This sets the exception handler to show the full exception on startup.
We don't think it does anything else, with respect to logging, for
instance. Everything else can be configured in the config file, and this
is plain reasonable to simply always enable in our view.
This commit fixes the requirement to provide a pruneOption to ensure
that the check command is run when a check option is set. This is useful
for check-only configurations. The option is implicitly set if checkOpts
are given by default.
This makes kwin_wayland successfully gain SCHED_RR priority, which
significantly improves the rendering lag, ie. cursor movement and
desktop animations, under heavy CPU load like on compilation.
This was already enabled for plasma5 but is not for plasma6 somehow.
These have been deprecated since NixOS 20.03.
Also fixes evaluation error caused by moving
the `defaultSessionFromLegacyOptions` variable
in 6be2bfcc32
This service performs operations that significantly increase the
performance of Nextcloud, can take a while. These are designed however
to not require maintenance mode and can be executed during normal
operation[1].
Make nextcloud-cron a simple unit instead of oneshot: otherwise we risk
that it'll be stopped by the startup timeout (oneshot executes ExecStart
while "activating") which can be an issue for very long running tasks or
if Nextcloud needs to catch up if one task was broken for a while.
[1] https://docs.nextcloud.com/server/29/admin_manual/maintenance/upgrade.html#long-running-migration-steps
Previously you needed to set an devRootTokenID when dev=true despite the option being optional
Caused by wrong default value and not allowing null as value
This reverts commit e827697fd3.
This seems to cause various issues during system activation, as reported
in https://github.com/NixOS/nixpkgs/issues/302771 for example.
Due to being close to branchoff, revert this for now.
We'll open a tracking issue to collect and sort out remaining issues.
Since https://github.com/redis/redis/pull/4001 included in 6.2.0
transparent hugepages works when being set to madvise which is the NixOS
and upstream recommended default.
> WARNING Memory overcommit must be enabled! Without it, a background save or replication may fail under low memory condition.
> Being disabled, it can also cause failures without low memory condition, see https://github.com/jemalloc/jemalloc/issues/1328.
> To fix this issue add 'vm.overcommit_memory = 1' to /etc/sysctl.conf and then reboot or run the command 'sysctl vm.overcommit_memory=1' for this to take effect.
Currently there is an issue with $PATH & parallel causing build errors.
It’s probably best to just remove the dependency where bash forking is
good enough here.
Before the startup, the matrix-appservice-irc service sets up the
registration file such that it can be used by matrix-synapse. Part of
that setup requires us to change the group of said file so that the home
server can read it. Consequently, we need CAP_CHOWN and require that the
@chown system calls are allowed.
While we supposedly set up both of these, the setup of system calls is
broken as we have both an allow and a deny list of syscalls. But while
the allow list contains "@chown", the deny list contains "@privileged"
which contains "@chown" itself. So ultimately, we end up denying
"@chown".
Fix this issue by specifying "@chown" after the deny list.
Currently the installWrapper warning is issued if sudo (and sudo-rs)
aren't installed. This is fine, except we get the warning even if we
explicitly turn off installWrapper -- say, for this very reason!
Rather than warning on every build until either sudo is installed or
Akkoma is uninstalled, only warn if cfg.installWrapper is true.
Yall won't miss me. The packages I leave orphaned are trivially updated as dependents need the new versions.
But passively endorsing the direction this organization and its leadership is something I can't do.
To those who still have faith in turning this around, you da real MVP 🖖
Previously evaluaton would fail if `smtp.to` was not set, since the
default case was not handled.
With the current versions of the python elasticsearch libraries, any
configuration containing the modules default would fail since it's not a
valid URL.
The issue adressed in #279068 is also adressed, thuse closes#279068.
* PHP 8.3 seems supported, so let's go for it!
* The conditions for which Nextcloud will be the default were bogus: for
<24.11 I'd suggest to go for nextcloud29 already. The people on
unstable relying on the condition were on nextcloud28 so the upgrade
will work fine.
Also, it's unstable, so such upgrades are to be expected IMHO.
* Update the release notes to reflect that the new default is Nextcloud
29 and warn that only one major upgrade at a time can be done.
Fixes issues described in #208242 for this part of the nixpkgs tree.
There are no behavioral changes in this, it only adjusts the code so
that it is easier to understand.
Also updates my information and contact info.
I no longer use The Hedgehog as my github username or online presence
username, so this fixes that. It also matches my github username, so it
should be easier for others to mention me if needed.
Since with the completion of the docbook migration) it seems unclear
what relevance editing xml in generall and docbook in particular with
Emacs still has to NixOS at all, and people interested in the topic
will presumably look to other resources elsewhere (e.g. to the nXML
mode's actual documenation).
- `mount-nvidia-binaries`: this option allows users to avoid mounting
nvidia binaries on the container.
- `mount-nvidia-docker-1-directories`: this option allows users to
avoid mounting `/usr/local/nvidia/lib{,64}` on containers.
