| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The test is not essential for running any real NixOS machines and the
tests just checks whether the runInMachine function for building
derivations inside a VM works.
I was very much reluctant to just remove this test before I actually
fixed the issue upstream because this would mean that it's highly likely
that I'm never going to do it.
However, since all the channels on our side are currently blocking on
this test, which isn't relevant either, I'm going to risk possibly never
fixing runInMachine upstream since I'd have removed that test anyway for
the aforementioned reason.
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
| |
This is incredibly useful if you don't want to leave the terminal.
|
| |
|
|
|
|
| |
Neat. Like cat, just better in certain circumstances.
|
|
|
|
|
|
|
| |
Well, that was a bit too hasty, I forgot that MPD runs as its own user
and therefore can't talk to pulseaudio if this isn't set.
This reverts commit dc4882e92f84c157d682b860365644ffd1f5c0d6.
|
|
|
|
|
| |
This seems like an immensely useful tool in conjunction with
`taskwarrior`.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The upstream Hydra channels are based on the "tested" job in
release-combined.nix, which already contain a bunch of tests that must
succeed for the channel to advance.
Since evaluation time has skyrocketed for us within the last few months,
we now even don't get *any* evaluation since a while since we hit a
timeout of 10 hours.
I have no idea which changes resulted in this spike of evaluation time
but right now I'm too busy with other things to do a proper
investigation.
To hopefully mitigate this, I decided to base our vuizvui jobset on the
"nixos-unstable" channel and we now only run *additional* tests not
already tested by the upstream channel.
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
|
| |
Reat that muesli made this lovely little tool, so naturally had to add
it.
|
|
|
|
|
| |
We've finally arrived in the 21st century and don't require a proxy
anymore. Hurrah!
|
|
|
|
|
|
|
|
|
|
| |
IWD seems to have an issue with the re-configured eduroam I frequently
use. Every other OS has no issues with the new configuration, only IWD
does. They did some background work on the CA cert, and even gave me the
public CA PEM to integrate in the configuration directly, and yet IWD
complains of `bad_certificate`. I don't have time to dig deeper, so I'm
switching to `wpa_supplicant` again until I have time to test it again
with a newer version of IWD.
|
|
|
|
|
|
|
|
|
| |
alock is really bad locking software. It is slow, it becomes
transparent, it even crashes sometimes. All the things a good lock
screen should never do.
i3lock is the trusted “standard” featureless locksceen, so this switch
has been procrastinated for way too long.
|
| |
|
| |
|
|
|
|
| |
Co-Authored-By: midzer <midzer@gmail.com>
|
|
|
|
|
|
|
| |
It was only used for a single use-case, by now all browsers support
this functionality. It’s a pretty heavy dependency for a single use :)
Co-Authored-By: midzer <midzer@gmail.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Version 1.10.X switches the default to new MIME-based format which
breaks interoperability with pass and other implementations.
I'd gladly switch to that format, but the 1.10.0 changelog also
announces plans to remove support for GnuPG and Git in the long term:
> The goal is to remove the support for multiple backends and any
> external dependencies, including git and gpg binaries.
GnuPG and Git support is the reason why I started using pass and
ultimately switched to gopass. If the latter stops being a viable
password manager, switching back to pass will be much harder with the
new MIME format.
There is also an upstream issue[1] about this and while I haven't read
through *all* the comments, other people seem to have similar reasons
for switching to gopass.
So far however I'm not convinced that my use case will have a future
with gopass, so I'll stay at 1.9.x until I've had the time to properly
research other options or maybe even stay with gopass (and go all-in
with the new MIME format).
[1]: https://github.com/gopasspw/gopass/issues/1365
Signed-off-by: aszlig <aszlig@nix.build>
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This was one of the places where types.loaOf was still in place and it
got removed a while[1] ago and this in turn causes evaluation to fail
for quite a few machines:
The option value `boot.initrd.luks.devices' in `...' is not of type
`attribute set of submodules'.
I've not only changed all the machines to use attribute sets but also
fixed the check in core/tests.nix, because comparing against a list when
the actual type is an attribute set will result in all the LUKS tests to
be part of *all* channels, no matter whether you're actually using LUKS.
[1]: https://github.com/NixOS/nixpkgs/commit/20d491a317d9956ddca80913f07
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @Profpatsch
Cc: @sternenseemann
|
|
|
|
|
|
|
|
| |
I've had this laying around in my configuration.nix since quite a while,
but today is one of those days where I can't stand all the cruft piling
up there anymore and decided to add it here.
