1 .TH "unbound.conf" "5" "Mar 21, 2013" "NLnet Labs" "unbound 1.4.20"
3 .\" unbound.conf.5 -- unbound.conf manual
5 .\" Copyright (c) 2007, NLnet Labs. All rights reserved.
7 .\" See LICENSE for the license.
13 \- Unbound configuration file.
22 The file format has attributes and values. Some attributes have attributes inside them.
23 The notation is: attribute: value.
25 Comments start with # and last to the end of line. Empty lines are
26 ignored as is whitespace at the beginning of a line.
29 \fIunbound\-checkconf\fR(8)
30 can be used to check unbound.conf prior to usage.
32 An example config file is shown below. Copy this to /etc/unbound/unbound.conf
33 and start the server with:
36 $ unbound \-c /etc/unbound/unbound.conf
39 Most settings are the defaults. Stop the server with:
42 $ kill `cat /etc/unbound/unbound.pid`
45 Below is a minimal config file. The source distribution contains an extensive
46 example.conf file with all the options.
49 # unbound.conf(5) config file for unbound(8).
51 directory: "/etc/unbound"
53 # make sure unbound can access entropy from inside the chroot.
54 # e.g. on linux the use these commands (on BSD, devfs(8) is used):
55 # mount \-\-bind \-n /dev/random /etc/unbound/dev/random
56 # and mount \-\-bind \-n /dev/log /etc/unbound/dev/log
57 chroot: "/etc/unbound"
58 # logfile: "/etc/unbound/unbound.log" #uncomment to use logfile.
59 pidfile: "/etc/unbound/unbound.pid"
60 # verbosity: 1 # uncomment and increase to get more logging.
61 # listen on all interfaces, answer queries from the local subnet.
64 access\-control: 10.0.0.0/8 allow
65 access\-control: 2001:DB8::/64 allow
69 There must be whitespace between keywords. Attribute keywords end with a colon ':'. An attribute
70 is followed by its containing attributes, or a value.
72 Files can be included using the
74 directive. It can appear anywhere, it accepts a single file name as argument.
75 Processing continues as if the text from the included file was copied into
76 the config file at that point. If also using chroot, using full path names
77 for the included files works, relative pathnames for the included names work
78 if the directory where the daemon is started equals its chroot/working
79 directory. Wildcards can be used to include multiple files, see \fIglob\fR(7).
81 These options are part of the
85 .B verbosity: \fI<number>
86 The verbosity number, level 0 means no verbosity, only errors. Level 1
87 gives operational information. Level 2 gives detailed operational
88 information. Level 3 gives query level information, output per query.
89 Level 4 gives algorithm level information. Level 5 logs client
90 identification for cache misses. Default is level 1.
91 The verbosity can also be increased from the commandline, see \fIunbound\fR(8).
93 .B statistics\-interval: \fI<seconds>
94 The number of seconds between printing statistics to the log for every thread.
95 Disable with value 0 or "". Default is disabled. The histogram statistics
96 are only printed if replies were sent during the statistics interval,
97 requestlist statistics are printed for every interval (but can be 0).
98 This is because the median calculation requires data to be present.
100 .B statistics\-cumulative: \fI<yes or no>
101 If enabled, statistics are cumulative since starting unbound, without clearing
102 the statistics counters after logging the statistics. Default is no.
104 .B extended\-statistics: \fI<yes or no>
105 If enabled, extended statistics are printed from \fIunbound\-control\fR(8).
106 Default is off, because keeping track of more statistics takes time. The
107 counters are listed in \fIunbound\-control\fR(8).
109 .B num\-threads: \fI<number>
110 The number of threads to create to serve clients. Use 1 for no threading.
112 .B port: \fI<port number>
113 The port number, default 53, on which the server responds to queries.
115 .B interface: \fI<ip address[@port]>
116 Interface to use to connect to the network. This interface is listened to
117 for queries from clients, and answers to clients are given from it.
118 Can be given multiple times to work on several interfaces. If none are
119 given the default is to listen to localhost.
120 The interfaces are not changed on a reload (kill \-HUP) but only on restart.
121 A port number can be specified with @port (without spaces between
122 interface and port number), if not specified the default port (from
125 .B interface\-automatic: \fI<yes or no>
126 Detect source interface on UDP queries and copy them to replies. This
127 feature is experimental, and needs support in your OS for particular socket
128 options. Default value is no.
130 .B outgoing\-interface: \fI<ip address>
131 Interface to use to connect to the network. This interface is used to send
132 queries to authoritative servers and receive their replies. Can be given
133 multiple times to work on several interfaces. If none are given the
134 default (all) is used. You can specify the same interfaces in
137 .B outgoing\-interface:
138 lines, the interfaces are then used for both purposes. Outgoing queries are
139 sent via a random outgoing interface to counter spoofing.
141 .B outgoing\-range: \fI<number>
142 Number of ports to open. This number of file descriptors can be opened per
143 thread. Must be at least 1. Default depends on compile options. Larger
144 numbers need extra resources from the operating system. For performance a
145 a very large value is best, use libevent to make this possible.
147 .B outgoing\-port\-permit: \fI<port number or range>
148 Permit unbound to open this port or range of ports for use to send queries.
149 A larger number of permitted outgoing ports increases resilience against
150 spoofing attempts. Make sure these ports are not needed by other daemons.
151 By default only ports above 1024 that have not been assigned by IANA are used.
152 Give a port number or a range of the form "low\-high", without spaces.
154 The \fBoutgoing\-port\-permit\fR and \fBoutgoing\-port\-avoid\fR statements
155 are processed in the line order of the config file, adding the permitted ports
156 and subtracting the avoided ports from the set of allowed ports. The
157 processing starts with the non IANA allocated ports above 1024 in the set
160 .B outgoing\-port\-avoid: \fI<port number or range>
161 Do not permit unbound to open this port or range of ports for use to send
162 queries. Use this to make sure unbound does not grab a port that another
163 daemon needs. The port is avoided on all outgoing interfaces, both IP4 and IP6.
