1unbound.conf(5) unbound 1.10.0 unbound.conf(5)
2
3
4
6 unbound.conf - Unbound configuration file.
7
9 unbound.conf
10
12 unbound.conf is used to configure unbound(8). The file format has
13 attributes and values. Some attributes have attributes inside them.
14 The notation is: attribute: value.
15
16 Comments start with # and last to the end of line. Empty lines are
17 ignored as is whitespace at the beginning of a line.
18
19 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
24 /etc/unbound/unbound.conf and start the server with:
25
26 $ unbound -c /etc/unbound/unbound.conf
27
28 Most settings are the defaults. Stop the server with:
29
30 $ kill `cat /etc/unbound/unbound.pid`
31
32 Below is a minimal config file. The source distribution contains an
33 extensive example.conf file with all the options.
34
35 # 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
60 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
64 included files works, relative pathnames for the included names work if
65 the directory where the daemon is started equals its chroot/working
66 directory or is specified before the include statement with directory:
67 dir. Wildcards can be used to include multiple files, see glob(7).
68
69 Server Options
70 These options are part of the server: clause.
71
72 verbosity: <number>
73 The verbosity number, level 0 means no verbosity, only errors.
74 Level 1 gives operational information. Level 2 gives detailed
75 operational information. Level 3 gives query level information,
76 output per query. Level 4 gives algorithm level information.
77 Level 5 logs client identification for cache misses. Default is
78 level 1. The verbosity can also be increased from the command‐
79 line, see unbound(8).
80
81 statistics-interval: <seconds>
82 The number of seconds between printing statistics to the log for
83 every thread. Disable with value 0 or "". Default is disabled.
84 The histogram statistics are only printed if replies were sent
85 during the statistics interval, requestlist statistics are
86 printed for every interval (but can be 0). This is because the
87 median calculation requires data to be present.
88
89 statistics-cumulative: <yes or no>
90 If enabled, statistics are cumulative since starting unbound,
91 without clearing the statistics counters after logging the sta‐
92 tistics. Default is no.
93
94 extended-statistics: <yes or no>
95 If enabled, extended statistics are printed from unbound-con‐
96 trol(8). Default is off, because keeping track of more statis‐
97 tics takes time. The counters are listed in unbound-control(8).
98
99 num-threads: <number>
100 The number of threads to create to serve clients. Use 1 for no
101 threading.
102
103 port: <port number>
104 The port number, default 53, on which the server responds to
105 queries.
106
107 interface: <ip address[@port]>
108 Interface to use to connect to the network. This interface is
109 listened to for queries from clients, and answers to clients are
110 given from it. Can be given multiple times to work on several
111 interfaces. If none are given the default is to listen to local‐
112 host. The interfaces are not changed on a reload (kill -HUP)
113 but only on restart. A port number can be specified with @port
114 (without spaces between interface and port number), if not spec‐
115 ified the default port (from port) is used.
116
117 ip-address: <ip address[@port]>
118 Same as interface: (for ease of compatibility with nsd.conf).
119
120 interface-automatic: <yes or no>
121 Detect source interface on UDP queries and copy them to replies.
122 This feature is experimental, and needs support in your OS for
123 particular socket options. Default value is no.
124
125 outgoing-interface: <ip address or ip6 netblock>
126 Interface to use to connect to the network. This interface is
127 used to send queries to authoritative servers and receive their
128 replies. Can be given multiple times to work on several inter‐
129 faces. If none are given the default (all) is used. You can
130 specify the same interfaces in interface: and outgoing-inter‐
131 face: lines, the interfaces are then used for both purposes.
132 Outgoing queries are sent via a random outgoing interface to
133 counter spoofing.
134
135 If an IPv6 netblock is specified instead of an individual IPv6
136 address, outgoing UDP queries will use a randomised source
137 address taken from the netblock to counter spoofing. Requires
138 the IPv6 netblock to be routed to the host running unbound, and
139 requires OS support for unprivileged non-local binds (currently
140 only supported on Linux). Several netblocks may be specified
141 with multiple outgoing-interface: options, but do not specify
142 both an individual IPv6 address and an IPv6 netblock, or the
143 randomisation will be compromised. Consider combining with pre‐
144 fer-ip6: yes to increase the likelihood of IPv6 nameservers
145 being selected for queries. On Linux you need these two com‐
146 mands to be able to use the freebind socket option to receive
147 traffic for the ip6 netblock: ip -6 addr add mynetblock/64 dev
148 lo && ip -6 route add local mynetblock/64 dev lo
149
150 outgoing-range: <number>
151 Number of ports to open. This number of file descriptors can be
152 opened per thread. Must be at least 1. Default depends on com‐
153 pile options. Larger numbers need extra resources from the oper‐
154 ating system. For performance a very large value is best, use
155 libevent to make this possible.
156
157 outgoing-port-permit: <port number or range>
158 Permit unbound to open this port or range of ports for use to
159 send queries. A larger number of permitted outgoing ports
160 increases resilience against spoofing attempts. Make sure these
161 ports are not needed by other daemons. By default only ports
162 above 1024 that have not been assigned by IANA are used. Give a
163 port number or a range of the form "low-high", without spaces.
164
165 The outgoing-port-permit and outgoing-port-avoid statements are
166 processed in the line order of the config file, adding the per‐
167 mitted ports and subtracting the avoided ports from the set of
168 allowed ports. The processing starts with the non IANA allo‐
169 cated ports above 1024 in the set of allowed ports.
170
171 outgoing-port-avoid: <port number or range>
172 Do not permit unbound to open this port or range of ports for
173 use to send queries. Use this to make sure unbound does not grab
174 a port that another daemon needs. The port is avoided on all
175 outgoing interfaces, both IP4 and IP6. By default only ports
176 above 1024 that have not been assigned by IANA are used. Give a
177 port number or a range of the form "low-high", without spaces.
178
179 outgoing-num-tcp: <number>
180 Number of outgoing TCP buffers to allocate per thread. Default
181 is 10. If set to 0, or if do-tcp is "no", no TCP queries to
182 authoritative servers are done. For larger installations
183 increasing this value is a good idea.
184
185 incoming-num-tcp: <number>
186 Number of incoming TCP buffers to allocate per thread. Default
187 is 10. If set to 0, or if do-tcp is "no", no TCP queries from
188 clients are accepted. For larger installations increasing this
189 value is a good idea.
190
191 edns-buffer-size: <number>
192 Number of bytes size to advertise as the EDNS reassembly buffer
193 size. This is the value put into datagrams over UDP towards
194 peers. The actual buffer size is determined by msg-buffer-size
195 (both for TCP and UDP). Do not set higher than that value.
196 Default is 4096 which is RFC recommended. If you have fragmen‐
197 tation reassembly problems, usually seen as timeouts, then a
198 value of 1472 can fix it. Setting to 512 bypasses even the most
199 stringent path MTU problems, but is seen as extreme, since the
200 amount of TCP fallback generated is excessive (probably also for
201 this resolver, consider tuning the outgoing tcp number).
202
203 max-udp-size: <number>
204 Maximum UDP response size (not applied to TCP response). 65536
205 disables the udp response size maximum, and uses the choice from
206 the client, always. Suggested values are 512 to 4096. Default
207 is 4096.
208
209 stream-wait-size: <number>
210 Number of bytes size maximum to use for waiting stream buffers.
211 Default is 4 megabytes. A plain number is in bytes, append 'k',
212 'm' or 'g' for kilobytes, megabytes or gigabytes (1024*1024
213 bytes in a megabyte). As TCP and TLS streams queue up multiple
214 results, the amount of memory used for these buffers does not
215 exceed this number, otherwise the responses are dropped. This
216 manages the total memory usage of the server (under heavy use),
217 the number of requests that can be queued up per connection is
218 also limited, with further requests waiting in TCP buffers.
219
220 msg-buffer-size: <number>
221 Number of bytes size of the message buffers. Default is 65552
222 bytes, enough for 64 Kb packets, the maximum DNS message size.
223 No message larger than this can be sent or received. Can be
224 reduced to use less memory, but some requests for DNS data, such
225 as for huge resource records, will result in a SERVFAIL reply to
226 the client.
227
228 msg-cache-size: <number>
229 Number of bytes size of the message cache. Default is 4
230 megabytes. A plain number is in bytes, append 'k', 'm' or 'g'
231 for kilobytes, megabytes or gigabytes (1024*1024 bytes in a
232 megabyte).
233
234 msg-cache-slabs: <number>
235 Number of slabs in the message cache. Slabs reduce lock con‐
236 tention by threads. Must be set to a power of 2. Setting
237 (close) to the number of cpus is a reasonable guess.
238
239 num-queries-per-thread: <number>
240 The number of queries that every thread will service simultane‐
241 ously. If more queries arrive that need servicing, and no
242 queries can be jostled out (see jostle-timeout), then the
243 queries are dropped. This forces the client to resend after a
244 timeout; allowing the server time to work on the existing
245 queries. Default depends on compile options, 512 or 1024.
246
247 jostle-timeout: <msec>
248 Timeout used when the server is very busy. Set to a value that
249 usually results in one roundtrip to the authority servers. If
250 too many queries arrive, then 50% of the queries are allowed to
251 run to completion, and the other 50% are replaced with the new
252 incoming query if they have already spent more than their
253 allowed time. This protects against denial of service by slow
254 queries or high query rates. Default 200 milliseconds. The
255 effect is that the qps for long-lasting queries is about (num‐
256 queriesperthread / 2) / (average time for such long queries)
257 qps. The qps for short queries can be about (numqueries‐
258 perthread / 2) / (jostletimeout in whole seconds) qps per
259 thread, about (1024/2)*5 = 2560 qps by default.
260
261 delay-close: <msec>
262 Extra delay for timeouted UDP ports before they are closed, in
263 msec. Default is 0, and that disables it. This prevents very
264 delayed answer packets from the upstream (recursive) servers
265 from bouncing against closed ports and setting off all sort of
266 close-port counters, with eg. 1500 msec. When timeouts happen
267 you need extra sockets, it checks the ID and remote IP of pack‐
268 ets, and unwanted packets are added to the unwanted packet
269 counter.
270
271 unknown-server-time-limit: <msec>
272 The wait time in msec for waiting for an unknown server to
273 reply. Increase this if you are behind a slow satellite link,
274 to eg. 1128. That would then avoid re-querying every initial
275 query because it times out. Default is 376 msec.
276
277 so-rcvbuf: <number>
278 If not 0, then set the SO_RCVBUF socket option to get more buf‐
279 fer space on UDP port 53 incoming queries. So that short spikes
280 on busy servers do not drop packets (see counter in netstat
281 -su). Default is 0 (use system value). Otherwise, the number
282 of bytes to ask for, try "4m" on a busy server. The OS caps it
283 at a maximum, on linux unbound needs root permission to bypass
284 the limit, or the admin can use sysctl net.core.rmem_max. On
285 BSD change kern.ipc.maxsockbuf in /etc/sysctl.conf. On OpenBSD
286 change header and recompile kernel. On Solaris ndd -set /dev/udp
287 udp_max_buf 8388608.
288
289 so-sndbuf: <number>
290 If not 0, then set the SO_SNDBUF socket option to get more buf‐
291 fer space on UDP port 53 outgoing queries. This for very busy
292 servers handles spikes in answer traffic, otherwise 'send:
293 resource temporarily unavailable' can get logged, the buffer
294 overrun is also visible by netstat -su. Default is 0 (use sys‐
295 tem value). Specify the number of bytes to ask for, try "4m" on
296 a very busy server. The OS caps it at a maximum, on linux
297 unbound needs root permission to bypass the limit, or the admin
298 can use sysctl net.core.wmem_max. On BSD, Solaris changes are
299 similar to so-rcvbuf.
300
301 so-reuseport: <yes or no>
302 If yes, then open dedicated listening sockets for incoming
303 queries for each thread and try to set the SO_REUSEPORT socket
304 option on each socket. May distribute incoming queries to
305 threads more evenly. Default is yes. On Linux it is supported
306 in kernels >= 3.9. On other systems, FreeBSD, OSX it may also
307 work. You can enable it (on any platform and kernel), it then
308 attempts to open the port and passes the option if it was avail‐
309 able at compile time, if that works it is used, if it fails, it
310 continues silently (unless verbosity 3) without the option. At
311 extreme load it could be better to turn it off to distribute the
312 queries evenly, reported for Linux systems (4.4.x).
313
314 ip-transparent: <yes or no>
315 If yes, then use IP_TRANSPARENT socket option on sockets where
316 unbound is listening for incoming traffic. Default no. Allows
317 you to bind to non-local interfaces. For example for non-exis‐
318 tent IP addresses that are going to exist later on, with host
319 failover configuration. This is a lot like interface-automatic,
320 but that one services all interfaces and with this option you
321 can select which (future) interfaces unbound provides service
322 on. This option needs unbound to be started with root permis‐
323 sions on some systems. The option uses IP_BINDANY on FreeBSD
324 systems and SO_BINDANY on OpenBSD systems.
