When dhcpcd instead of networkd is used, the network-online.target behaved
the same as network.target, resulting in broken services that need a working
network connectivity when being started.
This commit makes dhcpcd wait for a lease and makes it wanted by
network-online.target. In turn, network-online.target is now wanted by
multi-user.target, so it will be activated at every boot.
This is deliberate because using the taskd binary to configure
Taskserver has a good chance of messing up permissions.
The nixos-taskserver tool now can manage even manual configurations, so
there really is no need anymore to expose the taskd binary.
If people still want to use the taskd binary at their own risk they can
still add taskserver to systemPackages themselves.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Putting an include directive in the configuration file referencing a
store path with the real configuration file has the disavantage that
once we change the real configuration file the store path is also a
different one.
So we would have to replace that include directive with the new
configuration file, which is very much error-prone, because whenever
taskd modifies the configuration file on its own it generates a new one
with *only* the key/value options and without any include directives.
Another problem is that we only added the include directive on the first
initalization, so whenever there is *any* configuration change, it won't
affect anything.
We're now passing all the configuration options via command line,
because taskd treats everything in the form of --<name>=<value> to be a
configuration directive.
This also has the effect that we now no longer have extraConfig, because
configuration isn't a file anymore.
Instead we now have an attribute set that is mapped down to
configuration options.
Unfortunately this isn't so easy with the way taskd is configured,
because there is an option called "server" and also other options like
"server.cert", "server.key" and so on, which do not map very well to
attribute sets.
So we have an exception for the "server" option, which is now called
"server.listen", because it specifies the listening address.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Fixes: #22705
The helper tool so far was only intended for use in automatic PKI
handling, but it also is very useful if you have an existing CA.
One of the main advantages is that you don't need to specify the data
directory anymore and the right permissions are also handled as well.
Another advantage is that we now have an uniform management tool for
both automatic and manual config, so the documentation in the NixOS
manual now applies to the manual PKI config as well.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
The error message displays that a specific user doesn't exist in an
organisation, but uses the User object's name attribute to show which
user it was.
This is basically a very stupid chicken and egg problem and easily fixed
by using the user name provided on the command line.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
The GeoIP databases from MaxMind have no stable URLs and change every
month (or so). Our current method of packaging these database in Nix and
playing catch-up with ever-changing file hashes is a bad idea. For
instance, it makes it impossible to realize old NixOS configurations.
This patch adds a NixOS service that periodically updates the GeoIP
databases in /var/lib/geoip-databases. Moving NixOS modules over can be
done in later patches.
I tried adding MD5 check, but not all databases have them, so i skipped
it. We are downloading over HTTPS though, it should be good. I also
tried adding zip support, but the first zip file I extracted had a
different filename inside than the archive name, which breaks an
assumption in this service, so I skipped that too.
Changes v9 -> v10:
- Pass "--max-time" to curl to set upper bound on downloads (ensures
no indefinite hanging if there's problem with networking).
Timeout for network connectivity check: 60s.
Timeout for geoip database (each): 15m.
Changes v8 -> v9:
- Mention the random timer delay in the documentation for the
'interval' option.
Changes v7 -> v8:
- Add "RemainAfterExit=true" for the setup service, so it won't be
restarted needlessly. (Thanks @danbst!)
Changes v6 -> v7:
- Add --skip-existing flag to geoip-updater, which skips updating
existing database files. Pass that flag when we run the service on
boot (and on any NixOS configuration change).
(IMHO, this is somewhat a workaround for systemd persistent timers
not being triggered immediately when a timer has never expired
before. But it does have the nice side effect of ensuring that the
installed databases always correspond to the configured ones, since
the service is now always run after configuration changes.)
Changes v5 -> v6:
- Update database files atomically (per DB)
- If a database is removed from the configuration, it'll be removed
from /var/lib/geoip-databases too (on next run).
- Add NixOS module assertion so that if user inputs non- .gz or .xz
file there will be a build time error instead of runtime.
- Run updater as user "nobody" instead of "root".
- Rename NixOS service from "geoip-databases" to "geoip-updater".
- Drop RemainAfterExit, or else the timer won't trigger the unit.
- Bring back "curl --fail", or else we won't catch and log curl
failures.
Changes v4 -> v5:
- Add "GeoLite2-City.mmdb.gz" to default database list.
