Although double '/' in paths is not a problem for GRUB supplied with nixpkgs, sometimes NixOS's grub.conf read by external GRUB and there are versions of GRUB which fail
The switch from slim to lightdm in #30890 broke some nixos tests
because lightdm by default doesn't permit auto-login for root.
Override /etc/pam.d/lightdm-autologin to allow it.
The option was added in 1251b34b5b
with type `types.path` but default `null`, so eval failed with
the default setting. This broke the acme and certmgr tests.
cc: @vincentbernat @fpletz
This is the semantics as understood by `xdg-open`. Using these semantics
on a non-colon-separated variable works because it acts as if it was a
one element long list.
This fixes an issue where it would try to exec
`google-chrome-beta:google-chrome:chromium:firefox` on a system
configured with these semantics in mind.
The instructions to install nixos behind a proxy were not clear. While
one could guess that setting http_proxy variables can get the install
rolling, one could end up with an installed system where the proxy
settings for the nix-daemon are not configured.
This commit updates the documentation with
1. steps to install behind a proxy
2. configure the global proxy settings so that nix-daemon can access
internet.
3. Pointers to use nesting.clone in case one has to use different proxy
settings on different networks.
Switch from slim to lightdm as the display-manager.
If plasma5 is used as desktop-manager use sdddm.
If gnome3 is used as desktop-manager use gdm.
Based on #12516
The recommended TLS configuration comes with `ssl_stapling on` and
`ssl_stapling_verify on`. However, this last directive also requires
the use of `ssl_trusted_certificate` to verify the received answer.
When using `enableACME` or similar, we can help the user by providing
the correct value for the directive.
The result can be tested with:
openssl s_client -connect web.example.com:443 -status 2> /dev/null
Without OCSP stapling, we get:
OCSP response: no response sent
After this change, we get:
OCSP Response Data:
OCSP Response Status: successful (0x0)
Response Type: Basic OCSP Response
Version: 1 (0x0)
Responder Id: C = US, O = Let's Encrypt, CN = Let's Encrypt Authority X3
Produced At: Aug 30 20:46:00 2018 GMT
A shared exported guard `__NIXOS_SET_ENVIRONMENT_DONE` is introduced that can
be used to prevent child shells from sourcing `system.build.setEnvironment`
the second time.
This fixes e.g. `nix run derivation` when run from e.g. ZSH through the console or
ssh. Before this Bash would resource the common environment resetting the `PATH`
environment variable.
We also export `system.build.setEnvironment` to `/etc/set-environment` making it
easy to reset the common environment with `. /etc/set-environment` when
needed and to grep for environment variables in `/etc` (which was the
motivation of #30418).
This reverts changes made in b00a3fc6fd
(the original #30418).
This allows one to add rules which change a packet's routing table:
iptables -t raw -I PREROUTING 1 -m set --match-set myset src -j MARK --set-mark 2
ip rule add fwmark 2 table 1 priority 1000
ip route add default dev wg0 table 1
to the beginning of raw table PREROUTING chain, and still have rpfilter.
DefaultTimeoutStartSec is normally set to 90 seconds and works fine. But
when running NixOS tests on a very slow machine (like a VM without
nested virtualisation support) this default is to low and causes
systemd units to fail spuriously. One symptom of this issue are tests
at times failing with "timed out waiting for the VM to connect".
Since the VM connect timeout is 300 seconds I also set
DefaultTimeoutStartSec to this which is ridiculously high.
The background color option is self-explanatory.
The mode is either `normal` or `stretch`, they are as defined by GRUB,
where normal will put the image in the top-left corner of the menu, and
stretch is the default, where it stretches the image without
consideration for the aspect ratio.
* https://www.gnu.org/software/grub/manual/grub/grub.html#background_005fimage
The wallpaper used is *structurally compatible* with the other one,
meaning that the logo is at the same location, and not bigger.
It has one drawback: the logo is brighter, which clashes with the grub
usage. This is to be fixed with new options in grub.
- default coreutils is stripped of /share/ (11 -> 2 MiB)
- coreutils-full retains /share/ and adds openssl for faster *sum tools
- NixOS systemPackages contains coreutils-full
- *Support parameter defaults are moved inside
(it seemed confusing to have `? false` and "at once" with `? isLinux`)
Closure considerations:
+ typical build-time closure will get lighter by ~9 MiB
- typical closure of NixOS installation will grow by ~2 MiB,
due to referring to both versions. I think it would be possible to
re-use most of the utils between the two versions, but the expression
would get much more complex.
I considered having stdenv with minimal coreutils and the default
`coreutils` attribute being full, but it turned out there were too many
trivial references in nixpkgs, so it didn't seem easy to keep rebuild
impact of openssl from growing significantly.
The additions are:
- image/svg+xml for SVG images
- application/atom+xml for Atom feeds
These types are also present in mime.types. For better readability,
the list is sorted and formatted with one type per line.
This prevents issues when gitea adds new locales etc. And if they
change locale values in future versions. Or if you rollback to a
previous version of gitea it might be a good idea to use the previous
locale files.
This is a 277K (as of right now) addition that can greatly help in some
last recourse scenarios. The specific rEFInd setup will not be able to
boot the installer image, but this is not why it has been added. It has
been added to make use of its volumes scanning capabilities to boot
existing EFI images on the target computer, which is sometimes necessary
with buggy EFI. While is isn't NixOS's job to fix buggy EFI, shipping
this small bit with the installer will help the unlucky few.
Example scenario: two wildly different EFI implementation I have
encountered have fatal flaws in which they sometimes will lose all the
settings, this includes boot configuration. This is compounded by the
fact that the two specific and distinct implementation do not allow
manually adding ESP paths from their interface. The only recourse is to
let the EFI boot the default paths, EFI/boot/boot{platform}.efi, which
is not a default location used by the NixOS bootloaders. rEFInd is able
to scan the volumes and detect the existing efi bootloaders, and boot
them successfully.
Following up https://github.com/NixOS/nixpkgs/pull/23665
Bootable USB-drives are not limited to ISO-images, there can be "normal" MBR/GPT-partitioned disk connected via USB-rack.
Also, "uas" implies "usb-storage", so there is no need to mention both.
thermald has two modes: zero-config and manual. Sometimes it is useful
to manually configure thermald to achieve better thermal results or to give
thermald a hand when detecting possible cooling options.
I broke it:
in job ‘nixos.iso_graphical.x86_64-linux’:
The option `services.udisks2.enable' has conflicting definitions, in `/nix/store/bwcjw1ddj94q83vbbnq1nnrs5aisaw59-source/nixos/modules/profiles/installation-device.nix' and `/nix/store/bwcjw1ddj94q83vbbnq1nnrs5aisaw59-source/nixos/modules/services/x11/desktop-managers/plasma5.nix'.
And don't need to source the uevent files anymore either since $MAJOR
or $MINOR aren't used elsewhere.
[dezgeg: The reason these are no longer needed is that 0d27df280f
switched /tmp to a devtmpfs which automatically creates such device
nodes]
When rebuilding you have to manually run `systemctl --user
daemon-reload`. It gathers all authenticated users using
`loginctl list-user` and runs `daemon-reload` for each of them.
This is a first step towards a `nixos-rebuild` which is able to reload
user units from systemd. The entire task is fairly hard, however I
consider this patch usable as it allows to restart units without running
`daemon-reload` for each authenticated user.
This fixes an issue where setting both
`boot.loader.systemd-boot.editor` to `false` and
`boot.loader.systemd-boot.consoleMode` to any value would concatenate
the two configuration lines in the output, resulting in an invalid
`loader.conf`.
This allows the user to add `wpa_supplicant` config options not yet supported by Nix without having to write the entire `wpa_supplicant.conf` file manually.
Introduces an option `services.datadog-agent.extraIntegrations` that
can be set to include additional Datadog agent integrations from the
integrations-core repository.
Documentation and an example is provided with the change.
Relates to NixOS/nixpkgs#40399
Refactors the datadog-agent (i.e. V6) module to let users configure
arbitrary checks, not just a limited set, without having to resort to
linking the files manually and updating the systemd unit.
