1unbound.conf(5) unbound 1.17.1 unbound.conf(5)
2
6 unbound.conf - Unbound configuration file.
7
9 unbound.conf
10
12 unbound.conf is used to configure unbound(8). The file format has at‐
13 tributes and values. Some attributes have attributes inside them. The
14 notation is: attribute: value.
15 Comments start with # and last to the end of line. Empty lines are ig‐
17 nored as is whitespace at the beginning of a line.
18 The utility unbound-checkconf(8) can be used to check unbound.conf
20 prior to usage.
21
23 An example config file is shown below. Copy this to /etc/unbound/un‐
24 bound.conf and start the server with:
25 $ unbound -c /etc/unbound/unbound.conf
27 Most settings are the defaults. Stop the server with:
29 $ kill `cat /etc/unbound/unbound.pid`
31 Below is a minimal config file. The source distribution contains an ex‐
33 tensive example.conf file with all the options.
34 # unbound.conf(5) config file for unbound(8).
36 server:
37 directory: "/etc/unbound"
38 username: unbound
39 # make sure unbound can access entropy from inside the chroot.
40 # e.g. on linux the use these commands (on BSD, devfs(8) is used):
41 # mount --bind -n /dev/urandom /etc/unbound/dev/urandom
42 # and mount --bind -n /dev/log /etc/unbound/dev/log
43 chroot: "/etc/unbound"
44 # logfile: "/etc/unbound/unbound.log" #uncomment to use logfile.
45 pidfile: "/etc/unbound/unbound.pid"
46 # verbosity: 1 # uncomment and increase to get more logging.
47 # listen on all interfaces, answer queries from the local subnet.
48 interface: 0.0.0.0
49 interface: ::0
50 access-control: 10.0.0.0/8 allow
51 access-control: 2001:DB8::/64 allow
52
54 There must be whitespace between keywords. Attribute keywords end with
55 a colon ':'. An attribute is followed by a value, or its containing
56 attributes in which case it is referred to as a clause. Clauses can be
57 repeated throughout the file (or included files) to group attributes
58 under the same clause.
59 Files can be included using the include: directive. It can appear any‐
61 where, it accepts a single file name as argument. Processing continues
62 as if the text from the included file was copied into the config file
63 at that point. If also using chroot, using full path names for the in‐
64 cluded files works, relative pathnames for the included names work if
65 the directory where the daemon is started equals its chroot/working di‐
66 rectory or is specified before the include statement with directory:
67 dir. Wildcards can be used to include multiple files, see glob(7).
68 For a more structural include option, the include-toplevel: directive
70 can be used. This closes whatever clause is currently active (if any)
71 and forces the use of clauses in the included files and right after
72 this directive.
73 Server Options
75 These options are part of the server: clause.
76 verbosity: <number>
78 The verbosity number, level 0 means no verbosity, only errors.
79 Level 1 gives operational information. Level 2 gives detailed
80 operational information including short information per query.
81 Level 3 gives query level information, output per query. Level
82 4 gives algorithm level information. Level 5 logs client iden‐
83 tification for cache misses. Default is level 1. The verbosity
84 can also be increased from the commandline, see unbound(8).
85 statistics-interval: <seconds>
87 The number of seconds between printing statistics to the log for
88 every thread. Disable with value 0 or "". Default is disabled.
89 The histogram statistics are only printed if replies were sent
90 during the statistics interval, requestlist statistics are
91 printed for every interval (but can be 0). This is because the
92 median calculation requires data to be present.
93 statistics-cumulative: <yes or no>
95 If enabled, statistics are cumulative since starting Unbound,
96 without clearing the statistics counters after logging the sta‐
97 tistics. Default is no.
98 extended-statistics: <yes or no>
100 If enabled, extended statistics are printed from unbound-con‐
101 trol(8). Default is off, because keeping track of more statis‐
102 tics takes time. The counters are listed in unbound-control(8).
103 statistics-inhibit-zero: <yes or no>
105 If enabled, selected extended statistics with a value of 0 are
106 inhibited from printing with unbound-control(8). These are
107 query types, query classes, query opcodes, answer rcodes (except
108 NOERROR, FORMERR, SERVFAIL, NXDOMAIN, NOTIMPL, REFUSED) and RPZ
109 actions. Default is on.
110 num-threads: <number>
112 The number of threads to create to serve clients. Use 1 for no
113 threading.
114 port: <port number>
116 The port number, default 53, on which the server responds to
117 queries.
118 interface: <ip address or interface name [@port]>
120 Interface to use to connect to the network. This interface is
121 listened to for queries from clients, and answers to clients are
122 given from it. Can be given multiple times to work on several
123 interfaces. If none are given the default is to listen to local‐
124 host. If an interface name is used instead of an ip address,
125 the list of ip addresses on that interface are used. The inter‐
126 faces are not changed on a reload (kill -HUP) but only on
127 restart. A port number can be specified with @port (without
128 spaces between interface and port number), if not specified the
129 default port (from port) is used.
130 ip-address: <ip address or interface name [@port]>
132 Same as interface: (for ease of compatibility with nsd.conf).
133 interface-automatic: <yes or no>
135 Listen on all addresses on all (current and future) interfaces,
136 detect the source interface on UDP queries and copy them to
137 replies. This is a lot like ip-transparent, but this option
138 services all interfaces whilst with ip-transparent you can se‐
139 lect which (future) interfaces Unbound provides service on.
140 This feature is experimental, and needs support in your OS for
141 particular socket options. Default value is no.
142 interface-automatic-ports: <string>
144 List the port numbers that interface-automatic listens on. If
145 empty, the default port is listened on. The port numbers are
146 separated by spaces in the string. Default is "".
147 This can be used to have interface automatic to deal with the
149 interface, and listen on the normal port number, by including it
150 in the list, and also https or dns over tls port numbers by
151 putting them in the list as well.
152 outgoing-interface: <ip address or ip6 netblock>
154 Interface to use to connect to the network. This interface is
155 used to send queries to authoritative servers and receive their
156 replies. Can be given multiple times to work on several inter‐
157 faces. If none are given the default (all) is used. You can
158 specify the same interfaces in interface: and outgoing-inter‐
159 face: lines, the interfaces are then used for both purposes.
160 Outgoing queries are sent via a random outgoing interface to
161 counter spoofing.
162 If an IPv6 netblock is specified instead of an individual IPv6
164 address, outgoing UDP queries will use a randomised source ad‐
165 dress taken from the netblock to counter spoofing. Requires the
166 IPv6 netblock to be routed to the host running Unbound, and re‐
167 quires OS support for unprivileged non-local binds (currently
168 only supported on Linux). Several netblocks may be specified
169 with multiple outgoing-interface: options, but do not specify
170 both an individual IPv6 address and an IPv6 netblock, or the
171 randomisation will be compromised. Consider combining with pre‐
172 fer-ip6: yes to increase the likelihood of IPv6 nameservers be‐
173 ing selected for queries. On Linux you need these two commands
174 to be able to use the freebind socket option to receive traffic
175 for the ip6 netblock: ip -6 addr add mynetblock/64 dev lo && ip
176 -6 route add local mynetblock/64 dev lo
177 outgoing-range: <number>
179 Number of ports to open. This number of file descriptors can be
180 opened per thread. Must be at least 1. Default depends on com‐
181 pile options. Larger numbers need extra resources from the oper‐
182 ating system. For performance a very large value is best, use
183 libevent to make this possible.
184 outgoing-port-permit: <port number or range>
186 Permit Unbound to open this port or range of ports for use to
187 send queries. A larger number of permitted outgoing ports in‐
188 creases resilience against spoofing attempts. Make sure these
189 ports are not needed by other daemons. By default only ports
190 above 1024 that have not been assigned by IANA are used. Give a
191 port number or a range of the form "low-high", without spaces.
192 The outgoing-port-permit and outgoing-port-avoid statements are
194 processed in the line order of the config file, adding the per‐
195 mitted ports and subtracting the avoided ports from the set of
196 allowed ports. The processing starts with the non IANA allo‐
197 cated ports above 1024 in the set of allowed ports.
198 outgoing-port-avoid: <port number or range>
200 Do not permit Unbound to open this port or range of ports for
201 use to send queries. Use this to make sure Unbound does not grab
202 a port that another daemon needs. The port is avoided on all
203 outgoing interfaces, both IP4 and IP6. By default only ports
204 above 1024 that have not been assigned by IANA are used. Give a
205 port number or a range of the form "low-high", without spaces.
206 outgoing-num-tcp: <number>
208 Number of outgoing TCP buffers to allocate per thread. Default
209 is 10. If set to 0, or if do-tcp is "no", no TCP queries to au‐
210 thoritative servers are done. For larger installations increas‐
211 ing this value is a good idea.
212 incoming-num-tcp: <number>
214 Number of incoming TCP buffers to allocate per thread. Default
215 is 10. If set to 0, or if do-tcp is "no", no TCP queries from
216 clients are accepted. For larger installations increasing this
217 value is a good idea.
218 edns-buffer-size: <number>
220 Number of bytes size to advertise as the EDNS reassembly buffer
221 size. This is the value put into datagrams over UDP towards
222 peers. The actual buffer size is determined by msg-buffer-size
223 (both for TCP and UDP). Do not set higher than that value. De‐
224 fault is 1232 which is the DNS Flag Day 2020 recommendation.
225 Setting to 512 bypasses even the most stringent path MTU prob‐
226 lems, but is seen as extreme, since the amount of TCP fallback
227 generated is excessive (probably also for this resolver, con‐
228 sider tuning the outgoing tcp number).
229 max-udp-size: <number>
231 Maximum UDP response size (not applied to TCP response). 65536
232 disables the udp response size maximum, and uses the choice from
233 the client, always. Suggested values are 512 to 4096. Default
234 is 4096.
235 stream-wait-size: <number>
237 Number of bytes size maximum to use for waiting stream buffers.
238 Default is 4 megabytes. A plain number is in bytes, append 'k',
239 'm' or 'g' for kilobytes, megabytes or gigabytes (1024*1024
240 bytes in a megabyte). As TCP and TLS streams queue up multiple
241 results, the amount of memory used for these buffers does not
242 exceed this number, otherwise the responses are dropped. This
243 manages the total memory usage of the server (under heavy use),
244 the number of requests that can be queued up per connection is
245 also limited, with further requests waiting in TCP buffers.
246 msg-buffer-size: <number>
248 Number of bytes size of the message buffers. Default is 65552
249 bytes, enough for 64 Kb packets, the maximum DNS message size.
250 No message larger than this can be sent or received. Can be re‐
251 duced to use less memory, but some requests for DNS data, such
252 as for huge resource records, will result in a SERVFAIL reply to
253 the client.
254 msg-cache-size: <number>
256 Number of bytes size of the message cache. Default is 4
257 megabytes. A plain number is in bytes, append 'k', 'm' or 'g'
258 for kilobytes, megabytes or gigabytes (1024*1024 bytes in a
259 megabyte).
260 msg-cache-slabs: <number>
262 Number of slabs in the message cache. Slabs reduce lock con‐
263 tention by threads. Must be set to a power of 2. Setting
264 (close) to the number of cpus is a reasonable guess.
265 num-queries-per-thread: <number>
267 The number of queries that every thread will service simultane‐
268 ously. If more queries arrive that need servicing, and no
269 queries can be jostled out (see jostle-timeout), then the
270 queries are dropped. This forces the client to resend after a
271 timeout; allowing the server time to work on the existing
272 queries. Default depends on compile options, 512 or 1024.
273 jostle-timeout: <msec>
275 Timeout used when the server is very busy. Set to a value that
276 usually results in one roundtrip to the authority servers. If
277 too many queries arrive, then 50% of the queries are allowed to
278 run to completion, and the other 50% are replaced with the new
279 incoming query if they have already spent more than their al‐
280 lowed time. This protects against denial of service by slow
281 queries or high query rates. Default 200 milliseconds. The ef‐
282 fect is that the qps for long-lasting queries is about (num‐
283 queriesperthread / 2) / (average time for such long queries)
284 qps. The qps for short queries can be about (numqueries‐
285 perthread / 2) / (jostletimeout in whole seconds) qps per
286 thread, about (1024/2)*5 = 2560 qps by default.
287 delay-close: <msec>
289 Extra delay for timeouted UDP ports before they are closed, in
290 msec. Default is 0, and that disables it. This prevents very
291 delayed answer packets from the upstream (recursive) servers
292 from bouncing against closed ports and setting off all sort of
293 close-port counters, with eg. 1500 msec. When timeouts happen
294 you need extra sockets, it checks the ID and remote IP of pack‐
295 ets, and unwanted packets are added to the unwanted packet
296 counter.
297 udp-connect: <yes or no>
299 Perform connect for UDP sockets that mitigates ICMP side channel
300 leakage. Default is yes.
301 unknown-server-time-limit: <msec>
303 The wait time in msec for waiting for an unknown server to re‐
304 ply. Increase this if you are behind a slow satellite link, to
305 eg. 1128. That would then avoid re-querying every initial query
306 because it times out. Default is 376 msec.
307 so-rcvbuf: <number>
309 If not 0, then set the SO_RCVBUF socket option to get more buf‐
310 fer space on UDP port 53 incoming queries. So that short spikes
311 on busy servers do not drop packets (see counter in netstat
312 -su). Default is 0 (use system value). Otherwise, the number
313 of bytes to ask for, try "4m" on a busy server. The OS caps it
314 at a maximum, on linux Unbound needs root permission to bypass
315 the limit, or the admin can use sysctl net.core.rmem_max. On
316 BSD change kern.ipc.maxsockbuf in /etc/sysctl.conf. On OpenBSD
317 change header and recompile kernel. On Solaris ndd -set /dev/udp
318 udp_max_buf 8388608.
319 so-sndbuf: <number>
321 If not 0, then set the SO_SNDBUF socket option to get more buf‐
322 fer space on UDP port 53 outgoing queries. This for very busy
323 servers handles spikes in answer traffic, otherwise 'send: re‐
324 source temporarily unavailable' can get logged, the buffer over‐
325 run is also visible by netstat -su. Default is 0 (use system
326 value). Specify the number of bytes to ask for, try "4m" on a
327 very busy server. The OS caps it at a maximum, on linux Unbound
328 needs root permission to bypass the limit, or the admin can use
329 sysctl net.core.wmem_max. On BSD, Solaris changes are similar
330 to so-rcvbuf.
331 so-reuseport: <yes or no>
333 If yes, then open dedicated listening sockets for incoming
334 queries for each thread and try to set the SO_REUSEPORT socket
335 option on each socket. May distribute incoming queries to
336 threads more evenly. Default is yes. On Linux it is supported
337 in kernels >= 3.9. On other systems, FreeBSD, OSX it may also
338 work. You can enable it (on any platform and kernel), it then
339 attempts to open the port and passes the option if it was avail‐
340 able at compile time, if that works it is used, if it fails, it
341 continues silently (unless verbosity 3) without the option. At
342 extreme load it could be better to turn it off to distribute the
343 queries evenly, reported for Linux systems (4.4.x).
344 ip-transparent: <yes or no>
346 If yes, then use IP_TRANSPARENT socket option on sockets where
347 Unbound is listening for incoming traffic. Default no. Allows
348 you to bind to non-local interfaces. For example for non-exis‐
349 tent IP addresses that are going to exist later on, with host
350 failover configuration. This is a lot like interface-automatic,
351 but that one services all interfaces and with this option you
352 can select which (future) interfaces Unbound provides service
353 on. This option needs Unbound to be started with root permis‐
354 sions on some systems. The option uses IP_BINDANY on FreeBSD
355 systems and SO_BINDANY on OpenBSD systems.
356 ip-freebind: <yes or no>
358 If yes, then use IP_FREEBIND socket option on sockets where Un‐
359 bound is listening to incoming traffic. Default no. Allows you
360 to bind to IP addresses that are nonlocal or do not exist, like
361 when the network interface or IP address is down. Exists only
362 on Linux, where the similar ip-transparent option is also avail‐
363 able.
364 ip-dscp: <number>
366 The value of the Differentiated Services Codepoint (DSCP) in the
367 differentiated services field (DS) of the outgoing IP packet
368 headers. The field replaces the outdated IPv4 Type-Of-Service
369 field and the IPv6 traffic class field.