Changes v3 -> v4:
- Remove unneeded geoip-updater-setup.service after adding
'wantedBy = [ "multi-user.target" ]' directly to
geoip-updater.service
- Drop unneeded "Service" name from service descriptions.
Changes v2 -> v3:
- Network may be down when starting from a cold boot, so try a few
times. Possibly, if using systemd-networkd, it'll pass on the first
try. But with default DHCP on NixOS, the service is started before
hostnames can be resolved and thus we need a few extra seconds.
- Add error handling and mark service as failed if fatal error.
- Add proper syslog log levels.
- Add RandomizedDelaySec=3600 to the timer to not put high load on the
MaxMind servers. Suggested by @Mic92.
- Set RemainAfterExit on geoip-updater.service instead of
geoip-updater-setup.service. (The latter is only a proxy that pulls
in the former service).
Changes v1 -> v2:
From Данило Глинський (Danylo Hlynskyi) <abcz2.uprola@gmail.com>:
nixos/geoip-databases: add `databases` option and fix initial setup
There were two great issues when using this service:
- When you just enable service, databases aren't downloaded, they are
downloaded when timer triggers. Fixed this with automatic download on
first system activation.
- When there is no internet, updater outputs nothing to logs, which is
IMO misbehavior. Fixed this with removing `--fail` option, better be
explicit here.
Recent versions of libreswan seem to omit this file, but it may be added/changed in the future. It is silly to have the service fail because a file is missing that only enriches the environment.
This fixes an issue where `nixops deploy` wouldn't restart the chrony
service when the chrony configuration changed, because it wouldn't
detect that `/etc/chrony.conf` was a dependency of the chrony service.
From Postfix documentation:
With this setting, the Postfix SMTP server will not reject mail with "User
unknown in local recipient table". Don't do this on systems that receive mail
directly from the Internet. With today's worms and viruses, Postfix will become
a backscatter source: it accepts mail for non-existent recipients and then
tries to return that mail as "undeliverable" to the often forged sender
address.
20e81f7c0d prevented key generation in
`preStart`, leaving the service broken for the case where the user has
no pre-existing key.
Eventually, we ought to store the state elsewhere so that `/etc` can be
read-only but for now we fix this the easy way.
Pushing to gogs only works if the `gogs` user's shell is `bash`. For error and
solution, refer to [this SO thread](http://stackoverflow.com/a/22315659)
DBus daemon now loads its config from /run/current-system/dbus.
Reloading the daemon makes it re-read that file and catch the updates
after a system upgrade.
Offline mode: When adding a lot of data, start this service.
It will will not flood the DHT since it only exposes the API.
When you are done simply reverse the process.
This allows overriding the `server_name` attribute of virtual
hosts. By doing so it is possible to have multiple virtualHost
definitions that share the same `server_name`. This is useful in
particular when you need a HTTP as well as a HTTPS virtualhost: same
server_name, different port.
A default of 0 means that if you deploy two NixOS boxes with the default
configuration, the second will fail because the brokerId was already in
use. Using -1 instead tells it to pick one automatically at first start.
This provides a default console_cmd for the slim display-manager.
When the user enters "console" as the user name, slim will run this
command.
Having a default is rather important; the virtual terminals don't work
with some display drivers, so having a broken X session can leave you
locked out of your machine.
Without it, the following error is shown in the "Add Printer" window:
Failed to group devices: 'The name org.fedoraproject.Config.Printing was not provided by any .service files'
* Fix the FW names
FW_REFUSE was removed and nixos-fw-input was renamed to nixos-fw.
* Update the comment (documentation) at the top
Order the chains of the main table alphabetically (like in the rest of
the file) and add nixos-fw-rpfilter (from the raw table) and nixos-drop
(used while reloading the firewall).
* Refactor the module (mainly comments)
- Move some attributes to the top for better visibility (that should
hopefully make it easier to read and understand this module without
jumping around too much).
- Add some missing examples and improve some descriptions.
- Reorder the mkOption attributes for consistency.
- Wrap lines at 72 characters.
- Use two spaces between sentences.
The reason being less mental overhead when reading upstream
documentation. Examples can be pasted right into the configuration
instead of translating to Nix attrset first.
Make sure that the output of the sieve compiler produces files that
have a newer time stamp than the source sieve script. Otherwise you
get errors in the logs about Dovecot not being able to compile do to a
permission issue.
