
of more name servers. This has been addressed by placing
limits on the number of levels of recursion
<command>named</command> will allow (default 7), and
on the number of queries that it will send before
terminating a recursive query (default 50).
</para>
<para>
The recursion depth limit is configured via the
<option>max-recursion-depth</option> option, and the query limit
via the <option>max-recursion-queries</option> option.
</para>
<para>
The flaw was discovered by Florian Maury of ANSSI, and is
disclosed in CVE-2014-8500. [RT #37580]
</para>
</listitem>
<listitem>
<para>
Two separate problems were identified in BIND's GeoIP code that
could lead to an assertion failure. One was triggered by use of
both IPv4 and IPv6 address families, the other by referencing
not installed. Both are covered by CVE-2014-8680. [RT #37672]
[RT #37679]
</para>
<para>
A less serious security flaw was also found in GeoIP: changes
to the <command>geoip-directory</command> option in
<command>rndc reconfig</command>. In theory, this could allow
<command>named</command> to allow access to unintended clients.
</para>
</listitem>
<listitem>
<para>
Specific APL data could trigger an INSIST. This flaw
is disclosed in CVE-2015-8704. [RT #41396]
</para>
</listitem>
<listitem>
<para>
Certain errors that could be encountered when printing out
or logging an OPT record containing a CLIENT-SUBNET option
could be mishandled, resulting in an assertion failure.
This flaw is disclosed in CVE-2015-8705. [RT #41397]
</para>
</listitem>
<listitem>
<para>
Malformed control messages can trigger assertions in named
and rndc. This flaw is disclosed in CVE-2016-1285. [RT
#41666]
</para>
</listitem>
<listitem>
<para>
The resolver could abort with an assertion failure due to
improper DNAME handling when parsing fetch reply
messages. This flaw is disclosed in CVE-2016-1286. [RT #41753]
</para>
</listitem>
</itemizedlist>
</section>
<section xml:id="relnotes_features"><info><title>New Features</title></info>
<itemizedlist>
<listitem>
<para>
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.)
</para>
<para>
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.
</para>
<para>
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.
</para>
</listitem>
<listitem>
<para>
New quotas have been added to limit the queries that are
sent by recursive resolvers to authoritative servers
experiencing denial-of-service attacks. When configured,
these options can both reduce the harm done to authoritative
servers and also avoid the resource exhaustion that can be
experienced by recursives when they are being used as a
vehicle for such an attack.
</para>
<itemizedlist>
<listitem>
<para>
<option>fetches-per-server</option> 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
<option>fetch-quota-params</option> option.
</para>
</listitem>
<listitem>
<para>
<option>fetches-per-zone</option> 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.)
</para>
</listitem>
</itemizedlist>
<para>
Statistics counters have also been added to track the number
of queries affected by these quotas.
</para>
</listitem>
<listitem>
<para>
Added support for <command>dnstap</command>, a fast,
flexible method for capturing and logging DNS traffic,
developed by Robert Edmonds at Farsight Security, Inc.,
whose assistance is gratefully acknowledged.
</para>
<para>
To enable <command>dnstap</command> at compile time,
the <command>fstrm</command> and <command>protobuf-c</command>
libraries must be available, and BIND must be configured with
<option>--enable-dnstap</option>.
</para>
<para>
A new utility <command>dnstap-read</command> has been added
to allow <command>dnstap</command> data to be presented in
a human-readable format.
</para>
<para>
For more information on <command>dnstap</command>, see
<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://dnstap.info">http://dnstap.info</link>.
</para>
</listitem>
<listitem>
<para>
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,
<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://localhost:8888/xml/v3/traffic">http://localhost:8888/xml/v3/traffic</link>
or
<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://localhost:8888/json/v1/traffic">http://localhost:8888/json/v1/traffic</link>.
</para>
</listitem>
<listitem>
<para>
The serial number of a dynamically updatable zone can
now be set using
<command>rndc signing -serial <replaceable>number</replaceable> <replaceable>zonename</replaceable></command>.
This is particularly useful with <option>inline-signing</option>
zones that have been reset. Setting the serial number to a value
larger than that on the slaves will trigger an AXFR-style
transfer.
</para>
</listitem>
<listitem>
<para>
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 serviers. The SERVFAIL cache timeout is controlled
by <option>servfail-ttl</option>, which defaults to 1 second
and has an upper limit of 30.
</para>
</listitem>
<listitem>
<para>
The new <command>rndc nta</command> 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 <option>nta-lifetime</option> in
file (<filename><replaceable>viewname</replaceable>.nta</filename>)
in order to persist across restarts of the <command>named</command> server.
</para>
</listitem>
<listitem>
<para>
The EDNS Client Subnet (ECS) option is now supported for
authoritative servers; if a query contains an ECS option then
ACLs containing <option>geoip</option> or <option>ecs</option>
elements can match against the 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.