370 rrset-cache-size: <number>
372 Number of bytes size of the RRset cache. Default is 4 megabytes.
373 A plain number is in bytes, append 'k', 'm' or 'g' for kilo‐
374 bytes, megabytes or gigabytes (1024*1024 bytes in a megabyte).
375 rrset-cache-slabs: <number>
377 Number of slabs in the RRset cache. Slabs reduce lock contention
378 by threads. Must be set to a power of 2.
379 cache-max-ttl: <seconds>
381 Time to live maximum for RRsets and messages in the cache. De‐
382 fault is 86400 seconds (1 day). When the TTL expires, the cache
383 item has expired. Can be set lower to force the resolver to
384 query for data often, and not trust (very large) TTL values.
385 Downstream clients also see the lower TTL.
386 cache-min-ttl: <seconds>
388 Time to live minimum for RRsets and messages in the cache. De‐
389 fault is 0. If the minimum kicks in, the data is cached for
390 longer than the domain owner intended, and thus less queries are
391 made to look up the data. Zero makes sure the data in the cache
392 is as the domain owner intended, higher values, especially more
393 than an hour or so, can lead to trouble as the data in the cache
394 does not match up with the actual data any more.
395 cache-max-negative-ttl: <seconds>
397 Time to live maximum for negative responses, these have a SOA in
398 the authority section that is limited in time. Default is 3600.
399 This applies to nxdomain and nodata answers.
400 infra-host-ttl: <seconds>
402 Time to live for entries in the host cache. The host cache con‐
403 tains roundtrip timing, lameness and EDNS support information.
404 Default is 900.
405 infra-cache-slabs: <number>
407 Number of slabs in the infrastructure cache. Slabs reduce lock
408 contention by threads. Must be set to a power of 2.
409 infra-cache-numhosts: <number>
411 Number of hosts for which information is cached. Default is
412 10000.
413 infra-cache-min-rtt: <msec>
415 Lower limit for dynamic retransmit timeout calculation in infra‐
416 structure cache. Default is 50 milliseconds. Increase this value
417 if using forwarders needing more time to do recursive name reso‐
418 lution.
419 infra-cache-max-rtt: <msec>
421 Upper limit for dynamic retransmit timeout calculation in infra‐
422 structure cache. Default is 2 minutes.
423 infra-keep-probing: <yes or no>
425 If enabled the server keeps probing hosts that are down, in the
426 one probe at a time regime. Default is no. Hosts that are
427 down, eg. they did not respond during the one probe at a time
428 period, are marked as down and it may take infra-host-ttl time
429 to get probed again.
430 define-tag: <"list of tags">
432 Define the tags that can be used with local-zone and access-con‐
433 trol. Enclose the list between quotes ("") and put spaces be‐
434 tween tags.
435 do-ip4: <yes or no>
437 Enable or disable whether ip4 queries are answered or issued.
438 Default is yes.
439 do-ip6: <yes or no>
441 Enable or disable whether ip6 queries are answered or issued.
442 Default is yes. If disabled, queries are not answered on IPv6,
443 and queries are not sent on IPv6 to the internet nameservers.
444 With this option you can disable the IPv6 transport for sending
445 DNS traffic, it does not impact the contents of the DNS traffic,
446 which may have ip4 and ip6 addresses in it.
447 prefer-ip4: <yes or no>
449 If enabled, prefer IPv4 transport for sending DNS queries to in‐
450 ternet nameservers. Default is no. Useful if the IPv6 netblock
451 the server has, the entire /64 of that is not owned by one oper‐
452 ator and the reputation of the netblock /64 is an issue, using
453 IPv4 then uses the IPv4 filters that the upstream servers have.
454 prefer-ip6: <yes or no>
456 If enabled, prefer IPv6 transport for sending DNS queries to in‐
457 ternet nameservers. Default is no.
458 do-udp: <yes or no>
460 Enable or disable whether UDP queries are answered or issued.
461 Default is yes.
462 do-tcp: <yes or no>
464 Enable or disable whether TCP queries are answered or issued.
465 Default is yes.
466 tcp-mss: <number>
468 Maximum segment size (MSS) of TCP socket on which the server re‐
469 sponds to queries. Value lower than common MSS on Ethernet (1220
470 for example) will address path MTU problem. Note that not all
471 platform supports socket option to set MSS (TCP_MAXSEG). De‐
472 fault is system default MSS determined by interface MTU and ne‐
473 gotiation between server and client.
474 outgoing-tcp-mss: <number>
476 Maximum segment size (MSS) of TCP socket for outgoing queries
477 (from Unbound to other servers). Value lower than common MSS on
478 Ethernet (1220 for example) will address path MTU problem. Note
479 that not all platform supports socket option to set MSS
480 (TCP_MAXSEG). Default is system default MSS determined by in‐
481 terface MTU and negotiation between Unbound and other servers.
482 tcp-idle-timeout: <msec>
484 The period Unbound will wait for a query on a TCP connection.
485 If this timeout expires Unbound closes the connection. This op‐
486 tion defaults to 30000 milliseconds. When the number of free
487 incoming TCP buffers falls below 50% of the total number config‐
488 ured, the option value used is progressively reduced, first to
489 1% of the configured value, then to 0.2% of the configured value
490 if the number of free buffers falls below 35% of the total num‐
491 ber configured, and finally to 0 if the number of free buffers
492 falls below 20% of the total number configured. A minimum time‐
493 out of 200 milliseconds is observed regardless of the option
494 value used.
495 tcp-reuse-timeout: <msec>
497 The period Unbound will keep TCP persistent connections open to
498 authority servers. This option defaults to 60000 milliseconds.
499 max-reuse-tcp-queries: <number>
501 The maximum number of queries that can be sent on a persistent
502 TCP connection. This option defaults to 200 queries.
503 tcp-auth-query-timeout: <number>
505 Timeout in milliseconds for TCP queries to auth servers. This
506 option defaults to 3000 milliseconds.
507 edns-tcp-keepalive: <yes or no>
509 Enable or disable EDNS TCP Keepalive. Default is no.
510 edns-tcp-keepalive-timeout: <msec>
512 The period Unbound will wait for a query on a TCP connection
513 when EDNS TCP Keepalive is active. If this timeout expires Un‐
514 bound closes the connection. If the client supports the EDNS TCP
515 Keepalive option, Unbound sends the timeout value to the client
516 to encourage it to close the connection before the server times
517 out. This option defaults to 120000 milliseconds. When the
518 number of free incoming TCP buffers falls below 50% of the total
519 number configured, the advertised timeout is progressively re‐
520 duced to 1% of the configured value, then to 0.2% of the config‐
521 ured value if the number of free buffers falls below 35% of the
522 total number configured, and finally to 0 if the number of free
523 buffers falls below 20% of the total number configured. A mini‐
524 mum actual timeout of 200 milliseconds is observed regardless of
525 the advertised timeout.
526 tcp-upstream: <yes or no>
528 Enable or disable whether the upstream queries use TCP only for
529 transport. Default is no. Useful in tunneling scenarios. If
530 set to no you can specify TCP transport only for selected for‐
531 ward or stub zones using forward-tcp-upstream or stub-tcp-up‐
532 stream respectively.
533 udp-upstream-without-downstream: <yes or no>
535 Enable udp upstream even if do-udp is no. Default is no, and
536 this does not change anything. Useful for TLS service
537 providers, that want no udp downstream but use udp to fetch data
538 upstream.
539 tls-upstream: <yes or no>
541 Enabled or disable whether the upstream queries use TLS only for
542 transport. Default is no. Useful in tunneling scenarios. The
543 TLS contains plain DNS in TCP wireformat. The other server must
544 support this (see tls-service-key). If you enable this, also
545 configure a tls-cert-bundle or use tls-win-cert or tls-sys‐
546 tem-cert to load CA certs, otherwise the connections cannot be
547 authenticated. This option enables TLS for all of them, but if
548 you do not set this you can configure TLS specifically for some
549 forward zones with forward-tls-upstream. And also with
550 stub-tls-upstream.
551 ssl-upstream: <yes or no>
553 Alternate syntax for tls-upstream. If both are present in the
554 config file the last is used.
555 tls-service-key: <file>
557 If enabled, the server provides DNS-over-TLS or DNS-over-HTTPS
558 service on the TCP ports marked implicitly or explicitly for
559 these services with tls-port or https-port. The file must con‐
560 tain the private key for the TLS session, the public certificate
561 is in the tls-service-pem file and it must also be specified if
562 tls-service-key is specified. The default is "", turned off.
563 Enabling or disabling this service requires a restart (a reload
564 is not enough), because the key is read while root permissions
565 are held and before chroot (if any). The ports enabled implic‐
566 itly or explicitly via tls-port: and https-port: do not provide
567 normal DNS TCP service. Unbound needs to be compiled with lib‐
568 nghttp2 in order to provide DNS-over-HTTPS.
569 ssl-service-key: <file>
571 Alternate syntax for tls-service-key.
572 tls-service-pem: <file>
574 The public key certificate pem file for the tls service. De‐
575 fault is "", turned off.
576 ssl-service-pem: <file>
578 Alternate syntax for tls-service-pem.
579 tls-port: <number>
581 The port number on which to provide TCP TLS service, default
582 853, only interfaces configured with that port number as @number
583 get the TLS service.
584 ssl-port: <number>
586 Alternate syntax for tls-port.
587 tls-cert-bundle: <file>
589 If null or "", no file is used. Set it to the certificate bun‐
590 dle file, for example "/etc/pki/tls/certs/ca-bundle.crt". These
591 certificates are used for authenticating connections made to
592 outside peers. For example auth-zone urls, and also DNS over
593 TLS connections. It is read at start up before permission drop
594 and chroot.
595 ssl-cert-bundle: <file>
597 Alternate syntax for tls-cert-bundle.
598 tls-win-cert: <yes or no>
600 Add the system certificates to the cert bundle certificates for
601 authentication. If no cert bundle, it uses only these certifi‐
602 cates. Default is no. On windows this option uses the certifi‐
603 cates from the cert store. Use the tls-cert-bundle option on
604 other systems. On other systems, this option enables the system
605 certificates.
606 tls-system-cert: <yes or no>
608 This the same setting as the tls-win-cert setting, under a dif‐
609 ferent name. Because it is not windows specific.
610 tls-additional-port: <portnr>
612 List portnumbers as tls-additional-port, and when interfaces are
613 defined, eg. with the @port suffix, as this port number, they
614 provide dns over TLS service. Can list multiple, each on a new
615 statement.
616 tls-session-ticket-keys: <file>
618 If not "", lists files with 80 bytes of random contents that are
619 used to perform TLS session resumption for clients using the Un‐
620 bound server. These files contain the secret key for the TLS
621 session tickets. First key use to encrypt and decrypt TLS ses‐
622 sion tickets. Other keys use to decrypt only. With this you
623 can roll over to new keys, by generating a new first file and
624 allowing decrypt of the old file by listing it after the first
625 file for some time, after the wait clients are not using the old
626 key any more and the old key can be removed. One way to create
627 the file is dd if=/dev/random bs=1 count=80 of=ticket.dat The
628 first 16 bytes should be different from the old one if you cre‐
629 ate a second key, that is the name used to identify the key.
630 Then there is 32 bytes random data for an AES key and then 32
631 bytes random data for the HMAC key.
632 tls-ciphers: <string with cipher list>
634 Set the list of ciphers to allow when serving TLS. Use "" for
635 defaults, and that is the default.
636 tls-ciphersuites: <string with ciphersuites list>
638 Set the list of ciphersuites to allow when serving TLS. This is
639 for newer TLS 1.3 connections. Use "" for defaults, and that is
640 the default.
641 pad-responses: <yes or no>
643 If enabled, TLS serviced queries that contained an EDNS Padding
644 option will cause responses padded to the closest multiple of
645 the size specified in pad-responses-block-size. Default is yes.
646 pad-responses-block-size: <number>
648 The block size with which to pad responses serviced over TLS.
649 Only responses to padded queries will be padded. Default is
650 468.
651 pad-queries: <yes or no>
653 If enabled, all queries sent over TLS upstreams will be padded
654 to the closest multiple of the size specified in
655 pad-queries-block-size. Default is yes.
656 pad-queries-block-size: <number>
658 The block size with which to pad queries sent over TLS up‐
659 streams. Default is 128.
660 tls-use-sni: <yes or no>
662 Enable or disable sending the SNI extension on TLS connections.
663 Default is yes. Changing the value requires a reload.
664 https-port: <number>
666 The port number on which to provide DNS-over-HTTPS service, de‐
667 fault 443, only interfaces configured with that port number as
668 @number get the HTTPS service.
669 http-endpoint: <endpoint string>
671 The HTTP endpoint to provide DNS-over-HTTPS service on. Default
672 "/dns-query".
673 http-max-streams: <number of streams>
675 Number used in the SETTINGS_MAX_CONCURRENT_STREAMS parameter in
676 the HTTP/2 SETTINGS frame for DNS-over-HTTPS connections. De‐
677 fault 100.
678 http-query-buffer-size: <size in bytes>
680 Maximum number of bytes used for all HTTP/2 query buffers com‐
681 bined. These buffers contain (partial) DNS queries waiting for
682 request stream completion. An RST_STREAM frame will be send to
683 streams exceeding this limit. Default is 4 megabytes. A plain
684 number is in bytes, append 'k', 'm' or 'g' for kilobytes,
685 megabytes or gigabytes (1024*1024 bytes in a megabyte).
686 http-response-buffer-size: <size in bytes>
688 Maximum number of bytes used for all HTTP/2 response buffers
689 combined. These buffers contain DNS responses waiting to be
690 written back to the clients. An RST_STREAM frame will be send
691 to streams exceeding this limit. Default is 4 megabytes. A plain
692 number is in bytes, append 'k', 'm' or 'g' for kilobytes,
693 megabytes or gigabytes (1024*1024 bytes in a megabyte).
694 http-nodelay: <yes or no>
696 Set TCP_NODELAY socket option on sockets used to provide DNS-
697 over-HTTPS service. Ignored if the option is not available. De‐
698 fault is yes.
699 http-notls-downstream: <yes or no>
701 Disable use of TLS for the downstream DNS-over-HTTP connections.
702 Useful for local back end servers. Default is no.
703 proxy-protocol-port: <portnr>
705 List port numbers as proxy-protocol-port, and when interfaces
706 are defined, eg. with the @port suffix, as this port number,
707 they support and expect PROXYv2. In this case the proxy address
708 will only be used for the network communication and initial ACL
709 (check if the proxy itself is denied/refused by configuration).
710 The proxied address (if any) will then be used as the true
711 client address and will be used where applicable for logging,
712 ACL, DNSTAP, RPZ and IP ratelimiting. PROXYv2 is supported for
713 UDP and TCP/TLS listening interfaces. There is no support for
714 PROXYv2 on a DoH or DNSCrypt listening interface. Can list mul‐
715 tiple, each on a new statement.
716 use-systemd: <yes or no>
718 Enable or disable systemd socket activation. Default is no.
719 do-daemonize: <yes or no>
721 Enable or disable whether the Unbound server forks into the
722 background as a daemon. Set the value to no when Unbound runs
723 as systemd service. Default is yes.
724 tcp-connection-limit: <IP netblock> <limit>
726 Allow up to limit simultaneous TCP connections from the given
727 netblock. When at the limit, further connections are accepted
728 but closed immediately. This option is experimental at this
729 time.
730 access-control: <IP netblock> <action>
732 The netblock is given as an IP4 or IP6 address with /size ap‐
733 pended for a classless network block. The action can be deny,
734 refuse, allow, allow_setrd, allow_snoop, deny_non_local or
735 refuse_non_local. The most specific netblock match is used, if
736 none match refuse is used. The order of the access-control
737 statements therefore does not matter.
738 The action deny stops queries from hosts from that netblock.
740 The action refuse stops queries too, but sends a DNS rcode RE‐
742 FUSED error message back.
743 The action allow gives access to clients from that netblock. It
745 gives only access for recursion clients (which is what almost
746 all clients need). Nonrecursive queries are refused.
747 The allow action does allow nonrecursive queries to access the
749 local-data that is configured. The reason is that this does not
750 involve the Unbound server recursive lookup algorithm, and
751 static data is served in the reply. This supports normal opera‐
752 tions where nonrecursive queries are made for the authoritative
753 data. For nonrecursive queries any replies from the dynamic
754 cache are refused.
755 The allow_setrd action ignores the recursion desired (RD) bit
757 and treats all requests as if the recursion desired bit is set.