Checks are now configured via a `services.datadog-agent.checks` option
which takes an attribute set in which the keys refer directly to
Datadog check names, and the values are attribute sets representing
Datadog's configuration structure.
With this mechanism users can configure arbitrary integrations, for
example for the `ntp`-check, simply by saying:
services.datadog-agent.checks.ntp = {
init_config = null;
# ... other check configuration options as per Datadog
# documentation
};
The previous check-specific configuration options for non-default
checks have been removed. Disk & network check configuration options
have been kept rather than making them a `default`-value of the
`checks`-option because they will be overridden by user-configurations
in that case.
Relates to NixOS/nixpkgs#40399.
From reading the source I'm pretty sure it doesn't support multiple Yubikeys, hence
those options are useless.
Also, I'm pretty sure nobody actually uses this feature, because enabling it causes
extra utils' checks to fail (even before applying any patches of this branch).
As I don't have the hardware to test this, I'm too lazy to fix the utils, but
I did test that with extra utils checks commented out and Yubikey
enabled the resulting script still passes the syntax check.
Also reuse common cryptsetup invocation subexpressions.
- Passphrase reading is done via the shell now, not by cryptsetup.
This way the same passphrase can be reused between cryptsetup
invocations, which this module now tries to do by default (can be
disabled).
- Number of retries is now infinity, it makes no sense to make users
reboot when they fail to type in their passphrase.
Some modules of cloud-init can cope with a network not immediately
available (notably, the EC2 module), but some others won't retry if
network is not available (notably, the Cloudstack module).
network.target doesn't give much guarantee about the network
availability. Applications not able to start without a fully
configured network should be ordered after network-online.target.
Also see #44573 and #44524.
We override the ESP mount point in the config file /etc/fwupd/uefi.conf
(available since version 1.0.6), as it is set to a path in the nix store
during build time.
Tests are disabled as it needs /etc/os-release, which is not available
when building with sandboxing enabled.
In the last year `programs.oh-my-zsh` gained more complexity and since
the introduction of features like `customPkgs` which builds a
`ZSH_CUSTOM` path from a sequence of derivation a documentation may be
fairly helpful to make the knowledge how to use the module and how to
package new ZSH plugins visible.
See https://github.com/NixOS/nixpkgs/pull/43282#issuecomment-410770432
First change is to override the nm-dispatcher systemd service so that
it puts coreutils (wc/env/...) and iproute in PATH.
Second change is to make sure userscripts have the execute bit.
This allows a developer to better identify in which snippet the
failure happened. Furthermore, users seeking help will have more
information available about the failure.
This reverts a change applied in PR #18491. When interfaces are
configured by DHCP (typical in a cloud environment), ordering after
network.target cause trouble to applications expecting some network to
be present on boot (for example, cloud-init is quite brittle when
network hasn't been configured for `cloud-init.service`) and on
shutdown (for example, collectd needs to flush metrics on shutdown).
When ordering after network.target, we ensure applications relying on
network.target won't have any network reachability on boot and
potentially on shutdown.
Therefore, I think ordering before network.target is better.
If multiple third-party modules shall be used for `oh-my-zsh` it has to
be possible to create another env which composes all the packages.
Now it can be done like this:
```
{ pkgs, ... }:
{
programs.zsh.enable = true;
programs.zsh.ohMyZsh = {
enable = true;
customPkgs = with pkgs; [
lambda-mod-zsh-theme
nix-zsh-completions
];
theme = "lambda-mod";
plugins = [ "nix" ];
};
}
```
Please keep in mind that this is not compatible with
`programs.zsh.ohMyZsh.custom`, only one of these options can be used
ATM.
Each package should store its outputs into
`$out/share/zsh/<output-name>`. Completions (and ZSH-only) extensions
should live in the `fpath` (`$out/share/zsh/site-functions`), plugins in
`.../plugins` and themes in `.../themes` (please refer to
fdb6bf6ed68c2f089ae6c729dfeaa3eddea2ce6a and 406d64aad162b3a4881747be4e24705fb5182573).
All scripts in `customPkgs` will be linked together using `linkFarm` to
provide a single directory for all scripts from all derivations in
`customPkgs` as suggested in https://github.com/NixOS/nixpkgs/pull/43282#issuecomment-410396365.
The web_access.patch would no longer apply.
It disabled a check that required the static files
for the web UI to be owned by the user the daemon runs as
(not root, so it doesn't work well with nix).
Besides updating netdata, this commit removes that patch,
changes the netdata service config to set the "web files owner/group"
option to "root" and adds a test that checks that the web UI is being served.
This allows the web files to be owned by root without patching.
Broke evaluation of the nixos options.
The option `services.dysnomia' defined in `.../nixos/modules/rename.nix' does not exist.
This reverts commit 5c897b4eff.
Use nixos-fw chain instead of INPUT so that the rules don't keep
stacking everytime the firewall is reloaded.
This also adds a comment to each rule about the associated exporter.
- based on module originally written by @srhb
- complies with available options in cfssl v1.3.2
- uid and gid 299 reserved in ids.nix
- added simple nixos test case
Fixes#30891
* Upgrade `graphite-web`, `carbon` and `whisper` from 1.0.2 -> 1.1.3.
* Replaced the deprecated `pythonPackages.graphite_influxdb` with
`pythonPackages.influxgraph.`
* Renamed `pythonPackages.graphite_web` to `pythonPackages.graphite-web`
to be consistent with the Python package name.
* Replaced the unmaintained `pythonPackages.graphite_pager` with
`pythonPackages.graphitepager`
* Moved all new packages from `python-packages.nix` to
`pkgs/development/python-modules`
when the parent interface of a vlan interface is not up (yet), ip link cannot bring the vlan interface up
the vlan interface will be automatically brought up when the parent interface is up later
fixNixOS/nixpkgs#28620
`ocserv` is a VPN server which follows the openconnect protocol
(https://github.com/openconnect/protocol). The packaging is slightly
inspired by the AUR version
(https://aur.archlinux.org/packages/ocserv/).
This patch initializes the package written in C, the man pages and a
module for a simple systemd unit to run the VPN server. The package
supports the following authentication methods for the server:
* `plain` (mostly username/password)
* `pam`
The third method (`radius`) is currently not supported since `nixpkgs`
misses a packaged client.
The module can be used like this:
``` nix
{
services.ocserv = {
enable = true;
config = ''
...
'';
};
}
```
The option `services.ocserv.config` is required on purpose to
ensure that nobody just enables the service and experiences unexpected
side-effects on the system. For a full reference, please refer to the
man pages, the online docs or the example value.
The docs recommend to simply use `nobody` as user, so no extra user has
been added to the internal user list. Instead a configuration like
this can be used:
```
run-as-user = nobody
run-as-group = nogroup
```
/cc @tenten8401
Fixes#42594
The default session might be found in `extraSessionFilePackages`, but it's not
viable to detect at evaluation time, so emit a warning.
In LightDM instead of checking `defaultSessionName` against
`displayManager.session.names` we rely on the assertions in
`desktopManager` and `windowMananger` and just check that there's at least one
default set. The second assertion could never actually be triggered.
This makes it easier to support a wider variety of .desktop session files. In
particular this makes it possible to use both the «legacy» sessions and upstream
session files.
We separate `xsession` into two parts, `xsessionWrapper` and `xsession`.
`xsessionWrapper` sets up the correct environment and then lauches the session's
Exec command (from the .desktop file), falling back to launching the default
window/desktopManager through the `xsession` script (required by at least some
nixos tests).
`xsession` then _only_ handles launching desktop-managers/window-managers defined
through `services.xserver.desktopManager.session`.
Pass gnome-session to extraSessionFilePackages, remove unnecessary environment variables, move the rest out of old session option, and then drop the option.
GPaste GNOME Shell extension uses GPaste library generated via introspection. Previously, we added the gpaste package to services.xserver.desktopManager.gnome3.sessionPath option, which
added its typelib directory to GI_TYPELIB_PATH environment variable globally, in order for GNOME Shell to be able to find it. This is not very Nix-y, though, so we have decided to patch the code to
append the path to the GI repository search path.
Additionally, the code relies on GPaste’s GSettings schemas, so we had to hard-code the paths to them as well. We ignored the GNOME Shell’s schemas, since they will already be available for the
extension inside GNOME Shell program.
