1SYSTEMD-RESOLVED.SERVICE(8)systemd-resolved.serviceSYSTEMD-RESOLVED.SERVICE(8)
2
3
4
6 systemd-resolved.service, systemd-resolved - Network Name Resolution
7 manager
8
10 systemd-resolved.service
11
12 /usr/lib/systemd/systemd-resolved
13
15 systemd-resolved is a system service that provides network name
16 resolution to local applications. It implements a caching and
17 validating DNS/DNSSEC stub resolver, as well as an LLMNR and
18 MulticastDNS resolver and responder. Local applications may submit
19 network name resolution requests via three interfaces:
20
21 • The native, fully-featured API systemd-resolved exposes on the bus,
22 see org.freedesktop.resolve1(5) and org.freedesktop.LogControl1(5)
23 for details. Usage of this API is generally recommended to clients
24 as it is asynchronous and fully featured (for example, properly
25 returns DNSSEC validation status and interface scope for addresses
26 as necessary for supporting link-local networking).
27
28 • The glibc getaddrinfo(3) API as defined by RFC3493[1] and its
29 related resolver functions, including gethostbyname(3). This API is
30 widely supported, including beyond the Linux platform. In its
31 current form it does not expose DNSSEC validation status
32 information however, and is synchronous only. This API is backed by
33 the glibc Name Service Switch (nss(5)). Usage of the glibc NSS
34 module nss-resolve(8) is required in order to allow glibc's NSS
35 resolver functions to resolve hostnames via systemd-resolved.
36
37 • Additionally, systemd-resolved provides a local DNS stub listener
38 on IP address 127.0.0.53 on the local loopback interface. Programs
39 issuing DNS requests directly, bypassing any local API may be
40 directed to this stub, in order to connect them to
41 systemd-resolved. Note however that it is strongly recommended that
42 local programs use the glibc NSS or bus APIs instead (as described
43 above), as various network resolution concepts (such as link-local
44 addressing, or LLMNR Unicode domains) cannot be mapped to the
45 unicast DNS protocol.
46
47 The DNS servers contacted are determined from the global settings in
48 /etc/systemd/resolved.conf, the per-link static settings in
49 /etc/systemd/network/*.network files (in case systemd-
50 networkd.service(8) is used), the per-link dynamic settings received
51 over DHCP, information provided via resolvectl(1), and any DNS server
52 information made available by other system services. See
53 resolved.conf(5) and systemd.network(5) for details about systemd's own
54 configuration files for DNS servers. To improve compatibility,
55 /etc/resolv.conf is read in order to discover configured system DNS
56 servers, but only if it is not a symlink to
57 /run/systemd/resolve/stub-resolv.conf, /usr/lib/systemd/resolv.conf or
58 /run/systemd/resolve/resolv.conf (see below).
59
61 systemd-resolved synthesizes DNS resource records (RRs) for the
62 following cases:
63
64 • The local, configured hostname is resolved to all locally
65 configured IP addresses ordered by their scope, or — if none are
66 configured — the IPv4 address 127.0.0.2 (which is on the local
67 loopback interface) and the IPv6 address ::1 (which is the local
68 host).
69
70 • The hostnames "localhost" and "localhost.localdomain" as well as
71 any hostname ending in ".localhost" or ".localhost.localdomain" are
72 resolved to the IP addresses 127.0.0.1 and ::1.
73
74 • The hostname "_gateway" is resolved to all current default routing
75 gateway addresses, ordered by their metric. This assigns a stable
76 hostname to the current gateway, useful for referencing it
77 independently of the current network configuration state.
78
79 • The mappings defined in /etc/hosts are resolved to their configured
80 addresses and back, but they will not affect lookups for
81 non-address types (like MX). Support for /etc/hosts may be disabled
82 with ReadEtcHosts=no, see resolved.conf(5).
83
85 The lookup requests that systemd-resolved.service receives are routed
86 to the available DNS servers, LLMNR, and MulticastDNS interfaces
87 according to the following rules:
88
89 • Names for which synthetic records are generated (the local
90 hostname, "localhost" and "localdomain", local gateway, as listed
91 in the previous section) and addresses configured in /etc/hosts are
92 never routed to the network and a reply is sent immediately.