325
326 ip-freebind: <yes or no>
327 If yes, then use IP_FREEBIND socket option on sockets where
328 unbound is listening to incoming traffic. Default no. Allows
329 you to bind to IP addresses that are nonlocal or do not exist,
330 like when the network interface or IP address is down. Exists
331 only on Linux, where the similar ip-transparent option is also
332 available.
333
334 rrset-cache-size: <number>
335 Number of bytes size of the RRset cache. Default is 4 megabytes.
336 A plain number is in bytes, append 'k', 'm' or 'g' for kilo‐
337 bytes, megabytes or gigabytes (1024*1024 bytes in a megabyte).
338
339 rrset-cache-slabs: <number>
340 Number of slabs in the RRset cache. Slabs reduce lock contention
341 by threads. Must be set to a power of 2.
342
343 cache-max-ttl: <seconds>
344 Time to live maximum for RRsets and messages in the cache.
345 Default is 86400 seconds (1 day). When the TTL expires, the
346 cache item has expired. Can be set lower to force the resolver
347 to query for data often, and not trust (very large) TTL values.
348 Downstream clients also see the lower TTL.
349
350 cache-min-ttl: <seconds>
351 Time to live minimum for RRsets and messages in the cache.
352 Default is 0. If the minimum kicks in, the data is cached for
353 longer than the domain owner intended, and thus less queries are
354 made to look up the data. Zero makes sure the data in the cache
355 is as the domain owner intended, higher values, especially more
356 than an hour or so, can lead to trouble as the data in the cache
357 does not match up with the actual data any more.
358
359 cache-max-negative-ttl: <seconds>
360 Time to live maximum for negative responses, these have a SOA in
361 the authority section that is limited in time. Default is 3600.
362 This applies to nxdomain and nodata answers.
363
364 infra-host-ttl: <seconds>
365 Time to live for entries in the host cache. The host cache con‐
366 tains roundtrip timing, lameness and EDNS support information.
367 Default is 900.
368
369 infra-cache-slabs: <number>
370 Number of slabs in the infrastructure cache. Slabs reduce lock
371 contention by threads. Must be set to a power of 2.
372
373 infra-cache-numhosts: <number>
374 Number of hosts for which information is cached. Default is
375 10000.
376
377 infra-cache-min-rtt: <msec>
378 Lower limit for dynamic retransmit timeout calculation in infra‐
379 structure cache. Default is 50 milliseconds. Increase this value
380 if using forwarders needing more time to do recursive name reso‐
381 lution.
382
383 define-tag: <"list of tags">
384 Define the tags that can be used with local-zone and access-con‐
385 trol. Enclose the list between quotes ("") and put spaces
386 between tags.
387
388 do-ip4: <yes or no>
389 Enable or disable whether ip4 queries are answered or issued.
390 Default is yes.
391
392 do-ip6: <yes or no>
393 Enable or disable whether ip6 queries are answered or issued.
394 Default is yes. If disabled, queries are not answered on IPv6,
395 and queries are not sent on IPv6 to the internet nameservers.
396 With this option you can disable the ipv6 transport for sending
397 DNS traffic, it does not impact the contents of the DNS traffic,
398 which may have ip4 and ip6 addresses in it.
399
400 prefer-ip6: <yes or no>
401 If enabled, prefer IPv6 transport for sending DNS queries to
402 internet nameservers. Default is no.
403
404 do-udp: <yes or no>
405 Enable or disable whether UDP queries are answered or issued.
406 Default is yes.
407
408 do-tcp: <yes or no>
409 Enable or disable whether TCP queries are answered or issued.
410 Default is yes.
411
412 tcp-mss: <number>
413 Maximum segment size (MSS) of TCP socket on which the server
414 responds to queries. Value lower than common MSS on Ethernet
415 (1220 for example) will address path MTU problem. Note that not
416 all platform supports socket option to set MSS (TCP_MAXSEG).
417 Default is system default MSS determined by interface MTU and
418 negotiation between server and client.
419
420 outgoing-tcp-mss: <number>
421 Maximum segment size (MSS) of TCP socket for outgoing queries
422 (from Unbound to other servers). Value lower than common MSS on
423 Ethernet (1220 for example) will address path MTU problem. Note
424 that not all platform supports socket option to set MSS
425 (TCP_MAXSEG). Default is system default MSS determined by
426 interface MTU and negotiation between Unbound and other servers.
427
428 tcp-idle-timeout: <msec>
429 The period Unbound will wait for a query on a TCP connection.
430 If this timeout expires Unbound closes the connection. This
431 option defaults to 30000 milliseconds. When the number of free
432 incoming TCP buffers falls below 50% of the total number config‐
433 ured, the option value used is progressively reduced, first to
434 1% of the configured value, then to 0.2% of the configured value
435 if the number of free buffers falls below 35% of the total num‐
436 ber configured, and finally to 0 if the number of free buffers
437 falls below 20% of the total number configured. A minimum time‐
438 out of 200 milliseconds is observed regardless of the option
439 value used.
440
441 edns-tcp-keepalive: <yes or no>
442 Enable or disable EDNS TCP Keepalive. Default is no.
443
444 edns-tcp-keepalive-timeout: <msec>
445 The period Unbound will wait for a query on a TCP connection
446 when EDNS TCP Keepalive is active. If this timeout expires
447 Unbound closes the connection. If the client supports the EDNS
448 TCP Keepalive option, Unbound sends the timeout value to the
449 client to encourage it to close the connection before the server
450 times out. This option defaults to 120000 milliseconds. When
451 the number of free incoming TCP buffers falls below 50% of the
452 total number configured, the advertised timeout is progressively
453 reduced to 1% of the configured value, then to 0.2% of the con‐
454 figured value if the number of free buffers falls below 35% of
455 the total number configured, and finally to 0 if the number of
456 free buffers falls below 20% of the total number configured. A
457 minimum actual timeout of 200 milliseconds is observed regard‐
458 less of the advertised timeout.
459
460 tcp-upstream: <yes or no>
461 Enable or disable whether the upstream queries use TCP only for
462 transport. Default is no. Useful in tunneling scenarios.
463
464 udp-upstream-without-downstream: <yes or no>
465 Enable udp upstream even if do-udp is no. Default is no, and
466 this does not change anything. Useful for TLS service
467 providers, that want no udp downstream but use udp to fetch data
468 upstream.
469
470 tls-upstream: <yes or no>
471 Enabled or disable whether the upstream queries use TLS only for
472 transport. Default is no. Useful in tunneling scenarios. The
473 TLS contains plain DNS in TCP wireformat. The other server must
474 support this (see tls-service-key). If you enable this, also
475 configure a tls-cert-bundle or use tls-win-cert to load CA
476 certs, otherwise the connections cannot be authenticated. This
477 option enables TLS for all of them, but if you do not set this
478 you can configure TLS specifically for some forward zones with
479 forward-tls-upstream. And also with stub-tls-upstream.
480
481 ssl-upstream: <yes or no>
482 Alternate syntax for tls-upstream. If both are present in the
483 config file the last is used.
484
485 tls-service-key: <file>
486 If enabled, the server provides TLS service on the TCP ports
487 marked implicitly or explicitly for TLS service with tls-port.
488 The file must contain the private key for the TLS session, the
489 public certificate is in the tls-service-pem file and it must
490 also be specified if tls-service-key is specified. The default
491 is "", turned off. Enabling or disabling this service requires
492 a restart (a reload is not enough), because the key is read
493 while root permissions are held and before chroot (if any). The
494 ports enabled implicitly or explicitly via tls-port: do not pro‐
495 vide normal DNS TCP service.
496
497 ssl-service-key: <file>
498 Alternate syntax for tls-service-key.
499
500 tls-service-pem: <file>
501 The public key certificate pem file for the tls service.
502 Default is "", turned off.
503
504 ssl-service-pem: <file>
505 Alternate syntax for tls-service-pem.
506
507 tls-port: <number>
508 The port number on which to provide TCP TLS service, default
509 853, only interfaces configured with that port number as @number
510 get the TLS service.
511
512 ssl-port: <number>
513 Alternate syntax for tls-port.
514
515 tls-cert-bundle: <file>
516 If null or "", no file is used. Set it to the certificate bun‐
517 dle file, for example "/etc/pki/tls/certs/ca-bundle.crt". These
518 certificates are used for authenticating connections made to
519 outside peers. For example auth-zone urls, and also DNS over
520 TLS connections.
521
522 ssl-cert-bundle: <file>
523 Alternate syntax for tls-cert-bundle.
524
525 tls-win-cert: <yes or no>
526 Add the system certificates to the cert bundle certificates for
527 authentication. If no cert bundle, it uses only these certifi‐
528 cates. Default is no. On windows this option uses the certifi‐
529 cates from the cert store. Use the tls-cert-bundle option on
530 other systems.
531
532 tls-additional-port: <portnr>
533 List portnumbers as tls-additional-port, and when interfaces are
534 defined, eg. with the @port suffix, as this port number, they
535 provide dns over TLS service. Can list multiple, each on a new
536 statement.
537
538 tls-session-ticket-keys: <file>
539 If not "", lists files with 80 bytes of random contents that are
540 used to perform TLS session resumption for clients using the
541 unbound server. These files contain the secret key for the TLS
542 session tickets. First key use to encrypt and decrypt TLS ses‐
543 sion tickets. Other keys use to decrypt only. With this you
544 can roll over to new keys, by generating a new first file and
545 allowing decrypt of the old file by listing it after the first
546 file for some time, after the wait clients are not using the old
547 key any more and the old key can be removed. One way to create
548 the file is dd if=/dev/random bs=1 count=80 of=ticket.dat The
549 first 16 bytes should be different from the old one if you cre‐
550 ate a second key, that is the name used to identify the key.
551 Then there is 32 bytes random data for an AES key and then 32
552 bytes random data for the HMAC key.
553
554 tls-ciphers: <string with cipher list>
555 Set the list of ciphers to allow when serving TLS. Use "" for
556 defaults, and that is the default.
557
558 tls-ciphersuites: <string with ciphersuites list>
559 Set the list of ciphersuites to allow when serving TLS. This is
560 for newer TLS 1.3 connections. Use "" for defaults, and that is
561 the default.
562
563 use-systemd: <yes or no>
564 Enable or disable systemd socket activation. Default is no.
565
566 do-daemonize: <yes or no>
567 Enable or disable whether the unbound server forks into the
568 background as a daemon. Set the value to no when unbound runs
569 as systemd service. Default is yes.
570
571 tcp-connection-limit: <IP netblock> <limit>
572 Allow up to limit simultaneous TCP connections from the given
573 netblock. When at the limit, further connections are accepted
574 but closed immediately. This option is experimental at this
575 time.
576
577 access-control: <IP netblock> <action>
578 The netblock is given as an IP4 or IP6 address with /size
579 appended for a classless network block. The action can be deny,
580 refuse, allow, allow_setrd, allow_snoop, deny_non_local or
581 refuse_non_local. The most specific netblock match is used, if
582 none match deny is used. The order of the access-control state‐
583 ments therefore does not matter.
584
585 The action deny stops queries from hosts from that netblock.
586
587 The action refuse stops queries too, but sends a DNS rcode
588 REFUSED error message back.
589
590 The action allow gives access to clients from that netblock. It
591 gives only access for recursion clients (which is what almost
592 all clients need). Nonrecursive queries are refused.
593
594 The allow action does allow nonrecursive queries to access the
595 local-data that is configured. The reason is that this does not
596 involve the unbound server recursive lookup algorithm, and
597 static data is served in the reply. This supports normal opera‐
598 tions where nonrecursive queries are made for the authoritative
599 data. For nonrecursive queries any replies from the dynamic
600 cache are refused.
601
602 The allow_setrd action ignores the recursion desired (RD) bit
603 and treats all requests as if the recursion desired bit is set.
604 Note that this behavior violates RFC 1034 which states that a
605 name server should never perform recursive service unless asked
606 via the RD bit since this interferes with trouble shooting of
607 name servers and their databases. This prohibited behavior may
608 be useful if another DNS server must forward requests for spe‐
609 cific zones to a resolver DNS server, but only supports stub
610 domains and sends queries to the resolver DNS server with the RD
611 bit cleared.
612
613 The action allow_snoop gives nonrecursive access too. This give
614 both recursive and non recursive access. The name allow_snoop
615 refers to cache snooping, a technique to use nonrecursive
616 queries to examine the cache contents (for malicious acts).
617 However, nonrecursive queries can also be a valuable debugging
618 tool (when you want to examine the cache contents). In that case
619 use allow_snoop for your administration host.