164 By default only ports above 1024 that have not been assigned by IANA are used.
165 Give a port number or a range of the form "low\-high", without spaces.
167 .B outgoing\-num\-tcp: \fI<number>
168 Number of outgoing TCP buffers to allocate per thread. Default is 10. If set
169 to 0, or if do_tcp is "no", no TCP queries to authoritative servers are done.
171 .B incoming\-num\-tcp: \fI<number>
172 Number of incoming TCP buffers to allocate per thread. Default is 10. If set
173 to 0, or if do_tcp is "no", no TCP queries from clients are accepted.
175 .B edns\-buffer\-size: \fI<number>
176 Number of bytes size to advertise as the EDNS reassembly buffer size.
177 This is the value put into datagrams over UDP towards peers. The actual
178 buffer size is determined by msg\-buffer\-size (both for TCP and UDP). Do
179 not set higher than that value. Default is 4096 which is RFC recommended.
180 If you have fragmentation reassembly problems, usually seen as timeouts,
181 then a value of 1480 can fix it. Setting to 512 bypasses even the most
182 stringent path MTU problems, but is seen as extreme, since the amount
183 of TCP fallback generated is excessive (probably also for this resolver,
184 consider tuning the outgoing tcp number).
186 .B msg\-buffer\-size: \fI<number>
187 Number of bytes size of the message buffers. Default is 65552 bytes, enough
188 for 64 Kb packets, the maximum DNS message size. No message larger than this
189 can be sent or received. Can be reduced to use less memory, but some requests
190 for DNS data, such as for huge resource records, will result in a SERVFAIL
193 .B msg\-cache\-size: \fI<number>
194 Number of bytes size of the message cache. Default is 4 megabytes.
195 A plain number is in bytes, append 'k', 'm' or 'g' for kilobytes, megabytes
196 or gigabytes (1024*1024 bytes in a megabyte).
198 .B msg\-cache\-slabs: \fI<number>
199 Number of slabs in the message cache. Slabs reduce lock contention by threads.
200 Must be set to a power of 2. Setting (close) to the number of cpus is a
203 .B num\-queries\-per\-thread: \fI<number>
204 The number of queries that every thread will service simultaneously.
205 If more queries arrive that need servicing, and no queries can be jostled out
206 (see \fIjostle\-timeout\fR), then the queries are dropped. This forces
207 the client to resend after a timeout; allowing the server time to work on
208 the existing queries. Default depends on compile options, 512 or 1024.
210 .B jostle\-timeout: \fI<msec>
211 Timeout used when the server is very busy. Set to a value that usually
212 results in one roundtrip to the authority servers. If too many queries
213 arrive, then 50% of the queries are allowed to run to completion, and
214 the other 50% are replaced with the new incoming query if they have already
215 spent more than their allowed time. This protects against denial of
216 service by slow queries or high query rates. Default 200 milliseconds.
217 The effect is that the qps for long-lasting queries is about
218 (numqueriesperthread / 2) / (average time for such long queries) qps.
219 The qps for short queries can be about (numqueriesperthread / 2)
220 / (jostletimeout in whole seconds) qps per thread, about (1024/2)*5 = 2560
223 .B so\-rcvbuf: \fI<number>
224 If not 0, then set the SO_RCVBUF socket option to get more buffer
225 space on UDP port 53 incoming queries. So that short spikes on busy
226 servers do not drop packets (see counter in netstat \-su). Default is
227 0 (use system value). Otherwise, the number of bytes to ask for, try
228 "4m" on a busy server. The OS caps it at a maximum, on linux unbound
229 needs root permission to bypass the limit, or the admin can use sysctl
230 net.core.rmem_max. On BSD change kern.ipc.maxsockbuf in /etc/sysctl.conf.
231 On OpenBSD change header and recompile kernel. On Solaris ndd \-set
232 /dev/udp udp_max_buf 8388608.
234 .B so\-sndbuf: \fI<number>
235 If not 0, then set the SO_SNDBUF socket option to get more buffer space on
236 UDP port 53 outgoing queries. This for very busy servers handles spikes
237 in answer traffic, otherwise 'send: resource temporarily unavailable'
238 can get logged, the buffer overrun is also visible by netstat \-su.
239 Default is 0 (use system value). Specify the number of bytes to ask
240 for, try "4m" on a very busy server. The OS caps it at a maximum, on
241 linux unbound needs root permission to bypass the limit, or the admin
242 can use sysctl net.core.wmem_max. On BSD, Solaris changes are similar
245 .B rrset\-cache\-size: \fI<number>
246 Number of bytes size of the RRset cache. Default is 4 megabytes.
247 A plain number is in bytes, append 'k', 'm' or 'g' for kilobytes, megabytes
248 or gigabytes (1024*1024 bytes in a megabyte).
250 .B rrset\-cache\-slabs: \fI<number>
251 Number of slabs in the RRset cache. Slabs reduce lock contention by threads.
252 Must be set to a power of 2.
254 .B cache\-max\-ttl: \fI<seconds>
255 Time to live maximum for RRsets and messages in the cache. Default is
256 86400 seconds (1 day). If the maximum kicks in, responses to clients
257 still get decrementing TTLs based on the original (larger) values.
258 When the internal TTL expires, the cache item has expired.
259 Can be set lower to force the resolver to query for data often, and not
260 trust (very large) TTL values.
262 .B cache\-min\-ttl: \fI<seconds>
263 Time to live minimum for RRsets and messages in the cache. Default is 0.
264 If the the minimum kicks in, the data is cached for longer than the domain
265 owner intended, and thus less queries are made to look up the data.
266 Zero makes sure the data in the cache is as the domain owner intended,
267 higher values, especially more than an hour or so, can lead to trouble as
268 the data in the cache does not match up with the actual data any more.
270 .B infra\-host\-ttl: \fI<seconds>
271 Time to live for entries in the host cache. The host cache contains
272 roundtrip timing, lameness and EDNS support information. Default is 900.
274 .B infra\-cache\-slabs: \fI<number>
275 Number of slabs in the infrastructure cache. Slabs reduce lock contention
276 by threads. Must be set to a power of 2.