758 Note that this behavior violates RFC 1034 which states that a
759 name server should never perform recursive service unless asked
760 via the RD bit since this interferes with trouble shooting of
761 name servers and their databases. This prohibited behavior may
762 be useful if another DNS server must forward requests for spe‐
763 cific zones to a resolver DNS server, but only supports stub do‐
764 mains and sends queries to the resolver DNS server with the RD
765 bit cleared.
766 The action allow_snoop gives nonrecursive access too. This give
768 both recursive and non recursive access. The name allow_snoop
769 refers to cache snooping, a technique to use nonrecursive
770 queries to examine the cache contents (for malicious acts).
771 However, nonrecursive queries can also be a valuable debugging
772 tool (when you want to examine the cache contents). In that case
773 use allow_snoop for your administration host.
774 By default only localhost is allowed, the rest is refused. The
776 default is refused, because that is protocol-friendly. The DNS
777 protocol is not designed to handle dropped packets due to pol‐
778 icy, and dropping may result in (possibly excessive) retried
779 queries.
780 The deny_non_local and refuse_non_local settings are for hosts
782 that are only allowed to query for the authoritative local-data,
783 they are not allowed full recursion but only the static data.
784 With deny_non_local, messages that are disallowed are dropped,
785 with refuse_non_local they receive error code REFUSED.
786 access-control-tag: <IP netblock> <"list of tags">
788 Assign tags to access-control elements. Clients using this ac‐
789 cess control element use localzones that are tagged with one of
790 these tags. Tags must be defined in define-tags. Enclose list
791 of tags in quotes ("") and put spaces between tags. If ac‐
792 cess-control-tag is configured for a netblock that does not have
793 an access-control, an access-control element with action allow
794 is configured for this netblock.
795 access-control-tag-action: <IP netblock> <tag> <action>
797 Set action for particular tag for given access control element.
798 If you have multiple tag values, the tag used to lookup the ac‐
799 tion is the first tag match between access-control-tag and lo‐
800 cal-zone-tag where "first" comes from the order of the define-
801 tag values.
802 access-control-tag-data: <IP netblock> <tag> <"resource record string">
804 Set redirect data for particular tag for given access control
805 element.
806 access-control-view: <IP netblock> <view name>
808 Set view for given access control element.
809 interface-action: <ip address or interface name [@port]> <action>
811 Similar to access-control: but for interfaces.
812 The action is the same as the ones defined under access-con‐
814 trol:. Interfaces are refused by default. By default only lo‐
815 calhost (the IP netblock, not the loopback interface) is allowed
816 through the default access-control: behavior.
817 Note that the interface needs to be already specified with in‐
819 terface: and that any access-control*: setting overrides all in‐
820 terface-*: settings for targeted clients.
821 interface-tag: <ip address or interface name [@port]> <"list of tags">
823 Similar to access-control-tag: but for interfaces.
824 Note that the interface needs to be already specified with in‐
826 terface: and that any access-control*: setting overrides all in‐
827 terface-*: settings for targeted clients.
828 interface-tag-action: <ip address or interface name [@port]> <tag> <ac‐
830 tion>
831 Similar to access-control-tag-action: but for interfaces.
832 Note that the interface needs to be already specified with in‐
834 terface: and that any access-control*: setting overrides all in‐
835 terface-*: settings for targeted clients.
836 interface-tag-data: <ip address or interface name [@port]> <tag> <"re‐
838 source record string">
839 Similar to access-control-tag-data: but for interfaces.
840 Note that the interface needs to be already specified with in‐
842 terface: and that any access-control*: setting overrides all in‐
843 terface-*: settings for targeted clients.
844 interface-view: <ip address or interface name [@port]> <view name>
846 Similar to access-control-view: but for interfaces.
847 Note that the interface needs to be already specified with in‐
849 terface: and that any access-control*: setting overrides all in‐
850 terface-*: settings for targeted clients.
851 chroot: <directory>
853 If chroot is enabled, you should pass the configfile (from the
854 commandline) as a full path from the original root. After the
855 chroot has been performed the now defunct portion of the config
856 file path is removed to be able to reread the config after a
857 reload.
858 All other file paths (working dir, logfile, roothints, and key
860 files) can be specified in several ways: as an absolute path
861 relative to the new root, as a relative path to the working di‐
862 rectory, or as an absolute path relative to the original root.
863 In the last case the path is adjusted to remove the unused por‐
864 tion.
865 The pidfile can be either a relative path to the working direc‐
867 tory, or an absolute path relative to the original root. It is
868 written just prior to chroot and dropping permissions. This al‐
869 lows the pidfile to be /var/run/unbound.pid and the chroot to be
870 /var/unbound, for example. Note that Unbound is not able to re‐
871 move the pidfile after termination when it is located outside of
872 the chroot directory.
873 Additionally, Unbound may need to access /dev/urandom (for en‐
875 tropy) from inside the chroot.
876 If given a chroot is done to the given directory. By default ch‐
878 root is enabled and the default is "/etc/unbound". If you give
879 "" no chroot is performed.
880 username: <name>
882 If given, after binding the port the user privileges are
883 dropped. Default is "unbound". If you give username: "" no user
884 change is performed.
885 If this user is not capable of binding the port, reloads (by
887 signal HUP) will still retain the opened ports. If you change
888 the port number in the config file, and that new port number re‐
889 quires privileges, then a reload will fail; a restart is needed.
890 directory: <directory>
892 Sets the working directory for the program. Default is "/etc/un‐
893 bound". On Windows the string "%EXECUTABLE%" tries to change to
894 the directory that unbound.exe resides in. If you give a
895 server: directory: dir before include: file statements then
896 those includes can be relative to the working directory.
897 logfile: <filename>
899 If "" is given, logging goes to stderr, or nowhere once daemo‐
900 nized. The logfile is appended to, in the following format:
901 [seconds since 1970] unbound[pid:tid]: type: message.
902 If this option is given, the use-syslog is option is set to
903 "no". The logfile is reopened (for append) when the config file
904 is reread, on SIGHUP.
905 use-syslog: <yes or no>
907 Sets Unbound to send log messages to the syslogd, using sys‐
908 log(3). The log facility LOG_DAEMON is used, with identity "un‐
909 bound". The logfile setting is overridden when use-syslog is
910 turned on. The default is to log to syslog.
911 log-identity: <string>
913 If "" is given (default), then the name of the executable, usu‐
914 ally "unbound" is used to report to the log. Enter a string to
915 override it with that, which is useful on systems that run more
916 than one instance of Unbound, with different configurations, so
917 that the logs can be easily distinguished against.
918 log-time-ascii: <yes or no>
920 Sets logfile lines to use a timestamp in UTC ascii. Default is
921 no, which prints the seconds since 1970 in brackets. No effect
922 if using syslog, in that case syslog formats the timestamp
923 printed into the log files.
924 log-queries: <yes or no>
926 Prints one line per query to the log, with the log timestamp and
927 IP address, name, type and class. Default is no. Note that it
928 takes time to print these lines which makes the server (signifi‐
929 cantly) slower. Odd (nonprintable) characters in names are
930 printed as '?'.
931 log-replies: <yes or no>
933 Prints one line per reply to the log, with the log timestamp and
934 IP address, name, type, class, return code, time to resolve,
935 from cache and response size. Default is no. Note that it
936 takes time to print these lines which makes the server (signifi‐
937 cantly) slower. Odd (nonprintable) characters in names are
938 printed as '?'.
939 log-tag-queryreply: <yes or no>
941 Prints the word 'query' and 'reply' with log-queries and
942 log-replies. This makes filtering logs easier. The default is
943 off (for backwards compatibility).
944 log-local-actions: <yes or no>
946 Print log lines to inform about local zone actions. These lines
947 are like the local-zone type inform prints out, but they are
948 also printed for the other types of local zones.
949 log-servfail: <yes or no>
951 Print log lines that say why queries return SERVFAIL to clients.
952 This is separate from the verbosity debug logs, much smaller,
953 and printed at the error level, not the info level of debug info
954 from verbosity.
955 pidfile: <filename>
957 The process id is written to the file. Default is "/run/un‐
958 bound/unbound.pid". So,
959 kill -HUP `cat /run/unbound/unbound.pid`
960 triggers a reload,
961 kill -TERM `cat /run/unbound/unbound.pid`
962 gracefully terminates.
963 root-hints: <filename>
965 Read the root hints from this file. Default is nothing, using
966 builtin hints for the IN class. The file has the format of zone
967 files, with root nameserver names and addresses only. The de‐
968 fault may become outdated, when servers change, therefore it is
969 good practice to use a root-hints file.
970 hide-identity: <yes or no>
972 If enabled id.server and hostname.bind queries are refused.
973 identity: <string>
975 Set the identity to report. If set to "", the default, then the
976 hostname of the server is returned.
977 hide-version: <yes or no>
979 If enabled version.server and version.bind queries are refused.
980 version: <string>
982 Set the version to report. If set to "", the default, then the
983 package version is returned.
984 hide-http-user-agent: <yes or no>
986 If enabled the HTTP header User-Agent is not set. Use with cau‐
987 tion as some webserver configurations may reject HTTP requests
988 lacking this header. If needed, it is better to explicitly set
989 the http-user-agent below.
990 http-user-agent: <string>
992 Set the HTTP User-Agent header for outgoing HTTP requests. If
993 set to "", the default, then the package name and version are
994 used.
995 nsid: <string>
997 Add the specified nsid to the EDNS section of the answer when
998 queried with an NSID EDNS enabled packet. As a sequence of hex
999 characters or with ascii_ prefix and then an ascii string.
1000 hide-trustanchor: <yes or no>
1002 If enabled trustanchor.unbound queries are refused.
1003 target-fetch-policy: <"list of numbers">
1005 Set the target fetch policy used by Unbound to determine if it
1006 should fetch nameserver target addresses opportunistically. The
1007 policy is described per dependency depth.
1008 The number of values determines the maximum dependency depth
1010 that Unbound will pursue in answering a query. A value of -1
1011 means to fetch all targets opportunistically for that dependency
1012 depth. A value of 0 means to fetch on demand only. A positive
1013 value fetches that many targets opportunistically.
1014 Enclose the list between quotes ("") and put spaces between num‐
1016 bers. The default is "3 2 1 0 0". Setting all zeroes, "0 0 0 0
1017 0" gives behaviour closer to that of BIND 9, while setting "-1
1018 -1 -1 -1 -1" gives behaviour rumoured to be closer to that of
1019 BIND 8.
1020 harden-short-bufsize: <yes or no>
1022 Very small EDNS buffer sizes from queries are ignored. Default
1023 is on, as described in the standard.
1024 harden-large-queries: <yes or no>
1026 Very large queries are ignored. Default is off, since it is le‐
1027 gal protocol wise to send these, and could be necessary for op‐
1028 eration if TSIG or EDNS payload is very large.
1029 harden-glue: <yes or no>
1031 Will trust glue only if it is within the servers authority. De‐
1032 fault is yes.
1033 harden-dnssec-stripped: <yes or no>
1035 Require DNSSEC data for trust-anchored zones, if such data is
1036 absent, the zone becomes bogus. If turned off, and no DNSSEC
1037 data is received (or the DNSKEY data fails to validate), then
1038 the zone is made insecure, this behaves like there is no trust
1039 anchor. You could turn this off if you are sometimes behind an
1040 intrusive firewall (of some sort) that removes DNSSEC data from
1041 packets, or a zone changes from signed to unsigned to badly
1042 signed often. If turned off you run the risk of a downgrade at‐
1043 tack that disables security for a zone. Default is yes.
1044 harden-below-nxdomain: <yes or no>
1046 From RFC 8020 (with title "NXDOMAIN: There Really Is Nothing Un‐
1047 derneath"), returns nxdomain to queries for a name below another
1048 name that is already known to be nxdomain. DNSSEC mandates no‐
1049 error for empty nonterminals, hence this is possible. Very old
1050 software might return nxdomain for empty nonterminals (that usu‐
1051 ally happen for reverse IP address lookups), and thus may be in‐
1052 compatible with this. To try to avoid this only DNSSEC-secure
1053 nxdomains are used, because the old software does not have
1054 DNSSEC. Default is yes. The nxdomain must be secure, this
1055 means nsec3 with optout is insufficient.
1056 harden-referral-path: <yes or no>
1058 Harden the referral path by performing additional queries for
1059 infrastructure data. Validates the replies if trust anchors are
1060 configured and the zones are signed. This enforces DNSSEC vali‐
1061 dation on nameserver NS sets and the nameserver addresses that
1062 are encountered on the referral path to the answer. Default no,
1063 because it burdens the authority servers, and it is not RFC
1064 standard, and could lead to performance problems because of the
1065 extra query load that is generated. Experimental option. If
1066 you enable it consider adding more numbers after the tar‐
1067 get-fetch-policy to increase the max depth that is checked to.
1068 harden-algo-downgrade: <yes or no>
1070 Harden against algorithm downgrade when multiple algorithms are
1071 advertised in the DS record. If no, allows the weakest algo‐
1072 rithm to validate the zone. Default is no. Zone signers must
1073 produce zones that allow this feature to work, but sometimes
1074 they do not, and turning this option off avoids that validation
1075 failure.
1076 use-caps-for-id: <yes or no>
1078 Use 0x20-encoded random bits in the query to foil spoof at‐
1079 tempts. This perturbs the lowercase and uppercase of query
1080 names sent to authority servers and checks if the reply still
1081 has the correct casing. Disabled by default. This feature is
1082 an experimental implementation of draft dns-0x20.
1083 caps-exempt: <domain>
1085 Exempt the domain so that it does not receive caps-for-id per‐
1086 turbed queries. For domains that do not support 0x20 and also
1087 fail with fallback because they keep sending different answers,
1088 like some load balancers. Can be given multiple times, for dif‐
1089 ferent domains.
1090 caps-whitelist: <yes or no>
1092 Alternate syntax for caps-exempt.
1093 qname-minimisation: <yes or no>
1095 Send minimum amount of information to upstream servers to en‐
1096 hance privacy. Only send minimum required labels of the QNAME
1097 and set QTYPE to A when possible. Best effort approach; full
1098 QNAME and original QTYPE will be sent when upstream replies with
1099 a RCODE other than NOERROR, except when receiving NXDOMAIN from
1100 a DNSSEC signed zone. Default is yes.
1101 qname-minimisation-strict: <yes or no>
1103 QNAME minimisation in strict mode. Do not fall-back to sending
1104 full QNAME to potentially broken nameservers. A lot of domains
1105 will not be resolvable when this option in enabled. Only use if
1106 you know what you are doing. This option only has effect when
1107 qname-minimisation is enabled. Default is no.
1108 aggressive-nsec: <yes or no>
1110 Aggressive NSEC uses the DNSSEC NSEC chain to synthesize NXDO‐
1111 MAIN and other denials, using information from previous NXDO‐
1112 MAINs answers. Default is yes. It helps to reduce the query
1113 rate towards targets that get a very high nonexistent name
1114 lookup rate.
1115 private-address: <IP address or subnet>
1117 Give IPv4 of IPv6 addresses or classless subnets. These are ad‐
1118 dresses on your private network, and are not allowed to be re‐
1119 turned for public internet names. Any occurrence of such ad‐
1120 dresses are removed from DNS answers. Additionally, the DNSSEC
1121 validator may mark the answers bogus. This protects against
1122 so-called DNS Rebinding, where a user browser is turned into a
1123 network proxy, allowing remote access through the browser to
1124 other parts of your private network. Some names can be allowed
1125 to contain your private addresses, by default all the local-data
1126 that you configured is allowed to, and you can specify addi‐
1127 tional names using private-domain. No private addresses are en‐
1128 abled by default. We consider to enable this for the RFC1918
1129 private IP address space by default in later releases. That
1130 would enable private addresses for 10.0.0.0/8 172.16.0.0/12
1131 192.168.0.0/16 169.254.0.0/16 fd00::/8 and fe80::/10, since the
1132 RFC standards say these addresses should not be visible on the
1133 public internet. Turning on 127.0.0.0/8 would hinder many spam‐
1134 blocklists as they use that. Adding ::ffff:0:0/96 stops
1135 IPv4-mapped IPv6 addresses from bypassing the filter.
1136 private-domain: <domain name>
1138 Allow this domain, and all its subdomains to contain private ad‐
1139 dresses. Give multiple times to allow multiple domain names to
1140 contain private addresses. Default is none.