620
621 By default only localhost is allowed, the rest is refused. The
622 default is refused, because that is protocol-friendly. The DNS
623 protocol is not designed to handle dropped packets due to pol‐
624 icy, and dropping may result in (possibly excessive) retried
625 queries.
626
627 The deny_non_local and refuse_non_local settings are for hosts
628 that are only allowed to query for the authoritative local-data,
629 they are not allowed full recursion but only the static data.
630 With deny_non_local, messages that are disallowed are dropped,
631 with refuse_non_local they receive error code REFUSED.
632
633 access-control-tag: <IP netblock> <"list of tags">
634 Assign tags to access-control elements. Clients using this
635 access control element use localzones that are tagged with one
636 of these tags. Tags must be defined in define-tags. Enclose
637 list of tags in quotes ("") and put spaces between tags. If
638 access-control-tag is configured for a netblock that does not
639 have an access-control, an access-control element with action
640 allow is configured for this netblock.
641
642 access-control-tag-action: <IP netblock> <tag> <action>
643 Set action for particular tag for given access control element.
644 If you have multiple tag values, the tag used to lookup the
645 action is the first tag match between access-control-tag and
646 local-zone-tag where "first" comes from the order of the define-
647 tag values.
648
649 access-control-tag-data: <IP netblock> <tag> <"resource record string">
650 Set redirect data for particular tag for given access control
651 element.
652
653 access-control-view: <IP netblock> <view name>
654 Set view for given access control element.
655
656 chroot: <directory>
657 If chroot is enabled, you should pass the configfile (from the
658 commandline) as a full path from the original root. After the
659 chroot has been performed the now defunct portion of the config
660 file path is removed to be able to reread the config after a
661 reload.
662
663 All other file paths (working dir, logfile, roothints, and key
664 files) can be specified in several ways: as an absolute path
665 relative to the new root, as a relative path to the working
666 directory, or as an absolute path relative to the original root.
667 In the last case the path is adjusted to remove the unused por‐
668 tion.
669
670 The pidfile can be either a relative path to the working direc‐
671 tory, or an absolute path relative to the original root. It is
672 written just prior to chroot and dropping permissions. This
673 allows the pidfile to be /var/run/unbound.pid and the chroot to
674 be /var/unbound, for example. Note that Unbound is not able to
675 remove the pidfile after termination when it is located outside
676 of the chroot directory.
677
678 Additionally, unbound may need to access /dev/urandom (for
679 entropy) from inside the chroot.
680
681 If given a chroot is done to the given directory. By default
682 chroot is enabled and the default is "/etc/unbound". If you give
683 "" no chroot is performed.
684
685 username: <name>
686 If given, after binding the port the user privileges are
687 dropped. Default is "unbound". If you give username: "" no user
688 change is performed.
689
690 If this user is not capable of binding the port, reloads (by
691 signal HUP) will still retain the opened ports. If you change
692 the port number in the config file, and that new port number
693 requires privileges, then a reload will fail; a restart is
694 needed.
695
696 directory: <directory>
697 Sets the working directory for the program. Default is
698 "/etc/unbound". On Windows the string "%EXECUTABLE%" tries to
699 change to the directory that unbound.exe resides in. If you
700 give a server: directory: dir before include: file statements
701 then those includes can be relative to the working directory.
702
703 logfile: <filename>
704 If "" is given, logging goes to stderr, or nowhere once daemo‐
705 nized. The logfile is appended to, in the following format:
706 [seconds since 1970] unbound[pid:tid]: type: message.
707 If this option is given, the use-syslog is option is set to
708 "no". The logfile is reopened (for append) when the config file
709 is reread, on SIGHUP.
710
711 use-syslog: <yes or no>
712 Sets unbound to send log messages to the syslogd, using sys‐
713 log(3). The log facility LOG_DAEMON is used, with identity
714 "unbound". The logfile setting is overridden when use-syslog is
715 turned on. The default is to log to syslog.
716
717 log-identity: <string>
718 If "" is given (default), then the name of the executable, usu‐
719 ally "unbound" is used to report to the log. Enter a string to
720 override it with that, which is useful on systems that run more
721 than one instance of unbound, with different configurations, so
722 that the logs can be easily distinguished against.
723
724 log-time-ascii: <yes or no>
725 Sets logfile lines to use a timestamp in UTC ascii. Default is
726 no, which prints the seconds since 1970 in brackets. No effect
727 if using syslog, in that case syslog formats the timestamp
728 printed into the log files.
729
730 log-queries: <yes or no>
731 Prints one line per query to the log, with the log timestamp and
732 IP address, name, type and class. Default is no. Note that it
733 takes time to print these lines which makes the server (signifi‐
734 cantly) slower. Odd (nonprintable) characters in names are
735 printed as '?'.
736
737 log-replies: <yes or no>
738 Prints one line per reply to the log, with the log timestamp and
739 IP address, name, type, class, return code, time to resolve,
740 from cache and response size. Default is no. Note that it
741 takes time to print these lines which makes the server (signifi‐
742 cantly) slower. Odd (nonprintable) characters in names are
743 printed as '?'.
744
745 log-tag-queryreply: <yes or no>
746 Prints the word 'query' and 'reply' with log-queries and
747 log-replies. This makes filtering logs easier. The default is
748 off (for backwards compatibility).
749
750 log-local-actions: <yes or no>
751 Print log lines to inform about local zone actions. These lines
752 are like the local-zone type inform prints out, but they are
753 also printed for the other types of local zones.
754
755 log-servfail: <yes or no>
756 Print log lines that say why queries return SERVFAIL to clients.
757 This is separate from the verbosity debug logs, much smaller,
758 and printed at the error level, not the info level of debug info
759 from verbosity.
760
761 pidfile: <filename>
762 The process id is written to the file. Default is
763 "/run/unbound/unbound.pid". So,
764 kill -HUP `cat /run/unbound/unbound.pid`
765 triggers a reload,
766 kill -TERM `cat /run/unbound/unbound.pid`
767 gracefully terminates.
768
769 root-hints: <filename>
770 Read the root hints from this file. Default is nothing, using
771 builtin hints for the IN class. The file has the format of zone
772 files, with root nameserver names and addresses only. The
773 default may become outdated, when servers change, therefore it
774 is good practice to use a root-hints file.
775
776 hide-identity: <yes or no>
777 If enabled id.server and hostname.bind queries are refused.
778
779 identity: <string>
780 Set the identity to report. If set to "", the default, then the
781 hostname of the server is returned.
782
783 hide-version: <yes or no>
784 If enabled version.server and version.bind queries are refused.
785
786 version: <string>
787 Set the version to report. If set to "", the default, then the
788 package version is returned.
789
790 hide-trustanchor: <yes or no>
791 If enabled trustanchor.unbound queries are refused.
792
793 target-fetch-policy: <"list of numbers">
794 Set the target fetch policy used by unbound to determine if it
795 should fetch nameserver target addresses opportunistically. The
796 policy is described per dependency depth.
797
798 The number of values determines the maximum dependency depth
799 that unbound will pursue in answering a query. A value of -1
800 means to fetch all targets opportunistically for that dependency
801 depth. A value of 0 means to fetch on demand only. A positive
802 value fetches that many targets opportunistically.
803
804 Enclose the list between quotes ("") and put spaces between num‐
805 bers. The default is "3 2 1 0 0". Setting all zeroes, "0 0 0 0
806 0" gives behaviour closer to that of BIND 9, while setting "-1
807 -1 -1 -1 -1" gives behaviour rumoured to be closer to that of
808 BIND 8.
809
810 harden-short-bufsize: <yes or no>
811 Very small EDNS buffer sizes from queries are ignored. Default
812 is off, since it is legal protocol wise to send these, and
813 unbound tries to give very small answers to these queries, where
814 possible.
815
816 harden-large-queries: <yes or no>
817 Very large queries are ignored. Default is off, since it is
818 legal protocol wise to send these, and could be necessary for
819 operation if TSIG or EDNS payload is very large.
820
821 harden-glue: <yes or no>
822 Will trust glue only if it is within the servers authority.
823 Default is yes.
824
825 harden-dnssec-stripped: <yes or no>
826 Require DNSSEC data for trust-anchored zones, if such data is
827 absent, the zone becomes bogus. If turned off, and no DNSSEC
828 data is received (or the DNSKEY data fails to validate), then
829 the zone is made insecure, this behaves like there is no trust
830 anchor. You could turn this off if you are sometimes behind an
831 intrusive firewall (of some sort) that removes DNSSEC data from
832 packets, or a zone changes from signed to unsigned to badly
833 signed often. If turned off you run the risk of a downgrade
834 attack that disables security for a zone. Default is yes.
835
836 harden-below-nxdomain: <yes or no>
837 From RFC 8020 (with title "NXDOMAIN: There Really Is Nothing
838 Underneath"), returns nxdomain to queries for a name below
839 another name that is already known to be nxdomain. DNSSEC man‐
840 dates noerror for empty nonterminals, hence this is possible.
841 Very old software might return nxdomain for empty nonterminals
842 (that usually happen for reverse IP address lookups), and thus
843 may be incompatible with this. To try to avoid this only
844 DNSSEC-secure nxdomains are used, because the old software does
845 not have DNSSEC. Default is yes. The nxdomain must be secure,
846 this means nsec3 with optout is insufficient.
847
848 harden-referral-path: <yes or no>
849 Harden the referral path by performing additional queries for
850 infrastructure data. Validates the replies if trust anchors are
851 configured and the zones are signed. This enforces DNSSEC vali‐
852 dation on nameserver NS sets and the nameserver addresses that
853 are encountered on the referral path to the answer. Default no,
854 because it burdens the authority servers, and it is not RFC
855 standard, and could lead to performance problems because of the
856 extra query load that is generated. Experimental option. If
857 you enable it consider adding more numbers after the tar‐
858 get-fetch-policy to increase the max depth that is checked to.
859
860 harden-algo-downgrade: <yes or no>
861 Harden against algorithm downgrade when multiple algorithms are
862 advertised in the DS record. If no, allows the weakest algo‐
863 rithm to validate the zone. Default is no. Zone signers must
864 produce zones that allow this feature to work, but sometimes
865 they do not, and turning this option off avoids that validation
866 failure.
867
868 use-caps-for-id: <yes or no>
869 Use 0x20-encoded random bits in the query to foil spoof
870 attempts. This perturbs the lowercase and uppercase of query
871 names sent to authority servers and checks if the reply still
872 has the correct casing. Disabled by default. This feature is
873 an experimental implementation of draft dns-0x20.
874
875 caps-whitelist: <domain>
876 Whitelist the domain so that it does not receive caps-for-id
877 perturbed queries. For domains that do not support 0x20 and
878 also fail with fallback because they keep sending different
879 answers, like some load balancers. Can be given multiple times,
880 for different domains.
881
882 qname-minimisation: <yes or no>
883 Send minimum amount of information to upstream servers to
884 enhance privacy. Only send minimum required labels of the QNAME
885 and set QTYPE to A when possible. Best effort approach; full
886 QNAME and original QTYPE will be sent when upstream replies with
887 a RCODE other than NOERROR, except when receiving NXDOMAIN from
888 a DNSSEC signed zone. Default is yes.
889
890 qname-minimisation-strict: <yes or no>
891 QNAME minimisation in strict mode. Do not fall-back to sending
892 full QNAME to potentially broken nameservers. A lot of domains
893 will not be resolvable when this option in enabled. Only use if
894 you know what you are doing. This option only has effect when
895 qname-minimisation is enabled. Default is off.
896
897 aggressive-nsec: <yes or no>
898 Aggressive NSEC uses the DNSSEC NSEC chain to synthesize NXDO‐
899 MAIN and other denials, using information from previous NXDO‐
900 MAINs answers. Default is no. It helps to reduce the query
901 rate towards targets that get a very high nonexistent name
902 lookup rate.
903
904 private-address: <IP address or subnet>
905 Give IPv4 of IPv6 addresses or classless subnets. These are
906 addresses on your private network, and are not allowed to be
907 returned for public internet names. Any occurrence of such
908 addresses are removed from DNS answers. Additionally, the DNSSEC
909 validator may mark the answers bogus. This protects against
910 so-called DNS Rebinding, where a user browser is turned into a
911 network proxy, allowing remote access through the browser to
912 other parts of your private network. Some names can be allowed
913 to contain your private addresses, by default all the local-data
914 that you configured is allowed to, and you can specify addi‐
915 tional names using private-domain. No private addresses are
916 enabled by default. We consider to enable this for the RFC1918
917 private IP address space by default in later releases. That
918 would enable private addresses for 10.0.0.0/8 172.16.0.0/12
919 192.168.0.0/16 169.254.0.0/16 fd00::/8 and fe80::/10, since the
920 RFC standards say these addresses should not be visible on the
921 public internet. Turning on 127.0.0.0/8 would hinder many spam‐
922 blocklists as they use that. Adding ::ffff:0:0/96 stops
923 IPv4-mapped IPv6 addresses from bypassing the filter.
924
925 private-domain: <domain name>
926 Allow this domain, and all its subdomains to contain private
927 addresses. Give multiple times to allow multiple domain names
928 to contain private addresses. Default is none.