278 .B infra\-cache\-numhosts: \fI<number>
279 Number of hosts for which information is cached. Default is 10000.
281 .B do\-ip4: \fI<yes or no>
282 Enable or disable whether ip4 queries are answered or issued. Default is yes.
284 .B do\-ip6: \fI<yes or no>
285 Enable or disable whether ip6 queries are answered or issued. Default is yes.
286 If disabled, queries are not answered on IPv6, and queries are not sent on
287 IPv6 to the internet nameservers.
289 .B do\-udp: \fI<yes or no>
290 Enable or disable whether UDP queries are answered or issued. Default is yes.
292 .B do\-tcp: \fI<yes or no>
293 Enable or disable whether TCP queries are answered or issued. Default is yes.
295 .B tcp\-upstream: \fI<yes or no>
296 Enable or disable whether the upstream queries use TCP only for transport.
297 Default is no. Useful in tunneling scenarios.
299 .B ssl\-upstream: \fI<yes or no>
300 Enabled or disable whether the upstream queries use SSL only for transport.
301 Default is no. Useful in tunneling scenarios. The SSL contains plain DNS in
302 TCP wireformat. The other server must support this (see \fBssl\-service\-key\fR).
304 .B ssl\-service-key: \fI<file>
305 If enabled, the server provider SSL service on its TCP sockets. The clients
306 have to use ssl\-upstream: yes. The file is the private key for the TLS
307 session. The public certificate is in the ssl\-service\-pem file. Default
308 is "", turned off. Requires a restart (a reload is not enough) if changed,
309 because the private key is read while root permissions are held and before
310 chroot (if any). Normal DNS TCP service is not provided and gives errors,
311 this service is best run with a different \fBport:\fR config or \fI@port\fR
312 suffixes in the \fBinterface\fR config.
314 .B ssl\-service\-pem: \fI<file>
315 The public key certificate pem file for the ssl service. Default is "",
318 .B ssl\-port: \fI<number>
319 The port number on which to provide TCP SSL service, default 443, only
320 interfaces configured with that port number as @number get the SSL service.
322 .B do\-daemonize: \fI<yes or no>
323 Enable or disable whether the unbound server forks into the background as
324 a daemon. Default is yes.
326 .B access\-control: \fI<IP netblock> <action>
327 The netblock is given as an IP4 or IP6 address with /size appended for a
328 classless network block. The action can be \fIdeny\fR, \fIrefuse\fR,
329 \fIallow\fR or \fIallow_snoop\fR.
331 The action \fIdeny\fR stops queries from hosts from that netblock.
333 The action \fIrefuse\fR stops queries too, but sends a DNS rcode REFUSED
336 The action \fIallow\fR gives access to clients from that netblock.
337 It gives only access for recursion clients (which is
338 what almost all clients need). Nonrecursive queries are refused.
340 The \fIallow\fR action does allow nonrecursive queries to access the
341 local\-data that is configured. The reason is that this does not involve
342 the unbound server recursive lookup algorithm, and static data is served
343 in the reply. This supports normal operations where nonrecursive queries
344 are made for the authoritative data. For nonrecursive queries any replies
345 from the dynamic cache are refused.
347 The action \fIallow_snoop\fR gives nonrecursive access too. This give
348 both recursive and non recursive access. The name \fIallow_snoop\fR refers
349 to cache snooping, a technique to use nonrecursive queries to examine
350 the cache contents (for malicious acts). However, nonrecursive queries can
351 also be a valuable debugging tool (when you want to examine the cache
352 contents). In that case use \fIallow_snoop\fR for your administration host.
354 By default only localhost is \fIallow\fRed, the rest is \fIrefuse\fRd.
355 The default is \fIrefuse\fRd, because that is protocol\-friendly. The DNS
356 protocol is not designed to handle dropped packets due to policy, and
357 dropping may result in (possibly excessive) retried queries.
359 .B chroot: \fI<directory>
360 If chroot is enabled, you should pass the configfile (from the
361 commandline) as a full path from the original root. After the
362 chroot has been performed the now defunct portion of the config
363 file path is removed to be able to reread the config after a reload.
365 All other file paths (working dir, logfile, roothints, and
366 key files) can be specified in several ways:
367 as an absolute path relative to the new root,
368 as a relative path to the working directory, or
369 as an absolute path relative to the original root.
370 In the last case the path is adjusted to remove the unused portion.
372 The pidfile can be either a relative path to the working directory, or
373 an absolute path relative to the original root. It is written just prior
374 to chroot and dropping permissions. This allows the pidfile to be
375 /var/run/unbound.pid and the chroot to be /var/unbound, for example.
377 Additionally, unbound may need to access /dev/random (for entropy)
378 from inside the chroot.
380 If given a chroot is done to the given directory. The default is
381 "/var/unbound". If you give "" no chroot is performed.
383 .B username: \fI<name>
384 If given, after binding the port the user privileges are dropped. Default is
385 "unbound". If you give username: "" no user change is performed.
387 If this user is not capable of binding the
388 port, reloads (by signal HUP) will still retain the opened ports.
389 If you change the port number in the config file, and that new port number
390 requires privileges, then a reload will fail; a restart is needed.
392 .B directory: \fI<directory>
393 Sets the working directory for the program. Default is "/var/unbound".
395 .B logfile: \fI<filename>
396 If "" is given, logging goes to stderr, or nowhere once daemonized.
397 The logfile is appended to, in the following format:
399 [seconds since 1970] unbound[pid:tid]: type: message.
401 If this option is given, the use\-syslog is option is set to "no".
402 The logfile is reopened (for append) when the config file is reread, on
405 .B use\-syslog: \fI<yes or no>
406 Sets unbound to send log messages to the syslogd, using
408 The log facility LOG_DAEMON is used, with identity "unbound".
409 The logfile setting is overridden when use\-syslog is turned on.
410 The default is to log to syslog.
412 .B log\-time\-ascii: \fI<yes or no>
413 Sets logfile lines to use a timestamp in UTC ascii. Default is no, which
414 prints the seconds since 1970 in brackets. No effect if using syslog, in
415 that case syslog formats the timestamp printed into the log files.
