This commit fixes a papercut in nixos-rebuild where people wanting to
switch to a specialisation (or test one) were forced to manually figure
out the specialisation's path and run its activation script - since now,
there's a dedicated option to do just that.
This is a backwards-compatible change which doesn't affect the existing
behavior, which - to be fair - might still be considered sus by some
people, the painful scenario here being:
- you boot into specialisation `foo`,
- you run `nixos-rebuild switch`,
- whoops, you're no longer at specialisation `foo`, but you're rather
brought back to the base system.
(it's especially painful for cases where specialisation is used to load
extra drivers, e.g. Nvidia, since then launching `nixos-rebuild switch`,
while forgetting that you're inside a specialisation, can cause some
parts of your system to get accidentally unloaded.)
I've tried to mitigate that by improving specialisations so that they
create a dedicated file somewhere in `/run/current-system` containing
the specialisation's name (which `nixos-rebuild` could then use as the
default value for `--specialisation`), but I haven't been able to come
up with anything working (plus it would be a breaking change then).
Closes https://github.com/NixOS/nixpkgs/issues/174065
This provides an easy way to specify exclude patterns in config. It was
already possible via extraBackupOptions; this change creates a simpler,
similar to other backup services, way to specify them.
This commit also moves the indicator files out of the directory that's
being backed up, so that the directory remains static throughout the
backup operation.
Added the RFC42-style added the posibility to use
`services.dokuwiki.sites.<name>.settings' instead of passing a plain
string to `<name>.extraConfig`. ´<name>.pluginsConfig` now also accepts
structured configuration.
Also added two "quality of life" tests to ensure customisations to the
dokuiwki package are not being discarded and both webserver
configurations handle rewriting correctly.
As a follow up to f9d1f80045, we should
add the ability to test explicit versions of the wordpress derivation.
Since we are currently only supporting wordpress6_1 in unstable, this
change is a noop.
Updates #209051
The nixOS test failed sporadically with a timeout.
This is due to a race condition in the startup of
the scheduler vs the task-queue.
The scheduler runs the migration scripts in "pre-start" and
celery isn't available, yet. The celery worker (paperless-task-queue)
was already started by systemd but was unable to connect
(as the migration scripts from "pre-start" still ran).
This fix adds the necessary "after" condition in the systemd
worker unit and adds a test to "paperless"
Signed-off-by: Florian Brandes <florian.brandes@posteo.de>
It's better to utilize the boot process and systemd mechanisms to test
these zfs features, rather than manually simulating the same behavior
with testScript.
When test-input-reader runs, it's standard input exists and will
be buffered, so by the time the file exists, the standard input
can already be written to.
I have no reason to believe that a terminal emulator would start
accepting input _after_ launching the command.
I've tested this for hours in a loop without a single failure or
timeout.
This commit upgrades headscale to the newest version, 0.17.0 and updates
the module with the current breaking config changes.
In addition, the module is rewritten to conform with RFC0042 to try to
prevent some drift between the module and the upstream.
A new maintainer, Misterio77, is added as maintainer.
Signed-off-by: Kristoffer Dalby <kristoffer@tailscale.com>
Co-authored-by: Gabriel Fontes <hi@m7.rs>
Co-authored-by: Geoffrey Huntley <ghuntley@ghuntley.com>
On x86_64-linux only because bootspec is for NixOS (for the moment?),
and NixOS is really only a Linux concept (for the moment?).
Not on aarch64-linux because it fails for whatever reason 🤷
Nginx breaks at runtime when duplicate modules are added. To detect
this, add a `name` key to all modules.
Also remove the outdated modsecurity v2 module and unify `modsecurity`
and `modsecurity-nginx`.
We can't actually get metrics for a virtual disk drive so the exporter
fails to start with 0.9.x.
Instead let's just make sure it said that /dev/vda was unavailable.
This splits the tests into two: one where cups.socket is started
normally, the order with socket activation.
Why? It's almost impossible to follow the test with 4 different
machines printing at the same time. It should also be more efficient
because only two VMs at a time were needed anyway.
Adds a new option for backup jobs `inhibitsSleep` which prevents
the system from going to sleep while a backup is in progress.
Uses `systemd-inhibit`, which holds a "lock" that prevents the
system from sleeping while the process it invokes is running.
This did require wrapping the existing backup script using
`writeShellScript` so that it could be run by `systemd-inhibit`.
Changes sgx-psw to append `aesm` to `LD_LIBRARY_PATH`:
- Append instead of prepend to allow for overriding in service config
- As we already add a wrapper to add `aesm` to `LD_LIBRARY_PATH` it is
not necessary to also set in `LD_LIBRARY_PATH` of the systemd service.
Co-authored-by: Vincent Haupert <mail@vincent-haupert.de>
- The default cipher is BF-CBC, which openvpn refuses to use by default.
Switched to AES-256-CBC.
- openvpn does not require an external "ip" executable anymore, and does
not support the "ipconfig" option by default, so remove that option.
This is a follow up to #200815 and #184634.