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The package attributes for GSstreamer 0.x were dropped[1] a while ago,
so evaluating the packages no longer succeeds and instead we get an
error like this on Hydra:
error: undefined variable 'gstreamer' at .../profiles/packages.nix:129:7
Initially I was tempted to change this to use pkgs.gst_all_1.gstreamer,
but looking at the differences between the old pkgs.gstreamer, the new
package only contains the relevant libraries while the old one contains
a few binaries, namely gst-feedback, gst-inspect, gst-launch,
gst-typefind, gst-xmlinspect, gst-xmllaunch and all of them again with a
"-0.10" suffix.
All of these tools are development tools and should not be used in real
applications, for example from the manual[2] on gst-launch-1.0:
> Please note that gst-launch-1.0 is primarily a debugging tool. You
> should not build applications on top of it. For applications, use the
> gst_parse_launch() function of the GStreamer API as an easy way to
> construct pipelines from pipeline descriptions.
While environment.systemPackages does link other directories than just
$out/bin, the primary reason why you'd want to put something in there is
to make it available in the system's PATH.
When asking @devhell about this he didn't remember the exact reason why
he put gstreamer in systemPackages, so I can only *assume* it was
because of tools like gst-launch, which unfortunately even I can
remember seeing recommended (instead of eg. ffmpeg) in some dark corners
of the web.
So if this would be really about gst-launch, we would need to put in
pkgs.gst_all_1.gstreamer.dev into systemPackages, since the development
tools are now where they belong.
Given that @devhell doesn't know the reason anymore and *also* told me
to remove it *and* it's also just a development tool which is not
supposed to be used in production, I'm hereby removing the package.
Should there really be applications which rely on this, they should
hopefully break after this change so they can be fixed to not rely on
these development tools.
[1]: https://github.com/NixOS/nixpkgs/commit/4a4e642abaaa026b55f42248a7b
[2]: https://gstreamer.freedesktop.org/documentation/tools/gst-launch.html
Signed-off-by: aszlig <aszlig@nix.build>
Cc: @devhell
|
|
|
|
|
|
|
|
|
|
| |
If we don't have a PID namespace, we're not allowed to mount a new
procfs instance and subsequently get an error (EPERM).
To cope with this, we're now bind-mounting /proc just like the other
pseudo file systems IFF we're not using the CLONE_NEWPID flag.
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
While the sandbox was initially written for games, I now use the
implementation for other things, such as sandboxing database management
systems inside "nix develop".
However, both MariaDB and PostgreSQL do not like it very much if for
example IPC is too restricted and if the PID file contains the PID of
the process inside the namespace.
Additionally I always wanted to have a way to enable network namespaces
for games as well, so this is a good occasion to make them configurable.
Of course, since we need the mount and user namespaces to implement our
sandbox in the first place, we can't allow users to disable these
namespaces, but for everything else, we now have a new "namespaces"
attribute.
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A recent change[1] has removed the lib output from the systemd package,
so our GnuPG agent wrapper no longer compiles.
Using getLib falls back to the "out" output if the "lib" output is
unavailable and should be backwards- and forwards-compatible (in case
the "lib" output is added back someday).
[1]: https://github.com/NixOS/nixpkgs/commit/b68bddfbda2092c5fde2c4cece2
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Unfortunately, most of the functionality can't be easily exposed via Nix
Flakes, so this will be a very limited Flake with only our NixOS modules
and our packages as an overlay.
One of those things that are very hard to expose are our packaged games,
since the availability of them are dependent on the configuration (eg.
GOG, itch.io or HIB credentials).
Even when it comes to buildSandbox - which is the main reason why I'm
adding a flake.nix - I am not so sure what's the best way to expose it.
For example on one side, this could be exposed as
vuizvui.lib.buildSandbox, which in turn expects a pkgs argument, but on
the other side, we want to have certain other parts that *do* depend on
a locked version of nixpkgs. Using pkgs as part of a "lib" attribute
also sounds a bit weird to me, since lib usually doesn't contain
anything depending on stdenv.
So until we have a better way of integration, I'll just expose the
overlay and our NixOS modules.
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Using "with lib;" over such a broad scope makes it more difficult to
detect early evaluation errors since we need to evaluate the whole
system to check whether a non-existing attribute set is actually using
the "lib" fallback or whether it really exists in the current scope.