1141 unwanted-reply-threshold: <number>
1143 If set, a total number of unwanted replies is kept track of in
1144 every thread. When it reaches the threshold, a defensive action
1145 is taken and a warning is printed to the log. The defensive ac‐
1146 tion is to clear the rrset and message caches, hopefully flush‐
1147 ing away any poison. A value of 10 million is suggested. De‐
1148 fault is 0 (turned off).
1149 do-not-query-address: <IP address>
1151 Do not query the given IP address. Can be IP4 or IP6. Append
1152 /num to indicate a classless delegation netblock, for example
1153 like 10.2.3.4/24 or 2001::11/64.
1154 do-not-query-localhost: <yes or no>
1156 If yes, localhost is added to the do-not-query-address entries,
1157 both IP6 ::1 and IP4 127.0.0.1/8. If no, then localhost can be
1158 used to send queries to. Default is yes.
1159 prefetch: <yes or no>
1161 If yes, message cache elements are prefetched before they expire
1162 to keep the cache up to date. Default is no. Turning it on
1163 gives about 10 percent more traffic and load on the machine, but
1164 popular items do not expire from the cache.
1165 prefetch-key: <yes or no>
1167 If yes, fetch the DNSKEYs earlier in the validation process,
1168 when a DS record is encountered. This lowers the latency of re‐
1169 quests. It does use a little more CPU. Also if the cache is
1170 set to 0, it is no use. Default is no.
1171 deny-any: <yes or no>
1173 If yes, deny queries of type ANY with an empty response. De‐
1174 fault is no. If disabled, Unbound responds with a short list of
1175 resource records if some can be found in the cache and makes the
1176 upstream type ANY query if there are none.
1177 rrset-roundrobin: <yes or no>
1179 If yes, Unbound rotates RRSet order in response (the random num‐
1180 ber is taken from the query ID, for speed and thread safety).
1181 Default is yes.
1182 minimal-responses: <yes or no>
1184 If yes, Unbound does not insert authority/additional sections
1185 into response messages when those sections are not required.
1186 This reduces response size significantly, and may avoid TCP
1187 fallback for some responses. This may cause a slight speedup.
1188 The default is yes, even though the DNS protocol RFCs mandate
1189 these sections, and the additional content could be of use and
1190 save roundtrips for clients. Because they are not used, and the
1191 saved roundtrips are easier saved with prefetch, whilst this is
1192 faster.
1193 disable-dnssec-lame-check: <yes or no>
1195 If true, disables the DNSSEC lameness check in the iterator.
1196 This check sees if RRSIGs are present in the answer, when dnssec
1197 is expected, and retries another authority if RRSIGs are unex‐
1198 pectedly missing. The validator will insist in RRSIGs for
1199 DNSSEC signed domains regardless of this setting, if a trust an‐
1200 chor is loaded.
1201 module-config: <"module names">
1203 Module configuration, a list of module names separated by spa‐
1204 ces, surround the string with quotes (""). The modules can be
1205 respip, validator, or iterator (and possibly more, see below).
1206 Setting this to just "iterator" will result in a non-validating
1207 server. Setting this to "validator iterator" will turn on
1208 DNSSEC validation. The ordering of the modules is significant,
1209 the order decides the order of processing. You must also set
1210 trust-anchors for validation to be useful. Adding respip to the
1211 front will cause RPZ processing to be done on all queries. The
1212 default is "validator iterator".
1213 When the server is built with EDNS client subnet support the de‐
1215 fault is "subnetcache validator iterator". Most modules that
1216 need to be listed here have to be listed at the beginning of the
1217 line. The subnetcachedb module has to be listed just before the
1218 iterator. The python module can be listed in different places,
1219 it then processes the output of the module it is just before.
1220 The dynlib module can be listed pretty much anywhere, it is only
1221 a very thin wrapper that allows dynamic libraries to run in its
1222 place.
1223 trust-anchor-file: <filename>
1225 File with trusted keys for validation. Both DS and DNSKEY en‐
1226 tries can appear in the file. The format of the file is the
1227 standard DNS Zone file format. Default is "", or no trust an‐
1228 chor file.
1229 auto-trust-anchor-file: <filename>
1231 File with trust anchor for one zone, which is tracked with
1232 RFC5011 probes. The probes are run several times per month,
1233 thus the machine must be online frequently. The initial file
1234 can be one with contents as described in trust-anchor-file. The
1235 file is written to when the anchor is updated, so the Unbound
1236 user must have write permission. Write permission to the file,
1237 but also to the directory it is in (to create a temporary file,
1238 which is necessary to deal with filesystem full events), it must
1239 also be inside the chroot (if that is used).
1240 trust-anchor: <"Resource Record">
1242 A DS or DNSKEY RR for a key to use for validation. Multiple en‐
1243 tries can be given to specify multiple trusted keys, in addition
1244 to the trust-anchor-files. The resource record is entered in
1245 the same format as 'dig' or 'drill' prints them, the same format
1246 as in the zone file. Has to be on a single line, with "" around
1247 it. A TTL can be specified for ease of cut and paste, but is ig‐
1248 nored. A class can be specified, but class IN is default.
1249 trusted-keys-file: <filename>
1251 File with trusted keys for validation. Specify more than one
1252 file with several entries, one file per entry. Like trust-an‐
1253 chor-file but has a different file format. Format is BIND-9
1254 style format, the trusted-keys { name flag proto algo "key"; };
1255 clauses are read. It is possible to use wildcards with this
1256 statement, the wildcard is expanded on start and on reload.
1257 trust-anchor-signaling: <yes or no>
1259 Send RFC8145 key tag query after trust anchor priming. Default
1260 is yes.
1261 root-key-sentinel: <yes or no>
1263 Root key trust anchor sentinel. Default is yes.
1264 domain-insecure: <domain name>
1266 Sets domain name to be insecure, DNSSEC chain of trust is ig‐
1267 nored towards the domain name. So a trust anchor above the do‐
1268 main name can not make the domain secure with a DS record, such
1269 a DS record is then ignored. Can be given multiple times to
1270 specify multiple domains that are treated as if unsigned. If
1271 you set trust anchors for the domain they override this setting
1272 (and the domain is secured).
1273 This can be useful if you want to make sure a trust anchor for
1275 external lookups does not affect an (unsigned) internal domain.
1276 A DS record externally can create validation failures for that
1277 internal domain.
1278 val-override-date: <rrsig-style date spec>
1280 Default is "" or "0", which disables this debugging feature. If
1281 enabled by giving a RRSIG style date, that date is used for ver‐
1282 ifying RRSIG inception and expiration dates, instead of the cur‐
1283 rent date. Do not set this unless you are debugging signature
1284 inception and expiration. The value -1 ignores the date alto‐
1285 gether, useful for some special applications.
1286 val-sig-skew-min: <seconds>
1288 Minimum number of seconds of clock skew to apply to validated
1289 signatures. A value of 10% of the signature lifetime (expira‐
1290 tion - inception) is used, capped by this setting. Default is
1291 3600 (1 hour) which allows for daylight savings differences.
1292 Lower this value for more strict checking of short lived signa‐
1293 tures.
1294 val-sig-skew-max: <seconds>
1296 Maximum number of seconds of clock skew to apply to validated
1297 signatures. A value of 10% of the signature lifetime (expira‐
1298 tion - inception) is used, capped by this setting. Default is
1299 86400 (24 hours) which allows for timezone setting problems in
1300 stable domains. Setting both min and max very low disables the
1301 clock skew allowances. Setting both min and max very high makes
1302 the validator check the signature timestamps less strictly.
1303 val-max-restart: <number>
1305 The maximum number the validator should restart validation with
1306 another authority in case of failed validation. Default is 5.
1307 val-bogus-ttl: <number>
1309 The time to live for bogus data. This is data that has failed
1310 validation; due to invalid signatures or other checks. The TTL
1311 from that data cannot be trusted, and this value is used in‐
1312 stead. The value is in seconds, default 60. The time interval
1313 prevents repeated revalidation of bogus data.
1314 val-clean-additional: <yes or no>
1316 Instruct the validator to remove data from the additional sec‐
1317 tion of secure messages that are not signed properly. Messages
1318 that are insecure, bogus, indeterminate or unchecked are not af‐
1319 fected. Default is yes. Use this setting to protect the users
1320 that rely on this validator for authentication from potentially
1321 bad data in the additional section.
1322 val-log-level: <number>
1324 Have the validator print validation failures to the log. Re‐
1325 gardless of the verbosity setting. Default is 0, off. At 1,
1326 for every user query that fails a line is printed to the logs.
1327 This way you can monitor what happens with validation. Use a
1328 diagnosis tool, such as dig or drill, to find out why validation
1329 is failing for these queries. At 2, not only the query that
1330 failed is printed but also the reason why Unbound thought it was
1331 wrong and which server sent the faulty data.
1332 val-permissive-mode: <yes or no>
1334 Instruct the validator to mark bogus messages as indeterminate.
1335 The security checks are performed, but if the result is bogus
1336 (failed security), the reply is not withheld from the client
1337 with SERVFAIL as usual. The client receives the bogus data. For
1338 messages that are found to be secure the AD bit is set in
1339 replies. Also logging is performed as for full validation. The
1340 default value is "no".
1341 ignore-cd-flag: <yes or no>
1343 Instruct Unbound to ignore the CD flag from clients and refuse
1344 to return bogus answers to them. Thus, the CD (Checking Dis‐
1345 abled) flag does not disable checking any more. This is useful
1346 if legacy (w2008) servers that set the CD flag but cannot vali‐
1347 date DNSSEC themselves are the clients, and then Unbound pro‐
1348 vides them with DNSSEC protection. The default value is "no".
1349 serve-expired: <yes or no>
1351 If enabled, Unbound attempts to serve old responses from cache
1352 with a TTL of serve-expired-reply-ttl in the response without
1353 waiting for the actual resolution to finish. The actual resolu‐
1354 tion answer ends up in the cache later on. Default is "no".
1355 serve-expired-ttl: <seconds>
1357 Limit serving of expired responses to configured seconds after
1358 expiration. 0 disables the limit. This option only applies when
1359 serve-expired is enabled. A suggested value per RFC 8767 is be‐
1360 tween 86400 (1 day) and 259200 (3 days). The default is 0.
1361 serve-expired-ttl-reset: <yes or no>
1363 Set the TTL of expired records to the serve-expired-ttl value
1364 after a failed attempt to retrieve the record from upstream.
1365 This makes sure that the expired records will be served as long
1366 as there are queries for it. Default is "no".
1367 serve-expired-reply-ttl: <seconds>
1369 TTL value to use when replying with expired data. If serve-ex‐
1370 pired-client-timeout is also used then it is RECOMMENDED to use
1371 30 as the value (RFC 8767). The default is 30.
1372 serve-expired-client-timeout: <msec>
1374 Time in milliseconds before replying to the client with expired
1375 data. This essentially enables the serve-stale behavior as
1376 specified in RFC 8767 that first tries to resolve before immedi‐
1377 ately responding with expired data. A recommended value per RFC
1378 8767 is 1800. Setting this to 0 will disable this behavior.
1379 Default is 0.
1380 serve-original-ttl: <yes or no>
1382 If enabled, Unbound will always return the original TTL as re‐
1383 ceived from the upstream name server rather than the decrement‐
1384 ing TTL as stored in the cache. This feature may be useful if
1385 Unbound serves as a front-end to a hidden authoritative name
1386 server. Enabling this feature does not impact cache expiry, it
1387 only changes the TTL Unbound embeds in responses to queries.
1388 Note that enabling this feature implicitly disables enforcement
1389 of the configured minimum and maximum TTL, as it is assumed
1390 users who enable this feature do not want Unbound to change the
1391 TTL obtained from an upstream server. Thus, the values set us‐
1392 ing cache-min-ttl and cache-max-ttl are ignored. Default is
1393 "no".
1394 val-nsec3-keysize-iterations: <"list of values">
1396 List of keysize and iteration count values, separated by spaces,
1397 surrounded by quotes. Default is "1024 150 2048 150 4096 150".
1398 This determines the maximum allowed NSEC3 iteration count before
1399 a message is simply marked insecure instead of performing the
1400 many hashing iterations. The list must be in ascending order and
1401 have at least one entry. If you set it to "1024 65535" there is
1402 no restriction to NSEC3 iteration values. This table must be
1403 kept short; a very long list could cause slower operation.
1404 zonemd-permissive-mode: <yes or no>
1406 If enabled the ZONEMD verification failures are only logged and
1407 do not cause the zone to be blocked and only return servfail.
1408 Useful for testing out if it works, or if the operator only
1409 wants to be notified of a problem without disrupting service.
1410 Default is no.
1411 add-holddown: <seconds>
1413 Instruct the auto-trust-anchor-file probe mechanism for RFC5011
1414 autotrust updates to add new trust anchors only after they have
1415 been visible for this time. Default is 30 days as per the RFC.
1416 del-holddown: <seconds>
1418 Instruct the auto-trust-anchor-file probe mechanism for RFC5011
1419 autotrust updates to remove revoked trust anchors after they
1420 have been kept in the revoked list for this long. Default is 30
1421 days as per the RFC.
1422 keep-missing: <seconds>
1424 Instruct the auto-trust-anchor-file probe mechanism for RFC5011
1425 autotrust updates to remove missing trust anchors after they
1426 have been unseen for this long. This cleans up the state file
1427 if the target zone does not perform trust anchor revocation, so
1428 this makes the auto probe mechanism work with zones that perform
1429 regular (non-5011) rollovers. The default is 366 days. The
1430 value 0 does not remove missing anchors, as per the RFC.
1431 permit-small-holddown: <yes or no>
1433 Debug option that allows the autotrust 5011 rollover timers to
1434 assume very small values. Default is no.
1435 key-cache-size: <number>
1437 Number of bytes size of the key cache. Default is 4 megabytes.
1438 A plain number is in bytes, append 'k', 'm' or 'g' for kilo‐
1439 bytes, megabytes or gigabytes (1024*1024 bytes in a megabyte).
1440 key-cache-slabs: <number>
1442 Number of slabs in the key cache. Slabs reduce lock contention
1443 by threads. Must be set to a power of 2. Setting (close) to the
1444 number of cpus is a reasonable guess.
1445 neg-cache-size: <number>
1447 Number of bytes size of the aggressive negative cache. Default
1448 is 1 megabyte. A plain number is in bytes, append 'k', 'm' or
1449 'g' for kilobytes, megabytes or gigabytes (1024*1024 bytes in a
1450 megabyte).
1451 unblock-lan-zones: <yes or no>
1453 Default is disabled. If enabled, then for private address
1454 space, the reverse lookups are no longer filtered. This allows
1455 Unbound when running as dns service on a host where it provides
1456 service for that host, to put out all of the queries for the
1457 'lan' upstream. When enabled, only localhost, 127.0.0.1 reverse
1458 and ::1 reverse zones are configured with default local zones.
1459 Disable the option when Unbound is running as a (DHCP-) DNS net‐
1460 work resolver for a group of machines, where such lookups should
1461 be filtered (RFC compliance), this also stops potential data
1462 leakage about the local network to the upstream DNS servers.
1463 insecure-lan-zones: <yes or no>
1465 Default is disabled. If enabled, then reverse lookups in pri‐
1466 vate address space are not validated. This is usually required
1467 whenever unblock-lan-zones is used.
1468 local-zone: <zone> <type>
1470 Configure a local zone. The type determines the answer to give
1471 if there is no match from local-data. The types are deny,
1472 refuse, static, transparent, redirect, nodefault, typetranspar‐
1473 ent, inform, inform_deny, inform_redirect, always_transparent,
1474 always_refuse, always_nxdomain, always_null, noview, and are ex‐
1475 plained below. After that the default settings are listed. Use
1476 local-data: to enter data into the local zone. Answers for local
1477 zones are authoritative DNS answers. By default the zones are
1478 class IN.
1479 If you need more complicated authoritative data, with referrals,
1481 wildcards, CNAME/DNAME support, or DNSSEC authoritative service,
1482 setup a stub-zone for it as detailed in the stub zone section
1483 below. A stub-zone can be used to have unbound send queries to
1484 another server, an authoritative server, to fetch the informa‐
1485 tion. With a forward-zone, unbound sends queries to a server
1486 that is a recursive server to fetch the information. With an
1487 auth-zone a zone can be loaded from file and used, it can be
1488 used like a local-zone for users downstream, or the auth-zone
1489 information can be used to fetch information from when resolving
1490 like it is an upstream server. The forward-zone and auth-zone
1491 options are described in their sections below. If you want to
1492 perform filtering of the information that the users can fetch,
1493 the local-zone and local-data statements allow for this, but
1494 also the rpz functionality can be used, described in the RPZ
1495 section.