Previously, the mkDesktops function produced a flat package containing
session files in the top level. As a preparation for introduction of
Wayland sessions, the files will now be placed to $out/share/xsessions.
It seems like Gitlab doesn't pick up GITLAB_UPLOADS_PATH. The internal uploads
folder is already symlinked to /run/gitlab/uploads by the gitlab package. Here
we symlink this further to ${statePath}/uploads, since /run is (usually) a tmpfs.
* The ELK stack is upgraded to 6.3.2.
* `elasticsearch6`, `logstash6` and `kibana6` now come with X-Pack which is
a suite of additional features. These are however licensed under the unfree
"Elastic License".
* Fortunately they also provide OSS versions which are now packaged
under: `elasticsearch6-oss`, `logstash6-oss` and `kibana6-oss`.
Note that the naming of the attributes is consistent with upstream.
* The test `nix-build nixos/tests/elk.nix -A ELK-6` will test the OSS
version by default. You can also run the test on the unfree ELK using:
`NIXPKGS_ALLOW_UNFREE=1 nix-build nixos/tests/elk.nix -A ELK-6 --arg enableUnfree true`
This reverts commit 095fe5b43d.
Pointless renames considered harmful. All they do is force people to
spend extra work updating their configs for no benefit, and hindering
the ability to switch between unstable and stable versions of NixOS.
Like, what was the value of having the "nixos." there? I mean, by
definition anything in a NixOS module has something to do with NixOS...
Having socket-activated epmd means that there always be only a single
instance managed centrally. Because Erlang also starts it
automatically if not available, and in worst case scenario 'epmd' can
be started by some Erlang application running under systemd. And then
restarting this application unit will cause complete loss of names in
'epmd' (if other Erlang system are also installed on this host).
E.g. see at which lengths RabbitMQ goes to recover from such
situations:
7741b37b1e/src/rabbit_epmd_monitor.erl (L36)
Having the only one socket-activated epmd completely solves this
problem.
This makes the command ‘nix-env -qa -f. --arg config '{skipAliases =
true;}'’ work in Nixpkgs.
Misc...
- qtikz: use libsForQt5.callPackage
This ensures we get the right poppler.
- rewrites:
docbook5_xsl -> docbook_xsl_ns
docbook_xml_xslt -> docbook_xsl
diffpdf: fixup
Problem: Restarting (stopping) system.slice would not only stop X11 but
also most system units/services. We obviously don't want this happening
to users when they switch from 18.03 to 18.09 or nixos-unstable.
Reason: The following change in systemd:
d8e5a93382
The commit adds system.slice to the perpetual units, which means
removing the unit file and adding it to the source code. This is done so
that system.slice can't be stopped anymore but in our case it ironically
would cause this script to stop system.slice because the unit file was
removed (and an older systemd version is still running).
Related issue: https://github.com/NixOS/nixpkgs/issues/39791
When running e.g. `nixos-option boot.kernelPackages` I get an output
like this on the current unstable channel (18.09pre144959.be1461fc0ab):
```
$ nixos-option boot.kernelPackages
Value:
*exit 1*
```
This is fairly counter-intuitive as I have no clue what might went
wrong. `strace` delivers an output like this:
```
read(3, "error: Package \342\200\230cryptodev-linu"..., 128) = 128
read(3, "ux/cryptodev/default.nix:22 is m"..., 128) = 128
read(3, "lowBroken = true; }\nin configura"..., 128) = 128
read(3, "you can add\n { allowBroken = tr"..., 128) = 128
read(3, "n)\n", 128) = 3
read(3, "", 128) = 0
```
`nixos-option` evaluates the system config using `nix-instantiate` which
might break when the evaluation fails (e.g. due to broken or unfree
packages that are prohibited to evaluate by default). The script aborts
due to the shebang `@shell@ -e`.
In order to ensure that no unexpected
behavior occurs due to removing `-e` from the interpreter the easiest
way to work around this was to wrap `nix-instantiate` in `evalNix()`
with a `set +e`. The function checks the success of the evaluation with
`$?` in the end. Additionally `evalNix` shouldn't break, if one
evaluation (e.g. the values that contain a package set by default) to
return additional information like a description.
With the change `nixos-option boot.kernelPackages` delivers the
following output for me:
```
Value:
error: Package ‘cryptodev-linux-1.9-4.14.52’ in /nix/store/47z2s8cwppymmgzw6n7pbcashikyk5jk-nixos/nixos/pkgs/os-specific/linux/cryptodev/default.nix:22 is marked as broken, refusing to evaluate.
Default:
{ __unfix__ = <LAMBDA>; acpi_call = <CODE>; amdgpu-pro = <CODE>; ati_drivers_x11 = <CODE>; batman_adv = <CODE>; bbswitch = <CODE>; bcc = <CODE>; beegfs-module = <CODE>; blcr = <CODE>; broadcom_sta = <CODE>; callPackage = <CODE>; cpupower = <CODE>; cryptodev = <CODE>; dpdk = <CODE>; e1000e = <CODE>; ena = <CODE>; evdi = <CODE>; exfat-nofuse = <CODE>; extend = <CODE>; facetimehd = <CODE>; fusionio-vsl = <CODE>; hyperv-daemons = <CODE>; ixgbevf = <CODE>; jool = <CODE>; kernel = <CODE>; lttng-modules = <CODE>; mba6x_bl = <CODE>; mwprocapture = <CODE>; mxu11x0 = <CODE>; ndiswrapper = <CODE>; netatop = <CODE>; nvidiaPackages = <CODE>; nvidia_x11 = <CODE>; nvidia_x11_beta = <CODE>; nvidia_x11_legacy304 = <CODE>; nvidia_x11_legacy340 = <CODE>; nvidiabl = <CODE>; odp-dpdk = <CODE>; openafs = <CODE>; openafs_1_8 = <CODE>; perf = <CODE>; phc-intel = <CODE>; pktgen = <CODE>; ply = <CODE>; prl-tools = <CODE>; recurseForDerivations = true; rtl8192eu = <CODE>; rtl8723bs = <CODE>; rtl8812au = <CODE>; rtl8814au = <CODE>; rtlwifi_new = <CODE>; sch_cake = <CODE>; spl = <CODE>; splLegacyCrypto = <CODE>; splStable = <CODE>; splUnstable = <CODE>; stdenv = <CODE>; sysdig = <CODE>; systemtap = <CODE>; tbs = <CODE>; tmon = <CODE>; tp_smapi = <CODE>; usbip = <CODE>; v4l2loopback = <CODE>; v86d = <CODE>; vhba = <CODE>; virtualbox = <CODE>; virtualboxGuestAdditions = <CODE>; wireguard = <CODE>; x86_energy_perf_policy = <CODE>; zfs = <CODE>; zfsLegacyCrypto = <CODE>; zfsStable = <CODE>; zfsUnstable = <CODE>; }
Example:
{ _type = "literalExample"; text = "pkgs.linuxPackages_2_6_25"; }
Description:
"This option allows you to override the Linux kernel used by\nNixOS. Since things like external kernel module packages are\ntied to the kernel you're using, it also overrides those.\nThis option is a function that takes Nixpkgs as an argument\n(as a convenience), and returns an attribute set containing at\nthe very least an attribute <varname>kernel</varname>.\nAdditional attributes may be needed depending on your\nconfiguration. For instance, if you use the NVIDIA X driver,\nthen it also needs to contain an attribute\n<varname>nvidia_x11</varname>.\n"
Declared by:
"/nix/var/nix/profiles/per-user/root/channels/nixos/nixpkgs/nixos/modules/system/boot/kernel.nix"
Defined by:
"/home/ma27/Projects/nixos-config/system/boot.nix"
```
* nixos/virtualbox: Adds more options to virtualbox-image.nix
Previously you could only set the size of the disk.
This change adds the ability to change the amount of memory
that the image gets, along with the name / derivation name /
file name for the VM.
* Incorporates some review feedback
This adds configuration options which automate the configuration of NVIDIA Optimus using PRIME. This allows using the NVIDIA proprietary driver on Optimus laptops, in order to render using the NVIDIA GPU while outputting to displays connected only to the integrated Intel GPU. It also adds an option for enabling kernel modesetting for the NVIDIA driver (via a kernel command line flag); this is particularly useful together with Optimus/PRIME because it fixes tearing on PRIME-connected screens.