929
930 unwanted-reply-threshold: <number>
931 If set, a total number of unwanted replies is kept track of in
932 every thread. When it reaches the threshold, a defensive action
933 is taken and a warning is printed to the log. The defensive
934 action is to clear the rrset and message caches, hopefully
935 flushing away any poison. A value of 10 million is suggested.
936 Default is 0 (turned off).
937
938 do-not-query-address: <IP address>
939 Do not query the given IP address. Can be IP4 or IP6. Append
940 /num to indicate a classless delegation netblock, for example
941 like 10.2.3.4/24 or 2001::11/64.
942
943 do-not-query-localhost: <yes or no>
944 If yes, localhost is added to the do-not-query-address entries,
945 both IP6 ::1 and IP4 127.0.0.1/8. If no, then localhost can be
946 used to send queries to. Default is yes.
947
948 prefetch: <yes or no>
949 If yes, message cache elements are prefetched before they expire
950 to keep the cache up to date. Default is no. Turning it on
951 gives about 10 percent more traffic and load on the machine, but
952 popular items do not expire from the cache.
953
954 prefetch-key: <yes or no>
955 If yes, fetch the DNSKEYs earlier in the validation process,
956 when a DS record is encountered. This lowers the latency of
957 requests. It does use a little more CPU. Also if the cache is
958 set to 0, it is no use. Default is no.
959
960 deny-any: <yes or no>
961 If yes, deny queries of type ANY with an empty response.
962 Default is no. If disabled, unbound responds with a short list
963 of resource records if some can be found in the cache and makes
964 the upstream type ANY query if there are none.
965
966 rrset-roundrobin: <yes or no>
967 If yes, Unbound rotates RRSet order in response (the random num‐
968 ber is taken from the query ID, for speed and thread safety).
969 Default is no.
970
971 minimal-responses: <yes or no>
972 If yes, Unbound doesn't insert authority/additional sections
973 into response messages when those sections are not required.
974 This reduces response size significantly, and may avoid TCP
975 fallback for some responses. This may cause a slight speedup.
976 The default is yes, even though the DNS protocol RFCs mandate
977 these sections, and the additional content could be of use and
978 save roundtrips for clients. Because they are not used, and the
979 saved roundtrips are easier saved with prefetch, whilst this is
980 faster.
981
982 disable-dnssec-lame-check: <yes or no>
983 If true, disables the DNSSEC lameness check in the iterator.
984 This check sees if RRSIGs are present in the answer, when dnssec
985 is expected, and retries another authority if RRSIGs are unex‐
986 pectedly missing. The validator will insist in RRSIGs for
987 DNSSEC signed domains regardless of this setting, if a trust
988 anchor is loaded.
989
990 module-config: <"module names">
991 Module configuration, a list of module names separated by spa‐
992 ces, surround the string with quotes (""). The modules can be
993 validator, iterator. Setting this to "iterator" will result in
994 a non-validating server. Setting this to "validator iterator"
995 will turn on DNSSEC validation. The ordering of the modules is
996 important. You must also set trust-anchors for validation to be
997 useful. The default is "validator iterator". When the server
998 is built with EDNS client subnet support the default is "subnet‐
999 cache validator iterator". Most modules that need to be listed
1000 here have to be listed at the beginning of the line. The
1001 cachedb module has to be listed just before the iterator. The
1002 python module can be listed in different places, it then pro‐
1003 cesses the output of the module it is just before.
1004
1005 trust-anchor-file: <filename>
1006 File with trusted keys for validation. Both DS and DNSKEY
1007 entries can appear in the file. The format of the file is the
1008 standard DNS Zone file format. Default is "", or no trust
1009 anchor file.
1010
1011 auto-trust-anchor-file: <filename>
1012 File with trust anchor for one zone, which is tracked with
1013 RFC5011 probes. The probes are run several times per month,
1014 thus the machine must be online frequently. The initial file
1015 can be one with contents as described in trust-anchor-file. The
1016 file is written to when the anchor is updated, so the unbound
1017 user must have write permission. Write permission to the file,
1018 but also to the directory it is in (to create a temporary file,
1019 which is necessary to deal with filesystem full events), it must
1020 also be inside the chroot (if that is used).
1021
1022 trust-anchor: <"Resource Record">
1023 A DS or DNSKEY RR for a key to use for validation. Multiple
1024 entries can be given to specify multiple trusted keys, in addi‐
1025 tion to the trust-anchor-files. The resource record is entered
1026 in the same format as 'dig' or 'drill' prints them, the same
1027 format as in the zone file. Has to be on a single line, with ""
1028 around it. A TTL can be specified for ease of cut and paste, but
1029 is ignored. A class can be specified, but class IN is default.
1030
1031 trusted-keys-file: <filename>
1032 File with trusted keys for validation. Specify more than one
1033 file with several entries, one file per entry. Like
1034 trust-anchor-file but has a different file format. Format is
1035 BIND-9 style format, the trusted-keys { name flag proto algo
1036 "key"; }; clauses are read. It is possible to use wildcards
1037 with this statement, the wildcard is expanded on start and on
1038 reload.
1039
1040 trust-anchor-signaling: <yes or no>
1041 Send RFC8145 key tag query after trust anchor priming. Default
1042 is yes.
1043
1044 root-key-sentinel: <yes or no>
1045 Root key trust anchor sentinel. Default is yes.
1046
1047 dlv-anchor-file: <filename>
1048 This option was used during early days DNSSEC deployment when no
1049 parent-side DS record registrations were easily available.
1050 Nowadays, it is best to have DS records registered with the par‐
1051 ent zone (many top level zones are signed). File with trusted
1052 keys for DLV (DNSSEC Lookaside Validation). Both DS and DNSKEY
1053 entries can be used in the file, in the same format as for
1054 trust-anchor-file: statements. Only one DLV can be configured,
1055 more would be slow. The DLV configured is used as a root trusted
1056 DLV, this means that it is a lookaside for the root. Default is
1057 "", or no dlv anchor file. DLV is going to be decommissioned.
1058 Please do not use it any more.
1059
1060 dlv-anchor: <"Resource Record">
1061 Much like trust-anchor, this is a DLV anchor with the DS or
1062 DNSKEY inline. DLV is going to be decommissioned. Please do
1063 not use it any more.
1064
1065 domain-insecure: <domain name>
1066 Sets domain name to be insecure, DNSSEC chain of trust is
1067 ignored towards the domain name. So a trust anchor above the
1068 domain name can not make the domain secure with a DS record,
1069 such a DS record is then ignored. Also keys from DLV are
1070 ignored for the domain. Can be given multiple times to specify
1071 multiple domains that are treated as if unsigned. If you set
1072 trust anchors for the domain they override this setting (and the
1073 domain is secured).
1074
1075 This can be useful if you want to make sure a trust anchor for
1076 external lookups does not affect an (unsigned) internal domain.
1077 A DS record externally can create validation failures for that
1078 internal domain.
1079
1080 val-override-date: <rrsig-style date spec>
1081 Default is "" or "0", which disables this debugging feature. If
1082 enabled by giving a RRSIG style date, that date is used for ver‐
1083 ifying RRSIG inception and expiration dates, instead of the cur‐
1084 rent date. Do not set this unless you are debugging signature
1085 inception and expiration. The value -1 ignores the date alto‐
1086 gether, useful for some special applications.
1087
1088 val-sig-skew-min: <seconds>
1089 Minimum number of seconds of clock skew to apply to validated
1090 signatures. A value of 10% of the signature lifetime (expira‐
1091 tion - inception) is used, capped by this setting. Default is
1092 3600 (1 hour) which allows for daylight savings differences.
1093 Lower this value for more strict checking of short lived signa‐
1094 tures.
1095
1096 val-sig-skew-max: <seconds>
1097 Maximum number of seconds of clock skew to apply to validated
1098 signatures. A value of 10% of the signature lifetime (expira‐
1099 tion - inception) is used, capped by this setting. Default is
1100 86400 (24 hours) which allows for timezone setting problems in
1101 stable domains. Setting both min and max very low disables the
1102 clock skew allowances. Setting both min and max very high makes
1103 the validator check the signature timestamps less strictly.
1104
1105 val-bogus-ttl: <number>
1106 The time to live for bogus data. This is data that has failed
1107 validation; due to invalid signatures or other checks. The TTL
1108 from that data cannot be trusted, and this value is used
1109 instead. The value is in seconds, default 60. The time interval
1110 prevents repeated revalidation of bogus data.
1111
1112 val-clean-additional: <yes or no>
1113 Instruct the validator to remove data from the additional sec‐
1114 tion of secure messages that are not signed properly. Messages
1115 that are insecure, bogus, indeterminate or unchecked are not
1116 affected. Default is yes. Use this setting to protect the users
1117 that rely on this validator for authentication from potentially
1118 bad data in the additional section.
1119
1120 val-log-level: <number>
1121 Have the validator print validation failures to the log.
1122 Regardless of the verbosity setting. Default is 0, off. At 1,
1123 for every user query that fails a line is printed to the logs.
1124 This way you can monitor what happens with validation. Use a
1125 diagnosis tool, such as dig or drill, to find out why validation
1126 is failing for these queries. At 2, not only the query that
1127 failed is printed but also the reason why unbound thought it was
1128 wrong and which server sent the faulty data.
1129
1130 val-permissive-mode: <yes or no>
1131 Instruct the validator to mark bogus messages as indeterminate.
1132 The security checks are performed, but if the result is bogus
1133 (failed security), the reply is not withheld from the client
1134 with SERVFAIL as usual. The client receives the bogus data. For
1135 messages that are found to be secure the AD bit is set in
1136 replies. Also logging is performed as for full validation. The
1137 default value is "no".
1138
1139 ignore-cd-flag: <yes or no>
1140 Instruct unbound to ignore the CD flag from clients and refuse
1141 to return bogus answers to them. Thus, the CD (Checking Dis‐
1142 abled) flag does not disable checking any more. This is useful
1143 if legacy (w2008) servers that set the CD flag but cannot vali‐
1144 date DNSSEC themselves are the clients, and then unbound pro‐
1145 vides them with DNSSEC protection. The default value is "no".
1146
1147 serve-expired: <yes or no>
1148 If enabled, unbound attempts to serve old responses from cache
1149 with a TTL of serve-expired-reply-ttl in the response without
1150 waiting for the actual resolution to finish. The actual resolu‐
1151 tion answer ends up in the cache later on. Default is "no".
1152
1153 serve-expired-ttl: <seconds>
1154 Limit serving of expired responses to configured seconds after
1155 expiration. 0 disables the limit. This option only applies when
1156 serve-expired is enabled. A suggested value per draft-ietf-
1157 dnsop-serve-stale-10 is between 86400 (1 day) and 259200 (3
1158 days). The default is 0.
1159
1160 serve-expired-ttl-reset: <yes or no>
1161 Set the TTL of expired records to the serve-expired-ttl value
1162 after a failed attempt to retrieve the record from upstream.
1163 This makes sure that the expired records will be served as long
1164 as there are queries for it. Default is "no".
1165
1166 serve-expired-reply-ttl: <seconds>
1167 TTL value to use when replying with expired data. If
1168 serve-expired-client-timeout is also used then it is RECOMMENDED
1169 to use 30 as the value (draft-ietf-dnsop-serve-stale-10). The
1170 default is 30.
1171
1172 serve-expired-client-timeout: <msec>
1173 Time in milliseconds before replying to the client with expired
1174 data. This essentially enables the serve-stale behavior as
1175 specified in draft-ietf-dnsop-serve-stale-10 that first tries to
1176 resolve before immediately responding with expired data. A rec‐
1177 ommended value per draft-ietf-dnsop-serve-stale-10 is 1800.
1178 Setting this to 0 will disable this behavior. Default is 0.
1179
1180 val-nsec3-keysize-iterations: <"list of values">
1181 List of keysize and iteration count values, separated by spaces,
1182 surrounded by quotes. Default is "1024 150 2048 500 4096 2500".
1183 This determines the maximum allowed NSEC3 iteration count before
1184 a message is simply marked insecure instead of performing the
1185 many hashing iterations. The list must be in ascending order and
1186 have at least one entry. If you set it to "1024 65535" there is
1187 no restriction to NSEC3 iteration values. This table must be
1188 kept short; a very long list could cause slower operation.
1189
1190 add-holddown: <seconds>
1191 Instruct the auto-trust-anchor-file probe mechanism for RFC5011
1192 autotrust updates to add new trust anchors only after they have
1193 been visible for this time. Default is 30 days as per the RFC.