</para>
</listitem>
<listitem>
<para>
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.
</para>
</listitem>
<listitem>
<para>
A new <option>masterfile-style</option> zone option controls
the formatting of text zone files: When set to
<literal>full</literal>, the zone file will dumped in
single-line-per-record format.
</para>
</listitem>
<listitem>
<para>
<command>dig +ednsopt</command> can now be used to set
arbitrary EDNS options in DNS requests.
</para>
</listitem>
<listitem>
<para>
<command>dig +ednsflags</command> can now be used to set
yet-to-be-defined EDNS flags in DNS requests.
</para>
</listitem>
<listitem>
<para>
<command>dig +[no]ednsnegotiation</command> can now be used enable /
disable EDNS version negotiation.
</para>
</listitem>
<listitem>
<para>
<command>dig +header-only</command> can now be used to send
queries without a question section.
</para>
</listitem>
<listitem>
<para>
<command>dig +ttlunits</command> causes <command>dig</command>
to print TTL values with time-unit suffixes: w, d, h, m, s for
weeks, days, hours, minutes, and seconds.
</para>
</listitem>
<listitem>
<para>
<command>dig +zflag</command> can be used to set the last
unassigned DNS header flag bit. This bit in normally zero.
</para>
</listitem>
<listitem>
<para>
<command>dig +dscp=<replaceable>value</replaceable></command>
can now be used to set the DSCP code point in outgoing query
packets.
</para>
</listitem>
<listitem>
<para>
<command>dig +mapped</command> can now be used to determine
if mapped IPv4 addresses can be used.
</para>
</listitem>
<listitem>
<para>
<option>serial-update-method</option> can now be set to
<literal>date</literal>. On update, the serial number will
be set to the current date in YYYYMMDDNN format.
</para>
</listitem>
<listitem>
<para>
<command>dnssec-signzone -N date</command> also sets the serial
number to YYYYMMDDNN.
</para>
</listitem>
<listitem>
<para>
<command>named -L <replaceable>filename</replaceable></command>
causes <command>named</command> to send log messages to the specified file by
default instead of to the system log.
</para>
</listitem>
<listitem>
<para>
The rate limiter configured by the
<option>serial-query-rate</option> option no longer covers
NOTIFY messages; those are now separately controlled by
<option>notify-rate</option> and
<option>startup-notify-rate</option> (the latter of which
controls the rate of NOTIFY messages sent when the server
is first started up or reconfigured).
</para>
</listitem>
<listitem>
<para>
The default number of tasks and client objects available
for serving lightweight resolver queries have been increased,
and are now configurable via the new <option>lwres-tasks</option>
and <option>lwres-clients</option> options in
</para>
</listitem>
<listitem>
<para>
Log output to files can now be buffered by specifying
<command>buffered yes;</command> when creating a channel.
</para>
</listitem>
<listitem>
<para>
<command>delv +tcp</command> will exclusively use TCP when
sending queries.
</para>
</listitem>
<listitem>
<para>
<command>named</command> 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 <option>lock-file</option> option or
the <command>-X</command> command line option. The
default lock file is
Specifying <literal>none</literal> will disable the lock
file check.
</para>
</listitem>
<listitem>
<para>
<command>rndc delzone</command> can now be applied to zones
it is no longer restricted to zones which were added by
<command>rndc addzone</command>. (Note, however, that
must be removed from the configuration or it will return
when <command>named</command> is restarted or reloaded.)
</para>
</listitem>
<listitem>
<para>
<command>rndc modzone</command> can be used to reconfigure
a zone, using similar syntax to <command>rndc addzone</command>.
</para>
</listitem>
<listitem>
<para>
<command>rndc showzone</command> displays the current
configuration for a specified zone.
</para>
</listitem>
<listitem>
<para>
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.
</para>
<para>
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; };".
</para>
</listitem>
<listitem>
<para>
The new <command>mdig</command> command is a version of
<command>dig</command> 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]
</para>
</listitem>
<listitem>
<para>
To enable better monitoring and troubleshooting of RFC 5011
trust anchor management, the new <command>rndc managed-keys</command>
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]
</para>
</listitem>
<listitem>
<para>
An <command>--enable-querytrace</command> configure switch is
now available to enable very verbose query tracelogging. 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]
</para>
</listitem>
<listitem>
<para>
A new <command>tcp-only</command> option can be specified
in <command>server</command> statements to force
<command>named</command> to connect to the specified
server via TCP. [RT #37800]
</para>
</listitem>
<listitem>
<para>
The <command>nxdomain-redirect</command> 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 <command>type redirect</command> zone, has
better average performance but is less flexible.) [RT #37989]
</para>
</listitem>
<listitem>
<para>
The following types have been implemented: CSYNC, NINFO, RKEY,
SINK, TA, TALINK.