1496 deny Do not send an answer, drop the query. If there is a match
1498 from local data, the query is answered.
1499 refuse
1501 Send an error message reply, with rcode REFUSED. If there is
1502 a match from local data, the query is answered.
1503 static
1505 If there is a match from local data, the query is answered.
1506 Otherwise, the query is answered with nodata or nxdomain.
1507 For a negative answer a SOA is included in the answer if
1508 present as local-data for the zone apex domain.
1509 transparent
1511 If there is a match from local data, the query is answered.
1512 Otherwise if the query has a different name, the query is re‐
1513 solved normally. If the query is for a name given in local‐
1514 data but no such type of data is given in localdata, then a
1515 noerror nodata answer is returned. If no local-zone is given
1516 local-data causes a transparent zone to be created by de‐
1517 fault.
1518 typetransparent
1520 If there is a match from local data, the query is answered.
1521 If the query is for a different name, or for the same name
1522 but for a different type, the query is resolved normally.
1523 So, similar to transparent but types that are not listed in
1524 local data are resolved normally, so if an A record is in the
1525 local data that does not cause a nodata reply for AAAA
1526 queries.
1527 redirect
1529 The query is answered from the local data for the zone name.
1530 There may be no local data beneath the zone name. This an‐
1531 swers queries for the zone, and all subdomains of the zone
1532 with the local data for the zone. It can be used to redirect
1533 a domain to return a different address record to the end
1534 user, with local-zone: "example.com." redirect and lo‐
1535 cal-data: "example.com. A 127.0.0.1" queries for www.exam‐
1536 ple.com and www.foo.example.com are redirected, so that users
1537 with web browsers cannot access sites with suffix exam‐
1538 ple.com.
1539 inform
1541 The query is answered normally, same as transparent. The
1542 client IP address (@portnumber) is printed to the logfile.
1543 The log message is: timestamp, unbound-pid, info: zonename
1544 inform IP@port queryname type class. This option can be used
1545 for normal resolution, but machines looking up infected names
1546 are logged, eg. to run antivirus on them.
1547 inform_deny
1549 The query is dropped, like 'deny', and logged, like 'inform'.
1550 Ie. find infected machines without answering the queries.
1551 inform_redirect
1553 The query is redirected, like 'redirect', and logged, like
1554 'inform'. Ie. answer queries with fixed data and also log
1555 the machines that ask.
1556 always_transparent
1558 Like transparent, but ignores local data and resolves nor‐
1559 mally.
1560 always_refuse
1562 Like refuse, but ignores local data and refuses the query.
1563 always_nxdomain
1565 Like static, but ignores local data and returns nxdomain for
1566 the query.
1567 always_nodata
1569 Like static, but ignores local data and returns nodata for
1570 the query.
1571 always_deny
1573 Like deny, but ignores local data and drops the query.
1574 always_null
1576 Always returns 0.0.0.0 or ::0 for every name in the zone.
1577 Like redirect with zero data for A and AAAA. Ignores local
1578 data in the zone. Used for some block lists.
1579 noview
1581 Breaks out of that view and moves towards the global local
1582 zones for answer to the query. If the view first is no,
1583 it'll resolve normally. If view first is enabled, it'll
1584 break perform that step and check the global answers. For
1585 when the view has view specific overrides but some zone has
1586 to be answered from global local zone contents.
1587 nodefault
1589 Used to turn off default contents for AS112 zones. The other
1590 types also turn off default contents for the zone. The 'node‐
1591 fault' option has no other effect than turning off default
1592 contents for the given zone. Use nodefault if you use ex‐
1593 actly that zone, if you want to use a subzone, use transpar‐
1594 ent.
1595 The default zones are localhost, reverse 127.0.0.1 and ::1, the
1597 home.arpa, the onion, test, invalid and the AS112 zones. The AS112
1598 zones are reverse DNS zones for private use and reserved IP addresses
1599 for which the servers on the internet cannot provide correct answers.
1600 They are configured by default to give nxdomain (no reverse informa‐
1601 tion) answers. The defaults can be turned off by specifying your own
1602 local-zone of that name, or using the 'nodefault' type. Below is a list
1603 of the default zone contents.
1604 localhost
1606 The IP4 and IP6 localhost information is given. NS and SOA
1607 records are provided for completeness and to satisfy some DNS
1608 update tools. Default content:
1609 local-zone: "localhost." redirect
1610 local-data: "localhost. 10800 IN NS localhost."
1611 local-data: "localhost. 10800 IN
1612 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1613 local-data: "localhost. 10800 IN A 127.0.0.1"
1614 local-data: "localhost. 10800 IN AAAA ::1"
1615 reverse IPv4 loopback
1617 Default content:
1618 local-zone: "127.in-addr.arpa." static
1619 local-data: "127.in-addr.arpa. 10800 IN NS localhost."
1620 local-data: "127.in-addr.arpa. 10800 IN
1621 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1622 local-data: "1.0.0.127.in-addr.arpa. 10800 IN
1623 PTR localhost."
1624 reverse IPv6 loopback
1626 Default content:
1627 local-zone: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
1628 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa." static
1629 local-data: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
1630 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa. 10800 IN
1631 NS localhost."
1632 local-data: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
1633 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa. 10800 IN
1634 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1635 local-data: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
1636 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa. 10800 IN
1637 PTR localhost."
1638 home.arpa (RFC 8375)
1640 Default content:
1641 local-zone: "home.arpa." static
1642 local-data: "home.arpa. 10800 IN NS localhost."
1643 local-data: "home.arpa. 10800 IN
1644 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1645 onion (RFC 7686)
1647 Default content:
1648 local-zone: "onion." static
1649 local-data: "onion. 10800 IN NS localhost."
1650 local-data: "onion. 10800 IN
1651 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1652 test (RFC 6761)
1654 Default content:
1655 local-zone: "test." static
1656 local-data: "test. 10800 IN NS localhost."
1657 local-data: "test. 10800 IN
1658 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1659 invalid (RFC 6761)
1661 Default content:
1662 local-zone: "invalid." static
1663 local-data: "invalid. 10800 IN NS localhost."
1664 local-data: "invalid. 10800 IN
1665 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1666 reverse RFC1918 local use zones
1668 Reverse data for zones 10.in-addr.arpa, 16.172.in-addr.arpa
1669 to 31.172.in-addr.arpa, 168.192.in-addr.arpa. The lo‐
1670 cal-zone: is set static and as local-data: SOA and NS records
1671 are provided.
1672 reverse RFC3330 IP4 this, link-local, testnet and broadcast
1674 Reverse data for zones 0.in-addr.arpa, 254.169.in-addr.arpa,
1675 2.0.192.in-addr.arpa (TEST NET 1), 100.51.198.in-addr.arpa
1676 (TEST NET 2), 113.0.203.in-addr.arpa (TEST NET 3),
1677 255.255.255.255.in-addr.arpa. And from 64.100.in-addr.arpa
1678 to 127.100.in-addr.arpa (Shared Address Space).
1679 reverse RFC4291 IP6 unspecified
1681 Reverse data for zone
1682 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
1683 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa.
1684 reverse RFC4193 IPv6 Locally Assigned Local Addresses
1686 Reverse data for zone D.F.ip6.arpa.
1687 reverse RFC4291 IPv6 Link Local Addresses
1689 Reverse data for zones 8.E.F.ip6.arpa to B.E.F.ip6.arpa.
1690 reverse IPv6 Example Prefix
1692 Reverse data for zone 8.B.D.0.1.0.0.2.ip6.arpa. This zone is
1693 used for tutorials and examples. You can remove the block on
1694 this zone with:
1695 local-zone: 8.B.D.0.1.0.0.2.ip6.arpa. nodefault
1696 You can also selectively unblock a part of the zone by making
1697 that part transparent with a local-zone statement. This also
1698 works with the other default zones.
1699 local-data: "<resource record string>"
1701 Configure local data, which is served in reply to queries for it.
1702 The query has to match exactly unless you configure the local-zone
1703 as redirect. If not matched exactly, the local-zone type deter‐
1704 mines further processing. If local-data is configured that is not
1705 a subdomain of a local-zone, a transparent local-zone is config‐
1706 ured. For record types such as TXT, use single quotes, as in lo‐
1707 cal-data: 'example. TXT "text"'.
1708 If you need more complicated authoritative data, with referrals,
1710 wildcards, CNAME/DNAME support, or DNSSEC authoritative service,
1711 setup a stub-zone for it as detailed in the stub zone section be‐
1712 low.
1713 local-data-ptr: "IPaddr name"
1715 Configure local data shorthand for a PTR record with the reversed
1716 IPv4 or IPv6 address and the host name. For example "192.0.2.4
1717 www.example.com". TTL can be inserted like this: "2001:DB8::4
1718 7200 www.example.com"
1719 local-zone-tag: <zone> <"list of tags">
1721 Assign tags to localzones. Tagged localzones will only be applied
1722 when the used access-control element has a matching tag. Tags must
1723 be defined in define-tags. Enclose list of tags in quotes ("")
1724 and put spaces between tags. When there are multiple tags it
1725 checks if the intersection of the list of tags for the query and
1726 local-zone-tag is non-empty.
1727 local-zone-override: <zone> <IP netblock> <type>
1729 Override the localzone type for queries from addresses matching
1730 netblock. Use this localzone type, regardless the type configured
1731 for the local-zone (both tagged and untagged) and regardless the
1732 type configured using access-control-tag-action.
1733 response-ip: <IP-netblock> <action>
1735 This requires use of the "respip" module.
1736 If the IP address in an AAAA or A RR in the answer section of a
1738 response matches the specified IP netblock, the specified action
1739 will apply. <action> has generally the same semantics as that for
1740 access-control-tag-action, but there are some exceptions.
1741 Actions for response-ip are different from those for local-zone in
1743 that in case of the former there is no point of such conditions as
1744 "the query matches it but there is no local data". Because of
1745 this difference, the semantics of response-ip actions are modified
1746 or simplified as follows: The static, refuse, transparent, type‐
1747 transparent, and nodefault actions are invalid for response-ip.
1748 Using any of these will cause the configuration to be rejected as
1749 faulty. The deny action is non-conditional, i.e. it always results
1750 in dropping the corresponding query. The resolution result before
1751 applying the deny action is still cached and can be used for other
1752 queries.
1753 response-ip-data: <IP-netblock> <"resource record string">
1755 This requires use of the "respip" module.
1756 This specifies the action data for response-ip with action being
1758 to redirect as specified by "resource record string". "Resource
1759 record string" is similar to that of access-control-tag-action,
1760 but it must be of either AAAA, A or CNAME types. If the IP-net‐
1761 block is an IPv6/IPv4 prefix, the record must be AAAA/A respec‐
1762 tively, unless it is a CNAME (which can be used for both versions
1763 of IP netblocks). If it is CNAME there must not be more than one
1764 response-ip-data for the same IP-netblock. Also, CNAME and other
1765 types of records must not coexist for the same IP-netblock, fol‐
1766 lowing the normal rules for CNAME records. The textual domain
1767 name for the CNAME does not have to be explicitly terminated with
1768 a dot ("."); the root name is assumed to be the origin for the
1769 name.
1770 response-ip-tag: <IP-netblock> <"list of tags">
1772 This requires use of the "respip" module.
1773 Assign tags to response IP-netblocks. If the IP address in an
1775 AAAA or A RR in the answer section of a response matches the spec‐
1776 ified IP-netblock, the specified tags are assigned to the IP ad‐
1777 dress. Then, if an access-control-tag is defined for the client
1778 and it includes one of the tags for the response IP, the corre‐
1779 sponding access-control-tag-action will apply. Tag matching rule
1780 is the same as that for access-control-tag and local-zones. Un‐
1781 like local-zone-tag, response-ip-tag can be defined for an IP-net‐
1782 block even if no response-ip is defined for that netblock. If
1783 multiple response-ip-tag options are specified for the same IP-
1784 netblock in different statements, all but the first will be ig‐
1785 nored. However, this will not be flagged as a configuration er‐
1786 ror, but the result is probably not what was intended.
1787 Actions specified in an access-control-tag-action that has a
1789 matching tag with response-ip-tag can be those that are "invalid"
1790 for response-ip listed above, since access-control-tag-actions can
1791 be shared with local zones. For these actions, if they behave
1792 differently depending on whether local data exists or not in case
1793 of local zones, the behavior for response-ip-data will generally
1794 result in NOERROR/NODATA instead of NXDOMAIN, since the response-
1795 ip data are inherently type specific, and non-existence of data
1796 does not indicate anything about the existence or non-existence of
1797 the qname itself. For example, if the matching tag action is
1798 static but there is no data for the corresponding response-ip con‐
1799 figuration, then the result will be NOERROR/NODATA. The only case
1800 where NXDOMAIN is returned is when an always_nxdomain action ap‐
1801 plies.
1802 ratelimit: <number or 0>
1804 Enable ratelimiting of queries sent to nameserver for performing
1805 recursion. If 0, the default, it is disabled. This option is ex‐
1806 perimental at this time. The ratelimit is in queries per second
1807 that are allowed. More queries are turned away with an error
1808 (servfail). This stops recursive floods, eg. random query names,
1809 but not spoofed reflection floods. Cached responses are not rate‐
1810 limited by this setting. The zone of the query is determined by
1811 examining the nameservers for it, the zone name is used to keep
1812 track of the rate. For example, 1000 may be a suitable value to
1813 stop the server from being overloaded with random names, and keeps
1814 Unbound from sending traffic to the nameservers for those zones.
1815 Configured forwarders are excluded from ratelimiting.
1816 ratelimit-size: <memory size>
1818 Give the size of the data structure in which the current ongoing
1819 rates are kept track in. Default 4m. In bytes or use m(mega),
1820 k(kilo), g(giga). The ratelimit structure is small, so this data
1821 structure likely does not need to be large.
1822 ratelimit-slabs: <number>
1824 Give power of 2 number of slabs, this is used to reduce lock con‐
1825 tention in the ratelimit tracking data structure. Close to the
1826 number of cpus is a fairly good setting.
1827 ratelimit-factor: <number>
1829 Set the amount of queries to rate limit when the limit is ex‐
1830 ceeded. If set to 0, all queries are dropped for domains where
1831 the limit is exceeded. If set to another value, 1 in that number
1832 is allowed through to complete. Default is 10, allowing 1/10
1833 traffic to flow normally. This can make ordinary queries complete
1834 (if repeatedly queried for), and enter the cache, whilst also mit‐
1835 igating the traffic flow by the factor given.
1836 ratelimit-backoff: <yes or no>
1838 If enabled, the ratelimit is treated as a hard failure instead of
1839 the default maximum allowed constant rate. When the limit is
1840 reached, traffic is ratelimited and demand continues to be kept
1841 track of for a 2 second rate window. No traffic is allowed, ex‐
1842 cept for ratelimit-factor, until demand decreases below the con‐
1843 figured ratelimit for a 2 second rate window. Useful to set rate‐
1844 limit to a suspicious rate to aggressively limit unusually high
1845 traffic. Default is off.
1846 ratelimit-for-domain: <domain> <number qps or 0>
1848 Override the global ratelimit for an exact match domain name with
1849 the listed number. You can give this for any number of names.
1850 For example, for a top-level-domain you may want to have a higher
1851 limit than other names. A value of 0 will disable ratelimiting
1852 for that domain.
1853 ratelimit-below-domain: <domain> <number qps or 0>
1855 Override the global ratelimit for a domain name that ends in this
1856 name. You can give this multiple times, it then describes differ‐
1857 ent settings in different parts of the namespace. The closest
1858 matching suffix is used to determine the qps limit. The rate for
1859 the exact matching domain name is not changed, use rate‐
1860 limit-for-domain to set that, you might want to use different set‐
1861 tings for a top-level-domain and subdomains. A value of 0 will
1862 disable ratelimiting for domain names that end in this name.
