The text was originally added [0] following an apparently incomplete
research on how everything plays together. In fact, Nix propagates
`outputs` to the corresponding nested derivations, and there is some
messy behavior in Nixpkgs that only seems to propagate
`meta.outputsToInstall` in `buildEnv`[1].
This change moves the hints on how to use NixOS specifics to NixOS
module documentation (which is hopefully easier to find through
search.nixos.org), describes the default behavior in Nixpkgs (updating
a the link to the source), and removes the confusing mention of
`nix-env`.
the last of them should not be there to begin with. we don't want
beginners to use `nix-env`, as this is known to run them into trouble
eventually.
[0]: https://github.com/NixOS/nixpkgs/pull/76794
[1]: 1774d07242/pkgs/build-support/buildenv/default.nix (L66)
Since garage 0.8.2, garage accepts environment variables for passing secrets,
e.g. `GARAGE_RPC_SECRET` or `GARAGE_ADMIN_TOKEN`. The added `environmentFile`
allows those secrets to not be present in the nix store.
Update wg-quick.nix such that a search for `WireGuard` in the `NixOS Options` section of search.nixos.org brings up the convenient `networking.wg-quick.interfaces.wg0.configFile` option.
While network.target only guarantees that network devices have been
created the `network-online.target` allows delaying service startup
until after a configurable network state has been reached.
This should resolve spurious failures, e.g. when synapse tries to load
the discovery information for its OIDC provider from a remote host.
This is just a quick fix based on pname,
as I have no idea how to use slicing in the module
We should instead use slicing to get the package for the host
nix-prefetch-git is either run as part of a build, usually sandboxed,
or outside a build, unsandboxed, to prefetch something that will later
be used in a build. It's important that the latter use produces
hashes that can be reproduced by the former.
One way that they can differ is if the user's git config does
something that changes the result of git clone. I ran into this,
because my global git config automatically enables git-lfs, whereas
nix-prefetch-git otherwise only uses git-lfs if specifically
requested. This led to very confusing hash mismatches.
A couple notes:
---------------
Adding invalid `console=` parameters is not an issue. Any invalid
console is unused. The kernel will use the "rightmost" (last) valid
`console=` parameter as the default output. Thus the SBBR-mandated AMA0
on A1, and ttyS0 on x86_64 as documented by Oracle.
`nvme_core.shutdown_timeout=10` was added as it was written this way in
the A1 images. Unclear whether `nvme.shutdown_timeout=10` is wrong. At
worst this is a no-op.
This new option, networking.wireguard.interfaces.NAME.metric, allows
increasing the metric of the routes, effectively lowering priority.
(I'm using high metric to allow having the Wireguard interface always
up, even when the client machines are on their home network. Before I
had to stop the interface when home to avoid packet routing issues.)
While working on #192270, I noticed that only some wait_for_* helper
functions make the timeout configurable. I think we should be able to
customize it in all cases
While reviewing other changes related to synapse I rediscovered the
`lib.findFirst (...) (lib.last resources)` hack to find a listener
supporting the `client` resource. We decided to keep it that way for now
a while ago to avoid scope-creep on the RFC42 refactoring[1]. I wanted
to take care of that and forgot about it.
Anyways, I'm pretty sure that this is bogus: to register a user, you
need the `client` API and not a random listener which happens to be the
last one in the list. Also, you need something which serves the `client`
API to have the entire synapse<->messenger interaction working (whereas
`federation` is for synapse<->synapse).
So I decided to error out if no `client` listener is found. A listener
serving `client` can be defined in either the main synapse process or
one of its workers via `services.matrix-synapse.workers`[2].
However it's generally nicer to use assertions for that because then
it's possible to display multiple configuration errors at once and one
doesn't have to chase one `throw` after another. I decided to also error
out when using the result from `findFirst` though because module
assertions aren't thrown necessarily when you evaluate a single config
attribute, e.g. `config.environment.systemPackages` which depends on an
existing client listener because of `registerNewMatrixUser`[3].
While at it I realized that if `settings.instance_map` is wrongly
configured, e.g. by
settings.instance_map = mkForce {
/* no `main` in here */
}
an `attribute ... missing` error will be thrown while evaluating the
worker assertion.
[1] https://github.com/NixOS/nixpkgs/pull/158605#discussion_r815500487
[2] This also means that `registerNewMatrixUser` will still work if you
offload the entire `client` traffic to a worker.
[3] And getting a useful error message is way better for debugging in such a
case than `value is null while a set was expected`.
