3
0
Fork 0
forked from mirrors/nixpkgs
nixpkgs/nixos/modules/services/networking/unbound.nix
pennae 2e751c0772 treewide: automatically md-convert option descriptions
the conversion procedure is simple:

 - find all things that look like options, ie calls to either `mkOption`
   or `lib.mkOption` that take an attrset. remember the attrset as the
   option
 - for all options, find a `description` attribute who's value is not a
   call to `mdDoc` or `lib.mdDoc`
 - textually convert the entire value of the attribute to MD with a few
   simple regexes (the set from mdize-module.sh)
 - if the change produced a change in the manual output, discard
 - if the change kept the manual unchanged, add some text to the
   description to make sure we've actually found an option. if the
   manual changes this time, keep the converted description

this procedure converts 80% of nixos options to markdown. around 2000
options remain to be inspected, but most of those fail the "does not
change the manual output check": currently the MD conversion process
does not faithfully convert docbook tags like <code> and <package>, so
any option using such tags will not be converted at all.
2022-07-30 15:16:34 +02:00

315 lines
10 KiB
Nix

{ config, lib, pkgs, ... }:
with lib;
let
cfg = config.services.unbound;
yesOrNo = v: if v then "yes" else "no";
toOption = indent: n: v: "${indent}${toString n}: ${v}";
toConf = indent: n: v:
if builtins.isFloat v then (toOption indent n (builtins.toJSON v))
else if isInt v then (toOption indent n (toString v))
else if isBool v then (toOption indent n (yesOrNo v))
else if isString v then (toOption indent n v)
else if isList v then (concatMapStringsSep "\n" (toConf indent n) v)
else if isAttrs v then (concatStringsSep "\n" (
["${indent}${n}:"] ++ (
mapAttrsToList (toConf "${indent} ") v
)
))
else throw (traceSeq v "services.unbound.settings: unexpected type");
confNoServer = concatStringsSep "\n" ((mapAttrsToList (toConf "") (builtins.removeAttrs cfg.settings [ "server" ])) ++ [""]);
confServer = concatStringsSep "\n" (mapAttrsToList (toConf " ") (builtins.removeAttrs cfg.settings.server [ "define-tag" ]));
confFile = pkgs.writeText "unbound.conf" ''
server:
${optionalString (cfg.settings.server.define-tag != "") (toOption " " "define-tag" cfg.settings.server.define-tag)}
${confServer}
${confNoServer}
'';
rootTrustAnchorFile = "${cfg.stateDir}/root.key";
in {
###### interface
options = {
services.unbound = {
enable = mkEnableOption "Unbound domain name server";
package = mkOption {
type = types.package;
default = pkgs.unbound-with-systemd;
defaultText = literalExpression "pkgs.unbound-with-systemd";
description = lib.mdDoc "The unbound package to use";
};
user = mkOption {
type = types.str;
default = "unbound";
description = lib.mdDoc "User account under which unbound runs.";
};
group = mkOption {
type = types.str;
default = "unbound";
description = lib.mdDoc "Group under which unbound runs.";
};
stateDir = mkOption {
type = types.path;
default = "/var/lib/unbound";
description = lib.mdDoc "Directory holding all state for unbound to run.";
};
resolveLocalQueries = mkOption {
type = types.bool;
default = true;
description = lib.mdDoc ''
Whether unbound should resolve local queries (i.e. add 127.0.0.1 to
/etc/resolv.conf).
'';
};
enableRootTrustAnchor = mkOption {
default = true;
type = types.bool;
description = lib.mdDoc "Use and update root trust anchor for DNSSEC validation.";
};
localControlSocketPath = mkOption {
default = null;
# FIXME: What is the proper type here so users can specify strings,
# paths and null?
# My guess would be `types.nullOr (types.either types.str types.path)`
# but I haven't verified yet.
type = types.nullOr types.str;
example = "/run/unbound/unbound.ctl";
description = lib.mdDoc ''
When not set to `null` this option defines the path
at which the unbound remote control socket should be created at. The
socket will be owned by the unbound user (`unbound`)
and group will be `nogroup`.
Users that should be permitted to access the socket must be in the
`config.services.unbound.group` group.
If this option is `null` remote control will not be
enabled. Unbounds default values apply.
'';
};
settings = mkOption {
default = {};
type = with types; submodule {
freeformType = let
validSettingsPrimitiveTypes = oneOf [ int str bool float ];
validSettingsTypes = oneOf [ validSettingsPrimitiveTypes (listOf validSettingsPrimitiveTypes) ];
settingsType = oneOf [ str (attrsOf validSettingsTypes) ];
in attrsOf (oneOf [ settingsType (listOf settingsType) ])
// { description = ''
unbound.conf configuration type. The format consist of an attribute
set of settings. Each settings can be either one value, a list of
values or an attribute set. The allowed values are integers,
strings, booleans or floats.
'';
};
options = {
remote-control.control-enable = mkOption {
type = bool;
default = false;