1194
1195 del-holddown: <seconds>
1196 Instruct the auto-trust-anchor-file probe mechanism for RFC5011
1197 autotrust updates to remove revoked trust anchors after they
1198 have been kept in the revoked list for this long. Default is 30
1199 days as per the RFC.
1200
1201 keep-missing: <seconds>
1202 Instruct the auto-trust-anchor-file probe mechanism for RFC5011
1203 autotrust updates to remove missing trust anchors after they
1204 have been unseen for this long. This cleans up the state file
1205 if the target zone does not perform trust anchor revocation, so
1206 this makes the auto probe mechanism work with zones that perform
1207 regular (non-5011) rollovers. The default is 366 days. The
1208 value 0 does not remove missing anchors, as per the RFC.
1209
1210 permit-small-holddown: <yes or no>
1211 Debug option that allows the autotrust 5011 rollover timers to
1212 assume very small values. Default is no.
1213
1214 key-cache-size: <number>
1215 Number of bytes size of the key cache. Default is 4 megabytes.
1216 A plain number is in bytes, append 'k', 'm' or 'g' for kilo‐
1217 bytes, megabytes or gigabytes (1024*1024 bytes in a megabyte).
1218
1219 key-cache-slabs: <number>
1220 Number of slabs in the key cache. Slabs reduce lock contention
1221 by threads. Must be set to a power of 2. Setting (close) to the
1222 number of cpus is a reasonable guess.
1223
1224 neg-cache-size: <number>
1225 Number of bytes size of the aggressive negative cache. Default
1226 is 1 megabyte. A plain number is in bytes, append 'k', 'm' or
1227 'g' for kilobytes, megabytes or gigabytes (1024*1024 bytes in a
1228 megabyte).
1229
1230 unblock-lan-zones: <yes or no>
1231 Default is disabled. If enabled, then for private address
1232 space, the reverse lookups are no longer filtered. This allows
1233 unbound when running as dns service on a host where it provides
1234 service for that host, to put out all of the queries for the
1235 'lan' upstream. When enabled, only localhost, 127.0.0.1 reverse
1236 and ::1 reverse zones are configured with default local zones.
1237 Disable the option when unbound is running as a (DHCP-) DNS net‐
1238 work resolver for a group of machines, where such lookups should
1239 be filtered (RFC compliance), this also stops potential data
1240 leakage about the local network to the upstream DNS servers.
1241
1242 insecure-lan-zones: <yes or no>
1243 Default is disabled. If enabled, then reverse lookups in pri‐
1244 vate address space are not validated. This is usually required
1245 whenever unblock-lan-zones is used.
1246
1247 local-zone: <zone> <type>
1248 Configure a local zone. The type determines the answer to give
1249 if there is no match from local-data. The types are deny,
1250 refuse, static, transparent, redirect, nodefault, typetranspar‐
1251 ent, inform, inform_deny, inform_redirect, always_transparent,
1252 always_refuse, always_nxdomain, noview, and are explained below.
1253 After that the default settings are listed. Use local-data: to
1254 enter data into the local zone. Answers for local zones are
1255 authoritative DNS answers. By default the zones are class IN.
1256
1257 If you need more complicated authoritative data, with referrals,
1258 wildcards, CNAME/DNAME support, or DNSSEC authoritative service,
1259 setup a stub-zone for it as detailed in the stub zone section
1260 below.
1261
1262 deny Do not send an answer, drop the query. If there is a match
1263 from local data, the query is answered.
1264
1265 refuse
1266 Send an error message reply, with rcode REFUSED. If there is
1267 a match from local data, the query is answered.
1268
1269 static
1270 If there is a match from local data, the query is answered.
1271 Otherwise, the query is answered with nodata or nxdomain.
1272 For a negative answer a SOA is included in the answer if
1273 present as local-data for the zone apex domain.
1274
1275 transparent
1276 If there is a match from local data, the query is answered.
1277 Otherwise if the query has a different name, the query is
1278 resolved normally. If the query is for a name given in
1279 localdata but no such type of data is given in localdata,
1280 then a noerror nodata answer is returned. If no local-zone
1281 is given local-data causes a transparent zone to be created
1282 by default.
1283
1284 typetransparent
1285 If there is a match from local data, the query is answered.
1286 If the query is for a different name, or for the same name
1287 but for a different type, the query is resolved normally.
1288 So, similar to transparent but types that are not listed in
1289 local data are resolved normally, so if an A record is in the
1290 local data that does not cause a nodata reply for AAAA
1291 queries.
1292
1293 redirect
1294 The query is answered from the local data for the zone name.
1295 There may be no local data beneath the zone name. This
1296 answers queries for the zone, and all subdomains of the zone
1297 with the local data for the zone. It can be used to redirect
1298 a domain to return a different address record to the end
1299 user, with local-zone: "example.com." redirect and
1300 local-data: "example.com. A 127.0.0.1" queries for www.exam‐
1301 ple.com and www.foo.example.com are redirected, so that users
1302 with web browsers cannot access sites with suffix exam‐
1303 ple.com.
1304
1305 inform
1306 The query is answered normally, same as transparent. The
1307 client IP address (@portnumber) is printed to the logfile.
1308 The log message is: timestamp, unbound-pid, info: zonename
1309 inform IP@port queryname type class. This option can be used
1310 for normal resolution, but machines looking up infected names
1311 are logged, eg. to run antivirus on them.
1312
1313 inform_deny
1314 The query is dropped, like 'deny', and logged, like 'inform'.
1315 Ie. find infected machines without answering the queries.
1316
1317 inform_redirect
1318 The query is redirected, like 'redirect', and logged, like
1319 'inform'. Ie. answer queries with fixed data and also log
1320 the machines that ask.
1321
1322 always_transparent
1323 Like transparent, but ignores local data and resolves nor‐
1324 mally.
1325
1326 always_refuse
1327 Like refuse, but ignores local data and refuses the query.
1328
1329 always_nxdomain
1330 Like static, but ignores local data and returns nxdomain for
1331 the query.
1332
1333 noview
1334 Breaks out of that view and moves towards the global local
1335 zones for answer to the query. If the view first is no,
1336 it'll resolve normally. If view first is enabled, it'll
1337 break perform that step and check the global answers. For
1338 when the view has view specific overrides but some zone has
1339 to be answered from global local zone contents.
1340
1341 nodefault
1342 Used to turn off default contents for AS112 zones. The other
1343 types also turn off default contents for the zone. The 'node‐
1344 fault' option has no other effect than turning off default
1345 contents for the given zone. Use nodefault if you use
1346 exactly that zone, if you want to use a subzone, use trans‐
1347 parent.
1348
1349 The default zones are localhost, reverse 127.0.0.1 and ::1, the onion,
1350 test, invalid and the AS112 zones. The AS112 zones are reverse DNS
1351 zones for private use and reserved IP addresses for which the servers
1352 on the internet cannot provide correct answers. They are configured by
1353 default to give nxdomain (no reverse information) answers. The defaults
1354 can be turned off by specifying your own local-zone of that name, or
1355 using the 'nodefault' type. Below is a list of the default zone con‐
1356 tents.
1357
1358 localhost
1359 The IP4 and IP6 localhost information is given. NS and SOA
1360 records are provided for completeness and to satisfy some DNS
1361 update tools. Default content:
1362 local-zone: "localhost." redirect
1363 local-data: "localhost. 10800 IN NS localhost."
1364 local-data: "localhost. 10800 IN
1365 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1366 local-data: "localhost. 10800 IN A 127.0.0.1"
1367 local-data: "localhost. 10800 IN AAAA ::1"
1368
1369 reverse IPv4 loopback
1370 Default content:
1371 local-zone: "127.in-addr.arpa." static
1372 local-data: "127.in-addr.arpa. 10800 IN NS localhost."
1373 local-data: "127.in-addr.arpa. 10800 IN
1374 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1375 local-data: "1.0.0.127.in-addr.arpa. 10800 IN
1376 PTR localhost."
1377
1378 reverse IPv6 loopback
1379 Default content:
1380 local-zone: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
1381 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa." static
1382 local-data: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
1383 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa. 10800 IN
1384 NS localhost."
1385 local-data: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
1386 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa. 10800 IN
1387 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1388 local-data: "1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
1389 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa. 10800 IN
1390 PTR localhost."
1391
1392 onion (RFC 7686)
1393 Default content:
1394 local-zone: "onion." static
1395 local-data: "onion. 10800 IN NS localhost."
1396 local-data: "onion. 10800 IN
1397 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1398
1399 test (RFC 6761)
1400 Default content:
1401 local-zone: "test." static
1402 local-data: "test. 10800 IN NS localhost."
1403 local-data: "test. 10800 IN
1404 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1405
1406 invalid (RFC 6761)
1407 Default content:
1408 local-zone: "invalid." static
1409 local-data: "invalid. 10800 IN NS localhost."
1410 local-data: "invalid. 10800 IN
1411 SOA localhost. nobody.invalid. 1 3600 1200 604800 10800"
1412
1413 reverse RFC1918 local use zones
1414 Reverse data for zones 10.in-addr.arpa, 16.172.in-addr.arpa
1415 to 31.172.in-addr.arpa, 168.192.in-addr.arpa. The
1416 local-zone: is set static and as local-data: SOA and NS
1417 records are provided.
1418
1419 reverse RFC3330 IP4 this, link-local, testnet and broadcast
1420 Reverse data for zones 0.in-addr.arpa, 254.169.in-addr.arpa,
1421 2.0.192.in-addr.arpa (TEST NET 1), 100.51.198.in-addr.arpa
1422 (TEST NET 2), 113.0.203.in-addr.arpa (TEST NET 3),
1423 255.255.255.255.in-addr.arpa. And from 64.100.in-addr.arpa
1424 to 127.100.in-addr.arpa (Shared Address Space).
1425
1426 reverse RFC4291 IP6 unspecified
1427 Reverse data for zone
1428 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
1429 0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa.
1430
1431 reverse RFC4193 IPv6 Locally Assigned Local Addresses
1432 Reverse data for zone D.F.ip6.arpa.
1433
1434 reverse RFC4291 IPv6 Link Local Addresses
1435 Reverse data for zones 8.E.F.ip6.arpa to B.E.F.ip6.arpa.
1436
1437 reverse IPv6 Example Prefix
1438 Reverse data for zone 8.B.D.0.1.0.0.2.ip6.arpa. This zone is
1439 used for tutorials and examples. You can remove the block on
1440 this zone with:
1441 local-zone: 8.B.D.0.1.0.0.2.ip6.arpa. nodefault
1442 You can also selectively unblock a part of the zone by making
1443 that part transparent with a local-zone statement. This also
1444 works with the other default zones.
1445
1446 local-data: "<resource record string>"
1447 Configure local data, which is served in reply to queries for it.
1448 The query has to match exactly unless you configure the local-zone
1449 as redirect. If not matched exactly, the local-zone type deter‐
1450 mines further processing. If local-data is configured that is not
1451 a subdomain of a local-zone, a transparent local-zone is config‐
1452 ured. For record types such as TXT, use single quotes, as in
1453 local-data: 'example. TXT "text"'.
1454
1455 If you need more complicated authoritative data, with referrals,
1456 wildcards, CNAME/DNAME support, or DNSSEC authoritative service,
1457 setup a stub-zone for it as detailed in the stub zone section
1458 below.
1459
1460 local-data-ptr: "IPaddr name"
1461 Configure local data shorthand for a PTR record with the reversed
1462 IPv4 or IPv6 address and the host name. For example "192.0.2.4
1463 www.example.com". TTL can be inserted like this: "2001:DB8::4
1464 7200 www.example.com"
1465
1466 local-zone-tag: <zone> <"list of tags">
1467 Assign tags to localzones. Tagged localzones will only be applied
1468 when the used access-control element has a matching tag. Tags must
1469 be defined in define-tags. Enclose list of tags in quotes ("")
1470 and put spaces between tags. When there are multiple tags it
1471 checks if the intersection of the list of tags for the query and
1472 local-zone-tag is non-empty.
1473
1474 local-zone-override: <zone> <IP netblock> <type>
1475 Override the localzone type for queries from addresses matching
1476 netblock. Use this localzone type, regardless the type configured
1477 for the local-zone (both tagged and untagged) and regardless the
1478 type configured using access-control-tag-action.
1479
1480 ratelimit: <number or 0>
1481 Enable ratelimiting of queries sent to nameserver for performing
1482 recursion. If 0, the default, it is disabled. This option is
1483 experimental at this time. The ratelimit is in queries per second
1484 that are allowed. More queries are turned away with an error
1485 (servfail). This stops recursive floods, eg. random query names,
1486 but not spoofed reflection floods. Cached responses are not rate‐
1487 limited by this setting. The zone of the query is determined by
1488 examining the nameservers for it, the zone name is used to keep
1489 track of the rate. For example, 1000 may be a suitable value to
1490 stop the server from being overloaded with random names, and keeps
1491 unbound from sending traffic to the nameservers for those zones.
