immich is quite demanding which leaves it out of reach if you have one
core (e.g. on a laptop, when my EPYC P core was fine).
Make the test look more like the test for immich.
Generate env.local at runtime so secrets such as APP_SECRET can be provided outside the Nix store.
Assisted-by: pi coding agent / Mika (OpenAI GPT-5.5)
Move mutable public media and uploads out of the Nix store and add nginx hardening for served media files.
Assisted-by: pi coding agent / Mika (OpenAI GPT-5.5)
Configuring a module from withing config is bad style. This makes what
gets configured not introspectable. Worse, it requires special care to
handle priorities correctly, so things will correctly overwrite or merge.
The `GIT_PATH` setting was removed, when Netbox migrated Git operations
to dulwhich. https://github.com/netbox-community/netbox/commit/b9cac97b73c1c08213db6272de11c03711491486
Furthermore, this reduces LDAP to just a single option, because once an
LDAP configuration is set, we can already make assumptions.
With this change secrets will be automatically generated if left at their
individual option defaults.
We now also allow configuration of multiple peppers, to allow proper
rotation as intended by upstream.
Finally, we now we prevent secrets from leaking into the store.
The patches in the vectorchord package are for 1.1.1 only and do not
cleanly apply to 1.0.0. Regardless, the latter does not need them to
successfully build and run.
I have effectively renounced on maintaining all these packages and I do
not plan to return them except if I'm forced to.
I am also fine with most of these packages being dropped for next
releases if no maintainer shows up.
Change-Id: I8d167c8029b6991181bd7a094af21c3313af2b51
Signed-off-by: Raito Bezarius <masterancpp@gmail.com>
While the config supported UNIX domain sockets, the runtime environment
did not. For this to work we need to relax the UMask, which is not great
and will be fixed once Rustical gains socket activation.
- account for the new __hid column introduced in the new version
- fix flakiness of the test
- capture traces in the test output
Signed-off-by: phanirithvij <phanirithvij2000@gmail.com>