1863 ip-ratelimit: <number or 0>
1865 Enable global ratelimiting of queries accepted per IP address. If
1866 0, the default, it is disabled. This option is experimental at
1867 this time. The ratelimit is in queries per second that are al‐
1868 lowed. More queries are completely dropped and will not receive a
1869 reply, SERVFAIL or otherwise. IP ratelimiting happens before
1870 looking in the cache. This may be useful for mitigating amplifica‐
1871 tion attacks.
1872 ip-ratelimit-size: <memory size>
1874 Give the size of the data structure in which the current ongoing
1875 rates are kept track in. Default 4m. In bytes or use m(mega),
1876 k(kilo), g(giga). The ip ratelimit structure is small, so this
1877 data structure likely does not need to be large.
1878 ip-ratelimit-slabs: <number>
1880 Give power of 2 number of slabs, this is used to reduce lock con‐
1881 tention in the ip ratelimit tracking data structure. Close to the
1882 number of cpus is a fairly good setting.
1883 ip-ratelimit-factor: <number>
1885 Set the amount of queries to rate limit when the limit is ex‐
1886 ceeded. If set to 0, all queries are dropped for addresses where
1887 the limit is exceeded. If set to another value, 1 in that number
1888 is allowed through to complete. Default is 10, allowing 1/10
1889 traffic to flow normally. This can make ordinary queries complete
1890 (if repeatedly queried for), and enter the cache, whilst also mit‐
1891 igating the traffic flow by the factor given.
1892 ip-ratelimit-backoff: <yes or no>
1894 If enabled, the ratelimit is treated as a hard failure instead of
1895 the default maximum allowed constant rate. When the limit is
1896 reached, traffic is ratelimited and demand continues to be kept
1897 track of for a 2 second rate window. No traffic is allowed, ex‐
1898 cept for ip-ratelimit-factor, until demand decreases below the
1899 configured ratelimit for a 2 second rate window. Useful to set
1900 ip-ratelimit to a suspicious rate to aggressively limit unusually
1901 high traffic. Default is off.
1902 outbound-msg-retry: <number>
1904 The number of retries, per upstream nameserver in a delegation,
1905 that Unbound will attempt in case a throwaway response is re‐
1906 ceived. No response (timeout) contributes to the retry counter.
1907 If a forward/stub zone is used, this is the number of retries per
1908 nameserver in the zone. Default is 5.
1909 max-sent-count: <number>
1911 Hard limit on the number of outgoing queries Unbound will make
1912 while resolving a name, making sure large NS sets do not loop.
1913 Results in SERVFAIL when reached. It resets on query restarts
1914 (e.g., CNAME) and referrals. Default is 32.
1915 max-query-restarts: <number>
1917 Hard limit on the number of times Unbound is allowed to restart a
1918 query upon encountering a CNAME record. Results in SERVFAIL when
1919 reached. Changing this value needs caution as it can allow long
1920 CNAME chains to be accepted, where Unbound needs to verify (re‐
1921 solve) each link individually. Default is 11.
1922 fast-server-permil: <number>
1924 Specify how many times out of 1000 to pick from the set of fastest
1925 servers. 0 turns the feature off. A value of 900 would pick from
1926 the fastest servers 90 percent of the time, and would perform nor‐
1927 mal exploration of random servers for the remaining time. When
1928 prefetch is enabled (or serve-expired), such prefetches are not
1929 sped up, because there is no one waiting for it, and it presents a
1930 good moment to perform server exploration. The fast-server-num op‐
1931 tion can be used to specify the size of the fastest servers set.
1932 The default for fast-server-permil is 0.
1933 fast-server-num: <number>
1935 Set the number of servers that should be used for fast server se‐
1936 lection. Only use the fastest specified number of servers with the
1937 fast-server-permil option, that turns this on or off. The default
1938 is to use the fastest 3 servers.
1939 edns-client-string: <IP netblock> <string>
1941 Include an EDNS0 option containing configured ascii string in
1942 queries with destination address matching the configured IP net‐
1943 block. This configuration option can be used multiple times. The
1944 most specific match will be used.
1945 edns-client-string-opcode: <opcode>
1947 EDNS0 option code for the edns-client-string option, from 0 to
1948 65535. A value from the `Reserved for Local/Experimental` range
1949 (65001-65534) should be used. Default is 65001.
1950 ede: <yes or no>
1952 If enabled, Unbound will respond with Extended DNS Error codes
1953 (RFC8914). These EDEs attach informative error messages to a re‐
1954 sponse for various errors. Default is "no".
1955 When the val-log-level option is also set to 2, responses with Ex‐
1957 tended DNS Errors concerning DNSSEC failures that are not served
1958 from cache, will also contain a descriptive text message about the
1959 reason for the failure.
1960 ede-serve-expired: <yes or no>
1962 If enabled, Unbound will attach an Extended DNS Error (RFC8914)
1963 Code 3 - Stale Answer as EDNS0 option to the expired response.
1964 Note that this will not attach the EDE code without setting the
1965 global ede option to "yes" as well. Default is "no".
1966 Remote Control Options
1968 In the remote-control: clause are the declarations for the remote con‐
1969 trol facility. If this is enabled, the unbound-control(8) utility can
1970 be used to send commands to the running Unbound server. The server
1971 uses these clauses to setup TLSv1 security for the connection. The un‐
1972 bound-control(8) utility also reads the remote-control section for op‐
1973 tions. To setup the correct self-signed certificates use the un‐
1974 bound-control-setup(8) utility.
1975 control-enable: <yes or no>
1977 The option is used to enable remote control, default is "no". If
1978 turned off, the server does not listen for control commands.
1979 control-interface: <ip address or interface name or path>
1981 Give IPv4 or IPv6 addresses or local socket path to listen on for
1982 control commands. If an interface name is used instead of an ip
1983 address, the list of ip addresses on that interface are used. By
1984 default localhost (127.0.0.1 and ::1) is listened to. Use 0.0.0.0
1985 and ::0 to listen to all interfaces. If you change this and per‐
1986 missions have been dropped, you must restart the server for the
1987 change to take effect.
1988 If you set it to an absolute path, a unix domain socket is used.
1990 This socket does not use the certificates and keys, so those files
1991 need not be present. To restrict access, Unbound sets permissions
1992 on the file to the user and group that is configured, the access
1993 bits are set to allow the group members to access the control
1994 socket file. Put users that need to access the socket in the that
1995 group. To restrict access further, create a directory to put the
1996 control socket in and restrict access to that directory.
1997 control-port: <port number>
1999 The port number to listen on for IPv4 or IPv6 control interfaces,
2000 default is 8953. If you change this and permissions have been
2001 dropped, you must restart the server for the change to take ef‐
2002 fect.
2003 control-use-cert: <yes or no>
2005 For localhost control-interface you can disable the use of TLS by
2006 setting this option to "no", default is "yes". For local sockets,
2007 TLS is disabled and the value of this option is ignored.
2008 server-key-file: <private key file>
2010 Path to the server private key, by default unbound_server.key.
2011 This file is generated by the unbound-control-setup utility. This
2012 file is used by the Unbound server, but not by unbound-control.
2013 server-cert-file: <certificate file.pem>
2015 Path to the server self signed certificate, by default un‐
2016 bound_server.pem. This file is generated by the unbound-con‐
2017 trol-setup utility. This file is used by the Unbound server, and
2018 also by unbound-control.
2019 control-key-file: <private key file>
2021 Path to the control client private key, by default unbound_con‐
2022 trol.key. This file is generated by the unbound-control-setup
2023 utility. This file is used by unbound-control.
2024 control-cert-file: <certificate file.pem>
2026 Path to the control client certificate, by default unbound_con‐
2027 trol.pem. This certificate has to be signed with the server cer‐
2028 tificate. This file is generated by the unbound-control-setup
2029 utility. This file is used by unbound-control.
2030 Stub Zone Options
2032 There may be multiple stub-zone: clauses. Each with a name: and zero or
2033 more hostnames or IP addresses. For the stub zone this list of name‐
2034 servers is used. Class IN is assumed. The servers should be authority
2035 servers, not recursors; Unbound performs the recursive processing it‐
2036 self for stub zones.
2037 The stub zone can be used to configure authoritative data to be used by
2039 the resolver that cannot be accessed using the public internet servers.
2040 This is useful for company-local data or private zones. Setup an au‐
2041 thoritative server on a different host (or different port). Enter a
2042 config entry for Unbound with stub-addr: <ip address of host[@port]>.
2043 The Unbound resolver can then access the data, without referring to the
2044 public internet for it.
2045 This setup allows DNSSEC signed zones to be served by that authorita‐
2047 tive server, in which case a trusted key entry with the public key can
2048 be put in config, so that Unbound can validate the data and set the AD
2049 bit on replies for the private zone (authoritative servers do not set
2050 the AD bit). This setup makes Unbound capable of answering queries for
2051 the private zone, and can even set the AD bit ('authentic'), but the AA
2052 ('authoritative') bit is not set on these replies.
2053 Consider adding server: statements for domain-insecure: and for lo‐
2055 cal-zone: name nodefault for the zone if it is a locally served zone.
2056 The insecure clause stops DNSSEC from invalidating the zone. The local
2057 zone nodefault (or transparent) clause makes the (reverse-) zone bypass
2058 Unbound's filtering of RFC1918 zones.
2059 name: <domain name>
2061 Name of the stub zone. This is the full domain name of the zone.
2062 stub-host: <domain name>
2064 Name of stub zone nameserver. Is itself resolved before it is
2065 used. To use a nondefault port for DNS communication append '@'
2066 with the port number. If tls is enabled, then you can append a
2067 '#' and a name, then it'll check the tls authentication certifi‐
2068 cates with that name. If you combine the '@' and '#', the '@'
2069 comes first. If only '#' is used the default port is the con‐
2070 figured tls-port.
2071 stub-addr: <IP address>
2073 IP address of stub zone nameserver. Can be IP 4 or IP 6. To use
2074 a nondefault port for DNS communication append '@' with the port
2075 number. If tls is enabled, then you can append a '#' and a
2076 name, then it'll check the tls authentication certificates with
2077 that name. If you combine the '@' and '#', the '@' comes first.
2078 If only '#' is used the default port is the configured tls-port.
2079 stub-prime: <yes or no>
2081 This option is by default no. If enabled it performs NS set
2082 priming, which is similar to root hints, where it starts using
2083 the list of nameservers currently published by the zone. Thus,
2084 if the hint list is slightly outdated, the resolver picks up a
2085 correct list online.
2086 stub-first: <yes or no>
2088 If enabled, a query is attempted without the stub clause if it
2089 fails. The data could not be retrieved and would have caused
2090 SERVFAIL because the servers are unreachable, instead it is
2091 tried without this clause. The default is no.
2092 stub-tls-upstream: <yes or no>
2094 Enabled or disable whether the queries to this stub use TLS for
2095 transport. Default is no.
2096 stub-ssl-upstream: <yes or no>
2098 Alternate syntax for stub-tls-upstream.
2099 stub-tcp-upstream: <yes or no>
2101 If it is set to "yes" then upstream queries use TCP only for
2102 transport regardless of global flag tcp-upstream. Default is
2103 no.
2104 stub-no-cache: <yes or no>
2106 Default is no. If enabled, data inside the stub is not cached.
2107 This is useful when you want immediate changes to be visible.
2108 Forward Zone Options
2110 There may be multiple forward-zone: clauses. Each with a name: and zero
2111 or more hostnames or IP addresses. For the forward zone this list of
2112 nameservers is used to forward the queries to. The servers listed as
2113 forward-host: and forward-addr: have to handle further recursion for
2114 the query. Thus, those servers are not authority servers, but are
2115 (just like Unbound is) recursive servers too; Unbound does not perform
2116 recursion itself for the forward zone, it lets the remote server do it.
2117 Class IN is assumed. CNAMEs are chased by Unbound itself, asking the
2118 remote server for every name in the indirection chain, to protect the
2119 local cache from illegal indirect referenced items. A forward-zone en‐
2120 try with name "." and a forward-addr target will forward all queries to
2121 that other server (unless it can answer from the cache).
2122 name: <domain name>
2124 Name of the forward zone. This is the full domain name of the
2125 zone.
2126 forward-host: <domain name>
2128 Name of server to forward to. Is itself resolved before it is
2129 used. To use a nondefault port for DNS communication append '@'
2130 with the port number. If tls is enabled, then you can append a
2131 '#' and a name, then it'll check the tls authentication certifi‐
2132 cates with that name. If you combine the '@' and '#', the '@'
2133 comes first. If only '#' is used the default port is the con‐
2134 figured tls-port.
2135 forward-addr: <IP address>
2137 IP address of server to forward to. Can be IP 4 or IP 6. To use
2138 a nondefault port for DNS communication append '@' with the port
2139 number. If tls is enabled, then you can append a '#' and a
2140 name, then it'll check the tls authentication certificates with
2141 that name. If you combine the '@' and '#', the '@' comes first.
2142 If only '#' is used the default port is the configured tls-port.
2143 At high verbosity it logs the TLS certificate, with TLS enabled.
2145 If you leave out the '#' and auth name from the forward-addr,
2146 any name is accepted. The cert must also match a CA from the
2147 tls-cert-bundle.
2148 forward-first: <yes or no>
2150 If a forwarded query is met with a SERVFAIL error, and this op‐
2151 tion is enabled, Unbound will fall back to normal recursive res‐
2152 olution for this query as if no query forwarding had been speci‐
2153 fied. The default is "no".
2154 forward-tls-upstream: <yes or no>
2156 Enabled or disable whether the queries to this forwarder use TLS
2157 for transport. Default is no. If you enable this, also config‐
2158 ure a tls-cert-bundle or use tls-win-cert to load CA certs, oth‐
2159 erwise the connections cannot be authenticated.
2160 forward-ssl-upstream: <yes or no>
2162 Alternate syntax for forward-tls-upstream.
2163 forward-tcp-upstream: <yes or no>
2165 If it is set to "yes" then upstream queries use TCP only for
2166 transport regardless of global flag tcp-upstream. Default is
2167 no.
2168 forward-no-cache: <yes or no>
2170 Default is no. If enabled, data inside the forward is not
2171 cached. This is useful when you want immediate changes to be
2172 visible.
2173 Authority Zone Options
2175 Authority zones are configured with auth-zone:, and each one must have
2176 a name:. There can be multiple ones, by listing multiple auth-zone
2177 clauses, each with a different name, pertaining to that part of the
2178 namespace. The authority zone with the name closest to the name looked
2179 up is used. Authority zones are processed after local-zones and before
2180 cache (for-downstream: yes), and when used in this manner make Unbound
2181 respond like an authority server. Authority zones are also processed
2182 after cache, just before going to the network to fetch information for
2183 recursion (for-upstream: yes), and when used in this manner provide a
2184 local copy of an authority server that speeds up lookups of that data.
2185 Authority zones can be read from zonefile. And can be kept updated via
2187 AXFR and IXFR. After update the zonefile is rewritten. The update
2188 mechanism uses the SOA timer values and performs SOA UDP queries to de‐
2189 tect zone changes.
2190 If the update fetch fails, the timers in the SOA record are used to
2192 time another fetch attempt. Until the SOA expiry timer is reached.
2193 Then the zone is expired. When a zone is expired, queries are SERV‐
2194 FAIL, and any new serial number is accepted from the primary (even if
2195 older), and if fallback is enabled, the fallback activates to fetch
2196 from the upstream instead of the SERVFAIL.
2197 name: <zone name>
2199 Name of the authority zone.
2200 primary: <IP address or host name>
2202 Where to download a copy of the zone from, with AXFR and IXFR.
2203 Multiple primaries can be specified. They are all tried if one
2204 fails. To use a nondefault port for DNS communication append
2205 '@' with the port number. You can append a '#' and a name, then
2206 AXFR over TLS can be used and the tls authentication certifi‐
2207 cates will be checked with that name. If you combine the '@'
2208 and '#', the '@' comes first. If you point it at another Un‐
2209 bound instance, it would not work because that does not support
2210 AXFR/IXFR for the zone, but if you used url: to download the
2211 zonefile as a text file from a webserver that would work. If
2212 you specify the hostname, you cannot use the domain from the
2213 zonefile, because it may not have that when retrieving that
2214 data, instead use a plain IP address to avoid a circular depen‐
2215 dency on retrieving that IP address.