417 .B log\-queries: \fI<yes or no>
418 Prints one line per query to the log, with the log timestamp and IP address,
419 name, type and class. Default is no. Note that it takes time to print these
420 lines which makes the server (significantly) slower. Odd (nonprintable)
421 characters in names are printed as '?'.
423 .B pidfile: \fI<filename>
424 The process id is written to the file. Default is "/var/unbound/unbound.pid".
427 kill \-HUP `cat /var/unbound/unbound.pid`
431 kill \-QUIT `cat /var/unbound/unbound.pid`
433 gracefully terminates.
435 .B root\-hints: \fI<filename>
436 Read the root hints from this file. Default is nothing, using builtin hints
437 for the IN class. The file has the format of zone files, with root
438 nameserver names and addresses only. The default may become outdated,
439 when servers change, therefore it is good practice to use a root\-hints file.
441 .B hide\-identity: \fI<yes or no>
442 If enabled id.server and hostname.bind queries are refused.
444 .B identity: \fI<string>
445 Set the identity to report. If set to "", the default, then the hostname
446 of the server is returned.
448 .B hide\-version: \fI<yes or no>
449 If enabled version.server and version.bind queries are refused.
451 .B version: \fI<string>
452 Set the version to report. If set to "", the default, then the package
455 .B target\-fetch\-policy: \fI<"list of numbers">
456 Set the target fetch policy used by unbound to determine if it should fetch
457 nameserver target addresses opportunistically. The policy is described per
460 The number of values determines the maximum dependency depth
461 that unbound will pursue in answering a query.
462 A value of \-1 means to fetch all targets opportunistically for that dependency
463 depth. A value of 0 means to fetch on demand only. A positive value fetches
464 that many targets opportunistically.
466 Enclose the list between quotes ("") and put spaces between numbers.
467 The default is "3 2 1 0 0". Setting all zeroes, "0 0 0 0 0" gives behaviour
468 closer to that of BIND 9, while setting "\-1 \-1 \-1 \-1 \-1" gives behaviour
469 rumoured to be closer to that of BIND 8.
471 .B harden\-short\-bufsize: \fI<yes or no>
472 Very small EDNS buffer sizes from queries are ignored. Default is off, since
473 it is legal protocol wise to send these, and unbound tries to give very
474 small answers to these queries, where possible.
476 .B harden\-large\-queries: \fI<yes or no>
477 Very large queries are ignored. Default is off, since it is legal protocol
478 wise to send these, and could be necessary for operation if TSIG or EDNS
479 payload is very large.
481 .B harden\-glue: \fI<yes or no>
482 Will trust glue only if it is within the servers authority. Default is on.
484 .B harden\-dnssec\-stripped: \fI<yes or no>
485 Require DNSSEC data for trust\-anchored zones, if such data is absent,
486 the zone becomes bogus. If turned off, and no DNSSEC data is received
487 (or the DNSKEY data fails to validate), then the zone is made insecure,
488 this behaves like there is no trust anchor. You could turn this off if
489 you are sometimes behind an intrusive firewall (of some sort) that
490 removes DNSSEC data from packets, or a zone changes from signed to
491 unsigned to badly signed often. If turned off you run the risk of a
492 downgrade attack that disables security for a zone. Default is on.
494 .B harden\-below\-nxdomain: \fI<yes or no>
495 From draft-vixie-dnsext-resimprove, returns nxdomain to queries for a name
496 below another name that is already known to be nxdomain. DNSSEC mandates
497 noerror for empty nonterminals, hence this is possible. Very old software
498 might return nxdomain for empty nonterminals (that usually happen for reverse
499 IP address lookups), and thus may be incompatible with this. To try to avoid
500 this only DNSSEC-secure nxdomains are used, because the old software does not
501 have DNSSEC. Default is off.
503 .B harden\-referral\-path: \fI<yes or no>
504 Harden the referral path by performing additional queries for
505 infrastructure data. Validates the replies if trust anchors are configured
506 and the zones are signed. This enforces DNSSEC validation on nameserver
507 NS sets and the nameserver addresses that are encountered on the referral
509 Default off, because it burdens the authority servers, and it is
510 not RFC standard, and could lead to performance problems because of the
511 extra query load that is generated. Experimental option.
512 If you enable it consider adding more numbers after the target\-fetch\-policy
513 to increase the max depth that is checked to.
515 .B use\-caps\-for\-id: \fI<yes or no>
516 Use 0x20\-encoded random bits in the query to foil spoof attempts.
517 This perturbs the lowercase and uppercase of query names sent to
518 authority servers and checks if the reply still has the correct casing.
520 This feature is an experimental implementation of draft dns\-0x20.
522 .B private\-address: \fI<IP address or subnet>
523 Give IPv4 of IPv6 addresses or classless subnets. These are addresses
524 on your private network, and are not allowed to be returned for public
525 internet names. Any occurence of such addresses are removed from
526 DNS answers. Additionally, the DNSSEC validator may mark the answers
527 bogus. This protects against so\-called DNS Rebinding, where a user browser
528 is turned into a network proxy, allowing remote access through the browser
529 to other parts of your private network. Some names can be allowed to
530 contain your private addresses, by default all the \fBlocal\-data\fR
531 that you configured is allowed to, and you can specify additional
532 names using \fBprivate\-domain\fR. No private addresses are enabled
533 by default. We consider to enable this for the RFC1918 private IP
534 address space by default in later releases. That would enable private
535 addresses for 10.0.0.0/8 172.16.0.0/12 192.168.0.0/16 169.254.0.0/16
536 fd00::/8 and fe80::/10, since the RFC standards say these addresses
537 should not be visible on the public internet. Turning on 127.0.0.0/8
538 would hinder many spamblocklists as they use that.
540 .B private\-domain: \fI<domain name>
541 Allow this domain, and all its subdomains to contain private addresses.
542 Give multiple times to allow multiple domain names to contain private
543 addresses. Default is none.
545 .B unwanted\-reply\-threshold: \fI<number>
546 If set, a total number of unwanted replies is kept track of in every thread.
547 When it reaches the threshold, a defensive action is taken and a warning
548 is printed to the log. The defensive action is to clear the rrset and
549 message caches, hopefully flushing away any poison. A value of 10 million
550 is suggested. Default is 0 (turned off).