The user still needs to enable the Optimus/PRIME feature and specify the bus IDs of the Intel and NVIDIA GPUs, but this is still much easier for users and more reliable. The implementation handles both the X configuration file as well as getting display managers to run certain necessary `xrandr` commands just after X has started.
Configuration of commands run after X startup is done using a new configuration option `services.xserver.displayManager.setupCommands`. Support for this option is implemented for LightDM, GDM and SDDM; all of these have been tested with this feature including logging into a Plasma session.
Note: support of `setupCommands` for GDM is implemented by making GDM run the session executable via a wrapper; the wrapper will run the `setupCommands` before execing. This seemed like the simplest and most reliable approach, and solves running these commands both for GDM's X server and user X servers (GDM starts separate X servers for itself and user sessions). An alternative approach would be with autostart files but that seems harder to set up and less reliable.
Note that some simple features for X configuration file generation (in `xserver.nix`) are added which are used in the implementation:
- `services.xserver.extraConfig`: Allows adding arbitrary new sections. This is used to add the Device section for the Intel GPU.
- `deviceSection` and `screenSection` within `services.xserver.drivers`. This allows the nvidia configuration module to add additional contents into the `Device` and `Screen` sections of the "nvidia" driver, and not into such sections for other drivers that may be enabled.
This allows non-privileged users to configure local DNS
entries by editing hosts files read by NetworkManager's dnsmasq
instance.
Cherry-picked from e6c3d5a507909c4e0c0a5013040684cce89c35ce and
5a566004a2b12c3d91bf0acdb704f1b40770c28f.
With a config like
{
networking.interfaces."lo".ip4 = [
{ address = "10.8.8.8"; prefixLength = 32; }
];
}
a nixos-rebuild switch would take a long time, and you'd see:
$ systemctl list-jobs
JOB UNIT TYPE STATE
734400 network-interfaces.target start waiting
734450 sys-subsystem-net-devices-lo.device start running
734449 network-link-lo.service start waiting
and:
systemd[1]: sys-subsystem-net-devices-lo.device: Job sys-subsystem-net-devices-lo.device/star>
systemd[1]: sys-subsystem-net-devices-lo.device: Job sys-subsystem-net-devices-lo.device/star>
systemd[1]: Timed out waiting for device sys-subsystem-net-devices-lo.device.
This removes the device dependency for `lo` and fixes this bug.
Closes#7227
The deep merge caused all the options to be unset when generating docs, unless quagga was enabled.
Using imports, instead, properly allows the documentation to be generated.
The `.service` file defining the `systemd` unit for `autorandr.service`
which is bundled with the package itself uses `--default default` in the
`ExecStart` section. This can be an issue when having multiple layouts
(e.g. `default` as workstation layout I mostly work on and `mobile` when
I go somewhere else).
When the service gets restarted and `--default` can't be applied,
however the current layout can't be detected (e.g. when working with an
unknown beamer) the service silently fails with a message like this:
```
Jun 22 18:44:46 hauptshuhle autorandr[3168]: /nix/store/h83b72ffm68nm8fyjnppljchp456a94r-xrandr-1.5.0/bin/xrandr: ca>
Jun 22 18:44:46 hauptshuhle autorandr[3168]: Failed to apply profile 'default' (line 718):
Jun 22 18:44:46 hauptshuhle autorandr[3168]: Command failed: /nix/store/h83b72ffm68nm8fyjnppljchp456a94r-xrandr-1.>
```
As discussed in the IRC (see https://botbot.me/freenode/nixos/2018-07-05/?msg=101791455&page=6)
it's a bad long-term solution in terms of maintenance to manually patch
the service file bundled with the derivation, instead the service shall
be configured declaratively. Additionally this makes possible overrides
from the user-space way easier.
The `udev` rule (in `$out/etc/udev/rules.d`) won't' be affected, it
simply runs `systemctl start autorandr.service` when e.g. a new display
is added, so now `udev` communicates with the NixOS systemd unit.
To update the plasma start menu `kbuildsyscoca5` needs to be executed.
There are several people complaining about missing applications in their
plasma start menu.
This patch adds a activationScript for plasma, that runs
`kbuildsyscoca5` for each user that has `isNormalUser` == `true`.
In fff5923686 all occurences of
users.extraUsers and users.extraGroups have been changed tree-wide to
users.users and users.group. In the meantime the hadoop modules were
introduced via #41381 (060a98e9f4).
Unfortunately those modules still use users.extraUsers, which has been
renamed a long time ago (14321ae243, about
three years from now), so let's actually rename it accordingly as well.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @matthewbauer, @aespinosa
The order of sudoers entries is significant. The man page for sudoers(5)
notes:
Where there are multiple matches, the last match is used (which is not
necessarily the most specific match).
This module adds a rule for group "wheel" matching all commands. If you
wanted to add a more specific rule allowing members of the "wheel" group
to run command `foo` without a password, you'd need to use mkAfter to
ensure your rule comes after the more general rule.
extraRules = lib.mkAfter [
{
groups = [ "wheel" ];
commands = [
{
command = "${pkgs.foo}/bin/foo";
options = [ "NOPASSWD" "SETENV" ];
}
]
}
];
Otherwise, when configuration options are merged, if the general rule
ends up after the specific rule, it will dictate the behavior even when
running the `foo` command.
- Introduce new "server" output holding the server binaries
- Adapt tsmbac.patch to new build environment
- Adapt openafs nixos server module accordingly
- Update upstream CellServDB: 2017-03-14 -> 2018-05-14
- Introduce package attributes to refer to the openafs packages to use for
server, programs and kernel module
Rather than special-casing the dns options in networkmanager.nix, use
the module system to let unbound and systemd-resolved contribute to
the newtorkmanager config.
fixes#41838
At the moment it works fine for "file://" keys, but does not work for
dataPools with "prompt" keys, because the passphrase cannot be entered
(yet).
Commit 401370287a introduced a small error
where the closing tag of <literal/> was an opening tag instead.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @basvandijk, @xeji
Forcibly restarting NSCD is unnecessary and breaks setups that use SSSD for
authentication. NSCD is capable of detecting changes to /etc/resolv.conf and
invalidating its caches internally. Restarting NSCD/SSSD breaks user name and
UID resolution.
Fixes issue #33231 and makes it possible to enable Plasma and KDE at the same time.
Previously, this worked like this:
- The gdk-pixbuf package comes with a cache file covering the modules bundled
with gdk-pixbuf.
- The librsvg package comes with a cache covering modules from gdk-pixbuf as
well as librsvg.
- plasma5 and xfce modules set the environment variable GDK_PIXBUF_MODULE_FILE
to the one from librsvg, so that SVG was supported in addition to the
formats supported by gdk-pixbuf. However if both were enabled a configuration
conflict would result (despite setting to the same value).
While this sort of worked (ignoring the conflict which perhaps could be hacked
around), it is unscalable and a hack, as there would be a real problem when one
wanted to add a third package that supports additional image formats.
A new NixOS module (gdk-pixbuf) is added with a configuration option
(modulePackages) that other modules use to request specific packages to be
included in the loaders cache. When any package is present in the list, the
module generates a system-wide loaders cache which includes the requested
packages (and always gdk-pixbuf itself), and sets the environment variable
GDK_PIXBUF_MODULE_FILE to point to the generated cache file.
The plasma5 and xfce modules are updated to add librsvg to modulePackages
instead of setting GDK_PIXBUF_MODULE_FILE.
Note that many packages create wrappers that set GDK_PIXBUF_MODULE_FILE,
some directly to the one from librsvg. Therefore this change does not
change the existing hack in the librsvg package which ensures that
file is generated. This change aims only to solve the conflict in the
global environent variable configuration.
The freeradius service was merged with #34587
but the module was not added to module-list.
This commit fixes that and enables the use of
services.freeradius in nixos configuration.
