| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
| |
This reverts commit 71d24b28c16fbae58d1ecfeaecf5a26c7ec015da.
Evaluation has been fixed upstream at NixOS/nixpkgs@c4de45b and the test
should succeed after NixOS/nixpkgs@5caa8ab as well.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
The Quake 3 demo data have been marked as "unfreeRedistributable" in
NixOS/nixpkgs@6b447a3, which causes this test to not evaluate unless we
set config.allowUnfree to true.
I guess the best solution to fix this upstream is to use something like
OpenArena for the Quake 3 tests, because I didn't manage to find
anything regarding the demo files and/or whether those are
redistributable or not.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
I've added that test in NixOS/nixpkgs@4c61faa and it fails at the
moment, still leaving the channels in a broken state in regards to
networkd.
The reason I'm doing this anyway is to ensure I won't forget to add that
test here as well. We can't easily go back to the "not broken" state
anymore, except if we roll back to a version with systemd < 228.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
|
|
|
|
|
| |
This was just pushed a few minutes ago as commit
NixOS/nixpkgs@565707c57abc6ee9e81d4f6c9bd5fed5bbac2d40.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Using an equality test on an attribute set is actually a very bad idea,
because it forces strict evaluation.
In this case a simple test on the kernel version between the defined one
and the available one should be sufficient.
And if we have a false-positive in test inclusions it doesn't hurt
anyway.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
|
|
|
|
|
|
|
|
| |
This time I've gone through the hoops of almost running out of memory,
because evaluating the channel expression now eats lots of RAM.
Meaning: I've now actually tested it on my system without just throwing
it at the Hydra box which has lots of RAM... unlike my local machine :-(
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
|
|
|
|
| |
Forgot to include pkgs to module args.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
|
|
|
|
|
|
|
|
|
| |
I've added all tests that are not related to packages in
environment.systemPackages or are otherwise tricky to include or to
distinguish (for example NFSv3 vs. NFSv4).
So we now should have a pretty good test coverage for all of our
machines.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
|
|
|
|
|
| |
That way we have everything regarding the vuizvui testing system in one
module instead of two.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
|
|
|
|
|
| |
We should emit the actual derivations but instead just the main
attribute "nixos.networking" was added to vuizvui.requiredTests.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
|
|
|
|
|
|
|
|
|
| |
Unfortunately, we went into trouble and the Hydra machine not only
consumed lots of ram during evaluation but it also triggered a bunch of
evaluation errors because the available tests were strictly evaluated.
Using attribute paths consisting of plain strings doesn't have the same
problem, even though they look a bit uglier.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
This is basically just to get started and for testing whether our idea
will actually work out as planned. My gut feeling is that Hydra
explicitly needs jobs for these tests, but we'll see.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