This makes "nix-instantiate --parse" feasible for detecting typos early
on.
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
| |
A bit more cleanup.
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is a step towards a saner deployment for these machines.
The NIX_PATH is fixed, so that everything on the running system always
uses exactly the nixpkgs version the system was deployed with.
The deployment is done by copying the system closure and switching to
it via the `bin/switch-to-configuration` script.
Uses the reference to pkgs.path, and applies filterSourceGitignore on
the directory, since I often deploy from a local checkout which is
half a GB without the gitignore filter.
|
|
|
|
| |
Improvements to filterSourceGitignore were made.
|
|
|
|
| |
used for access to /data/seeding
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Since the removal[1] of the Perl test driver, our tests will no longer
run or even evaluate.
Fortunately, the test API is more or less the same, so the transition to
Python was not very involved.
However, I did add a "# fmt: off" on top of every testScript, since
formatting with black not only has issues with parameterised
antiquotations but is also plain ugly to mix 2 spaces of indentation
with 4 spaces of indentation.
Additionally, I'd like to have a maximum line length of 79 characters in
my Nix expressions while black on the other side even *insists* of using
longer lines.
[1]: https://github.com/NixOS/nixpkgs/commit/0620184f3f94f1bf8de014ab168
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
| |
I've grown quite fond of this colorscheme. At least for now.
|
|
|
|
| |
This is really useful.
|
|
|
|
|
| |
At some point compton was renamed to picom, we have an alias for it, but
I feel better if it's using the new name.
|
|
|
|
|
|
| |
Explicitely set default fonts for (sans)serif, monospace and emoji.
Add Noto Color Emoji as emoji font.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
So far we directly start a mosh-daemon when the user connects.
This breaks down in situations where UDP is blocked (e.g. some
hotspots). In that case, ssh can be used directly:
Example:
ssh -t weechat@legosi ssh
The ssh argument tells it to connect directly.
Note the `-t`, which forces a pseudo-tty, otherwise tmux will
complain that it can’t find a terminal.
|
| |
|
| |
|
|
|
|
|
|
|
| |
This has also been removed [1] upstream and therefore needs to be
removed here.
[1]: https://github.com/NixOS/nixpkgs/commit/b2f3bbd3fb958601a7357e39d66f226e065d76c1
|
|
|
|
|
|
|
| |
This package, the module and its test have been removed [1], so to
unbreak Hydra we need to remove it as well.
[1]: https://github.com/NixOS/nixpkgs/commit/aebf9a4709215c230e5841d60e2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The option has been removed upstream[1] and it only has been true by
default for a very small time frame in 2014[2] and I believe even
earlier (before the nixos -> nixpkgs merge) there was another occasion
where it defaulted to true.
However, with the option gone, this is now no longer necessary.
[1]: https://github.com/NixOS/nixpkgs/commit/aebf9a4709215c230e5841d60e2
[2]: https://github.com/NixOS/nixpkgs/commit/b792394119b8ffc4a2fd34a6704
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
|
|
|
|
|
|
|
|
| |
For translations there is a nice web interface called Pontoon[1], which
should it make conventient to edit translations.
However, for developers like me I'd call this quite inconvenient, so I
need a Vim plugin to make it at least more pleasing to look at :-)
[1]: https://github.com/mozilla/pontoon
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
|
|
| |
Similar to aszlig, I have fond memories of these fonts. I've not been
able to use them though because of my work. Maybe these will work
better, and v2 of these fonts will hopefully land in nixpkgs soon too.
|
|
|
|
|
| |
Same look, but written in rust and the binary is smaller too, what's not
to like?
|
|
|
|
|
|
|
|
|
|
|
|
| |
While being at it, let's bump the Nix syntax/indentation plugin to the
latest version, since there are a few quirks that were annoying me since
quite a while but I was too lazy to actually fix.
I haven't checked whether the latest version fixes these quirks since I
don't know them on top of my head, but if it doesn't I surely will
stumble on them soon enough.
Signed-off-by: aszlig <aszlig@nix.build>
|
|
|
|
|
|
|
|
|
|
| |
So far I had termencoding set to "ascii", because my terminal was only
really able to display ASCII characters and nothing else.
Since this is no longer the case, we can also get rid of this
restriction in my Vim configuration.
Signed-off-by: aszlig <aszlig@nix.build>
|