552 .B do\-not\-query\-address: \fI<IP address>
553 Do not query the given IP address. Can be IP4 or IP6. Append /num to
554 indicate a classless delegation netblock, for example like
555 10.2.3.4/24 or 2001::11/64.
557 .B do\-not\-query\-localhost: \fI<yes or no>
558 If yes, localhost is added to the do\-not\-query\-address entries, both
559 IP6 ::1 and IP4 127.0.0.1/8. If no, then localhost can be used to send
560 queries to. Default is yes.
562 .B prefetch: \fI<yes or no>
563 If yes, message cache elements are prefetched before they expire to
564 keep the cache up to date. Default is no. Turning it on gives about
565 10 percent more traffic and load on the machine, but popular items do
566 not expire from the cache.
568 .B prefetch-key: \fI<yes or no>
569 If yes, fetch the DNSKEYs earlier in the validation process, when a DS
570 record is encountered. This lowers the latency of requests. It does use
571 a little more CPU. Also if the cache is set to 0, it is no use. Default is no.
573 .B rrset-roundrobin: \fI<yes or no>
574 If yes, Unbound rotates RRSet order in response (the random number is taken
575 from the query ID, for speed and thread safety). Default is no.
577 .B minimal-responses: \fI<yes or no>
578 If yes, Unbound doesn't insert authority/additional sections into response
579 messages when those sections are not required. This reduces response
580 size significantly, and may avoid TCP fallback for some responses.
581 This may cause a slight speedup. The default is no, because the DNS
582 protocol RFCs mandate these sections, and the additional content could
583 be of use and save roundtrips for clients.
585 .B module\-config: \fI<"module names">
586 Module configuration, a list of module names separated by spaces, surround
587 the string with quotes (""). The modules can be validator, iterator.
588 Setting this to "iterator" will result in a non\-validating server.
589 Setting this to "validator iterator" will turn on DNSSEC validation.
590 The ordering of the modules is important.
591 You must also set trust\-anchors for validation to be useful.
593 .B trust\-anchor\-file: \fI<filename>
594 File with trusted keys for validation. Both DS and DNSKEY entries can appear
595 in the file. The format of the file is the standard DNS Zone file format.
596 Default is "", or no trust anchor file.
598 .B auto\-trust\-anchor\-file: \fI<filename>
599 File with trust anchor for one zone, which is tracked with RFC5011 probes.
600 The probes are several times per month, thus the machine must be online
601 frequently. The initial file can be one with contents as described in
602 \fBtrust\-anchor\-file\fR. The file is written to when the anchor is updated,
603 so the unbound user must have write permission.
605 .B trust\-anchor: \fI<"Resource Record">
606 A DS or DNSKEY RR for a key to use for validation. Multiple entries can be
607 given to specify multiple trusted keys, in addition to the trust\-anchor\-files.
608 The resource record is entered in the same format as 'dig' or 'drill' prints
609 them, the same format as in the zone file. Has to be on a single line, with
610 "" around it. A TTL can be specified for ease of cut and paste, but is ignored.
611 A class can be specified, but class IN is default.
613 .B trusted\-keys\-file: \fI<filename>
614 File with trusted keys for validation. Specify more than one file
615 with several entries, one file per entry. Like \fBtrust\-anchor\-file\fR
616 but has a different file format. Format is BIND\-9 style format,
617 the trusted\-keys { name flag proto algo "key"; }; clauses are read.
618 It is possible to use wildcards with this statement, the wildcard is
619 expanded on start and on reload.
621 .B dlv\-anchor\-file: \fI<filename>
622 File with trusted keys for DLV (DNSSEC Lookaside Validation). Both DS and
623 DNSKEY entries can be used in the file, in the same format as for
624 \fItrust\-anchor\-file:\fR statements. Only one DLV can be configured, more
625 would be slow. The DLV configured is used as a root trusted DLV, this
626 means that it is a lookaside for the root. Default is "", or no dlv anchor file.
628 .B dlv\-anchor: \fI<"Resource Record">
629 Much like trust\-anchor, this is a DLV anchor with the DS or DNSKEY inline.
631 .B domain\-insecure: \fI<domain name>
632 Sets domain name to be insecure, DNSSEC chain of trust is ignored towards
633 the domain name. So a trust anchor above the domain name can not make the
634 domain secure with a DS record, such a DS record is then ignored.
635 Also keys from DLV are ignored for the domain. Can be given multiple times
636 to specify multiple domains that are treated as if unsigned. If you set
637 trust anchors for the domain they override this setting (and the domain
640 This can be useful if you want to make sure a trust anchor for external
641 lookups does not affect an (unsigned) internal domain. A DS record
642 externally can create validation failures for that internal domain.
644 .B val\-override\-date: \fI<rrsig\-style date spec>
645 Default is "" or "0", which disables this debugging feature. If enabled by
646 giving a RRSIG style date, that date is used for verifying RRSIG inception
647 and expiration dates, instead of the current date. Do not set this unless
648 you are debugging signature inception and expiration. The value \-1 ignores
649 the date altogether, useful for some special applications.
651 .B val\-sig\-skew\-min: \fI<seconds>
652 Minimum number of seconds of clock skew to apply to validated signatures.
653 A value of 10% of the signature lifetime (expiration \- inception) is
654 used, capped by this setting. Default is 3600 (1 hour) which allows for
655 daylight savings differences. Lower this value for more strict checking
656 of short lived signatures.
658 .B val\-sig\-skew\-max: \fI<seconds>
659 Maximum number of seconds of clock skew to apply to validated signatures.
660 A value of 10% of the signature lifetime (expiration \- inception)
661 is used, capped by this setting. Default is 86400 (24 hours) which
662 allows for timezone setting problems in stable domains. Setting both
663 min and max very low disables the clock skew allowances. Setting both
664 min and max very high makes the validator check the signature timestamps
667 .B val\-bogus\-ttl: \fI<number>
668 The time to live for bogus data. This is data that has failed validation;
669 due to invalid signatures or other checks. The TTL from that data cannot be
670 trusted, and this value is used instead. The value is in seconds, default 60.