</para>
</listitem>
<listitem>
<para>
A new <command>message-compression</command> option can be
used to specify whether or not to use name compression when
answering queries. Setting this to <userinput>no</userinput>
results in larger responses, but reduces CPU consumption and
may improve throughput. The default is <userinput>yes</userinput>.
</para>
</listitem>
<listitem>
<para>
A "read-only" clause is now available for non-destructive
control channel access. In such cases, a restricted set of
rndc commands are allowed for querying information from named.
By default, control channel access is read-write.
</para>
</listitem>
<listitem>
<para>
When loading managed signed zones detect if the RRSIG's
inception time is in the future and regenerate the RRSIG
immediately. This helps when the system's clock needs to
be reset backwards.
</para>
</listitem>
</itemizedlist>
</section>
<section xml:id="relnotes_changes"><info><title>Feature Changes</title></info>
<itemizedlist>
<listitem>
<para>
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]
</para>
</listitem>
<listitem>
<para>
Updated the compiled in addresses for H.ROOT-SERVERS.NET.
</para>
</listitem>
<listitem>
<para>
ACLs containing <command>geoip asnum</command> elements were
not correctly matched unless the full organization name was
specified in the ACL (as in
<command>geoip asnum "AS1234 Example, Inc.";</command>).
They can now match against the AS number alone (as in
<command>geoip asnum "AS1234";</command>).
</para>
</listitem>
<listitem>
<para>
When using native PKCS#11 cryptography (i.e.,
<command>configure --enable-native-pkcs11</command>) HSM PINs
of up to 256 characters can now be used.
</para>
</listitem>
<listitem>
<para>
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.)
</para>
</listitem>
<listitem>
<para>
Update forwarding performance has been improved by allowing
a single TCP connection to be shared between multiple updates.
</para>
</listitem>
<listitem>
<para>
By default, <command>nsupdate</command> 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 <command>check-names no</command>.
</para>
</listitem>
<listitem>
<para>
Added support for OPENPGPKEY type.
</para>
</listitem>
<listitem>
<para>
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
consistent behavior when upgrading, if a file using the old
name format is found to exist, it will continue to be used.
</para>
</listitem>
<listitem>
<para>
"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.)
</para>
</listitem>
<listitem>
<para>
Errors reported when running <command>rndc addzone</command>
(e.g., when a zone file cannot be loaded) have been clarified
to make it easier to diagnose problems.
</para>
</listitem>
<listitem>
<para>
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.
</para>
</listitem>
<listitem>
<para>
If <command>named</command> 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.
</para>
</listitem>
<listitem>
<para>
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 <replaceable>number</replaceable>;".
[RT #37927]
</para>
</listitem>
<listitem>
<para>
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 <command>named</command> and
<command>dig</command>.
</para>
<para>
The SIT-related named.conf options have been marked as
obsolete, and are otherwise ignored.
</para>
</listitem>
<listitem>
<para>
When <command>dig</command> 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]
</para>
</listitem>
<listitem>
<para>
A alternative NXDOMAIN redirect method (nxdomain-redirect)
which allows the redirect information to be looked up from
a namespace on the Internet rather than requiring a zone
to be configured on the server is now available.
</para>
</listitem>
<listitem>
<para>
Retrieving the local port range from net.ipv4.ip_local_port_range
on Linux is now supported.
</para>
</listitem>
<listitem>
<para>
Within the <option>response-policy</option> option, it is now
possible to configure RPZ rewrite logging on a per-zone basis
using the <option>log</option> clause.
</para>
</listitem>
<listitem>
<para>
The default preferred glue is now the address type of the
transport the query was received over.
</para>
</listitem>
<listitem>
<para>
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.
</para>
</listitem>
<listitem>
<para>
Zone transfers now use smaller message sizes to improve
message compression. This results in reduced network usage.
</para>
</listitem>
<listitem>
<para>
Added support for the type AVC.
</para>
</listitem>
</itemizedlist>
</section>
<section xml:id="relnotes_port"><info><title>Porting Changes</title></info>
<itemizedlist>
<listitem>
<para>
None.
</para>
</listitem>
</itemizedlist>
</section>
<section xml:id="relnotes_bugs"><info><title>Bug Fixes</title></info>
<itemizedlist>
<listitem>
<para>
None.
</para>
</listitem>
</itemizedlist>
</section>
<section xml:id="end_of_life"><info><title>End of Life</title></info>
<para>
The end of life for BIND 9.11 is yet to be determined but
will not be before BIND 9.13.0 has been released for 6 months.
<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://www.isc.org/downloads/software-support-policy/">https://www.isc.org/downloads/software-support-policy/</link>
</para>
</section>
<section xml:id="relnotes_thanks"><info><title>Thank You</title></info>
<para>
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
<link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://www.isc.org/donate/">http://www.isc.org/donate/</link>.
</para>
</section>
</section>