1492
1493 ratelimit-size: <memory size>
1494 Give the size of the data structure in which the current ongoing
1495 rates are kept track in. Default 4m. In bytes or use m(mega),
1496 k(kilo), g(giga). The ratelimit structure is small, so this data
1497 structure likely does not need to be large.
1498
1499 ratelimit-slabs: <number>
1500 Give power of 2 number of slabs, this is used to reduce lock con‐
1501 tention in the ratelimit tracking data structure. Close to the
1502 number of cpus is a fairly good setting.
1503
1504 ratelimit-factor: <number>
1505 Set the amount of queries to rate limit when the limit is
1506 exceeded. If set to 0, all queries are dropped for domains where
1507 the limit is exceeded. If set to another value, 1 in that number
1508 is allowed through to complete. Default is 10, allowing 1/10
1509 traffic to flow normally. This can make ordinary queries complete
1510 (if repeatedly queried for), and enter the cache, whilst also mit‐
1511 igating the traffic flow by the factor given.
1512
1513 ratelimit-for-domain: <domain> <number qps or 0>
1514 Override the global ratelimit for an exact match domain name with
1515 the listed number. You can give this for any number of names.
1516 For example, for a top-level-domain you may want to have a higher
1517 limit than other names. A value of 0 will disable ratelimiting
1518 for that domain.
1519
1520 ratelimit-below-domain: <domain> <number qps or 0>
1521 Override the global ratelimit for a domain name that ends in this
1522 name. You can give this multiple times, it then describes differ‐
1523 ent settings in different parts of the namespace. The closest
1524 matching suffix is used to determine the qps limit. The rate for
1525 the exact matching domain name is not changed, use rate‐
1526 limit-for-domain to set that, you might want to use different set‐
1527 tings for a top-level-domain and subdomains. A value of 0 will
1528 disable ratelimiting for domain names that end in this name.
1529
1530 ip-ratelimit: <number or 0>
1531 Enable global ratelimiting of queries accepted per ip address. If
1532 0, the default, it is disabled. This option is experimental at
1533 this time. The ratelimit is in queries per second that are
1534 allowed. More queries are completely dropped and will not receive
1535 a reply, SERVFAIL or otherwise. IP ratelimiting happens before
1536 looking in the cache. This may be useful for mitigating amplifica‐
1537 tion attacks.
1538
1539 ip-ratelimit-size: <memory size>
1540 Give the size of the data structure in which the current ongoing
1541 rates are kept track in. Default 4m. In bytes or use m(mega),
1542 k(kilo), g(giga). The ip ratelimit structure is small, so this
1543 data structure likely does not need to be large.
1544
1545 ip-ratelimit-slabs: <number>
1546 Give power of 2 number of slabs, this is used to reduce lock con‐
1547 tention in the ip ratelimit tracking data structure. Close to the
1548 number of cpus is a fairly good setting.
1549
1550 ip-ratelimit-factor: <number>
1551 Set the amount of queries to rate limit when the limit is
1552 exceeded. If set to 0, all queries are dropped for addresses
1553 where the limit is exceeded. If set to another value, 1 in that
1554 number is allowed through to complete. Default is 10, allowing
1555 1/10 traffic to flow normally. This can make ordinary queries
1556 complete (if repeatedly queried for), and enter the cache, whilst
1557 also mitigating the traffic flow by the factor given.
1558
1559 fast-server-permil: <number>
1560 Specify how many times out of 1000 to pick from the set of fastest
1561 servers. 0 turns the feature off. A value of 900 would pick from
1562 the fastest servers 90 percent of the time, and would perform nor‐
1563 mal exploration of random servers for the remaining time. When
1564 prefetch is enabled (or serve-expired), such prefetches are not
1565 sped up, because there is no one waiting for it, and it presents a
1566 good moment to perform server exploration. The fast-server-num
1567 option can be used to specify the size of the fastest servers set.
1568 The default for fast-server-permil is 0.
1569
1570 fast-server-num: <number>
1571 Set the number of servers that should be used for fast server
1572 selection. Only use the fastest specified number of servers with
1573 the fast-server-permil option, that turns this on or off. The
1574 default is to use the fastest 3 servers.
1575
1576 Remote Control Options
1577 In the remote-control: clause are the declarations for the remote con‐
1578 trol facility. If this is enabled, the unbound-control(8) utility can
1579 be used to send commands to the running unbound server. The server
1580 uses these clauses to setup TLSv1 security for the connection. The
1581 unbound-control(8) utility also reads the remote-control section for
1582 options. To setup the correct self-signed certificates use the
1583 unbound-control-setup(8) utility.
1584
1585 control-enable: <yes or no>
1586 The option is used to enable remote control, default is "no". If
1587 turned off, the server does not listen for control commands.
1588
1589 control-interface: <ip address or path>
1590 Give IPv4 or IPv6 addresses or local socket path to listen on for
1591 control commands. By default localhost (127.0.0.1 and ::1) is
1592 listened to. Use 0.0.0.0 and ::0 to listen to all interfaces. If
1593 you change this and permissions have been dropped, you must
1594 restart the server for the change to take effect.
1595
1596 If you set it to an absolute path, a local socket is used. The
1597 local socket does not use the certificates and keys, so those
1598 files need not be present. To restrict access, unbound sets per‐
1599 missions on the file to the user and group that is configured, the
1600 access bits are set to allow the group members to access the con‐
1601 trol socket file. Put users that need to access the socket in the
1602 that group. To restrict access further, create a directory to put
1603 the control socket in and restrict access to that directory.
1604
1605 control-port: <port number>
1606 The port number to listen on for IPv4 or IPv6 control interfaces,
1607 default is 8953. If you change this and permissions have been
1608 dropped, you must restart the server for the change to take
1609 effect.
1610
1611 control-use-cert: <yes or no>
1612 For localhost control-interface you can disable the use of TLS by
1613 setting this option to "no", default is "yes". For local sockets,
1614 TLS is disabled and the value of this option is ignored.
1615
1616 server-key-file: <private key file>
1617 Path to the server private key, by default unbound_server.key.
1618 This file is generated by the unbound-control-setup utility. This
1619 file is used by the unbound server, but not by unbound-control.
1620
1621 server-cert-file: <certificate file.pem>
1622 Path to the server self signed certificate, by default
1623 unbound_server.pem. This file is generated by the unbound-con‐
1624 trol-setup utility. This file is used by the unbound server, and
1625 also by unbound-control.
1626
1627 control-key-file: <private key file>
1628 Path to the control client private key, by default unbound_con‐
1629 trol.key. This file is generated by the unbound-control-setup
1630 utility. This file is used by unbound-control.
1631
1632 control-cert-file: <certificate file.pem>
1633 Path to the control client certificate, by default unbound_con‐
1634 trol.pem. This certificate has to be signed with the server cer‐
1635 tificate. This file is generated by the unbound-control-setup
1636 utility. This file is used by unbound-control.
1637
1638 Stub Zone Options
1639 There may be multiple stub-zone: clauses. Each with a name: and zero or
1640 more hostnames or IP addresses. For the stub zone this list of name‐
1641 servers is used. Class IN is assumed. The servers should be authority
1642 servers, not recursors; unbound performs the recursive processing
1643 itself for stub zones.
1644
1645 The stub zone can be used to configure authoritative data to be used by
1646 the resolver that cannot be accessed using the public internet servers.
1647 This is useful for company-local data or private zones. Setup an
1648 authoritative server on a different host (or different port). Enter a
1649 config entry for unbound with stub-addr: <ip address of host[@port]>.
1650 The unbound resolver can then access the data, without referring to the
1651 public internet for it.
1652
1653 This setup allows DNSSEC signed zones to be served by that authorita‐
1654 tive server, in which case a trusted key entry with the public key can
1655 be put in config, so that unbound can validate the data and set the AD
1656 bit on replies for the private zone (authoritative servers do not set
1657 the AD bit). This setup makes unbound capable of answering queries for
1658 the private zone, and can even set the AD bit ('authentic'), but the AA
1659 ('authoritative') bit is not set on these replies.
1660
1661 Consider adding server: statements for domain-insecure: and for
1662 local-zone: name nodefault for the zone if it is a locally served zone.
1663 The insecure clause stops DNSSEC from invalidating the zone. The local
1664 zone nodefault (or transparent) clause makes the (reverse-) zone bypass
1665 unbound's filtering of RFC1918 zones.
1666
1667 name: <domain name>
1668 Name of the stub zone.
1669
1670 stub-host: <domain name>
1671 Name of stub zone nameserver. Is itself resolved before it is
1672 used.
1673
1674 stub-addr: <IP address>
1675 IP address of stub zone nameserver. Can be IP 4 or IP 6. To use
1676 a nondefault port for DNS communication append '@' with the port
1677 number.
1678
1679 stub-prime: <yes or no>
1680 This option is by default no. If enabled it performs NS set
1681 priming, which is similar to root hints, where it starts using
1682 the list of nameservers currently published by the zone. Thus,
1683 if the hint list is slightly outdated, the resolver picks up a
1684 correct list online.
1685
1686 stub-first: <yes or no>
1687 If enabled, a query is attempted without the stub clause if it
1688 fails. The data could not be retrieved and would have caused
1689 SERVFAIL because the servers are unreachable, instead it is
1690 tried without this clause. The default is no.
1691
1692 stub-tls-upstream: <yes or no>
1693 Enabled or disable whether the queries to this stub use TLS for
1694 transport. Default is no.
1695
1696 stub-ssl-upstream: <yes or no>
1697 Alternate syntax for stub-tls-upstream.
1698
1699 stub-no-cache: <yes or no>
1700 Default is no. If enabled, data inside the stub is not cached.
1701 This is useful when you want immediate changes to be visible.
1702
1703 Forward Zone Options
1704 There may be multiple forward-zone: clauses. Each with a name: and zero
1705 or more hostnames or IP addresses. For the forward zone this list of
1706 nameservers is used to forward the queries to. The servers listed as
1707 forward-host: and forward-addr: have to handle further recursion for
1708 the query. Thus, those servers are not authority servers, but are
1709 (just like unbound is) recursive servers too; unbound does not perform
1710 recursion itself for the forward zone, it lets the remote server do it.
1711 Class IN is assumed. CNAMEs are chased by unbound itself, asking the
1712 remote server for every name in the indirection chain, to protect the
1713 local cache from illegal indirect referenced items. A forward-zone
1714 entry with name "." and a forward-addr target will forward all queries
1715 to that other server (unless it can answer from the cache).
1716
1717 name: <domain name>
1718 Name of the forward zone.
1719
1720 forward-host: <domain name>
1721 Name of server to forward to. Is itself resolved before it is
1722 used.
1723
1724 forward-addr: <IP address>
1725 IP address of server to forward to. Can be IP 4 or IP 6. To use
1726 a nondefault port for DNS communication append '@' with the port
1727 number. If tls is enabled, then you can append a '#' and a
1728 name, then it'll check the tls authentication certificates with
1729 that name. If you combine the '@' and '#', the '@' comes first.
1730
1731 At high verbosity it logs the TLS certificate, with TLS enabled.
1732 If you leave out the '#' and auth name from the forward-addr,
1733 any name is accepted. The cert must also match a CA from the
1734 tls-cert-bundle.
1735
1736 forward-first: <yes or no>
1737 If a forwarded query is met with a SERVFAIL error, and this
1738 option is enabled, unbound will fall back to normal recursive
1739 resolution for this query as if no query forwarding had been
1740 specified. The default is "no".
1741
1742 forward-tls-upstream: <yes or no>
1743 Enabled or disable whether the queries to this forwarder use TLS
1744 for transport. Default is no. If you enable this, also config‐
1745 ure a tls-cert-bundle or use tls-win-cert to load CA certs, oth‐
1746 erwise the connections cannot be authenticated.
1747
1748 forward-ssl-upstream: <yes or no>
1749 Alternate syntax for forward-tls-upstream.
1750
1751 forward-no-cache: <yes or no>
1752 Default is no. If enabled, data inside the forward is not
1753 cached. This is useful when you want immediate changes to be
1754 visible.
1755
1756 Authority Zone Options
1757 Authority zones are configured with auth-zone:, and each one must have
1758 a name:. There can be multiple ones, by listing multiple auth-zone
1759 clauses, each with a different name, pertaining to that part of the
1760 namespace. The authority zone with the name closest to the name looked
1761 up is used. Authority zones are processed after local-zones and before
1762 cache (for-downstream: yes), and when used in this manner make unbound
1763 respond like an authority server. Authority zones are also processed
1764 after cache, just before going to the network to fetch information for
1765 recursion (for-upstream: yes), and when used in this manner provide a
1766 local copy of an authority server that speeds up lookups of that data.