Add the NixOS option `hardware.nvidia-container-toolkit-cdi-generator.enable`.
This enables the ability to expose GPU's in containers for container
runtimes that support the Container Device Interface (CDI)
Remove `cdi.static` and `cdi.dynamic.nvidia.enable` attributes.
Add package to environment.systemPackages, services.dbus.packages, create gnome-remote-desktop user and group (fixes for GNOME 46)
This adds the `g-r-d` package to environment.systemPackages (allowing the usage of the `grdctl` command along with enabling `g-r-d`'s polkit rule), makes its dbus-related files recognizable to dbus, and creates the `gnome-remote-desktop` user and group necessary for systemd's running of the `gnome-remote-desktop-daemon` with the `--system` subcommand and enabling Remote Login.
https://gitlab.gnome.org/GNOME/gnome-remote-desktop/-/compare/45.1...46.0
In file included from ../src/grd-settings.c:28:
/nix/store/xxx-freerdp-3.4.0/lib/pkgconfig/../../include/freerdp3/freerdp/freerdp.h:25:10:
fatal error: winpr/stream.h: No such file or directory
25 | #include <winpr/stream.h>
| ^~~~~~~~~~~~~~~~
compilation terminated.
Ugh. So stuff I am aware of here:
- In freerdp3.pc, winpr3 is in Requires.private.
- In https://github.com/FreeRDP/FreeRDP/blob/3.4.0/include/freerdp/freerdp.h#L25 <winpr/stream.h>
is included.
- In GNOME/gnome-remote-desktop@d29909a
<freerdp/freerdp.h> is included in src/grd-settings.c.
- We patched pkg-config in NixOS to not include Requires.private in --cflags according to
mate-desktop/atril issue 351.
- According to https://gitlab.gnome.org/GNOME/gjs/-/issues/571, Requires.private is probably correct
if no data types are exposed in public API.
So to fix this somewhere, if src/grd-settings.c has direct usage of winpr, we can PR to g-r-d declaring
the dep. If freerdp/freerdp.h exposes winpr data types we PR to freerdp and move winpr to Requires.
Probably someone can help me do the check, I am committing this simply to unbreak the build for now.
Changelog-Reviewed-By: Maxine Aubrey <max@ine.dev>
Follow-up to #282377. #282377 broke `environment.etc."wireplumber<...>"`,
however WirePlumber did not yet have `extraConfig` style options for
configuring it ergonomically outside of `environment.etc`. This has
caused issues for people who had custom config files for WirePlumber, as
having to create a config package just to edit some settings is not as
ergonomic or discoverable as with a proper `extraConfig` style option.
This commit fixes this issue by adding the `extraConfig` option for
additional config file and the `extraScripts` option for additional
scripts to be used by config files.
With WirePlumber 0.5 it is possible to supply config files and scripts
via the `XDG_DATA_DIRS` variable to the WirePlumber daemon. This is how
the new options and with this change also the `configPackages` option
expose their files to the daemon. This way
`environment.etc."wireplumber"` works again for user configuration and
breakage of old configs from 23.11 to 24.05 should be limited to those
caused by the change in the config format from WirePlumber 0.4 to 0.5.
In the replacement arg of gsub() the & symbol is a special character
that need to be escaped. To avoid this, and further issues due to the
variable name possibly being interpreted as a regex, we do a normal
substring replacement.
This fixes issues #279803.
If allowAuxiliaryImperativeNetworks is enabled, the wpa_supplicant
daemon complains fails to start if /etc/wpa_supplicant.conf does not
exist. As this can be created using wpa_cli (or similar) later, it
shouldn't matter, so let's create an empty one.
This fixes issues #157537, #299466.
It should be curl -L (follow redirects), not curl -l (FTP directory
listing option). I know because it's my mistake.
Fixes: d4b989cafc ("nixos/deconz: delay signalling service readiness until it's actually up")
Closes#267442
$ nix path-info -Sh /nix/store/qj1dm7wfw5m3mxf1gn3fdm0az9y1h5ny-linux-firmware-20240312-xz
/nix/store/qj1dm7wfw5m3mxf1gn3fdm0az9y1h5ny-linux-firmware-20240312-xz 440.3M
$ nix path-info -Sh /nix/store/c3szcjxb3g990dbiz7llwmkaf0bi98j2-linux-firmware-20240312-zstd
/nix/store/c3szcjxb3g990dbiz7llwmkaf0bi98j2-linux-firmware-20240312-zstd 460.6M
This is an increase of 4.4%, but OTOH zstd has a significantly higher
decompression speed[1].
[1] https://gregoryszorc.com/blog/2017/03/07/better-compression-with-zstandard/
With the changes introduced in #303388 tesseract would only be compiled
with the languages defined in `PAPERLESS_OCR_LANGUAGE`. However, english
is always required, making tesseract fail to build when only non-english
languages are defined in tesseract:
```
eng.traineddata must be present in tessdata for Tesseract to work
```