2216 master: <IP address or host name>
2218 Alternate syntax for primary.
2219 url: <url to zonefile>
2221 Where to download a zonefile for the zone. With http or https.
2222 An example for the url is "http://www.example.com/exam‐
2223 ple.org.zone". Multiple url statements can be given, they are
2224 tried in turn. If only urls are given the SOA refresh timer is
2225 used to wait for making new downloads. If also primaries are
2226 listed, the primaries are first probed with UDP SOA queries to
2227 see if the SOA serial number has changed, reducing the number of
2228 downloads. If none of the urls work, the primaries are tried
2229 with IXFR and AXFR. For https, the tls-cert-bundle and the
2230 hostname from the url are used to authenticate the connection.
2231 If you specify a hostname in the URL, you cannot use the domain
2232 from the zonefile, because it may not have that when retrieving
2233 that data, instead use a plain IP address to avoid a circular
2234 dependency on retrieving that IP address. Avoid dependencies on
2235 name lookups by using a notation like "http://192.0.2.1/unbound-
2236 primaries/example.com.zone", with an explicit IP address.
2237 allow-notify: <IP address or host name or netblockIP/prefix>
2239 With allow-notify you can specify additional sources of noti‐
2240 fies. When notified, the server attempts to first probe and
2241 then zone transfer. If the notify is from a primary, it first
2242 attempts that primary. Otherwise other primaries are attempted.
2243 If there are no primaries, but only urls, the file is downloaded
2244 when notified. The primaries from primary: and url: statements
2245 are allowed notify by default.
2246 fallback-enabled: <yes or no>
2248 Default no. If enabled, Unbound falls back to querying the in‐
2249 ternet as a resolver for this zone when lookups fail. For exam‐
2250 ple for DNSSEC validation failures.
2251 for-downstream: <yes or no>
2253 Default yes. If enabled, Unbound serves authority responses to
2254 downstream clients for this zone. This option makes Unbound be‐
2255 have, for the queries with names in this zone, like one of the
2256 authority servers for that zone. Turn it off if you want Un‐
2257 bound to provide recursion for the zone but have a local copy of
2258 zone data. If for-downstream is no and for-upstream is yes,
2259 then Unbound will DNSSEC validate the contents of the zone be‐
2260 fore serving the zone contents to clients and store validation
2261 results in the cache.
2262 for-upstream: <yes or no>
2264 Default yes. If enabled, Unbound fetches data from this data
2265 collection for answering recursion queries. Instead of sending
2266 queries over the internet to the authority servers for this
2267 zone, it'll fetch the data directly from the zone data. Turn it
2268 on when you want Unbound to provide recursion for downstream
2269 clients, and use the zone data as a local copy to speed up
2270 lookups.
2271 zonemd-check: <yes or no>
2273 Enable this option to check ZONEMD records in the zone. Default
2274 is disabled. The ZONEMD record is a checksum over the zone
2275 data. This includes glue in the zone and data from the zone
2276 file, and excludes comments from the zone file. When there is a
2277 DNSSEC chain of trust, DNSSEC signatures are checked too.
2278 zonemd-reject-absence: <yes or no>
2280 Enable this option to reject the absence of the ZONEMD record.
2281 Without it, when zonemd is not there it is not checked. It is
2282 useful to enable for a nonDNSSEC signed zone where the operator
2283 wants to require the verification of a ZONEMD, hence a missing
2284 ZONEMD is a failure. The action upon failure is controlled by
2285 the zonemd-permissive-mode option, for log only or also block
2286 the zone. The default is no.
2287 Without the option absence of a ZONEMD is only a failure when
2289 the zone is DNSSEC signed, and we have a trust anchor, and the
2290 DNSSEC verification of the absence of the ZONEMD fails. With
2291 the option enabled, the absence of a ZONEMD is always a failure,
2292 also for nonDNSSEC signed zones.
2293 zonefile: <filename>
2295 The filename where the zone is stored. If not given then no
2296 zonefile is used. If the file does not exist or is empty, Un‐
2297 bound will attempt to fetch zone data (eg. from the primary
2298 servers).
2299 View Options
2301 There may be multiple view: clauses. Each with a name: and zero or more
2302 local-zone and local-data elements. Views can also contain view-first,
2303 response-ip, response-ip-data and local-data-ptr elements. View can be
2304 mapped to requests by specifying the view name in an access-con‐
2305 trol-view element. Options from matching views will override global op‐
2306 tions. Global options will be used if no matching view is found, or
2307 when the matching view does not have the option specified.
2308 name: <view name>
2310 Name of the view. Must be unique. This name is used in ac‐
2311 cess-control-view elements.
2312 local-zone: <zone> <type>
2314 View specific local-zone elements. Has the same types and behav‐
2315 iour as the global local-zone elements. When there is at least
2316 one local-zone specified and view-first is no, the default lo‐
2317 cal-zones will be added to this view. Defaults can be disabled
2318 using the nodefault type. When view-first is yes or when a view
2319 does not have a local-zone, the global local-zone will be used
2320 including it's default zones.
2321 local-data: "<resource record string>"
2323 View specific local-data elements. Has the same behaviour as the
2324 global local-data elements.
2325 local-data-ptr: "IPaddr name"
2327 View specific local-data-ptr elements. Has the same behaviour as
2328 the global local-data-ptr elements.
2329 view-first: <yes or no>
2331 If enabled, it attempts to use the global local-zone and lo‐
2332 cal-data if there is no match in the view specific options. The
2333 default is no.
2334 Python Module Options
2336 The python: clause gives the settings for the python(1) script module.
2337 This module acts like the iterator and validator modules do, on queries
2338 and answers. To enable the script module it has to be compiled into
2339 the daemon, and the word "python" has to be put in the module-config:
2340 option (usually first, or between the validator and iterator). Multiple
2341 instances of the python module are supported by adding the word
2342 "python" more than once.
2343 If the chroot: option is enabled, you should make sure Python's library
2345 directory structure is bind mounted in the new root environment, see
2346 mount(8). Also the python-script: path should be specified as an abso‐
2347 lute path relative to the new root, or as a relative path to the work‐
2348 ing directory.
2349 python-script: <python file>
2351 The script file to load. Repeat this option for every python
2352 module instance added to the module-config: option.
2353 Dynamic Library Module Options
2355 The dynlib: clause gives the settings for the dynlib module. This mod‐
2356 ule is only a very small wrapper that allows dynamic modules to be
2357 loaded on runtime instead of being compiled into the application. To
2358 enable the dynlib module it has to be compiled into the daemon, and the
2359 word "dynlib" has to be put in the module-config: option. Multiple in‐
2360 stances of dynamic libraries are supported by adding the word "dynlib"
2361 more than once.
2362 The dynlib-file: path should be specified as an absolute path relative
2364 to the new path set by chroot: option, or as a relative path to the
2365 working directory.
2366 dynlib-file: <dynlib file>
2368 The dynamic library file to load. Repeat this option for every
2369 dynlib module instance added to the module-config: option.
2370 DNS64 Module Options
2372 The dns64 module must be configured in the module-config: "dns64 val‐
2373 idator iterator" directive and be compiled into the daemon to be en‐
2374 abled. These settings go in the server: section.
2375 dns64-prefix: <IPv6 prefix>
2377 This sets the DNS64 prefix to use to synthesize AAAA records
2378 with. It must be /96 or shorter. The default prefix is
2379 64:ff9b::/96.
2380 dns64-synthall: <yes or no>
2382 Debug option, default no. If enabled, synthesize all AAAA
2383 records despite the presence of actual AAAA records.
2384 dns64-ignore-aaaa: <name>
2386 List domain for which the AAAA records are ignored and the A
2387 record is used by dns64 processing instead. Can be entered mul‐
2388 tiple times, list a new domain for which it applies, one per
2389 line. Applies also to names underneath the name given.
2390 DNSCrypt Options
2392 The dnscrypt: clause gives the settings of the dnscrypt channel. While
2393 those options are available, they are only meaningful if Unbound was
2394 compiled with --enable-dnscrypt. Currently certificate and secret/pub‐
2395 lic keys cannot be generated by Unbound. You can use dnscrypt-wrapper
2396 to generate those: https://github.com/cofyc/dnscrypt-wrapper/blob/mas‐
2397 ter/README.md#usage
2398 dnscrypt-enable: <yes or no>
2400 Whether or not the dnscrypt config should be enabled. You may
2401 define configuration but not activate it. The default is no.
2402 dnscrypt-port: <port number>
2404 On which port should dnscrypt should be activated. Note that you
2405 should have a matching interface option defined in the server
2406 section for this port.
2407 dnscrypt-provider: <provider name>
2409 The provider name to use to distribute certificates. This is of
2410 the form: 2.dnscrypt-cert.example.com.. The name MUST end with a
2411 dot.
2412 dnscrypt-secret-key: <path to secret key file>
2414 Path to the time limited secret key file. This option may be
2415 specified multiple times.
2416 dnscrypt-provider-cert: <path to cert file>
2418 Path to the certificate related to the dnscrypt-secret-keys.
2419 This option may be specified multiple times.
2420 dnscrypt-provider-cert-rotated: <path to cert file>
2422 Path to a certificate that we should be able to serve existing
2423 connection from but do not want to advertise over
2424 dnscrypt-provider's TXT record certs distribution. A typical
2425 use case is when rotating certificates, existing clients may
2426 still use the client magic from the old cert in their queries
2427 until they fetch and update the new cert. Likewise, it would al‐
2428 low one to prime the new cert/key without distributing the new
2429 cert yet, this can be useful when using a network of servers us‐
2430 ing anycast and on which the configuration may not get updated
2431 at the exact same time. By priming the cert, the servers can
2432 handle both old and new certs traffic while distributing only
2433 one. This option may be specified multiple times.
2434 dnscrypt-shared-secret-cache-size: <memory size>
2436 Give the size of the data structure in which the shared secret
2437 keys are kept in. Default 4m. In bytes or use m(mega),
2438 k(kilo), g(giga). The shared secret cache is used when a same
2439 client is making multiple queries using the same public key. It
2440 saves a substantial amount of CPU.
2441 dnscrypt-shared-secret-cache-slabs: <number>
2443 Give power of 2 number of slabs, this is used to reduce lock
2444 contention in the dnscrypt shared secrets cache. Close to the
2445 number of cpus is a fairly good setting.
2446 dnscrypt-nonce-cache-size: <memory size>
2448 Give the size of the data structure in which the client nonces
2449 are kept in. Default 4m. In bytes or use m(mega), k(kilo),
2450 g(giga). The nonce cache is used to prevent dnscrypt message
2451 replaying. Client nonce should be unique for any pair of client
2452 pk/server sk.
2453 dnscrypt-nonce-cache-slabs: <number>
2455 Give power of 2 number of slabs, this is used to reduce lock
2456 contention in the dnscrypt nonce cache. Close to the number of
2457 cpus is a fairly good setting.
2458 EDNS Client Subnet Module Options
2460 The ECS module must be configured in the module-config: "subnetcache
2461 validator iterator" directive and be compiled into the daemon to be en‐
2462 abled. These settings go in the server: section.
2463 If the destination address is allowed in the configuration Unbound will
2465 add the EDNS0 option to the query containing the relevant part of the
2466 client's address. When an answer contains the ECS option the response
2467 and the option are placed in a specialized cache. If the authority in‐
2468 dicated no support, the response is stored in the regular cache.
2469 Additionally, when a client includes the option in its queries, Unbound
2471 will forward the option when sending the query to addresses that are
2472 explicitly allowed in the configuration using send-client-subnet. The
2473 option will always be forwarded, regardless the allowed addresses, if
2474 client-subnet-always-forward is set to yes. In this case the lookup in
2475 the regular cache is skipped.
2476 The maximum size of the ECS cache is controlled by 'msg-cache-size' in
2478 the configuration file. On top of that, for each query only 100 differ‐
2479 ent subnets are allowed to be stored for each address family. Exceeding
2480 that number, older entries will be purged from cache.
2481 This module does not interact with the serve-expired* and prefetch: op‐
2483 tions.
2484 send-client-subnet: <IP address>
2486 Send client source address to this authority. Append /num to in‐
2487 dicate a classless delegation netblock, for example like
2488 10.2.3.4/24 or 2001::11/64. Can be given multiple times. Author‐
2489 ities not listed will not receive edns-subnet information, un‐
2490 less domain in query is specified in client-subnet-zone.
2491 client-subnet-zone: <domain>
2493 Send client source address in queries for this domain and its
2494 subdomains. Can be given multiple times. Zones not listed will
2495 not receive edns-subnet information, unless hosted by authority
2496 specified in send-client-subnet.
2497 client-subnet-always-forward: <yes or no>
2499 Specify whether the ECS address check (configured using
2500 send-client-subnet) is applied for all queries, even if the
2501 triggering query contains an ECS record, or only for queries for
2502 which the ECS record is generated using the querier address (and
2503 therefore did not contain ECS data in the client query). If en‐
2504 abled, the address check is skipped when the client query con‐
2505 tains an ECS record. And the lookup in the regular cache is
2506 skipped. Default is no.
2507 max-client-subnet-ipv6: <number>
2509 Specifies the maximum prefix length of the client source address
2510 we are willing to expose to third parties for IPv6. Defaults to
2511 56.
2512 max-client-subnet-ipv4: <number>
2514 Specifies the maximum prefix length of the client source address
2515 we are willing to expose to third parties for IPv4. Defaults to
2516 24.
2517 min-client-subnet-ipv6: <number>
2519 Specifies the minimum prefix length of the IPv6 source mask we
2520 are willing to accept in queries. Shorter source masks result in
2521 REFUSED answers. Source mask of 0 is always accepted. Default is
2522 0.
2523 min-client-subnet-ipv4: <number>
2525 Specifies the minimum prefix length of the IPv4 source mask we
2526 are willing to accept in queries. Shorter source masks result in
2527 REFUSED answers. Source mask of 0 is always accepted. Default is
2528 0.
2529 max-ecs-tree-size-ipv4: <number>
2531 Specifies the maximum number of subnets ECS answers kept in the
2532 ECS radix tree. This number applies for each qname/qclass/qtype
2533 tuple. Defaults to 100.
2534 max-ecs-tree-size-ipv6: <number>
2536 Specifies the maximum number of subnets ECS answers kept in the
2537 ECS radix tree. This number applies for each qname/qclass/qtype
2538 tuple. Defaults to 100.
2539 Opportunistic IPsec Support Module Options
2541 The IPsec module must be configured in the module-config: "ipsecmod
2542 validator iterator" directive and be compiled into Unbound by using
2543 --enable-ipsecmod to be enabled. These settings go in the server: sec‐
2544 tion.
2545 When Unbound receives an A/AAAA query that is not in the cache and
2547 finds a valid answer, it will withhold returning the answer and instead
2548 will generate an IPSECKEY subquery for the same domain name. If an an‐
2549 swer was found, Unbound will call an external hook passing the follow‐
2550 ing arguments:
2551 QNAME
2553 Domain name of the A/AAAA and IPSECKEY query. In string for‐
2554 mat.
2555 IPSECKEY TTL
2557 TTL of the IPSECKEY RRset.
2558 A/AAAA
2560 String of space separated IP addresses present in the A/AAAA
2561 RRset. The IP addresses are in string format.
2562 IPSECKEY
2564 String of space separated IPSECKEY RDATA present in the
2565 IPSECKEY RRset. The IPSECKEY RDATA are in DNS presentation
2566 format.
2567 The A/AAAA answer is then cached and returned to the client. If the
2569 external hook was called the TTL changes to ensure it doesn't surpass
2570 ipsecmod-max-ttl.
2571 The same procedure is also followed when prefetch: is used, but the
2573 A/AAAA answer is given to the client before the hook is called. ipsec‐
2574 mod-max-ttl ensures that the A/AAAA answer given from cache is still
2575 relevant for opportunistic IPsec.
2576 ipsecmod-enabled: <yes or no>
2578 Specifies whether the IPsec module is enabled or not. The IPsec
2579 module still needs to be defined in the module-config: direc‐
2580 tive. This option facilitates turning on/off the module without
2581 restarting/reloading Unbound. Defaults to yes.
2582 ipsecmod-hook: <filename>
2584 Specifies the external hook that Unbound will call with sys‐
2585 tem(3). The file can be specified as an absolute/relative path.
2586 The file needs the proper permissions to be able to be executed
2587 by the same user that runs Unbound. It must be present when the
2588 IPsec module is defined in the module-config: directive.
2589 ipsecmod-strict: <yes or no>
2591 If enabled Unbound requires the external hook to return a suc‐
2592 cess value of 0. Failing to do so Unbound will reply with SERV‐
2593 FAIL. The A/AAAA answer will also not be cached. Defaults to
2594 no.