93
94 • Single-label names are resolved using LLMNR on all local interfaces
95 where LLMNR is enabled. Lookups for IPv4 addresses are only sent
96 via LLMNR on IPv4, and lookups for IPv6 addresses are only sent via
97 LLMNR on IPv6. Note that lookups for single-label synthesized names
98 are not routed to LLMNR, MulticastDNS or unicast DNS.
99
100 • Queries for the address records (A and AAAA) of single-label
101 non-synthesized names are resolved via unicast DNS using search
102 domains. For any interface which defines search domains, such
103 look-ups are routed to that interface, suffixed with each of the
104 search domains defined on that interface in turn. When global
105 search domains are defined, such look-ups are routed to all
106 interfaces, suffixed by each of the global search domains in turn.
107 Additionally, lookup of single-label names via unicast DNS may be
108 enabled with the ResolveUnicastSingleLabel=yes setting. The details
109 of which servers are queried and how the final reply is chosen are
110 described below. Note that this means that address queries for
111 single-label names are never sent out to remote DNS servers by
112 default, and resoulution is only possible if search domains are
113 defined.
114
115 • Multi-label names with the domain suffix ".local" are resolved
116 using MulticastDNS on all local interfaces where MulticastDNS is
117 enabled. As with LLMNR, IPv4 address lookups are sent via IPv4 and
118 IPv6 address lookups are sent via IPv6.
119
120 • Queries for multi-label names are routed via unicast DNS on local
121 interfaces that have a DNS server configured, plus the globally
122 configured DNS servers if there are any. Which interfaces are used
123 is determined by the routing logic based on search and route-only
124 domains, described below. Note that by default, lookups for domains
125 with the ".local" suffix are not routed to DNS servers, unless the
126 domain is specified explicitly as routing or search domain for the
127 DNS server and interface. This means that on networks where the
128 ".local" domain is defined in a site-specific DNS server, explicit
129 search or routing domains need to be configured to make lookups
130 work within this DNS domain. Note that these days, it's generally
131 recommended to avoid defining ".local" in a DNS server, as
132 RFC6762[2] reserves this domain for exclusive MulticastDNS use.
133
134 • Address lookups (reverse lookups) are routed similarly to
135 multi-label names, with the exception that addresses from the
136 link-local address range are never routed to unicast DNS and are
137 only resolved using LLMNR and MulticastDNS (when enabled).
138
139 If lookups are routed to multiple interfaces, the first successful
140 response is returned (thus effectively merging the lookup zones on all
141 matching interfaces). If the lookup failed on all interfaces, the last
142 failing response is returned.
143
144 Routing of lookups is determined by the per-interface routing domains
145 (search and route-only) and global search domains. See
146 systemd.network(5) and resolvectl(1) for a description how those
147 settings are set dynamically and the discussion of Domains= in
148 resolved.conf(5) for a description of globally configured DNS settings.
149
150 The following query routing logic applies for unicast DNS lookups
151 initiated by systemd-resolved.service:
152
153 • If a name to look up matches (that is: is equal to or has as
154 suffix) any of the configured routing domains (search or
155 route-only) of any link, or the globally configured DNS settings,
156 "best matching" routing domain is determined: the matching one with
157 the most labels. The query is then sent to all DNS servers of any
158 links or the globally configured DNS servers associated with this
159 "best matching" routing domain. (Note that more than one link might
160 have this same "best matching" routing domain configured, in which
161 case the query is sent to all of them in parallel).
162
163 In case of single-label names, when search domains are defined, the
164 same logic applies, except that the name is first suffixed by each
165 of the search domains in turn. Note that this search logic doesn't
166 apply to any names with at least one dot. Also see the discussion
167 about compatibility with the traditional glibc resolver below.
168
169 • If a query does not match any configured routing domain (either
170 per-link or global), it is sent to all DNS servers that are
171 configured on links with the DefaultRoute= option set, as well as
172 the globally configured DNS server.
173
174 • If there is no link configured as DefaultRoute= and no global DNS
175 server configured, one of the compiled-in fallback DNS servers is
176 used.
177
178 • Otherwise the unicast DNS query fails, as no suitable DNS servers
179 can be determined.