Peviously only the timesyncd systemd unit was disabled. This meant
that when you activate a system that has chronyd enabled the following
strange startup behaviour takes place:
systemd[1]: Starting chrony NTP daemon...
systemd[1]: Stopping Network Time Synchronization...
systemd[1]: Stopped chrony NTP daemon.
systemd[1]: Starting Network Time Synchronization...
find-libs is currently choking when it finds the dynamic linker
as a DT_NEEDED dependency (from glibc) and bails out like this
(as glibc doesn't have a RPATH):
Couldn't satisfy dependency ld-linux-x86-64.so.2
Actually the caller of find-libs ignores the exit status, so the issue
almost always goes unnoticed and happens to work by chance. But
additionally what happens is that indirect .so dependencies are
left out from the dependency closure calculation, which breaks
latest cryptsetup as libssl.so isn't found anymore.
Kubernetes dashboard currently has cluster admin permissions,
which is not recommended.
- Renamed option "services.kubernetes.addons.dashboard.enableRBAC" to "services.kubernetes.addons.dashboard.rbac.enable"
- Added option "services.kubernetes.addons.dashboard.rbac.clusterAdmin", default = false.
- Setting recommended minimal permissions for the dashboard in accordance with https://github.com/kubernetes/dashboard/wiki/Installation
- Updated release note for 18.09.
F2FS is used on Raspberry Pi-like devices to enhance SD card performance. Allowing F2FS resizing would help in automatic deploying of SD card images without a Linux box to resize the file system offline.
Adds a module for running the journaldriver log forwarding agent via
systemd.
The agent can be deployed on both GCP instances and machines hosted
elsewhere to forward all logs from journald to Stackdriver Logging.
Consult the module options and upstream documentation for more
information.
Implementation notes:
* The service unit is configured to use systemd's dynamic user feature
which will let systemd set up the state directory and appropriate
user configuration at unit launch time instead of hardcoding it.
* The module depends on `network-online.target` to prevent a situation
where journaldriver is failing and restarting multiple times before
the network is online.
- Added option 'cni.configDir' to allow for having CNI config outside of nix-store
Existing behavior (writing verbatim CNI conf-files to nix-store) is still available.
- Removed unused option 'apiserver.publicAddress' and changed 'apiserver.address' to 'bindAddress'
This conforms better to k8s docs and removes existing --bind-address hardcoding to 0.0.0.0
- Fixed c/p mistake in apiserver systemd unit description
- Updated 18.09 release notes to reflect changes to existing options
And fixed some typos from previous PR
- Make docker images for Kubernetes Dashboard and kube-dns configurable
The usage of nixpkgs.config.packageOverrides is deprecated and we do
have overlays since quite a while.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @edolstra
This reverts a part of 5bd12c694b.
Apparently there's no way to specify user for RuntimeDirectory in systemd
service file (it's always root) but tor won't create control socket if the dir
is owned by anybody except the tor user.
These hardenings were adopted from the upstream service file, checked
against systemd.service(5) and systemd.exec(5) manuals, and tested to
actually work with all the options enabled.
`PrivateDevices` implies `DevicePolicy=closed` according to systemd.exec(5),
removed.
`--RunAsDaemon 0` is the default value according to tor(5), removed.
The `zsh-autosuggestions` package provides several configuration options
such as a different highlight style (like `fg=cyan` which is easier to
read).
With `rename.nix` the old `programs.zsh.enableAutosuggestions` is still
functional, but yields the following warning like this during evaluation:
```
trace: warning: The option `programs.zsh.enableAutosuggestions' defined in `<unknown-file>' has been renamed to `programs.zsh.autosuggestions.enable'.
```
The module provides the most common `zsh-autosuggestions` (highlight
style and strategy) as options that will be written into the interactive
shell init (`/etc/zshrc` by default). Further configuration options can
be declared using the `extraConfig` attr set:
```
{
programs.zsh.autosuggestions.extraConfig = {
"ZSH_AUTOSUGGEST_BUFFER_MAX_SIZE" = "buffer_size";
};
}
```
A full list of available configuration options for `zsh-autosuggestions`
can be viewed here: https://github.com/zsh-users/zsh-autosuggestions/blob/v0.4.3/README.md
[x] Support transparent proxying. This means services behind sslh (Apache, sshd and so on) will see the external IP and ports as if the external world connected directly to them.
[x] Run sslh daemon as unprivileged user instead of root (it is not only for security, transparent proxying requires it)
[x] Removed pidFile support (it is not compatible with running sslh daemon as unprivileged user)
[x] listenAddress default changed from "config.networking.hostName" (which resolves to meaningless "127.0.0.1" as with current /etc/hosts production) to "0.0.0.0" (all addresses)
Adds programs.mosh.withUtempter (default: true).
The option enables -with-utempter for mosh, allowing it to write to
/var/run/utmp and thus making connected sessions appear in the output
of `who -a`.
For that, a guid-wrapper is required. Also, the path to the `utempter` was
hardcoded in the resulting binary until now (so it could never been found),
thus, libutempter was patched accordingly to point to
/run/wrappers/bin/utempter which at least works when the wrapper is
configured.
Currently minio logs with enhanced tty data and journalctl does not include anything useful as a result:
```
Jun 08 11:03:28 alpha minio[17813]: [78B blob data]
Jun 08 11:03:28 alpha minio[17813]: [49B blob data]
Jun 08 11:03:28 alpha minio[17813]: [19B blob data]
Jun 08 11:03:28 alpha minio[17813]: [88B blob data]
Jun 08 11:03:28 alpha minio[17813]: [45B blob data]
Jun 08 11:03:28 alpha minio[17813]: [44B blob data]
Jun 08 11:03:28 alpha minio[17813]: [57B blob data]
```
Indicating that it detected some binary output. With the `--json` flag it logs:
```
Jun 08 11:14:58 alpha minio[18573]: {"level":"FATAL","time":"2018-06-07T23:14:58.770637778Z","error":{"message":"--address input is invalid: address 127.0.0.1: missing port in address","source":["/build/go/src/github.com/minio/minio/cmd/server-main.go:121:cmd.serverHandleCmdArgs()"]}}
```
DBus seems to resolve user IDs directly via glibc, circumventing nscd. In more
advanced setups this leads to user's coming from LDAP or SSSD not being
resolved by the dbus system bus daemon. The effect for such users is, that all
access to the system bus (e.g. busctl or nmcli) is denied.
Adding the respective NSS modules to the service's environment solves the issue
the same way it does for nscd.
This has been reported by @qknight in his Stack Overflow question:
https://stackoverflow.com/q/50678639
The correct way to override a single value would be to use something
like this:
systemd.services.nagios.serviceConfig.Restart = lib.mkForce "no";
However, this doesn't work because the check is applied for the attrsOf
type and thus the attribute values might still contain the attribute set
created by mkOverride.
The unitOption type however did already account for this, but at this
stage it's already too late.
So now the actual value is unpacked while checking the values of the
attribute set, which should allow us to override values in
serviceConfig.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @edolstra, @qknight
* add freeipmi to get power meter readings
* readline support for scontrol
* libssh2 support for X11 supporta
* Add note to enableSrunX11 in module
* fix hwloc support (was detected by configure)
The nixos module adds a new derivation to
systemPackages to make sure that the binaries
get the generated config file. This derivation
did not contain the man pages so far.
Activating the module now makes the man pages
available in the system environment.
This change allows users to specify an alternative database method. For
example an mpd satellite setup where another mpd on the network shares
it's database with the local instance. The `dbFile` parameter must not be
configured in that case.
BIND doesn't allow the options section (or any section I'd guess) to be
defined more than once, so whenever you want to set an additional option
you're stuck using weird hacks like this:
services.bind.forwarders = lib.mkForce [ "}; empty-zones-enable no; #" ];
This basically exploits the fact that values coming from the module
options aren't escaped and thus works in a similar vain to how SQL
injection works.
Another option would be to just set configFile to a file that includes
all the options, including zones. That obviously makes the configuration
way less extensible and more awkward to use with the module system.
To make sure this change does work correctly I added a small test just
for that. The test could use some improvements, but better to have a
test rather than none at all. For a future improvement the test could be
merged with the NSD test, because both use the same zone file format.
This change has been reviewed in #40053 and after not getting any
opposition, I'm hereby adding this to master.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @peti, @edolstra
Closes: #40053
The hooks directory contains now one level deep subdirectories which
need to be updated as well.