internal = true;
};
};
};
example = literalExpression ''
{
server = {
interface = [ "127.0.0.1" ];
};
forward-zone = [
{
name = ".";
forward-addr = "1.1.1.1@853#cloudflare-dns.com";
}
{
name = "example.org.";
forward-addr = [
"1.1.1.1@853#cloudflare-dns.com"
"1.0.0.1@853#cloudflare-dns.com"
];
}
];
remote-control.control-enable = true;
};
'';
description = ''
Declarative Unbound configuration
See the <citerefentry><refentrytitle>unbound.conf</refentrytitle>
<manvolnum>5</manvolnum></citerefentry> manpage for a list of
available options.
'';
};
};
};
###### implementation
config = mkIf cfg.enable {
services.unbound.settings = {
server = {
directory = mkDefault cfg.stateDir;
username = cfg.user;
chroot = ''""'';
pidfile = ''""'';
# when running under systemd there is no need to daemonize
do-daemonize = false;
interface = mkDefault ([ "127.0.0.1" ] ++ (optional config.networking.enableIPv6 "::1"));
access-control = mkDefault ([ "127.0.0.0/8 allow" ] ++ (optional config.networking.enableIPv6 "::1/128 allow"));
auto-trust-anchor-file = mkIf cfg.enableRootTrustAnchor rootTrustAnchorFile;
tls-cert-bundle = mkDefault "/etc/ssl/certs/ca-certificates.crt";
# prevent race conditions on system startup when interfaces are not yet
# configured
ip-freebind = mkDefault true;
define-tag = mkDefault "";
};
remote-control = {
control-enable = mkDefault false;
control-interface = mkDefault ([ "127.0.0.1" ] ++ (optional config.networking.enableIPv6 "::1"));
server-key-file = mkDefault "${cfg.stateDir}/unbound_server.key";
server-cert-file = mkDefault "${cfg.stateDir}/unbound_server.pem";
control-key-file = mkDefault "${cfg.stateDir}/unbound_control.key";
control-cert-file = mkDefault "${cfg.stateDir}/unbound_control.pem";
} // optionalAttrs (cfg.localControlSocketPath != null) {
control-enable = true;
control-interface = cfg.localControlSocketPath;
};
};
environment.systemPackages = [ cfg.package ];
users.users = mkIf (cfg.user == "unbound") {
unbound = {
description = "unbound daemon user";
isSystemUser = true;
group = cfg.group;
};
};
users.groups = mkIf (cfg.group == "unbound") {
unbound = {};
};
networking = mkIf cfg.resolveLocalQueries {
resolvconf = {
useLocalResolver = mkDefault true;
};
networkmanager.dns = "unbound";
};
environment.etc."unbound/unbound.conf".source = confFile;
systemd.services.unbound = {
description = "Unbound recursive Domain Name Server";
after = [ "network.target" ];
before = [ "nss-lookup.target" ];
wantedBy = [ "multi-user.target" "nss-lookup.target" ];
path = mkIf cfg.settings.remote-control.control-enable [ pkgs.openssl ];
preStart = ''
${optionalString cfg.enableRootTrustAnchor ''
${cfg.package}/bin/unbound-anchor -a ${rootTrustAnchorFile} || echo "Root anchor updated!"
''}
${optionalString cfg.settings.remote-control.control-enable ''
${cfg.package}/bin/unbound-control-setup -d ${cfg.stateDir}
''}
'';
restartTriggers = [
confFile
];
serviceConfig = {
ExecStart = "${cfg.package}/bin/unbound -p -d -c /etc/unbound/unbound.conf";
ExecReload = "+/run/current-system/sw/bin/kill -HUP $MAINPID";
NotifyAccess = "main";
Type = "notify";
# FIXME: Which of these do we actualy need, can we drop the chroot flag?
AmbientCapabilities = [
"CAP_NET_BIND_SERVICE"
"CAP_NET_RAW"
"CAP_SETGID"
"CAP_SETUID"
"CAP_SYS_CHROOT"
"CAP_SYS_RESOURCE"
];
User = cfg.user;
Group = cfg.group;
MemoryDenyWriteExecute = true;
NoNewPrivileges = true;
PrivateDevices = true;
PrivateTmp = true;
ProtectHome = true;
ProtectControlGroups = true;
ProtectKernelModules = true;
ProtectSystem = "strict";
RuntimeDirectory = "unbound";
ConfigurationDirectory = "unbound";
StateDirectory = "unbound";
RestrictAddressFamilies = [ "AF_INET" "AF_INET6" "AF_NETLINK" "AF_UNIX" ];
RestrictRealtime = true;
SystemCallArchitectures = "native";
SystemCallFilter = [
"~@clock"
"@cpu-emulation"
"@debug"
"@keyring"
"@module"
"mount"
"@obsolete"
"@resources"
];
RestrictNamespaces = true;
LockPersonality = true;
RestrictSUIDSGID = true;
Restart = "on-failure";
RestartSec = "5s";
};
};
};
imports = [
(mkRenamedOptionModule [ "services" "unbound" "interfaces" ] [ "services" "unbound" "settings" "server" "interface" ])
(mkChangedOptionModule [ "services" "unbound" "allowedAccess" ] [ "services" "unbound" "settings" "server" "access-control" ] (
config: map (value: "${value} allow") (getAttrFromPath [ "services" "unbound" "allowedAccess" ] config)
))
(mkRemovedOptionModule [ "services" "unbound" "forwardAddresses" ] ''
Add a new setting:
services.unbound.settings.forward-zone = [{
name = ".";
forward-addr = [ # Your current services.unbound.forwardAddresses ];
}];
If any of those addresses are local addresses (127.0.0.1 or ::1), you must
also set services.unbound.settings.server.do-not-query-localhost to false.
'')
(mkRemovedOptionModule [ "services" "unbound" "extraConfig" ] ''
You can use services.unbound.settings to add any configuration you want.
'')
];
}