180
181 The DefaultRoute= option is a boolean setting configurable with
182 resolvectl or in .network files. If not set, it is implicitly
183 determined based on the configured DNS domains for a link: if there's a
184 route-only domain other than "~.", it defaults to false, otherwise to
185 true.
186
187 Effectively this means: in order to support single-label
188 non-synthesized names, define appropriate search domains. In order to
189 preferably route all DNS queries not explicitly matched by routing
190 domain configuration to a specific link, configure a "~." route-only
191 domain on it. This will ensure that other links will not be considered
192 for these queries (unless they too carry such a routing domain). In
193 order to route all such DNS queries to a specific link only if no other
194 link is preferred, set the DefaultRoute= option for the link to true
195 and do not configure a "~." route-only domain on it. Finally, in order
196 to ensure that a specific link never receives any DNS traffic not
197 matching any of its configured routing domains, set the DefaultRoute=
198 option for it to false.
199
200 See org.freedesktop.resolve1(5) for information about the D-Bus APIs
201 systemd-resolved provides.
202
204 This section provides a short summary of differences in the stub
205 resolver implemented by nss-resolve(8) together with systemd-resolved
206 and the traditional stub resolver implemented in nss-dns.
207
208 • Some names are always resolved internally (see Synthetic Records
209 above). Traditionally they would be resolved by nss-files if
210 provided in /etc/hosts. But note that the details of how a query is
211 constructed are under the control of the client library. nss-dns
212 will first try to resolve names using search domains and even if
213 those queries are routed to systemd-resolved, it will send them out
214 over the network using the usual rules for multi-label name routing
215 [3].
216
217 • Single-label names are not resolved for A and AAAA records using
218 unicast DNS (unless overridden with ResolveUnicastSingleLabel=, see
219 resolved.conf(5)). This is similar to the no-tld-query option being
220 set in resolv.conf(5).
221
222 • Search domains are not used for suffixing of multi-label names.
223 (Search domains are nevertheless used for lookup routing, for names
224 that were originally specified as single-label or multi-label.) Any
225 name with at least one dot is always interpreted as a FQDN.
226 nss-dns would resolve names both as relative (using search domains)
227 and absolute FQDN names. Some names would be resolved as relative
228 first, and after that query has failed, as absolute, while other
229 names would be resolved in opposite order. The ndots option in
230 /etc/resolv.conf was used to control how many dots the name needs
231 to have to be resolved as relative first. This stub resolver does
232 not implement this at all: multi-label names are only resolved as
233 FQDNs.[4]
234
235 • This resolver has a notion of the special ".local" domain used for
236 MulticastDNS, and will not route queries with that suffix to
237 unicast DNS servers unless explicitly configured, see above. Also,
238 reverse lookups for link-local addresses are not sent to unicast
239 DNS servers.
240
241 • This resolver reads and caches /etc/hosts internally. (In other
242 words, nss-resolve replaces nss-files in addition to nss-dns).
243 Entries in /etc/hosts have highest priority.
244
245 • This resolver also implements LLMNR and MulticastDNS in addition to
246 the classic unicast DNS protocol, and will resolve single-label
247 names using LLMNR (when enabled) and names ending in ".local" using
248 MulticastDNS (when enabled).
249
250 • Environment variables $LOCALDOMAIN and $RES_OPTIONS described in
251 resolv.conf(5) are not supported currently.
252
254 Four modes of handling /etc/resolv.conf (see resolv.conf(5)) are
255 supported:
256
257 • systemd-resolved maintains the
258 /run/systemd/resolve/stub-resolv.conf file for compatibility with
259 traditional Linux programs. This file may be symlinked from
260 /etc/resolv.conf. This file lists the 127.0.0.53 DNS stub (see
261 above) as the only DNS server. It also contains a list of search
262 domains that are in use by systemd-resolved. The list of search
263 domains is always kept up-to-date. Note that
264 /run/systemd/resolve/stub-resolv.conf should not be used directly
265 by applications, but only through a symlink from /etc/resolv.conf.
266 This file may be symlinked from /etc/resolv.conf in order to
267 connect all local clients that bypass local DNS APIs to
268 systemd-resolved with correct search domains settings. This mode of
269 operation is recommended.