If you use gitea via ssh, ~/.ssh/authorized_keys also needs to be
updated because of the hardcoded path to gitea in the "command" option.
As shipped with k8s 1.10.3.
Also:
- updated the definition jsons as they are distributed in k8s.
- updated the image uris as they are renamed in k8s
- added imageDigest param as per 736848723e
As shipped with k8s 1.10.3.
Also:
- updated the definition jsons as they are distributed in k8s.
- updated the image uris as they are renamed in k8s
- added imageDigest param as per 736848723e
1) Change start-type to ```notify``` when running MariaDB so that we don't have to busy-wait for the
socket to appear.
2) Do not manually create the directory under /run as we can get systemd to do
that for us. This opens up the possibility later for not having to launch as root.
The Datadog agent requires `gohai` to be available on its `$PATH` in
order to collect certain metrics.
It would previously start up and collect certain types of metrics, but
log errors related to the missing gohai binary.
This commit configures the systemd-unit to make gohai available at
runtime.
This fixes#39810.
It is no longer required to change the config your ipfs repo manually if you change
localDiscovery option in nixos configuration after ipfs repository initialization.
toPath has confusing semantics and is never necessary; it can always
either just be omitted or replaced by pre-concatenating `/.`. It has
been marked as "!!! obsolete?" for more than 10 years in a C++
comment, hopefully removing it will let us properly deprecate and,
eventually, remove it.
Ideally I'd like the whole `nixos`/`nixpkgs` channel distinction to disappear, but this is a step along that path. After a while being in this state, we can stop creating the magic `nixpkgs -> .` symlink inside our `nixos` channel tarballs and simplify that whole mess a bit.
Wireguard is now split into two pretty much independent packages:
`wireguard` (Linux-specific kernel module) and `wireguard-tools`,
which is cross-platform.
`xsslock` (which was originally packaged in 6cb1d1aaaf)
is a simple screensaver which connects a given screen locker (e.g.
`i3lock`) with `logind`. Whenever `loginctl lock-sessions` is invoked
the locker will be used to lock the screen. This works with its power
management features (e.g. `lid switch`) as well, so the PC can be locked
automatically when the lid is closed.
The module can be used like this:
```
{
services.xserver.enable = true;
programs.xss-lock.enable = true;
programs.xss-lock.lockerCommand = "i3lock";
}
```
My c6f7d43678 made the mistake of not
having enough defaults. Now both variables are default as the *explicit*
value of the other, or a fallback. The fallback of `system` is the
default of `localSystem.system`. The fallback of `localSystem` is not
the other default (projected), as that would cause a cycle, but `{
system = builtins.currentTime; }` just as nixpkgs itself does it.
* networking/stubby.nix: implementing systemd service module for stubby
This change implements stubby, the DNS-over-TLS stub resolver daemon.
The motivation for this change was the desire to use stubby's
DNS-over-TLS funcitonality in tandem with unbound, which requires
passing certain configuration parameters. This module implements those
config parameters by exposing them for use in configuration.nix.
* networking/stubby.nix: merging back module list
re-merging the module list to remove unecessary changes.
* networking/stubby.nix: removing unecessary capabilities flag
This change removes the unecessary flag for toggling the capabilities
which allows the daemon to bind to low ports.
* networking/stubby.nix: adding debug level logging bool
Adding the option to turn on debug logging.
* networking/stubby.nix: clarifying idleTimeout and adding systemd target
Improving docs to note that idleTimeout is expressed in ms. Adding the
nss-lookup `before' target to the systemd service definition.
* networking/stubby.nix: Restrict options with types.enum
This change restricts fallbackProtocol and authenticationMode to accept
only valid options instead of any list or str types (respectively). This
change also fixes typo in the CapabilityBoundingSet systemd setting.
* networking/stubby.nix: cleaning up documentation
Cleaning up docs, adding literal tags to settings, and removing
whitespace.
* networking/stubby.nix: fixing missing linebreak in comments
* networking/stubby.nix: cleaning errant comments
When doing source routing/multihoming, it's practical to give names to routing
tables. The absence of the rt_table file in /etc make this impossible.
This patch recreates these files on rebuild so that they can be modified
by the user see NixOS#38638.
iproute2 is modified to look into config.networking.iproute2.confDir instead of
/etc/iproute2.
The original `nexus` derivation required `/run/sonatype-work/nexus3`
which explicitly depended on the NixOS path structure.
This would break `nexus` for everyone using `nixpkgs` on a non-NixOS
system, additionally the module never created `/run/sonatype-work`, so
the systemd unit created in `services.nexus` fails as well. The issue
wasn't actively known as the `nixos/nexus` test wasn't registered in
Hydra (see #40257).
This patch contains the following changes:
* Adds `tests.nexus` to `release.nix` to run the test on Hydra.
* Makes JVM parameters configurable: by default all JVM options were located
in `result/bin/nexus.vmoptions` which made it quite hard to patch
these parameters. Now it's possible to override all parameters by
running `VM_OPTS_FILE=custom-nexus.vmoptions ./result/bin/nexus run`
(after patching the `nexus` shell script), additionally it's possible
to override these parameters with `services.nexus.vmoptions`.
* Bumped Nexus from 3.5.1 to 3.11.0
* Run the `nexus` test on Hydra with `callTest` in `nixos/release.nix`,
furthermore the test checks if the UI is available on the specified
port.
* Added myself as maintainer for the NixOS test and the package to have
some more people in case of further breakage.
* Added sufficient disk space to the `nexus` test, otherwise the service
fails with the following errors:
```
com.orientechnologies.orient.core.exception.ODatabaseException: Cannot create database 'accesslog'
com.orientechnologies.orient.core.exception.OLowDiskSpaceException: Error occurred while executing
a write operation to database 'accesslog' due to limited free space on the disk (242 MB). The database
is now working in read-only mode. Please close the database (or stop OrientDB), make room on your hard
drive and then reopen the database. The minimal required space is 256 MB. Required space is now set to
256MB (you can change it by setting parameter storage.diskCache.diskFreeSpaceLimit) .
```
/cc @ironpinguin @xeji
When a package contains a directory in one of the systemd directories
(like flatpak does), it is symlinked into the *-units derivation.
Then later, the derivation will try to create the directory, which
will fail:
mkdir: cannot create directory '/nix/store/…-user-units/dbus.service.d': File exists
builder for '/nix/store/…-user-units.drv' failed with exit code 1
Closes: #33233
GRUB 2.0 supports png, jpeg and tga. This will use the image's suffix to
load the right module.
As jpeg module is named jpeg, jpg is renamed jpeg.
If the user uses wrong image suffix for an image, it wouldn't work anyway.
This will leave up to two additional left-over files in /boot/ if user switches
through all the supported file formats. The module already left the png
image if the user disabled the splash image.
If docker is enabled, start mesos-slave.service after docker.service
to avoid a race condition that could result in mesos-slave to fail
with "Failed to create docker: Timed out getting docker version"
The pull request that added dhparams (#39507) was made at the time where
the dhparams module overhaul (#39526) wasn't done yet, so it's still
using the old mechanics of the module.
As stated in the release notes:
Module implementers should not set a specific bit size in order to let
users configure it by themselves if they want to have a different bit
size than the default (2048).
An example usage of this would be:
{ config, ... }:
{
security.dhparams.params.myservice = {};
environment.etc."myservice.conf".text = ''
dhparams = ${config.security.dhparams.params.myservice.path}
'';
}
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @qknight, @abbradar, @hrdinka, @leenaars
Regression introduced by d4468bedb5.
No systemd messages are shown anymore during VM test runs, which is not
very helpful if you want to find out about failures.
There is a bit of a conflict between testing and the change that
introduced the regression. While the mentioned commit makes sure that
the primary console is tty0 for virtualisation.graphics = false, our VM
tests need to have the serial console as primary console.
So in order to support both, I added a new virtualisation.qemu.consoles
option, which allows to specify those options using the module system.
The default of this option is to use the changes that were introduced
and in test-instrumentation.nix we use only the serial console the same
way as before.