671 The time interval prevents repeated revalidation of bogus data.
673 .B val\-clean\-additional: \fI<yes or no>
674 Instruct the validator to remove data from the additional section of secure
675 messages that are not signed properly. Messages that are insecure, bogus,
676 indeterminate or unchecked are not affected. Default is yes. Use this setting
677 to protect the users that rely on this validator for authentication from
678 protentially bad data in the additional section.
680 .B val\-log\-level: \fI<number>
681 Have the validator print validation failures to the log. Regardless of
682 the verbosity setting. Default is 0, off. At 1, for every user query
683 that fails a line is printed to the logs. This way you can monitor what
684 happens with validation. Use a diagnosis tool, such as dig or drill,
685 to find out why validation is failing for these queries. At 2, not only
686 the query that failed is printed but also the reason why unbound thought
687 it was wrong and which server sent the faulty data.
689 .B val\-permissive\-mode: \fI<yes or no>
690 Instruct the validator to mark bogus messages as indeterminate. The security
691 checks are performed, but if the result is bogus (failed security), the
692 reply is not withheld from the client with SERVFAIL as usual. The client
693 receives the bogus data. For messages that are found to be secure the AD bit
694 is set in replies. Also logging is performed as for full validation.
695 The default value is "no".
697 .B ignore\-cd\-flag: \fI<yes or no>
698 Instruct unbound to ignore the CD flag from clients and refuse to
699 return bogus answers to them. Thus, the CD (Checking Disabled) flag
700 does not disable checking any more. This is useful if legacy (w2008)
701 servers that set the CD flag but cannot validate DNSSEC themselves are
702 the clients, and then unbound provides them with DNSSEC protection.
703 The default value is "no".
705 .B val\-nsec3\-keysize\-iterations: \fI<"list of values">
706 List of keysize and iteration count values, separated by spaces, surrounded
707 by quotes. Default is "1024 150 2048 500 4096 2500". This determines the
708 maximum allowed NSEC3 iteration count before a message is simply marked
709 insecure instead of performing the many hashing iterations. The list must
710 be in ascending order and have at least one entry. If you set it to
711 "1024 65535" there is no restriction to NSEC3 iteration values.
712 This table must be kept short; a very long list could cause slower operation.
714 .B add\-holddown: \fI<seconds>
715 Instruct the \fBauto\-trust\-anchor\-file\fR probe mechanism for RFC5011
716 autotrust updates to add new trust anchors only after they have been
717 visible for this time. Default is 30 days as per the RFC.
719 .B del\-holddown: \fI<seconds>
720 Instruct the \fBauto\-trust\-anchor\-file\fR probe mechanism for RFC5011
721 autotrust updates to remove revoked trust anchors after they have been
722 kept in the revoked list for this long. Default is 30 days as per
725 .B keep\-missing: \fI<seconds>
726 Instruct the \fBauto\-trust\-anchor\-file\fR probe mechanism for RFC5011
727 autotrust updates to remove missing trust anchors after they have been
728 unseen for this long. This cleans up the state file if the target zone
729 does not perform trust anchor revocation, so this makes the auto probe
730 mechanism work with zones that perform regular (non\-5011) rollovers.
731 The default is 366 days. The value 0 does not remove missing anchors,
734 .B key\-cache\-size: \fI<number>
735 Number of bytes size of the key cache. Default is 4 megabytes.
736 A plain number is in bytes, append 'k', 'm' or 'g' for kilobytes, megabytes
737 or gigabytes (1024*1024 bytes in a megabyte).
739 .B key\-cache\-slabs: \fI<number>
740 Number of slabs in the key cache. Slabs reduce lock contention by threads.
741 Must be set to a power of 2. Setting (close) to the number of cpus is a
744 .B neg\-cache\-size: \fI<number>
745 Number of bytes size of the aggressive negative cache. Default is 1 megabyte.
746 A plain number is in bytes, append 'k', 'm' or 'g' for kilobytes, megabytes
747 or gigabytes (1024*1024 bytes in a megabyte).
749 .B local\-zone: \fI<zone> <type>
750 Configure a local zone. The type determines the answer to give if
751 there is no match from local\-data. The types are deny, refuse, static,
752 transparent, redirect, nodefault, typetransparent, and are explained
753 below. After that the default settings are listed. Use local\-data: to
754 enter data into the local zone. Answers for local zones are authoritative
755 DNS answers. By default the zones are class IN.
757 If you need more complicated authoritative data, with referrals, wildcards,
758 CNAME/DNAME support, or DNSSEC authoritative service, setup a stub\-zone for
759 it as detailed in the stub zone section below.
762 Do not send an answer, drop the query.
763 If there is a match from local data, the query is answered.
766 Send an error message reply, with rcode REFUSED.
767 If there is a match from local data, the query is answered.
770 If there is a match from local data, the query is answered.
771 Otherwise, the query is answered with nodata or nxdomain.
772 For a negative answer a SOA is included in the answer if present
773 as local\-data for the zone apex domain.
775 \h'5'\fItransparent\fR
776 If there is a match from local data, the query is answered.
777 Otherwise if the query has a different name, the query is resolved normally.
778 If the query is for a name given in localdata but no such type of data is
779 given in localdata, then a noerror nodata answer is returned.
780 If no local\-zone is given local\-data causes a transparent zone
781 to be created by default.
783 \h'5'\fItypetransparent\fR
784 If there is a match from local data, the query is answered. If the query
785 is for a different name, or for the same name but for a different type,
786 the query is resolved normally. So, similar to transparent but types
787 that are not listed in local data are resolved normally, so if an A record
788 is in the local data that does not cause a nodata reply for AAAA queries.
791 The query is answered from the local data for the zone name.
792 There may be no local data beneath the zone name.
793 This answers queries for the zone, and all subdomains of the zone
794 with the local data for the zone.
795 It can be used to redirect a domain to return a different address record
796 to the end user, with
797 local\-zone: "example.com." redirect and
798 local\-data: "example.com. A 127.0.0.1"
799 queries for www.example.com and www.foo.example.com are redirected, so
800 that users with web browsers cannot access sites with suffix example.com.