The PCRE2 JIT SEAlloc does not support the `fork()` as announced in
their README [0]:
> If you are enabling JIT under SELinux environment you may also want to add
> --enable-jit-sealloc, which enables the use of an executable memory allocator
> that is compatible with SELinux. Warning: this allocator is experimental!
> It does not support fork() operation and may crash when no disk space is
> available. This option has no effect if JIT is disabled.
As a result using it in PHP can break apps and tools, it can only be
enabled under very specific context where you have a full picture of
what the PHP code is doing.
This contribution disables again the PCRE2 JIT SEAlloc and extends the
existing PHP/PCRE2 tests to make sure we do not enable it again by
mistake.
[0] https://www.pcre.org/readme.txt
remove trailing whitespace
switch docs to markdown
use mdDoc
remove trailing whitespace
get rid of double space
add tests and update options to use submodule
remove whitespace
remove whitespace
use mdDoc
remove whitespace
make default a no-op
make ALTER ROLE a single sql statement
document null case
The tests TLS setup was bogus: the xmpp-send-message script was trying
to connect to the server through a bogus domain name. Injecting the
right one.
I'm a bit confused about that one. I know for sure this NixOS test
succeeded last time I checked it, but the TLS conf is bogus for sure.
I assume the slixmpp SNI validation was a bit too loose and was
tightened at some point.
The xmpp-sendmessage the slixmpp-powered python script tend to timeout
and block the nixos channels.
Adding a signal-based timeout making sure that whatever happens, the
script won't run for more than 2 minutes. That should be pleinty
enough time to finish regardless of the runner specs. As a data point,
it runs in about 10 secs on my desktop machine.
The hack with `either` had the side-effect that the sub-options of the
submodule didn't appear in the manual. I decided to remove this because
the "migration" isn't that hard, you just need to fix some module
declarations.
However, `mkRenamedOptionModule` wouldn't work here because it'd create
a "virtual" option for the deprecated path (i.e.
`services.grafana.provision.{datasources,dashboards}`), but that's the
already a new option, i.e. the submodule for the new stuff.
To make sure that you still get errors, I implemented a small hack using
`coercedTo` which throws an error if a list is specified (as it would be
done on 22.05) which explains what to do instead to make the migration
easier.
Also, I linkified the options in the manual now to make it easier to
navigate between those.
This commit fixes broken non-declarative configs by
making the assertions more relaxed.
It also allows to remove the forced configuration merge by making
`settings` `null`able (now the default).
Both cases (trivial non-declarative config and `null`able config) are
verified with additional tests.
Fixes#198665
fscrypt can automatically unlock directories with the user's login
password. To do this it ships a PAM module which reads the user's
password and loads the respective keys into the user's kernel keyring.
Significant inspiration was taken from the ecryptfs implementation.
Upon testing the change itself I realized that it doesn't build properly
because
* the `pname` of a php extension is `php-<name>`, not `<name>`.
* calling the extension `openssl-legacy` resulted in PHP trying to compile
`ext/openssl-legacy` which broke since it doesn't exist:
source root is php-8.1.12
setting SOURCE_DATE_EPOCH to timestamp 1666719000 of file php-8.1.12/win32/wsyslog.c
patching sources
cdToExtensionRootPhase
/nix/store/48mnkga4kh84xyiqwzx8v7iv090i7z66-stdenv-linux/setup: line 1399: cd: ext/openssl-legacy: No such file or directory
I didn't encounter that one before because I was mostly interested in
having a sane behavior for everyone not using this "feature" and the
documentation around this. My findings about the behavior with turning
openssl1.1 on/off are still valid because I tested this on `master` with
manually replacing `openssl` by `openssl_1_1` in `php-packages.nix`.
To work around the issue I had to slightly modify the extension
build-system for PHP:
* The attribute `extensionName` is now relevant to determine the output
paths (e.g. `lib/openssl.so`). This is not a behavioral change for
existing extensions because then `extensionName==name`.
However when specifying `extName` in `php-packages.nix` this value is
overridden and it is made sure that the extension called `extName` NOT
`name` (i.e. `openssl` vs `openssl-legacy`) is built and installed.
The `name` still has to be kept to keep the legacy openssl available
as `php.extensions.openssl-legacy`.
Additionally I implemented a small VM test to check the behavior with
server-side encryption:
* For `stateVersion` below 22.11, OpenSSL 1.1 is used (in `basic.nix`
it's checked that OpenSSL 3 is used). With that the "default"
behavior of the module is checked.
* It is ensured that the PHP interpreter for Nextcloud's php-fpm
actually loads the correct openssl extension.
* It is tested that (encrypted) files remain usable when (temporarily)
installing OpenSSL3 (of course then they're not decryptable, but on a
rollback that should still be possible).
Finally, a few more documentation changes:
* I also mentioned the issue in `nextcloud.xml` to make sure the issue
is at least mentioned in the manual section about Nextcloud. Not too
much detail here, but the relevant option `enableBrokenCiphersForSSE`
is referenced.
* I fixed a few minor wording issues to also give the full context
(we're talking about Nextcloud; we're talking about the PHP extension
**only**; please check if you really need this even though it's
enabled by default).