1767
1768 Authority zones can be read from zonefile. And can be kept updated via
1769 AXFR and IXFR. After update the zonefile is rewritten. The update
1770 mechanism uses the SOA timer values and performs SOA UDP queries to
1771 detect zone changes.
1772
1773 If the update fetch fails, the timers in the SOA record are used to
1774 time another fetch attempt. Until the SOA expiry timer is reached.
1775 Then the zone is expired. When a zone is expired, queries are SERV‐
1776 FAIL, and any new serial number is accepted from the master (even if
1777 older), and if fallback is enabled, the fallback activates to fetch
1778 from the upstream instead of the SERVFAIL.
1779
1780 name: <zone name>
1781 Name of the authority zone.
1782
1783 master: <IP address or host name>
1784 Where to download a copy of the zone from, with AXFR and IXFR.
1785 Multiple masters can be specified. They are all tried if one
1786 fails. With the "ip#name" notation a AXFR over TLS can be used.
1787 If you point it at another Unbound instance, it would not work
1788 because that does not support AXFR/IXFR for the zone, but if you
1789 used url: to download the zonefile as a text file from a web‐
1790 server that would work. If you specify the hostname, you cannot
1791 use the domain from the zonefile, because it may not have that
1792 when retrieving that data, instead use a plain IP address to
1793 avoid a circular dependency on retrieving that IP address.
1794
1795 url: <url to zonefile>
1796 Where to download a zonefile for the zone. With http or https.
1797 An example for the url is "http://www.example.com/exam‐
1798 ple.org.zone". Multiple url statements can be given, they are
1799 tried in turn. If only urls are given the SOA refresh timer is
1800 used to wait for making new downloads. If also masters are
1801 listed, the masters are first probed with UDP SOA queries to see
1802 if the SOA serial number has changed, reducing the number of
1803 downloads. If none of the urls work, the masters are tried with
1804 IXFR and AXFR. For https, the tls-cert-bundle and the hostname
1805 from the url are used to authenticate the connection. If you
1806 specify a hostname in the URL, you cannot use the domain from
1807 the zonefile, because it may not have that when retrieving that
1808 data, instead use a plain IP address to avoid a circular depen‐
1809 dency on retrieving that IP address. Avoid dependencies on name
1810 lookups by using a notation like "http://192.0.2.1/unbound-mas‐
1811 ter/example.com.zone", with an explicit IP address.
1812
1813 allow-notify: <IP address or host name or netblockIP/prefix>
1814 With allow-notify you can specify additional sources of noti‐
1815 fies. When notified, the server attempts to first probe and
1816 then zone transfer. If the notify is from a master, it first
1817 attempts that master. Otherwise other masters are attempted.
1818 If there are no masters, but only urls, the file is downloaded
1819 when notified. The masters from master: statements are allowed
1820 notify by default.
1821
1822 fallback-enabled: <yes or no>
1823 Default no. If enabled, unbound falls back to querying the
1824 internet as a resolver for this zone when lookups fail. For
1825 example for DNSSEC validation failures.
1826
1827 for-downstream: <yes or no>
1828 Default yes. If enabled, unbound serves authority responses to
1829 downstream clients for this zone. This option makes unbound
1830 behave, for the queries with names in this zone, like one of the
1831 authority servers for that zone. Turn it off if you want
1832 unbound to provide recursion for the zone but have a local copy
1833 of zone data. If for-downstream is no and for-upstream is yes,
1834 then unbound will DNSSEC validate the contents of the zone
1835 before serving the zone contents to clients and store validation
1836 results in the cache.
1837
1838 for-upstream: <yes or no>
1839 Default yes. If enabled, unbound fetches data from this data
1840 collection for answering recursion queries. Instead of sending
1841 queries over the internet to the authority servers for this
1842 zone, it'll fetch the data directly from the zone data. Turn it
1843 on when you want unbound to provide recursion for downstream
1844 clients, and use the zone data as a local copy to speed up
1845 lookups.
1846
1847 zonefile: <filename>
1848 The filename where the zone is stored. If not given then no
1849 zonefile is used. If the file does not exist or is empty,
1850 unbound will attempt to fetch zone data (eg. from the master
1851 servers).
1852
1853 View Options
1854 There may be multiple view: clauses. Each with a name: and zero or more
1855 local-zone and local-data elements. Views can also contain view-first,
1856 response-ip, response-ip-data and local-data-ptr elements. View can be
1857 mapped to requests by specifying the view name in an access-con‐
1858 trol-view element. Options from matching views will override global
1859 options. Global options will be used if no matching view is found, or
1860 when the matching view does not have the option specified.
1861
1862 name: <view name>
1863 Name of the view. Must be unique. This name is used in
1864 access-control-view elements.
1865
1866 local-zone: <zone> <type>
1867 View specific local-zone elements. Has the same types and behav‐
1868 iour as the global local-zone elements. When there is at least
1869 one local-zone specified and view-first is no, the default
1870 local-zones will be added to this view. Defaults can be dis‐
1871 abled using the nodefault type. When view-first is yes or when a
1872 view does not have a local-zone, the global local-zone will be
1873 used including it's default zones.
1874
1875 local-data: "<resource record string>"
1876 View specific local-data elements. Has the same behaviour as the
1877 global local-data elements.
1878
1879 local-data-ptr: "IPaddr name"
1880 View specific local-data-ptr elements. Has the same behaviour as
1881 the global local-data-ptr elements.
1882
1883 view-first: <yes or no>
1884 If enabled, it attempts to use the global local-zone and
1885 local-data if there is no match in the view specific options.
1886 The default is no.
1887
1888 Python Module Options
1889 The python: clause gives the settings for the python(1) script module.
1890 This module acts like the iterator and validator modules do, on queries
1891 and answers. To enable the script module it has to be compiled into
1892 the daemon, and the word "python" has to be put in the module-config:
1893 option (usually first, or between the validator and iterator). Multiple
1894 instances of the python module are supported by adding the word
1895 "python" more than once.
1896
1897 If the chroot: option is enabled, you should make sure Python's library
1898 directory structure is bind mounted in the new root environment, see
1899 mount(8). Also the python-script: path should be specified as an abso‐
1900 lute path relative to the new root, or as a relative path to the work‐
1901 ing directory.
1902
1903 python-script: <python file>
1904 The script file to load. Repeat this option for every python
1905 module instance added to the module-config: option.
1906
1907 DNS64 Module Options
1908 The dns64 module must be configured in the module-config: "dns64 val‐
1909 idator iterator" directive and be compiled into the daemon to be
1910 enabled. These settings go in the server: section.
1911
1912 dns64-prefix: <IPv6 prefix>
1913 This sets the DNS64 prefix to use to synthesize AAAA records
1914 with. It must be /96 or shorter. The default prefix is
1915 64:ff9b::/96.
1916
1917 dns64-synthall: <yes or no>
1918 Debug option, default no. If enabled, synthesize all AAAA
1919 records despite the presence of actual AAAA records.
1920
1921 dns64-ignore-aaaa: <name>
1922 List domain for which the AAAA records are ignored and the A
1923 record is used by dns64 processing instead. Can be entered mul‐
1924 tiple times, list a new domain for which it applies, one per
1925 line. Applies also to names underneath the name given.
1926
1927 DNSCrypt Options
1928 The dnscrypt: clause gives the settings of the dnscrypt channel. While
1929 those options are available, they are only meaningful if unbound was
1930 compiled with --enable-dnscrypt. Currently certificate and secret/pub‐
1931 lic keys cannot be generated by unbound. You can use dnscrypt-wrapper
1932 to generate those: https://github.com/cofyc/dnscrypt-wrapper/blob/mas‐
1933 ter/README.md#usage
1934
1935 dnscrypt-enable: <yes or no>
1936 Whether or not the dnscrypt config should be enabled. You may
1937 define configuration but not activate it. The default is no.
1938
1939 dnscrypt-port: <port number>
1940 On which port should dnscrypt should be activated. Note that you
1941 should have a matching interface option defined in the server
1942 section for this port.
1943
1944 dnscrypt-provider: <provider name>
1945 The provider name to use to distribute certificates. This is of
1946 the form: 2.dnscrypt-cert.example.com.. The name MUST end with a
1947 dot.
1948
1949 dnscrypt-secret-key: <path to secret key file>
1950 Path to the time limited secret key file. This option may be
1951 specified multiple times.
1952
1953 dnscrypt-provider-cert: <path to cert file>
1954 Path to the certificate related to the dnscrypt-secret-keys.
1955 This option may be specified multiple times.
1956
1957 dnscrypt-provider-cert-rotated: <path to cert file>
1958 Path to a certificate that we should be able to serve existing
1959 connection from but do not want to advertise over
1960 dnscrypt-provider's TXT record certs distribution. A typical
1961 use case is when rotating certificates, existing clients may
1962 still use the client magic from the old cert in their queries
1963 until they fetch and update the new cert. Likewise, it would
1964 allow one to prime the new cert/key without distributing the new
1965 cert yet, this can be useful when using a network of servers
1966 using anycast and on which the configuration may not get updated
1967 at the exact same time. By priming the cert, the servers can
1968 handle both old and new certs traffic while distributing only
1969 one. This option may be specified multiple times.
1970
1971 dnscrypt-shared-secret-cache-size: <memory size>
1972 Give the size of the data structure in which the shared secret
1973 keys are kept in. Default 4m. In bytes or use m(mega),
1974 k(kilo), g(giga). The shared secret cache is used when a same
1975 client is making multiple queries using the same public key. It
1976 saves a substantial amount of CPU.
1977
1978 dnscrypt-shared-secret-cache-slabs: <number>
1979 Give power of 2 number of slabs, this is used to reduce lock
1980 contention in the dnscrypt shared secrets cache. Close to the
1981 number of cpus is a fairly good setting.
1982
1983 dnscrypt-nonce-cache-size: <memory size>
1984 Give the size of the data structure in which the client nonces
1985 are kept in. Default 4m. In bytes or use m(mega), k(kilo),
1986 g(giga). The nonce cache is used to prevent dnscrypt message
1987 replaying. Client nonce should be unique for any pair of client
1988 pk/server sk.
1989
1990 dnscrypt-nonce-cache-slabs: <number>
1991 Give power of 2 number of slabs, this is used to reduce lock
1992 contention in the dnscrypt nonce cache. Close to the number of
1993 cpus is a fairly good setting.
1994
1995 EDNS Client Subnet Module Options
1996 The ECS module must be configured in the module-config: "subnetcache
1997 validator iterator" directive and be compiled into the daemon to be
1998 enabled. These settings go in the server: section.
1999
2000 If the destination address is whitelisted with Unbound will add the
2001 EDNS0 option to the query containing the relevant part of the client's
2002 address. When an answer contains the ECS option the response and the
2003 option are placed in a specialized cache. If the authority indicated no
2004 support, the response is stored in the regular cache.
2005
2006 Additionally, when a client includes the option in its queries, Unbound
2007 will forward the option to the authority if present in the whitelist,
2008 or client-subnet-always-forward is set to yes. In this case the lookup
2009 in the regular cache is skipped.
2010
2011 The maximum size of the ECS cache is controlled by 'msg-cache-size' in
2012 the configuration file. On top of that, for each query only 100 differ‐
2013 ent subnets are allowed to be stored for each address family. Exceeding
2014 that number, older entries will be purged from cache.
2015
2016 send-client-subnet: <IP address>
2017 Send client source address to this authority. Append /num to
2018 indicate a classless delegation netblock, for example like
2019 10.2.3.4/24 or 2001::11/64. Can be given multiple times. Author‐
2020 ities not listed will not receive edns-subnet information,
2021 unless domain in query is specified in client-subnet-zone.
2022
2023 client-subnet-zone: <domain>
2024 Send client source address in queries for this domain and its
2025 subdomains. Can be given multiple times. Zones not listed will
2026 not receive edns-subnet information, unless hosted by authority
2027 specified in send-client-subnet.
2028
2029 client-subnet-always-forward: <yes or no>
2030 Specify whether the ECS whitelist check (configured using
2031 send-client-subnet) is applied for all queries, even if the
2032 triggering query contains an ECS record, or only for queries for
2033 which the ECS record is generated using the querier address (and
2034 therefore did not contain ECS data in the client query). If
2035 enabled, the whitelist check is skipped when the client query
2036 contains an ECS record. Default is no.
2037
2038 max-client-subnet-ipv6: <number>
2039 Specifies the maximum prefix length of the client source address
2040 we are willing to expose to third parties for IPv6. Defaults to
2041 56.
2042
2043 max-client-subnet-ipv4: <number>
2044 Specifies the maximum prefix length of the client source address
2045 we are willing to expose to third parties for IPv4. Defaults to
2046 24.