803 Used to turn off default contents for AS112 zones. The other types
804 also turn off default contents for the zone. The 'nodefault' option
805 has no other effect than turning off default contents for the
808 The default zones are localhost, reverse 127.0.0.1 and ::1, and the AS112
809 zones. The AS112 zones are reverse DNS zones for private use and reserved
810 IP addresses for which the servers on the internet cannot provide correct
811 answers. They are configured by default to give nxdomain (no reverse
812 information) answers. The defaults can be turned off by specifying your
813 own local\-zone of that name, or using the 'nodefault' type. Below is a
814 list of the default zone contents.
817 The IP4 and IP6 localhost information is given. NS and SOA records are provided
818 for completeness and to satisfy some DNS update tools. Default content:
820 local\-zone: "localhost." static
821 local\-data: "localhost. 10800 IN NS localhost."
822 local\-data: "localhost. 10800 IN
823 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
824 local\-data: "localhost. 10800 IN A 127.0.0.1"
825 local\-data: "localhost. 10800 IN AAAA ::1"
828 \h'5'\fIreverse IPv4 loopback\fR
831 local\-zone: "127.in\-addr.arpa." static
832 local\-data: "127.in\-addr.arpa. 10800 IN NS localhost."
833 local\-data: "127.in\-addr.arpa. 10800 IN
834 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
835 local\-data: "1.0.0.127.in\-addr.arpa. 10800 IN
839 \h'5'\fIreverse IPv6 loopback\fR
842 local\-zone: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
843 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa." static
844 local\-data: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
845 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa. 10800 IN
847 local\-data: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
848 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa. 10800 IN
849 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
850 local\-data: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
851 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa. 10800 IN
855 \h'5'\fIreverse RFC1918 local use zones\fR
856 Reverse data for zones 10.in\-addr.arpa, 16.172.in\-addr.arpa to
857 31.172.in\-addr.arpa, 168.192.in\-addr.arpa.
858 The \fBlocal\-zone:\fR is set static and as \fBlocal\-data:\fR SOA and NS
859 records are provided.
861 \h'5'\fIreverse RFC3330 IP4 this, link\-local, testnet and broadcast\fR
862 Reverse data for zones 0.in\-addr.arpa, 254.169.in\-addr.arpa,
863 2.0.192.in\-addr.arpa (TEST NET 1), 100.51.198.in\-addr.arpa (TEST NET 2),
864 113.0.203.in\-addr.arpa (TEST NET 3), 255.255.255.255.in\-addr.arpa.
866 \h'5'\fIreverse RFC4291 IP6 unspecified\fR
867 Reverse data for zone
869 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
870 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa.
873 \h'5'\fIreverse RFC4193 IPv6 Locally Assigned Local Addresses\fR
874 Reverse data for zone D.F.ip6.arpa.
876 \h'5'\fIreverse RFC4291 IPv6 Link Local Addresses\fR
877 Reverse data for zones 8.E.F.ip6.arpa to B.E.F.ip6.arpa.
879 \h'5'\fIreverse IPv6 Example Prefix\fR
880 Reverse data for zone 8.B.D.0.1.0.0.2.ip6.arpa. This zone is used for
881 tutorials and examples. You can remove the block on this zone with:
883 local\-zone: 8.B.D.0.1.0.0.2.ip6.arpa. nodefault
885 You can also selectively unblock a part of the zone by making that part
886 transparent with a local\-zone statement.
887 This also works with the other default zones.
888 .\" End of local-zone listing.
890 .B local\-data: \fI"<resource record string>"
891 Configure local data, which is served in reply to queries for it.
892 The query has to match exactly unless you configure the local\-zone as
893 redirect. If not matched exactly, the local\-zone type determines
894 further processing. If local\-data is configured that is not a subdomain of
895 a local\-zone, a transparent local\-zone is configured.
896 For record types such as TXT, use single quotes, as in
897 local\-data: 'example. TXT "text"'.
899 If you need more complicated authoritative data, with referrals, wildcards,
900 CNAME/DNAME support, or DNSSEC authoritative service, setup a stub\-zone for
901 it as detailed in the stub zone section below.
903 .B local\-data\-ptr: \fI"IPaddr name"
904 Configure local data shorthand for a PTR record with the reversed IPv4 or
905 IPv6 address and the host name. For example "192.0.2.4 www.example.com".
906 TTL can be inserted like this: "2001:DB8::4 7200 www.example.com"
907 .SS "Remote Control Options"
910 clause are the declarations for the remote control facility. If this is
911 enabled, the \fIunbound\-control\fR(8) utility can be used to send
912 commands to the running unbound server. The server uses these clauses
913 to setup SSLv3 / TLSv1 security for the connection. The
914 \fIunbound\-control\fR(8) utility also reads the \fBremote\-control\fR
915 section for options. To setup the correct self\-signed certificates use the
916 \fIunbound\-control\-setup\fR(8) utility.
918 .B control\-enable: \fI<yes or no>
919 The option is used to enable remote control, default is "no".
920 If turned off, the server does not listen for control commands.
922 .B control\-interface: <ip address>
923 Give IPv4 or IPv6 addresses to listen on for control commands.
924 By default localhost (127.0.0.1 and ::1) is listened to.
925 Use 0.0.0.0 and ::0 to listen to all interfaces.
927 .B control\-port: <port number>
928 The port number to listen on for control commands, default is 8953.
929 If you change this port number, and permissions have been dropped,
930 a reload is not sufficient to open the port again, you must then restart.