This is because I felt that sometimes it might be hard to understand
what's going on when e.g. an eval-warning appears without telling where
exactly it comes from.
Previously we did socket-activation but this breaks the autostart
feature since upstream expects libvirtd to be started unconditionally on
boot.
Fixes#171623.
* s/NextCloud/Nextcloud/g
* `enableBrokenCiphersForSSE` should be enabled by default for any NixOS
installation from before 22.11 to make sure existing installations
don't run into the issue. Not the other way round.
* Update release notes to reflect on that.
* Improve wording of the warning a bit: explain which option to change
to get rid of it.
* Ensure that basic tests w/o `enableBrokenCiphersForSSE` run with
OpenSSL 3.
I need to fix copying the chrome://gpu content to the clipboard (Ctrl+a doesn't
work anymore so we have to click the button) but we can at least test the font
rendering for now.
This is a small smoke test of each piece (setuid, setgid, caps) of
wrappers' functionality. It doesn't try to check for combinations of
functionalities or anything more complicated.
The new defaults allows jenkins-job-builder to reload the configuration
out-of-the-box, whereas the previous defaults required users to manually
reload/restart jenkins, or configure accessUser/accessTokenFile
themselves.
(If `extraJavaOptions = [ "-Djenkins.install.runSetupWizard=false" ]`
then the initial admin user is *not* created and you have to use JCasC
or something else to bootstrap.)
This corrects the multi-node test after a couple recent changes which
resulted in it being broken.
The `lib.toString` change was an incorrect tree-wide refactor, and the
aarch64 change also introduced an error in python indentation/formatting
I believe.
- Fix hostname configuration on proxmox, which uses "hostname" in user-data
instead of "local-hostname" in meta-data.
- Allow setting resolv.conf through cloud-init
- Add tests for new changes
- Add timeouts to make tests fail faster
This will add `passthru.schema_version` to be used as default value for
the adguardhome module.
It will also update the `update.sh` to keep the `schema_version` in sync
with the version by inspecting the sourcecode.
This might break existing configs, if they use deprecated values that don't
appear in newer schema_versions and schema_version wasn't set explicitly.
Explicit declarations of schema_version always have higher priority.
This also removes the `host` and `config` settings in favour of using the
appropriate `settings`.
Fixes#173938
Co-authored-by: Sandro <sandro.jaeckel@gmail.com>
I haven't had time to look into this yet but it looks like opening chrome://gpu
doesn't work anymore without proper GPU rendering (we use software rendering
due to the virtualisation).
According to the console output the new window never opens (at least with
Google Chrome - I couldn't test it with Chromium yet due to the failing builds
for M107 and M108):
```
(finished: sending keys ‘chrome://gpu
‘, in 0.14 seconds)
machine: waiting for a window to appear
machine: must succeed: xwininfo -root -tree | sed 's/.*0x[0-9a-f]* \"\([^\"]*\)\".*/\1/; t; d'
(finished: must succeed: xwininfo -root -tree | sed 's/.*0x[0-9a-f]* \"\([^\"]*\)\".*/\1/; t; d', in 0.05 seconds)
machine # Error: eglChooseConfig returned zero configs
machine # at Create (../../third_party/dawn/src/dawn/native/opengl/ContextEGL.cpp:53)
machine #
machine: must succeed: xwininfo -root -tree | sed 's/.*0x[0-9a-f]* \"\([^\"]*\)\".*/\1/; t; d'
machine # WARNING: lavapipe is not a conformant vulkan implementation, testing use only.
(finished: must succeed: xwininfo -root -tree | sed 's/.*0x[0-9a-f]* \"\([^\"]*\)\".*/\1/; t; d', in 0.06 seconds)
machine: must succeed: xwininfo -root -tree | sed 's/.*0x[0-9a-f]* \"\([^\"]*\)\".*/\1/; t; d'
(finished: must succeed: xwininfo -root -tree | sed 's/.*0x[0-9a-f]* \"\([^\"]*\)\".*/\1/; t; d', in 0.09 seconds)
[...]
```
The meta attribute "timeout" is only set for Chromium (might still be required
due to the long build duration). The Google Chrome tests were failing with:
error: attribute 'timeout' missing
According to nixos/lib/testing/meta.nix "null values are filtered out by
`meta`" so `timeout = chromiumPkg.meta.timeout or null` might be fine as
well.
This commit refactors `services.grafana.provision.datasources` towards
the RFC42 style. To preserve backwards compatibility, we have to jump
through a ton of hoops, introducing esoteric type signatures and bizarre
structs. The Grafana module definition should hopefully become a lot
cleaner after a release cycle or two once the old configuration style is
completely deprecated.
This commit refactors `services.grafana.provision.dashboards` towards
the RFC42 style. To preserve backwards compatibility, we have to jump
through a ton of hoops, introducing esoteric type signatures and bizarre
structs. The Grafana module definition should hopefully become a lot
cleaner after a release cycle or two once the old configuration style is
completely deprecated.