2595 ipsecmod-max-ttl: <seconds>
2597 Time to live maximum for A/AAAA cached records after calling the
2598 external hook. Defaults to 3600.
2599 ipsecmod-ignore-bogus: <yes or no>
2601 Specifies the behaviour of Unbound when the IPSECKEY answer is
2602 bogus. If set to yes, the hook will be called and the A/AAAA
2603 answer will be returned to the client. If set to no, the hook
2604 will not be called and the answer to the A/AAAA query will be
2605 SERVFAIL. Mainly used for testing. Defaults to no.
2606 ipsecmod-allow: <domain>
2608 Allow the ipsecmod functionality for the domain so that the mod‐
2609 ule logic will be executed. Can be given multiple times, for
2610 different domains. If the option is not specified, all domains
2611 are treated as being allowed (default).
2612 ipsecmod-whitelist: <yes or no>
2614 Alternate syntax for ipsecmod-allow.
2615 Cache DB Module Options
2617 The Cache DB module must be configured in the module-config: "validator
2618 cachedb iterator" directive and be compiled into the daemon with --en‐
2619 able-cachedb. If this module is enabled and configured, the specified
2620 backend database works as a second level cache: When Unbound cannot
2621 find an answer to a query in its built-in in-memory cache, it consults
2622 the specified backend. If it finds a valid answer in the backend, Un‐
2623 bound uses it to respond to the query without performing iterative DNS
2624 resolution. If Unbound cannot even find an answer in the backend, it
2625 resolves the query as usual, and stores the answer in the backend.
2626 This module interacts with the serve-expired-* options and will reply
2628 with expired data if Unbound is configured for that. Currently the use
2629 of serve-expired-client-timeout: and serve-expired-reply-ttl: is not
2630 consistent for data originating from the external cache as these will
2631 result in a reply with 0 TTL without trying to update the data first,
2632 ignoring the configured values.
2633 If Unbound was built with --with-libhiredis on a system that has in‐
2635 stalled the hiredis C client library of Redis, then the "redis" backend
2636 can be used. This backend communicates with the specified Redis server
2637 over a TCP connection to store and retrieve cache data. It can be used
2638 as a persistent and/or shared cache backend. It should be noted that
2639 Unbound never removes data stored in the Redis server, even if some
2640 data have expired in terms of DNS TTL or the Redis server has cached
2641 too much data; if necessary the Redis server must be configured to
2642 limit the cache size, preferably with some kind of least-recently-used
2643 eviction policy. Additionally, the redis-expire-records option can be
2644 used in order to set the relative DNS TTL of the message as timeout to
2645 the Redis records; keep in mind that some additional memory is used per
2646 key and that the expire information is stored as absolute Unix time‐
2647 stamps in Redis (computer time must be stable). This backend uses syn‐
2648 chronous communication with the Redis server based on the assumption
2649 that the communication is stable and sufficiently fast. The thread
2650 waiting for a response from the Redis server cannot handle other DNS
2651 queries. Although the backend has the ability to reconnect to the
2652 server when the connection is closed unexpectedly and there is a con‐
2653 figurable timeout in case the server is overly slow or hangs up, these
2654 cases are assumed to be very rare. If connection close or timeout hap‐
2655 pens too often, Unbound will be effectively unusable with this backend.
2656 It's the administrator's responsibility to make the assumption hold.
2657 The cachedb: clause gives custom settings of the cache DB module.
2659 backend: <backend name>
2661 Specify the backend database name. The default database is the
2662 in-memory backend named "testframe", which, as the name sug‐
2663 gests, is not of any practical use. Depending on the build-time
2664 configuration, "redis" backend may also be used as described
2665 above.
2666 secret-seed: <"secret string">
2668 Specify a seed to calculate a hash value from query information.
2669 This value will be used as the key of the corresponding answer
2670 for the backend database and can be customized if the hash
2671 should not be predictable operationally. If the backend data‐
2672 base is shared by multiple Unbound instances, all instances must
2673 use the same secret seed. This option defaults to "default".
2674 The following cachedb options are specific to the redis backend.
2676 redis-server-host: <server address or name>
2678 The IP (either v6 or v4) address or domain name of the Redis
2679 server. In general an IP address should be specified as other‐
2680 wise Unbound will have to resolve the name of the server every
2681 time it establishes a connection to the server. This option de‐
2682 faults to "127.0.0.1".
2683 redis-server-port: <port number>
2685 The TCP port number of the Redis server. This option defaults
2686 to 6379.
2687 redis-timeout: <msec>
2689 The period until when Unbound waits for a response from the Re‐
2690 dis sever. If this timeout expires Unbound closes the connec‐
2691 tion, treats it as if the Redis server does not have the re‐
2692 quested data, and will try to re-establish a new connection
2693 later. This option defaults to 100 milliseconds.
2694 redis-expire-records: <yes or no>
2696 If Redis record expiration is enabled. If yes, Unbound sets
2697 timeout for Redis records so that Redis can evict keys that have
2698 expired automatically. If Unbound is configured with serve-ex‐
2699 pired and serve-expired-ttl is 0, this option is internally re‐
2700 verted to "no". Redis SETEX support is required for this option
2701 (Redis >= 2.0.0). This option defaults to no.
2702 DNSTAP Logging Options
2704 DNSTAP support, when compiled in by using --enable-dnstap, is enabled
2705 in the dnstap: section. This starts an extra thread (when compiled
2706 with threading) that writes the log information to the destination. If
2707 Unbound is compiled without threading it does not spawn a thread, but
2708 connects per-process to the destination.
2709 dnstap-enable: <yes or no>
2711 If dnstap is enabled. Default no. If yes, it connects to the
2712 dnstap server and if any of the dnstap-log-..-messages options
2713 is enabled it sends logs for those messages to the server.
2714 dnstap-bidirectional: <yes or no>
2716 Use frame streams in bidirectional mode to transfer DNSTAP mes‐
2717 sages. Default is yes.
2718 dnstap-socket-path: <file name>
2720 Sets the unix socket file name for connecting to the server that
2721 is listening on that socket. Default is "/etc/un‐
2722 bound/dnstap.sock".
2723 dnstap-ip: <IPaddress[@port]>
2725 If "", the unix socket is used, if set with an IP address (IPv4
2726 or IPv6) that address is used to connect to the server.
2727 dnstap-tls: <yes or no>
2729 Set this to use TLS to connect to the server specified in
2730 dnstap-ip. The default is yes. If set to no, TCP is used to
2731 connect to the server.
2732 dnstap-tls-server-name: <name of TLS authentication>
2734 The TLS server name to authenticate the server with. Used when
2735 dnstap-tls is enabled. If "" it is ignored, default "".
2736 dnstap-tls-cert-bundle: <file name of cert bundle>
2738 The pem file with certs to verify the TLS server certificate. If
2739 "" the server default cert bundle is used, or the windows cert
2740 bundle on windows. Default is "".
2741 dnstap-tls-client-key-file: <file name>
2743 The client key file for TLS client authentication. If "" client
2744 authentication is not used. Default is "".
2745 dnstap-tls-client-cert-file: <file name>
2747 The client cert file for TLS client authentication. Default is
2748 "".
2749 dnstap-send-identity: <yes or no>
2751 If enabled, the server identity is included in the log messages.
2752 Default is no.
2753 dnstap-send-version: <yes or no>
2755 If enabled, the server version if included in the log messages.
2756 Default is no.
2757 dnstap-identity: <string>
2759 The identity to send with messages, if "" the hostname is used.
2760 Default is "".
2761 dnstap-version: <string>
2763 The version to send with messages, if "" the package version is
2764 used. Default is "".
2765 dnstap-log-resolver-query-messages: <yes or no>
2767 Enable to log resolver query messages. Default is no. These
2768 are messages from Unbound to upstream servers.
2769 dnstap-log-resolver-response-messages: <yes or no>
2771 Enable to log resolver response messages. Default is no. These
2772 are replies from upstream servers to Unbound.
2773 dnstap-log-client-query-messages: <yes or no>
2775 Enable to log client query messages. Default is no. These are
2776 client queries to Unbound.
2777 dnstap-log-client-response-messages: <yes or no>
2779 Enable to log client response messages. Default is no. These
2780 are responses from Unbound to clients.
2781 dnstap-log-forwarder-query-messages: <yes or no>
2783 Enable to log forwarder query messages. Default is no.
2784 dnstap-log-forwarder-response-messages: <yes or no>
2786 Enable to log forwarder response messages. Default is no.
2787 Response Policy Zone Options
2789 Response Policy Zones are configured with rpz:, and each one must have
2790 a name:. There can be multiple ones, by listing multiple rpz clauses,
2791 each with a different name. RPZ clauses are applied in order of config‐
2792 uration. The respip module needs to be added to the module-config,
2793 e.g.: module-config: "respip validator iterator".
2794 QNAME, Response IP Address, nsdname, nsip and clientip triggers are
2796 supported. Supported actions are: NXDOMAIN, NODATA, PASSTHRU, DROP,
2797 Local Data, tcp-only and drop. RPZ QNAME triggers are applied after
2798 local-zones and before auth-zones.
2799 The rpz zone is formatted with a SOA start record as usual. The items
2801 in the zone are entries, that specify what to act on (the trigger) and
2802 what to do (the action). The trigger to act on is recorded in the
2803 name, the action to do is recorded as the resource record. The names
2804 all end in the zone name, so you could type the trigger names without a
2805 trailing dot in the zonefile.
2806 An example RPZ record, that answers example.com with NXDOMAIN
2808 example.com CNAME .
2809 The triggers are encoded in the name on the left
2811 name query name
2812 netblock.rpz-client-ip client IP address
2813 netblock.rpz-ip response IP address in the answer
2814 name.rpz-nsdname nameserver name
2815 netblock.rpz-nsip nameserver IP address
2816 The netblock is written as <netblocklen>.<ip address in reverse>. For
2817 IPv6 use 'zz' for '::'. Specify individual addresses with scope length
2818 of 32 or 128. For example, 24.10.100.51.198.rpz-ip is 198.51.100.10/24
2819 and 32.10.zz.db8.2001.rpz-ip is 2001:db8:0:0:0:0:0:10/32.
2820 The actions are specified with the record on the right
2822 CNAME . nxdomain reply
2823 CNAME *. nodata reply
2824 CNAME rpz-passthru. do nothing, allow to continue
2825 CNAME rpz-drop. the query is dropped
2826 CNAME rpz-tcp-only. answer over TCP
2827 A 192.0.2.1 answer with this IP address
2828 Other records like AAAA, TXT and other CNAMEs (not rpz-..) can also be
2829 used to answer queries with that content.
2830 The RPZ zones can be configured in the config file with these settings
2832 in the rpz: block.
2833 name: <zone name>
2835 Name of the authority zone.
2836 primary: <IP address or host name>
2838 Where to download a copy of the zone from, with AXFR and IXFR.
2839 Multiple primaries can be specified. They are all tried if one
2840 fails. To use a nondefault port for DNS communication append
2841 '@' with the port number. You can append a '#' and a name, then
2842 AXFR over TLS can be used and the tls authentication certifi‐
2843 cates will be checked with that name. If you combine the '@'
2844 and '#', the '@' comes first. If you point it at another Un‐
2845 bound instance, it would not work because that does not support
2846 AXFR/IXFR for the zone, but if you used url: to download the
2847 zonefile as a text file from a webserver that would work. If
2848 you specify the hostname, you cannot use the domain from the
2849 zonefile, because it may not have that when retrieving that
2850 data, instead use a plain IP address to avoid a circular depen‐
2851 dency on retrieving that IP address.
2852 master: <IP address or host name>
2854 Alternate syntax for primary.
2855 url: <url to zonefile>
2857 Where to download a zonefile for the zone. With http or https.
2858 An example for the url is "http://www.example.com/exam‐
2859 ple.org.zone". Multiple url statements can be given, they are
2860 tried in turn. If only urls are given the SOA refresh timer is
2861 used to wait for making new downloads. If also primaries are
2862 listed, the primaries are first probed with UDP SOA queries to
2863 see if the SOA serial number has changed, reducing the number of
2864 downloads. If none of the urls work, the primaries are tried
2865 with IXFR and AXFR. For https, the tls-cert-bundle and the
2866 hostname from the url are used to authenticate the connection.
2867 allow-notify: <IP address or host name or netblockIP/prefix>
2869 With allow-notify you can specify additional sources of noti‐
2870 fies. When notified, the server attempts to first probe and
2871 then zone transfer. If the notify is from a primary, it first
2872 attempts that primary. Otherwise other primaries are attempted.
2873 If there are no primaries, but only urls, the file is downloaded
2874 when notified. The primaries from primary: and url: statements
2875 are allowed notify by default.
2876 zonefile: <filename>
2878 The filename where the zone is stored. If not given then no
2879 zonefile is used. If the file does not exist or is empty, Un‐
2880 bound will attempt to fetch zone data (eg. from the primary
2881 servers).
2882 rpz-action-override: <action>
2884 Always use this RPZ action for matching triggers from this zone.
2885 Possible action are: nxdomain, nodata, passthru, drop, disabled
2886 and cname.
2887 rpz-cname-override: <domain>
2889 The CNAME target domain to use if the cname action is configured
2890 for rpz-action-override.
2891 rpz-log: <yes or no>
2893 Log all applied RPZ actions for this RPZ zone. Default is no.
2894 rpz-log-name: <name>
2896 Specify a string to be part of the log line, for easy referenc‐
2897 ing.
2898 rpz-signal-nxdomain-ra: <yes or no>
2900 Signal when a query is blocked by the RPZ with NXDOMAIN with an
2901 unset RA flag. This allows certain clients, like dnsmasq, to
2902 infer that the domain is externally blocked. Default is no.
2903 for-downstream: <yes or no>
2905 If enabled the zone is authoritatively answered for and queries
2906 for the RPZ zone information are answered to downstream clients.
2907 This is useful for monitoring scripts, that can then access the
2908 SOA information to check if the rpz information is up to date.
2909 Default is no.
2910 tags: <list of tags>
2912 Limit the policies from this RPZ clause to clients with a match‐
2913 ing tag. Tags need to be defined in define-tag and can be as‐
2914 signed to client addresses using access-control-tag. Enclose
2915 list of tags in quotes ("") and put spaces between tags. If no
2916 tags are specified the policies from this clause will be applied
2917 for all clients.
2918
2920 In the example config settings below memory usage is reduced. Some ser‐
2921 vice levels are lower, notable very large data and a high TCP load are
2922 no longer supported. Very large data and high TCP loads are exceptional
2923 for the DNS. DNSSEC validation is enabled, just add trust anchors. If
2924 you do not have to worry about programs using more than 3 Mb of memory,
2925 the below example is not for you. Use the defaults to receive full ser‐
2926 vice, which on BSD-32bit tops out at 30-40 Mb after heavy usage.
2927 # example settings that reduce memory usage
2929 server:
2930 num-threads: 1
2931 outgoing-num-tcp: 1 # this limits TCP service, uses less buffers.
2932 incoming-num-tcp: 1
2933 outgoing-range: 60 # uses less memory, but less performance.
2934 msg-buffer-size: 8192 # note this limits service, 'no huge stuff'.
2935 msg-cache-size: 100k
2936 msg-cache-slabs: 1
2937 rrset-cache-size: 100k
2938 rrset-cache-slabs: 1
2939 infra-cache-numhosts: 200
2940 infra-cache-slabs: 1
2941 key-cache-size: 100k
2942 key-cache-slabs: 1
2943 neg-cache-size: 10k
2944 num-queries-per-thread: 30
2945 target-fetch-policy: "2 1 0 0 0 0"
2946 harden-large-queries: "yes"
2947 harden-short-bufsize: "yes"
2948
2950 /etc/unbound
2951 default Unbound working directory.
2952 /etc/unbound
2954 default chroot(2) location.
2955 /etc/unbound/unbound.conf
2957 Unbound configuration file.
2958 /run/unbound/unbound.pid
2960 default Unbound pidfile with process ID of the running daemon.
2961 unbound.log
2963 Unbound log file. default is to log to syslog(3).
2964
2966 unbound(8), unbound-checkconf(8).
2967
2969 Unbound was written by NLnet Labs. Please see CREDITS file in the dis‐
2970 tribution for further details.
2971NLnet Labs Jan 12, 2023 unbound.conf(5)