932 .B server\-key\-file: "<private key file>"
933 Path to the server private key, by default unbound_server.key.
934 This file is generated by the \fIunbound\-control\-setup\fR utility.
935 This file is used by the unbound server, but not by \fIunbound\-control\fR.
937 .B server\-cert\-file: "<certificate file.pem>"
938 Path to the server self signed certificate, by default unbound_server.pem.
939 This file is generated by the \fIunbound\-control\-setup\fR utility.
940 This file is used by the unbound server, and also by \fIunbound\-control\fR.
942 .B control\-key\-file: "<private key file>"
943 Path to the control client private key, by default unbound_control.key.
944 This file is generated by the \fIunbound\-control\-setup\fR utility.
945 This file is used by \fIunbound\-control\fR.
947 .B control\-cert\-file: "<certificate file.pem>"
948 Path to the control client certificate, by default unbound_control.pem.
949 This certificate has to be signed with the server certificate.
950 This file is generated by the \fIunbound\-control\-setup\fR utility.
951 This file is used by \fIunbound\-control\fR.
952 .SS "Stub Zone Options"
954 There may be multiple
956 clauses. Each with a name: and zero or more hostnames or IP addresses.
957 For the stub zone this list of nameservers is used. Class IN is assumed.
958 The servers should be authority servers, not recursors; unbound performs
959 the recursive processing itself for stub zones.
961 The stub zone can be used to configure authoritative data to be used
962 by the resolver that cannot be accessed using the public internet servers.
963 This is useful for company\-local data or private zones. Setup an
964 authoritative server on a different host (or different port). Enter a config
965 entry for unbound with
967 <ip address of host[@port]>.
968 The unbound resolver can then access the data, without referring to the
969 public internet for it.
971 This setup allows DNSSEC signed zones to be served by that
972 authoritative server, in which case a trusted key entry with the public key
973 can be put in config, so that unbound can validate the data and set the AD
974 bit on replies for the private zone (authoritative servers do not set the
975 AD bit). This setup makes unbound capable of answering queries for the
976 private zone, and can even set the AD bit ('authentic'), but the AA
977 ('authoritative') bit is not set on these replies.
979 .B name: \fI<domain name>
980 Name of the stub zone.
982 .B stub\-host: \fI<domain name>
983 Name of stub zone nameserver. Is itself resolved before it is used.
985 .B stub\-addr: \fI<IP address>
986 IP address of stub zone nameserver. Can be IP 4 or IP 6.
987 To use a nondefault port for DNS communication append '@' with the port number.
989 .B stub\-prime: \fI<yes or no>
990 This option is by default off. If enabled it performs NS set priming,
991 which is similar to root hints, where it starts using the list of nameservers
992 currently published by the zone. Thus, if the hint list is slightly outdated,
993 the resolver picks up a correct list online.
995 .B stub\-first: \fI<yes or no>
996 If enabled, a query is attempted without the stub clause if it fails.
997 The data could not be retrieved and would have caused SERVFAIL because
998 the servers are unreachable, instead it is tried without this clause.
1000 .SS "Forward Zone Options"
1002 There may be multiple
1004 clauses. Each with a \fBname:\fR and zero or more hostnames or IP
1005 addresses. For the forward zone this list of nameservers is used to
1006 forward the queries to. The servers listed as \fBforward\-host:\fR and
1007 \fBforward\-addr:\fR have to handle further recursion for the query. Thus,
1008 those servers are not authority servers, but are (just like unbound is)
1009 recursive servers too; unbound does not perform recursion itself for the
1010 forward zone, it lets the remote server do it. Class IN is assumed.
1011 A forward\-zone entry with name "." and a forward\-addr target will
1012 forward all queries to that other server (unless it can answer from
1015 .B name: \fI<domain name>
1016 Name of the forward zone.
1018 .B forward\-host: \fI<domain name>
1019 Name of server to forward to. Is itself resolved before it is used.
1021 .B forward\-addr: \fI<IP address>
1022 IP address of server to forward to. Can be IP 4 or IP 6.
1023 To use a nondefault port for DNS communication append '@' with the port number.
1025 .B forward\-first: \fI<yes or no>
1026 If enabled, a query is attempted without the forward clause if it fails.
1027 The data could not be retrieved and would have caused SERVFAIL because
1028 the servers are unreachable, instead it is tried without this clause.
1030 .SS "Python Module Options"
1034 clause gives the settings for the \fIpython\fR(1) script module. This module
1035 acts like the iterator and validator modules do, on queries and answers.
1036 To enable the script module it has to be compiled into the daemon,
1037 and the word "python" has to be put in the \fBmodule\-config:\fR option
1038 (usually first, or between the validator and iterator).
1040 .B python\-script: \fI<python file>\fR
1041 The script file to load.
1042 .SH "MEMORY CONTROL EXAMPLE"
1043 In the example config settings below memory usage is reduced. Some service
1044 levels are lower, notable very large data and a high TCP load are no longer
1045 supported. Very large data and high TCP loads are exceptional for the DNS.
1046 DNSSEC validation is enabled, just add trust anchors.
1047 If you do not have to worry about programs using more than 3 Mb of memory,
1048 the below example is not for you. Use the defaults to receive full service,
1049 which on BSD\-32bit tops out at 30\-40 Mb after heavy usage.
1052 # example settings that reduce memory usage
1055 outgoing\-num\-tcp: 1 # this limits TCP service, uses less buffers.
1056 incoming\-num\-tcp: 1
1057 outgoing\-range: 60 # uses less memory, but less performance.
1058 msg\-buffer\-size: 8192 # note this limits service, 'no huge stuff'.
1059 msg\-cache\-size: 100k
1060 msg\-cache\-slabs: 1
1061 rrset\-cache\-size: 100k
1062 rrset\-cache\-slabs: 1
1063 infra\-cache\-numhosts: 200
1064 infra\-cache\-slabs: 1
1065 key\-cache\-size: 100k
1066 key\-cache\-slabs: 1
1067 neg\-cache\-size: 10k
1068 num\-queries\-per\-thread: 30
1069 target\-fetch\-policy: "2 1 0 0 0 0"
1070 harden\-large\-queries: "yes"
1071 harden\-short\-bufsize: "yes"
1076 default unbound working directory.
1083 .I /var/unbound/unbound.conf
1084 unbound configuration file.
1086 .I /var/unbound/unbound.pid
1087 default unbound pidfile with process ID of the running daemon.
1090 unbound log file. default is to log to
1094 \fIunbound\-checkconf\fR(8).
1097 was written by NLnet Labs. Please see CREDITS file
1098 in the distribution for further details.
projects / FreeBSD / releng / 10.0.git / blob