Using outputsToInstall the intended behaviour of including host and dnsutils
when bind is installed can be implemented instead of using symlinks to fix
installing all outputs individually with nix-env.
Fixes#19761.
It hides bugs and do you ever actually want to serve up an empty directory?
It was pretty confusing to me when it tried to write into a read-only store
path because I accidentally pointed it to the wrong store path.
perlPackages.TextWrapI18N: init at 0.06
perlPackages.Po4a: init at 0.47
jade: init at 1.2.1
ding-libs: init at 0.6.0
Switch nscd to no-caching mode if SSSD is enabled.
abbradar: disable jade parallel building.
Closes#21150
* fix/asterisk-module: use unix-group for asterisk-files
* fix/asterisk-module: add configOption to use some default config-files
* fix/asterisk-module: correction of skel copy
* fix/asterisk-module: use /etc/asterisk as configDir
* fix/asterisk-module: add reload; do not restart unit
* asterisk: 13.6.0 -> 14.1.2
* fix/asterisk: compile with lua, pjsip, format_mp3
* fix/asterisk: fix indentation
* fix/asterisk: remove broken flag
The current default value of listenAddress = null blows up:
$ nixos-rebuild build
error: cannot coerce null to a string, at
.../nixpkgs/nixos/modules/services/monitoring/prometheus/alertmanager.nix:97:16
With listenAddress = "" we use the same default as upstream and there is
no blow up :-)
...by providing a default value of "no labels" (an empty attrset).
Without this change we get
$ nixos-rebuild test -I nixpkgs=.
building Nix...
building the system configuration...
error: The option `services.prometheus.scrapeConfigs.[definition 1-entry 1].static_configs.[definition 1-entry 1].labels' is used but not defined.
which is unneeded, because labels _are_ optional.
All the new options in detail:
Enable docker in multi-user.target make container created with restart=always
to start. We still want socket activation as it decouples dependencies between
the existing of /var/run/docker.sock and the docker daemon. This means that
services can rely on the availability of this socket. Fixes#11478#21303
wantedBy = ["multi-user.target"];
This allows us to remove the postStart hack, as docker reports on its own when
it is ready.
Type=notify
The following will set unset some limits because overhead in kernel's ressource
accounting was observed. Note that these limit only apply to containerd.
Containers will have their own limit set.
LimitNPROC=infinity
LimitCORE=infinity
TasksMax=infinity
Upgrades may require schema migrations. This can delay the startup of dockerd.
TimeoutStartSec=0
Allows docker to create its own cgroup subhierarchy to apply ressource limits on
containers.
Delegate=true
When dockerd is killed, container should be not affected to allow
`live restore` to work.
KillMode=process
currently services.nginx does not start up if `networking.enableIPv6 = false`
the commit changes the nginx behavior to handle this case accordingly.
The commit resolves#21308
The structured options are incomplete compared to upstream and I think
it will be a maintenance burden to try to keep up. Instead, provide an
option for the raw config file contents (prometheus.yml).
Otherwise it starts way too early, only to fail and having to restart
until devices are available. It is less wasteful to simply wait until
there's a reasonable chance of success. This is consistent with
upstream.
- most nixos user only require time synchronisation,
while ntpd implements a battery-included ntp server (1,215 LOCs of C-Code vs 64,302)
- timesyncd support ntp server per interface (if configured through dhcp for instance)
- timesyncd is already included in the systemd package, switching to it would
save a little disk space (1,5M)
The collectd service runs as an unprivileged user by default, so it does
not leak more information to its data directory than any user can obtain
elsewhere by other means.
If people are running it as root and are worried about information leak,
we can add collectd group and set perms to 750.
CC @offlinehacker.
Fixes#21198.
Split packages in three categories, all of them going into the system
package list:
- pre-requisite packages
- core packages
- optional packages
Add a new configuration option 'environment.lxqt.excludePackages' to
specify optional LXQt packages that should be excluded from system
packages.
Add 'gvfs' as a pre-requisite package, needed by 'pcmanfm-qt' to
handle virtual places, like "Computer" and "Network".
even if cups rewrites its config file due to config changes made through
its web-based management UI, we need to keep the PATH pointing to
currently-live nix store directories. fixes#20806.