2047
2048 min-client-subnet-ipv6: <number>
2049 Specifies the minimum prefix length of the IPv6 source mask we
2050 are willing to accept in queries. Shorter source masks result in
2051 REFUSED answers. Source mask of 0 is always accepted. Default is
2052 0.
2053
2054 min-client-subnet-ipv4: <number>
2055 Specifies the minimum prefix length of the IPv4 source mask we
2056 are willing to accept in queries. Shorter source masks result in
2057 REFUSED answers. Source mask of 0 is always accepted. Default is
2058 0.
2059
2060 max-ecs-tree-size-ipv4: <number>
2061 Specifies the maximum number of subnets ECS answers kept in the
2062 ECS radix tree. This number applies for each qname/qclass/qtype
2063 tuple. Defaults to 100.
2064
2065 max-ecs-tree-size-ipv6: <number>
2066 Specifies the maximum number of subnets ECS answers kept in the
2067 ECS radix tree. This number applies for each qname/qclass/qtype
2068 tuple. Defaults to 100.
2069
2070 Opportunistic IPsec Support Module Options
2071 The IPsec module must be configured in the module-config: "ipsecmod
2072 validator iterator" directive and be compiled into the daemon to be
2073 enabled. These settings go in the server: section.
2074
2075 When unbound receives an A/AAAA query that is not in the cache and
2076 finds a valid answer, it will withhold returning the answer and instead
2077 will generate an IPSECKEY subquery for the same domain name. If an
2078 answer was found, unbound will call an external hook passing the fol‐
2079 lowing arguments:
2080
2081 QNAME
2082 Domain name of the A/AAAA and IPSECKEY query. In string for‐
2083 mat.
2084
2085 IPSECKEY TTL
2086 TTL of the IPSECKEY RRset.
2087
2088 A/AAAA
2089 String of space separated IP addresses present in the A/AAAA
2090 RRset. The IP addresses are in string format.
2091
2092 IPSECKEY
2093 String of space separated IPSECKEY RDATA present in the
2094 IPSECKEY RRset. The IPSECKEY RDATA are in DNS presentation
2095 format.
2096
2097 The A/AAAA answer is then cached and returned to the client. If the
2098 external hook was called the TTL changes to ensure it doesn't surpass
2099 ipsecmod-max-ttl.
2100
2101 The same procedure is also followed when prefetch: is used, but the
2102 A/AAAA answer is given to the client before the hook is called. ipsec‐
2103 mod-max-ttl ensures that the A/AAAA answer given from cache is still
2104 relevant for opportunistic IPsec.
2105
2106 ipsecmod-enabled: <yes or no>
2107 Specifies whether the IPsec module is enabled or not. The IPsec
2108 module still needs to be defined in the module-config: direc‐
2109 tive. This option facilitates turning on/off the module without
2110 restarting/reloading unbound. Defaults to yes.
2111
2112 ipsecmod-hook: <filename>
2113 Specifies the external hook that unbound will call with sys‐
2114 tem(3). The file can be specified as an absolute/relative path.
2115 The file needs the proper permissions to be able to be executed
2116 by the same user that runs unbound. It must be present when the
2117 IPsec module is defined in the module-config: directive.
2118
2119 ipsecmod-strict: <yes or no>
2120 If enabled unbound requires the external hook to return a suc‐
2121 cess value of 0. Failing to do so unbound will reply with SERV‐
2122 FAIL. The A/AAAA answer will also not be cached. Defaults to
2123 no.
2124
2125 ipsecmod-max-ttl: <seconds>
2126 Time to live maximum for A/AAAA cached records after calling the
2127 external hook. Defaults to 3600.
2128
2129 ipsecmod-ignore-bogus: <yes or no>
2130 Specifies the behaviour of unbound when the IPSECKEY answer is
2131 bogus. If set to yes, the hook will be called and the A/AAAA
2132 answer will be returned to the client. If set to no, the hook
2133 will not be called and the answer to the A/AAAA query will be
2134 SERVFAIL. Mainly used for testing. Defaults to no.
2135
2136 ipsecmod-whitelist: <domain>
2137 Whitelist the domain so that the module logic will be executed.
2138 Can be given multiple times, for different domains. If the
2139 option is not specified, all domains are treated as being
2140 whitelisted (default).
2141
2142 Cache DB Module Options
2143 The Cache DB module must be configured in the module-config: "validator
2144 cachedb iterator" directive and be compiled into the daemon with
2145 --enable-cachedb. If this module is enabled and configured, the speci‐
2146 fied backend database works as a second level cache: When Unbound can‐
2147 not find an answer to a query in its built-in in-memory cache, it con‐
2148 sults the specified backend. If it finds a valid answer in the back‐
2149 end, Unbound uses it to respond to the query without performing itera‐
2150 tive DNS resolution. If Unbound cannot even find an answer in the
2151 backend, it resolves the query as usual, and stores the answer in the
2152 backend.
2153
2154 This module interacts with the serve-expired-* options and will reply
2155 with expired data if unbound is configured for that. Currently the use
2156 of serve-expired-client-timeout: and serve-expired-reply-ttl: is not
2157 consistent for data originating from the external cache as these will
2158 result in a reply with 0 TTL without trying to update the data first,
2159 ignoring the configured values.
2160
2161 If Unbound was built with --with-libhiredis on a system that has
2162 installed the hiredis C client library of Redis, then the "redis" back‐
2163 end can be used. This backend communicates with the specified Redis
2164 server over a TCP connection to store and retrieve cache data. It can
2165 be used as a persistent and/or shared cache backend. It should be
2166 noted that Unbound never removes data stored in the Redis server, even
2167 if some data have expired in terms of DNS TTL or the Redis server has
2168 cached too much data; if necessary the Redis server must be configured
2169 to limit the cache size, preferably with some kind of least-recently-
2170 used eviction policy. This backend uses synchronous communication with
2171 the Redis server based on the assumption that the communication is sta‐
2172 ble and sufficiently fast. The thread waiting for a response from the
2173 Redis server cannot handle other DNS queries. Although the backend has
2174 the ability to reconnect to the server when the connection is closed
2175 unexpectedly and there is a configurable timeout in case the server is
2176 overly slow or hangs up, these cases are assumed to be very rare. If
2177 connection close or timeout happens too often, Unbound will be effec‐
2178 tively unusable with this backend. It's the administrator's responsi‐
2179 bility to make the assumption hold.
2180
2181 The cachedb: clause gives custom settings of the cache DB module.
2182
2183 backend: <backend name>
2184 Specify the backend database name. The default database is the
2185 in-memory backend named "testframe", which, as the name sug‐
2186 gests, is not of any practical use. Depending on the build-time
2187 configuration, "redis" backend may also be used as described
2188 above.
2189
2190 secret-seed: <"secret string">
2191 Specify a seed to calculate a hash value from query information.
2192 This value will be used as the key of the corresponding answer
2193 for the backend database and can be customized if the hash
2194 should not be predictable operationally. If the backend data‐
2195 base is shared by multiple Unbound instances, all instances must
2196 use the same secret seed. This option defaults to "default".
2197
2198 The following cachedb otions are specific to the redis backend.
2199
2200 redis-server-host: <server address or name>
2201 The IP (either v6 or v4) address or domain name of the Redis
2202 server. In general an IP address should be specified as other‐
2203 wise Unbound will have to resolve the name of the server every
2204 time it establishes a connection to the server. This option
2205 defaults to "127.0.0.1".
2206
2207 redis-server-port: <port number>
2208 The TCP port number of the Redis server. This option defaults
2209 to 6379.
2210
2211 redis-timeout: <msec>
2212 The period until when Unbound waits for a response from the
2213 Redis sever. If this timeout expires Unbound closes the connec‐
2214 tion, treats it as if the Redis server does not have the
2215 requested data, and will try to re-establish a new connection
2216 later. This option defaults to 100 milliseconds.
2217
2218 Response Policy Zone Options
2219 Response Policy Zones are configured with rpz:, and each one must have
2220 a name:. There can be multiple ones, by listing multiple rpz clauses,
2221 each with a different name. RPZ clauses are applied in order of config‐
2222 uration. The respip module needs to be added to the module-config,
2223 e.g.: module-config: "respip validator iterator".
2224
2225 Only the QNAME and Response IP Address triggers are supported. The sup‐
2226 ported RPZ actions are: NXDOMAIN, NODATA, PASSTHRU, DROP and Local
2227 Data. RPZ QNAME triggers are applied after local-zones and before auth-
2228 zones.
2229
2230 name: <zone name>
2231 Name of the authority zone.
2232
2233 master: <IP address or host name>
2234 Where to download a copy of the zone from, with AXFR and IXFR.
2235 Multiple masters can be specified. They are all tried if one
2236 fails.
2237
2238 url: <url to zonefile>
2239 Where to download a zonefile for the zone. With http or https.
2240 An example for the url is "http://www.example.com/exam‐
2241 ple.org.zone". Multiple url statements can be given, they are
2242 tried in turn. If only urls are given the SOA refresh timer is
2243 used to wait for making new downloads. If also masters are
2244 listed, the masters are first probed with UDP SOA queries to see
2245 if the SOA serial number has changed, reducing the number of
2246 downloads. If none of the urls work, the masters are tried with
2247 IXFR and AXFR. For https, the tls-cert-bundle and the hostname
2248 from the url are used to authenticate the connection.
2249
2250 allow-notify: <IP address or host name or netblockIP/prefix>
2251 With allow-notify you can specify additional sources of noti‐
2252 fies. When notified, the server attempts to first probe and
2253 then zone transfer. If the notify is from a master, it first
2254 attempts that master. Otherwise other masters are attempted.
2255 If there are no masters, but only urls, the file is downloaded
2256 when notified. The masters from master: statements are allowed
2257 notify by default.
2258
2259 zonefile: <filename>
2260 The filename where the zone is stored. If not given then no
2261 zonefile is used. If the file does not exist or is empty,
2262 unbound will attempt to fetch zone data (eg. from the master
2263 servers).
2264
2265 rpz-action-override: <action>
2266 Always use this RPZ action for matching triggers from this zone.
2267 Possible action are: nxdomain, nodata, passthru, drop, disabled
2268 and cname.
2269
2270 rpz-cname-override: <domain>
2271 The CNAME target domain to use if the cname action is configured
2272 for rpz-action-override.
2273
2274 rpz-log: <yes or no>
2275 Log all applied RPZ actions for this RPZ zone. Default is no.
2276
2277 rpz-log-name: <name>
2278 Specify a string to be part of the log line, for easy referenc‐
2279 ing.
2280
2281 tags: <list of tags>
2282 Limit the policies from this RPZ clause to clients with a match‐
2283 ing tag. Tags need to be defined in define-tag and can be
2284 assigned to client addresses using access-control-tag. Enclose
2285 list of tags in quotes ("") and put spaces between tags. If no
2286 tags are specified the policies from this clause will be applied
2287 for all clients.
2288
2290 In the example config settings below memory usage is reduced. Some ser‐
2291 vice levels are lower, notable very large data and a high TCP load are
2292 no longer supported. Very large data and high TCP loads are exceptional
2293 for the DNS. DNSSEC validation is enabled, just add trust anchors. If
2294 you do not have to worry about programs using more than 3 Mb of memory,
2295 the below example is not for you. Use the defaults to receive full ser‐
2296 vice, which on BSD-32bit tops out at 30-40 Mb after heavy usage.
2297
2298 # example settings that reduce memory usage
2299 server:
2300 num-threads: 1
2301 outgoing-num-tcp: 1 # this limits TCP service, uses less buffers.
2302 incoming-num-tcp: 1
2303 outgoing-range: 60 # uses less memory, but less performance.
2304 msg-buffer-size: 8192 # note this limits service, 'no huge stuff'.
2305 msg-cache-size: 100k
2306 msg-cache-slabs: 1
2307 rrset-cache-size: 100k
2308 rrset-cache-slabs: 1
2309 infra-cache-numhosts: 200
2310 infra-cache-slabs: 1
2311 key-cache-size: 100k
2312 key-cache-slabs: 1
2313 neg-cache-size: 10k
2314 num-queries-per-thread: 30
2315 target-fetch-policy: "2 1 0 0 0 0"
2316 harden-large-queries: "yes"
2317 harden-short-bufsize: "yes"
2318
2320 /etc/unbound
2321 default unbound working directory.
2322
2323 /etc/unbound
2324 default chroot(2) location.
2325
2326 /etc/unbound/unbound.conf
2327 unbound configuration file.
2328
2329 /run/unbound/unbound.pid
2330 default unbound pidfile with process ID of the running daemon.
2331
2332 unbound.log
2333 unbound log file. default is to log to syslog(3).
2334
2336 unbound(8), unbound-checkconf(8).
2337
2339 Unbound was written by NLnet Labs. Please see CREDITS file in the dis‐
2340 tribution for further details.
2341
2342
2343
2344NLnet Labs Feb 20, 2020 unbound.conf(5)