270
271 • A static file /usr/lib/systemd/resolv.conf is provided that lists
272 the 127.0.0.53 DNS stub (see above) as only DNS server. This file
273 may be symlinked from /etc/resolv.conf in order to connect all
274 local clients that bypass local DNS APIs to systemd-resolved. This
275 file does not contain any search domains.
276
277 • systemd-resolved maintains the /run/systemd/resolve/resolv.conf
278 file for compatibility with traditional Linux programs. This file
279 may be symlinked from /etc/resolv.conf and is always kept
280 up-to-date, containing information about all known DNS servers.
281 Note the file format's limitations: it does not know a concept of
282 per-interface DNS servers and hence only contains system-wide DNS
283 server definitions. Note that /run/systemd/resolve/resolv.conf
284 should not be used directly by applications, but only through a
285 symlink from /etc/resolv.conf. If this mode of operation is used
286 local clients that bypass any local DNS API will also bypass
287 systemd-resolved and will talk directly to the known DNS servers.
288
289 • Alternatively, /etc/resolv.conf may be managed by other packages,
290 in which case systemd-resolved will read it for DNS configuration
291 data. In this mode of operation systemd-resolved is consumer rather
292 than provider of this configuration file.
293
294 Note that the selected mode of operation for this file is detected
295 fully automatically, depending on whether /etc/resolv.conf is a symlink
296 to /run/systemd/resolve/resolv.conf or lists 127.0.0.53 as DNS server.
297
299 SIGUSR1
300 Upon reception of the SIGUSR1 process signal systemd-resolved will
301 dump the contents of all DNS resource record caches it maintains,
302 as well as all feature level information it learnt about configured
303 DNS servers into the system logs.
304
305 SIGUSR2
306 Upon reception of the SIGUSR2 process signal systemd-resolved will
307 flush all caches it maintains. Note that it should normally not be
308 necessary to request this explicitly – except for debugging
309 purposes – as systemd-resolved flushes the caches automatically
310 anyway any time the host's network configuration changes. Sending
311 this signal to systemd-resolved is equivalent to the resolvectl
312 flush-caches command, however the latter is recommended since it
313 operates in a synchronous way.
314
315 SIGRTMIN+1
316 Upon reception of the SIGRTMIN+1 process signal systemd-resolved
317 will forget everything it learnt about the configured DNS servers.
318 Specifically any information about server feature support is
319 flushed out, and the server feature probing logic is restarted on
320 the next request, starting with the most fully featured level. Note
321 that it should normally not be necessary to request this explicitly
322 – except for debugging purposes – as systemd-resolved automatically
323 forgets learnt information any time the DNS server configuration
324 changes. Sending this signal to systemd-resolved is equivalent to
325 the resolvectl reset-server-features command, however the latter is
326 recommended since it operates in a synchronous way.
327
329 systemd(1), resolved.conf(5), dnssec-trust-anchors.d(5), nss-
330 resolve(8), resolvectl(1), resolv.conf(5), hosts(5),
331 systemd.network(5), systemd-networkd.service(8)
332
334 1. RFC3493
335 https://tools.ietf.org/html/rfc3493
336
337 2. RFC6762
338 https://tools.ietf.org/html/rfc6762
339
340 3. For example, if /etc/nsswitch.conf has
341
342 nameserver 127.0.0.53
343 search foobar.com barbar.com
344
345
346 and we look up "localhost", nss-dns will send the following queries
347 to systemd-resolved listening on 127.0.0.53:53: first
348 "localhost.foobar.com", then "localhost.barbar.com", and finally
349 "localhost". If (hopefully) the first two queries fail, systemd-
350 resolved will synthesize an answer for the third query.
351
352 When using nss-dns with any search domains, it is thus crucial to
353 always configure nss-files with higher priority and provide
354 mappings for names that should not be resolved using search
355 domains.
356
357 4. There are currently more than 1500 top-level domain names defined,
358 and new ones are added regularly, often using "attractive" names
359 that are also likely to be used locally. Not looking up multi-label
360 names in this fashion avoids fragility in both directions: a valid
361 global name could be obscured by a local name, and resolution of a
362 relative local name could suddenly break when a new top-level
363 domain is created, or when a new subdomain of a top-level domain in
364 registered. Resolving any given name as either relative or absolute
365 avoids this ambiguity.)
366
367
368
369
370systemd 248 SYSTEMD-RESOLVED.SERVICE(8)