For test-instrumentation.nix I didn't add a baudrate to the serial
console because I can't find a reason on top of my head why it should
need it. There also wasn't a reason stated when that was introduced in
7499e4a5b9.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @flokli, @dezgeg, @edolstra
after seeing
`adjtime failed: Invalid argument` in my syslog, I tried using
`ntpd -s` but it would trigger
`/etc/ntpd.conf: No such file or directory`
see https://github.com/NixOS/nixpkgs/issues/31885
Instead of running the daemon with a specific config file, use the
standard file so that user are able to use the ntp executable without
having to look for the current config file.
Tracking scripts in particular, cannot be included in extraOpts, because script declaration has to be above script usage in keepalived.conf.
Changes are fully backward compatible.
This introduces an option that allows us to turn off stateful generation
of Diffie-Hellman parameters, which in some way is still "stateful" as
the generated DH params file is non-deterministic.
However what we can avoid with this is to have an increased surface for
failures during system startup, because generation of the parameters is
done during build-time.
Aside from adding a NixOS VM test it also restructures the type of the
security.dhparams.params option, so that it's a submodule.
A new defaultBitSize option is also there to allow users to set a
system-wide default.
I added a release notes entry that described what has changed and also
included a few notes for module developers using this module, as the
first usage already popped up in NixOS/nixpkgs#39507.
Thanks to @Ekleog and @abbradar for reviewing.
Always enable both tty and serial console, but set preferred console
depending on cfg.graphical.
Even in qemu graphical mode, you can switch to the serial console via
Ctrl+Alt+3.
With that being done, you also don't need to specify
`systemd.services."serial-getty@ttyS0".enable = true;` either as described in
https://nixos.wiki/wiki/Cheatsheet#Building_a_service_as_a_VM_.28for_testing.29,
as systemd automatically spawns a getty on consoles passwd via cmdline.
This also means, vms built by 'nixos-rebuild build-vm' can simply be run
properly in nographic mode by appending `-nographic` to `result/bin/run-*-vm`,
without the need to explicitly add platform-specific QEMU_KERNEL_PARAMS.
This allows to set the default bit size for all the Diffie-Hellman
parameters defined in security.dhparams.params and it's particularly
useful so that we can set it to a very low value in tests (so it doesn't
take ages to generate).
Regardless for the use in testing, this also has an impact in production
systems if the owner wants to set all of them to a different size than
2048, they don't need to set it individually for every params that are
set.
I've added a subtest to the "dhparams" NixOS test to ensure this is
working properly.
Signed-off-by: aszlig <aszlig@nix.build>
When trying to run NSD to serve the root zone, one gets the following
error message:
error: illegal name: '.'
This is because the name of the zone is used as the derivation name for
building the zone file. However, Nix doesn't allow derivation names
starting with a period.
So whenever the zone is "." now, the file name generated is "root"
instead of ".".
I also added an assertion that makes sure the user sets
services.nsd.rootServer, otherwise NSD will fail at runtime because it
prevents serving the root zone without an explicit compile-time option.
Tested this by adding a root zone to the "nsd" NixOS VM test.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @hrdinka, @qknight
Without fcrontab being setuid, every attempt by an user in the fcron
group to edit their own crontab (via `fcrontab -e`) results in the
following error:
```
2018-05-06 11:29:07 ERROR could not change euid to 273: Operation not permitted
2018-05-06 11:29:07 ERROR fcron child aborted: this does not affect the main fcron daemon, but this may prevent a job from being run or an email from being sent.
```
Adding setuid by hand has resolved this issue and aligns with the way
fcrontab is installed on other distributions.
lib.optional returns a singleton or an empty list. Therefore the
argument does not need to be wrapped in a list.
An alternative patch could have used lib.optionals but seems like no
more elements are going to be added to the optional list.
This is apparent from the service file directory in plymouth:
├── multi-user.target.wants
│ ├── plymouth-quit.service -> ../plymouth-quit.service
│ └── plymouth-quit-wait.service -> ../plymouth-quit-wait.service
Leaving it unspecified caused gdm-wayland to crash on boot, see #39615.
The change made other display managers not quit plymouth properly however. By
removing "multi-user.target" from `plymouth-quit.after` this is resolved.
The following changes have been applied:
- the property `http.headers.X-Content-Type-Options` must a list of
strings rather than a serialized list
- instead of `/etc/docker/registry/config.yml` the configuration will be
written with `pkgs.writeText` and the store path will be used to run
the registry. This reduces the risk of possible impurities by relying
on the Nix store only.
- cleaned up the property paths to easy readability and reduce the
verbosity.
- enhanced the testcase to ensure that digests can be deleted as well
- the `services.docker-registry.extraConfig` object will be merged with
`registryConfig`
/cc @ironpinguin
@Ekleog writes in https://github.com/NixOS/nixpkgs/pull/39526:
> I think a default of 4096 is maybe too much? See certbot/certbot#4973;
> Let's Encrypt supposedly know what they are doing and use a
> pre-generated 2048-bit DH params (and using the same DH params as
> others is quite bad, even compared to lower bit size, if I correctly
> remember the attacks available -- because it increases by as much the
> value of breaking the group).
> Basically I don't have anything personal against 4096, but fear it may
> re-start the arms race: people like having "more security" than their
> distributions, and having NixOS already having more security than is
> actually useful (I personally don't know whether a real-size quantum
> computer will come before or after our being able to break 2048-bit
> keys, let alone 3072-bit ones -- see wikipedia for some numbers).
> So basically, I'd have set it to 3072 in order to both decrease build
> time and avoid having people setting it to 8192 and complaining about
> how slow things are, but that's just my opinion. :)
While he suggests is 3072 I'm using 2048 now, because it's the default
of "openssl dhparam". If users want to have a higher value, they can
still change it.
Signed-off-by: aszlig <aszlig@nix.build>
The ability to specify "-drive if=scsi" has been removed in QEMU version
2.12 (introduced in 3e3b39f173).
Quote from https://wiki.qemu.org/ChangeLog/2.12#Incompatible_changes:
> The deprecated way of configuring SCSI devices with "-drive if=scsi"
> on x86 has been removed. Use an appropriate SCSI controller together
> "-device scsi-hd" or "-device scsi-cd" and a corresponding "-blockdev"
> parameter instead.
So whenever the diskInterface is "scsi" we use the new way to specify
the drive and fall back to the deprecated way for the time being. The
reason why I'm not using the new way for "virtio" and "ide" as well is
because there is no simple generic way anymore to specify these.
This also turns the type of the virtualisation.qemu.diskInterface option
to be an enum, so the user knows which values are allowed but we can
also make sure the right value is provided to prevent typos.
I've tested this against a few non-disk-related NixOS VM tests but also
the installer.grub1 test (because it uses "ide" as its drive interface),
the installer.simple test (just to be sure it still works with
"virtio") and all the tests in nixos/tests/boot.nix.
In order to be able to run the grub1 test I had to go back to
8b1cf100cd (which is a known commit where
that test still works) and apply the QEMU update and this very commit,
because right now the test is broken.
Apart from the tests here in nixpkgs, I also ran another[1] test in
another repository which uses the "scsi" disk interface as well (in
comparison to most of the installer tests, this one actually failed
prior to this commit).
All of them now succeed.
[1]: 9b5a119972/tests/system/kernel/bfq.nix
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @edostra, @grahamc, @dezgeg, @abbradar, @ts468
Previously the script would contain an empty `if` block (which is invalid
syntax) if both `data.activationDelay == null` and `data.postRun == ""`. Fix
this by adding a no-op `true`.
As suggested in https://github.com/NixOS/nixpkgs/pull/39416#discussion_r183845745
the versioning attributes in `lib` should be consistent to
`nixos/version` which implicates the following changes:
* `lib.trivial.version` -> `lib.trivial.release`
* `lib.trivial.suffix` -> `lib.trivial.versionSuffix`
* `lib.nixpkgsVersion` -> `lib.version`
As `lib.nixpkgsVersion` is referenced several times in `NixOS/nixpkgs`,
`NixOS/nix` and probably several user's setups. As the rename will cause
a notable impact it's better to keep `lib.nixpkgsVersion` as alias with
a warning yielded by `builtins.trace`.
mke2fs has this annoying property that it uses getrandom() to get random
numbers (for whatever purposes) which blocks until the kernel's secure
RNG has sufficient entropy, which it usually doesn't in the early boot
(except if your CPU supports RDRAND) where we may need to create the
root disk.
