| Age | Commit message (Collapse) | Author |
|---|
|
... with release-small.nix
Tests for small channel should be a subset of tests for full channel.
|
|
This reverts commit 1d0625499854b583c57267a744111ba8a1d0cfaf.
We now get a build and maybe have a workaround #39570; /cc #39476.
|
|
The chromium package will frequently time out . Note that this is unrelated to the test but a side effect because the test depends on chromium.
https://hydra.nixos.org/job/nixos/trunk-combined/nixos.tests.chromium.x86_64-linux
See also #39476
|
|
... in supportedSystems
|
|
Currently, when building NixOS from a git clone, Nix has to copy
the entire repo at >1GB into the store by default. That is not
necessary and causes a dumping large path message.
If you need the old behaviour for some reason, you will have to
specify it by passing the path to your repo explicitly as the
nixpkgs argument like this:
--arg nixpkgs '{outPath = ./.; revCount = 56789; shortRev = "gfedcba"; }'
|
|
In practice it already is since the channel update script barfs out if
the job fails:
https://github.com/NixOS/nixos-channel-scripts/blob/9f96e1e25d197de46d5dbe3c4f81b8d57b9c9821/mirror-nixos-branch.pl#L132
|
|
The existing callSubTests seems to already have special-cased code to
allow enabling subtests on a single specific system by looking at the
`system` attribute in the test arguments. Replace it with a new version
similar to the callTestOnTheseSystems because:
- It's consistent with the existing functions for creating
system-specific tests (though admittedly, the callSubTests special
case for `system` predates them)
- This approach allows limiting to multiple system types, the previous
one inherently allows only one system type.
- This also fixes the problem that if you pass in e.g.
supportedSystems = [ "aarch64-linux" ], you end up with a
tests.chromium job that silently runs on x86_64-linux.
- Finally, this causes renames of the jobs like:
tests.chromium -> tests.chromium.x86_64-linux to be consistent with
the rest of the tests.
|
|
Currently, even if you pass supportedSystems = [ "aarch64-linux" ] you
end up with e.g. `nixos.tests.docker` which actually silently runs on
x86_64-linux. Using the new callTestOnTheseSystems fixes that.
As a side-effect, this also causes a rename of
`nixos.tests.docker` -> `nixos.tests.docker.x86_64-linux`, which is IMHO
a good thing since it's makes them consistent with the rest of the
tests.
|
|
nixos/system: make switch-to-configuration script pure
|
|
nixos/tests: add a test for config.users.mutableUsers.
|
|
|
|
|
|
|
|
Lately failing i686 tests like firefox have been blocking channel
releases. We're still building the tests for systems with limited
support but won't delay a channel release if they fail.
|
|
Arguably, breaking linux-latest should not block a release. Also, booting
the kernel + basic sanity checking is implicitly exercised by every other
vm test.
|
|
|
|
|
|
- sysctl is new and never succeeded on i686-linux
> cannot stat /proc/sys/net/core/bpf_jit_enable: No such file or directory
- testing plasma5 on i686 would defeat part of the reason why we ended
supporting i686 (lots of stuff built on Hydra)
|
|
|
|
|
|
More basic system tests
|
|
docker: Package in module, tests
|
|
Add XMonad test
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Seems to be working fine, based on the history:
https://hydra.nixos.org/job/nixos/trunk-combined/nixos.tests.installer.simpleUefiGummiboot.i686-linux/all
https://hydra.nixos.org/job/nixos/trunk-combined/nixos.tests.installer.simpleUefiGummiboot.x86_64-linux/all
|
|
|
|
|
|
|
|
We really want to break channel updates whenever we break something like
this, because this actually will hit machines out there and can be very
much annoying (we had broken keymaps a few times which is why I
introduced these tests in the first place).
Just to be sure I don't break channel updates with this commit, I ran
all of the keymap tests and they all succeeded.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
This is currently our default display manager, so I'm adding this to the
"tested" job as well to ensure we don't ship broken revisions where X is
most likely not working.
The test uses a custom SLiM theme that's specifically tailored for good
OCR results (mainly white background and black fonts without anything
else), because our default NixOS theme has a very small contrast between
background and fonts in some places.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
Fixes #24240.
I do not know the cause of the test failures, but the tests pass now.
|
|
... to get a channel bump after a long time; see #24240.
|
|
This reverts commit b806e25d65421d7aa0a524ce1601f0e51099df11.
This seems to push Hydra's memory usage out of the roof fail nixos
evaluating with:
Too many heap sections: Increase MAXHINCR or MAX_HEAP_SECTS
Let's revert this for now. It's not a big deal at all since the
nixpkgs-unstable jobset is still building the packages.
|
|
|
|
NixOS: Plasma 5 tests and warnings
|
|
|
|
/cc #23107.
|
|
Excluding modules/programs/environment.nix for PATHand QT_PLUGIN_PATH to allow the programs to continue running.
|
|
(cherry picked from commit 701f02a6eed902835e3d2a27838723cd9a7bb66a)
|
|
We can probably drop NFSv3...
|
|
This reverts commit 9e6eec201b8b56ef493c60af3acb2e8ddd8be258.
Verified locally that cd063d7 fixes the problem.
cc @edolstra @aszlig
|
|
It's hanging and blocking the unstable channel.
http://hydra.nixos.org/build/38984676
@aszlig
|
|
|