So let's give the VM a virtio RNG to avoid the boot getting stuck at
mke2fs.
Ubiquiti has both a LTS and current version of their Unifi controller software.
The latter adds new features, but may drop support for some devices.
This adds the capability to use either for the unifi module but defaults
to the LTS version, which was the previous behavior.
Previously we indirectly suggested that the user use
services.printing.extraConf to set this, but this doesn't work with the
default merge ordering. Fix this by making it an independent option.
Fixes#39611.
@cleverca found this bug in the declarative hooks config. Any shell
variables referenced in a hook script would get expanded by the hooks
directory builder.
Prevent variable expansion by quoting the here doc limit string.
Allow out of band communication between qemu VMs and the host.
Useful to retrieve IPs of VMs from the host (for instance when libvirt can't analyze
DHCP requests because VMs are configured with static addresses or when
there is connectivity default).
First of all let's start with a clean up the multiline string
indentation for descriptions, because having two indentation levels
after description is a waste of screen estate.
A quick survey in the form of the following also reveals that the
majority of multiline strings in nixpkgs is starting the two beginning
quotes in the same line:
$ find -name '*.nix' -exec sed -n -e '/=$/ { n; /'\'\''/p }' {} + | wc -l
817
$ find -name '*.nix' -exec grep "= *'' *\$" {} + | wc -l
14818
The next point is to get the type, default and example attributes on top
of the description because that's the way it's rendered in the manual.
Most services have their enable option close to the beginning of the
file, so let's move it to the top.
Also, I found the script attribute for dhparams-init.service a bit hard
to read as it was using string concatenation to split a "for" loop.
Now for the more substantial clean ups rather than just code style:
* Remove the "with lib;" at the beginning of the module, because it
makes it easier to do a quick check with "nix-instantiate --parse".
* Use ConditionPathExists instead of test -e for checking whether we
need to generate the dhparams file. This avoids spawning a shell if
the file exists already and it's probably more common that it will
exist, except for the initial creation of course.
* When cleaning up old dhparams file, use RemainAfterExit so that the
unit won't be triggered again whenever we stop and start a service
depending on it.
* Capitalize systemd unit descriptions to be more in par with most
other unit descriptions (also see 0c5e837b66).
* Use "=" instead of "==" for conditionals using []. It's just a very
small nitpick though and it will only fail for POSIX shells. Bash on
the other side accepts it anyway.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @Ekleog
This option allows us to turn off stateful generation of Diffie-Hellman
parameters, which in some way is still stateful as the generated DH
params file is non-deterministic.
However what we can avoid with this is to have an increased surface for
failures during system startup, because generation of the parameters is
done during build-time.
Another advantage of this is that we no longer need to take care of
cleaning up the files that are no longer used and in my humble opinion I
would have preferred that #11505 (which puts the dhparams in the Nix
store) would have been merged instead of #22634 (which we have now).
Luckily we can still change that and this change gives the user the
option to put the dhparams into the Nix store.
Beside of the more obvious advantages pointed out here, this also
effects test runtime if more services are starting to use this (for
example see #39507 and #39288), because generating DH params could take
a long time depending on the bit size which adds up to test runtime.
If we generate the DH params in a separate derivation, subsequent test
runs won't need to wait for DH params generation during bootup.
Of course, tests could still mock this by force-disabling the service
and adding a service or activation script that places pre-generated DH
params in /var/lib/dhparams but this would make tests less readable and
the workaround would have to be made for each test affected.
Note that the 'stateful' option is still true by default so that we are
backwards-compatible with existing systems.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @Ekleog, @abbradar, @fpletz
We're going to implement an option which allows us to turn off stateful
handling of Diffie-Hellman parameter files by putting them into the Nix
store.
However, modules now might need a way to reference these files, so we
add a now path option to every param specified, which carries a
read-only value of the path where to find the corresponding DH params
file.
I've also improved the description of security.dhparams.params a bit so
that it uses <warning/> and <note/>.
The NixOS VM test also reflects this change and checks whether the old
way to specify the bit size still works.
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @Ekleog
Following legacy packing conventions, `isArm` was defined just for
32-bit ARM instruction set. This is confusing to non packagers though,
because Aarch64 is an ARM instruction set.
The official ARM overview for ARMv8[1] is surprisingly not confusing,
given the overall state of affairs for ARM naming conventions, and
offers us a solution. It divides the nomenclature into three levels:
```
ISA: ARMv8 {-A, -R, -M}
/ \
Mode: Aarch32 Aarch64
| / \
Encoding: A64 A32 T32
```
At the top is the overall v8 instruction set archicture. Second are the
two modes, defined by bitwidth but differing in other semantics too, and
buttom are the encodings, (hopefully?) isomorphic if they encode the
same mode.
The 32 bit encodings are mostly backwards compatible with previous
non-Thumb and Thumb encodings, and if so we can pun the mode names to
instead mean "sets of compatable or isomorphic encodings", and then
voilà we have nice names for 32-bit and 64-bit arm instruction sets
which do not use the word ARM so as to not confused either laymen or
experienced ARM packages.
[1]: https://developer.arm.com/products/architecture/a-profile
* checks using package providing the nix-daemon that we'll be using
* made optional (unlike some other config checks) "just in case":
since this requires running the new Nix on the builder, this
won't work in a few (AFAIK very uncommon) situations such as
cross-compiling NixOS or using `include` directives in nix.conf
This does rely on Nix2 but not by the builder.
Since we only offer Nix2+ in-tree this should be fine,
and may otherwise be required anyway.
I know that "devinfo" output does not currently exist, but so does "devman".
It is mentioned in the nixpkgs manual, but no derivation in nixpkgs actually uses it.
HA doesn't mind the configuration being JSON instead of YAML but since YAML is
the official language, use that as it allows users to easily exchange config
data with other parties in the community.
Additionally, some settings based on NixOS configuation is set via defaultConfig
which is then merged with the user provided configration.
For now that just means http port and time zone but others can easily be added.
This partially reverts a change from e88f28965a
which removed the `mount --rbind /sys`.
While true that the activation scripts will mount `sysfs` at `/sys`,
none of the mountpoints lower in the `/sys` tree are handled by the
activation script, which includes `efivarfs`.
This fixes#38477 since it ensures the presence of `efivarfs` in the
`/sys` tree, which is why the systemd-boot installation failed.
This is more in line with what other services do; also looks cleaner.
It changes configuration entries for pre-and post-hooks type to lines from
lists of strings which are more logical for them; coersion is provided for
backwards compatibility.
Finally, add several steps to improve robustness:
1. Load kernel module on start if not loaded;
2. Don't remove wireguard interface on start; it is removed on service stop. If
it's not something is wrong.
This is needed because simp_le expects two certificates in fullchain.pem, leading to error:
> Not enough PEM encoded messages were found in fullchain.pem; at least 2 were expected, found 1.
We now create a CA and sign the key with it instead, providing correct fullchain.pem.
Also cleanup service a bit -- use PATH and a private temporary directory (which
is more suitable).
Do cleanup of user-created additional rules.
Of course it'd be much better to just use iptables-{save,restore} for
declarative management, but as it's still not there...
Introduced in 286b007bd3 and then
in 2e6b796761.
This a proper fix for what 70c6f6572d tried to do.
Removing the "config" prefix triggers the bug on pure nixos too, not only
on nixops.
Nothing probably uses this, but let's be pedantic and have the
pre-included channel on the install media be as close as possible to
what 'nix-channel --update' will give them.
The only remaining difference is that the channel adds programs.sqlite,
which is fundamentally unfixable.
a) Some providers can update multiple domains - support that.
b) Make "zone" and "script" configurable. Some providers require these.
c) Instead of leaving the ddclient daemon running all the time, use a systemd
timer to kick it off.
d) Don't use a predefined user - run everything via DynamicUser
e) Add documentation
Because it improves out-of-the-box user experience a lot (IMHO).
(zsh completion is already on by default.)
Remove "programs.bash.enableCompletion = true" from
nixos-generate-config.pl, which feels superflous now.
