fix(#2860): make hyperhive.forge.url nullable instead of guessing a URL

The option had a `http://localhost:3000` default, which is only ever
correct when the forge shares the caller's network namespace — inside an
agent's netns `localhost` is the agent, and the forge may well be on
another host. Making it *required* instead was worse: the flake's own
container configs are what hive-c0re extends per agent, so the value
they needed in order to evaluate became a second definition on every
agent and collided with the real one.

`null` resolves both. It is not a URL, so nothing can quietly talk to
the wrong machine, and it needs no placeholder anywhere: the bases
evaluate as they are, so nothing deployment-shaped sits on the config
agents inherit from. The units that would consume the URL — tea-login
and forge-avatar-sync — are simply not generated without one, making an
absent forge an absent integration rather than a misdirected one.

hive-forge-notify is unaffected: it reads HIVE_FORGE_URL from the
forwarded global environment, not from this option.

Verified:
  agent-base/ruth evaluate with forge.url = null, zero failing assertions
  bare base:      tea_login_present = false, avatar_present = false,
                  notify_present = true
  extended with a rendered URL: FORGE_URL=http://forge.real.test

Refs #2860
This commit is contained in:
atlas 2026-08-01 00:36:09 +02:00
commit bcb9e837f7
2 changed files with 85 additions and 84 deletions

View file

@ -149,34 +149,20 @@
nixosConfigurations =
let
# Values the agent modules require but that only a real
# deployment can know. Real containers are built from the
# generated meta flake, where hive-c0re renders these per
# agent from the host's `HIVE_FORGE_URL` (see meta.rs's
# `SERVICE_URL_OPTIONS`) — they never evaluate through
# `self.nixosConfigurations`, so nothing here can reach a
# running agent. These two configs exist only to typecheck
# the modules and to pre-build the container closure
# (`system.extraDependencies`, see hive-c0re/default.nix).
#
# Deliberately a `.invalid` host (RFC 2606: guaranteed not to
# resolve) rather than something plausible like a loopback
# port. If this value ever *did* escape into a runtime path,
# it must fail loudly at DNS instead of quietly connecting to
# whatever happens to be listening — which is the entire
# point of removing the `http://localhost:3000` default this
# replaces.
evalOnlyPlaceholders = {
hyperhive.forge.url = "http://forge.invalid";
};
# These two configs are what hive-c0re extends per agent
# (meta.rs's `mkAgent` does `<base>.extendModules { … }`, so
# the built base and every container stay on one version).
# Nothing deployment-specific may be set here: it would be
# inherited by every container on the hive and collide with
# the per-agent values hive-c0re renders. The agent modules
# are written so a bare evaluation needs no such values —
# service URLs default to `null`, meaning "not configured",
# and the units that would use them simply aren't generated.
mkContainer =
module:
nixpkgs.lib.nixosSystem {
system = "x86_64-linux";
modules = [
module
evalOnlyPlaceholders
];
modules = [ module ];
};
in
{

View file

@ -20,7 +20,8 @@ let
in
{
options.hyperhive.forge.url = lib.mkOption {
type = lib.types.str;
type = lib.types.nullOr lib.types.str;
default = null;
example = "http://forge.internal:3000";
description = ''
Base URL of the hyperhive-managed Forgejo. Used at container
@ -31,28 +32,38 @@ in
forge-token file is missing (i.e. hive-forge isn't running on
the host).
**Required, deliberately undefaulted.** hive-c0re renders it into
every agent's config from the host's `HIVE_FORGE_URL`, which
`hive-c0re.nix` sets unconditionally --- the forge is mandatory.
A loopback default would be a guess: the forge may run on a
different host from the agents, and inside an agent's network
namespace `localhost` reaches the agent, not the forge. An
unevaluatable config is better than one that builds and then
talks to the wrong machine.
**`null` means "no forge", not "guess one".** There is deliberately
no loopback default: the forge may run on a different host from
the agents, and inside an agent's network namespace `localhost`
reaches the agent rather than the forge, so a default would be a
value that builds fine and then talks to the wrong machine.
When this is `null` the tea-login and avatar-sync units are not
generated at all --- an absent integration, never a misdirected
one.
On a real hive it is always set: hive-c0re renders it into every
agent's config from the host's `HIVE_FORGE_URL` (which
`hive-c0re.nix` sets unconditionally) and refuses to render a meta
flake without it, so `null` only survives where the modules are
evaluated outside a hive --- exactly the case that has no forge.
'';
};
config = {
assertions = [
# The empty string is the one value the type permits that cannot
# be a URL, and it is what a caller supplies when they have
# nothing --- exactly the case the removed loopback default used
# to paper over. Reject it here so the failure names the option.
# Only a *set* value is constrained. `null` is the legitimate
# "no forge here" state (see the option doc) and is handled by
# not generating the units below, so it must not trip this.
# The empty string, by contrast, is the one non-null value the
# type permits that cannot be a URL --- it is what a caller
# supplies when they have nothing, which is precisely what `null`
# is for, so reject it and name the option.
{
assertion =
lib.hasPrefix "http://" config.hyperhive.forge.url
config.hyperhive.forge.url == null
|| lib.hasPrefix "http://" config.hyperhive.forge.url
|| lib.hasPrefix "https://" config.hyperhive.forge.url;
message = "hyperhive.forge.url must be an http:// or https:// URL (got: \"${config.hyperhive.forge.url}\")";
message = "hyperhive.forge.url must be an http:// or https:// URL, or null for no forge (got: \"${toString config.hyperhive.forge.url}\")";
}
];
@ -105,7 +116,9 @@ in
# One-shot: tea config.yml from the seeded forge token. Shape
# contract (always exit 0, no set -e, skip-silently, re-runnable):
# docs/conventions.md::Best-effort oneshot services.
systemd.services.tea-login = {
# Not generated at all when no forge is configured: an absent
# integration rather than one pointed at a guessed address.
systemd.services.tea-login = lib.mkIf (config.hyperhive.forge.url != null) {
description = "configure tea CLI from hive-forge token (best-effort)";
wantedBy = [ "multi-user.target" ];
after = [ "local-fs.target" ];
@ -188,49 +201,51 @@ in
# avatar sync), so the unit only exists when an icon is configured
# and needs no librsvg at runtime — Forgejo's Go image library
# can't decode SVG, hence PNG.
systemd.services.forge-avatar-sync = lib.mkIf (config.hyperhive.icon != null) {
description = "sync agent icon to Forgejo user avatar (best-effort)";
wantedBy = [ "multi-user.target" ];
after = [ "tea-login.service" ];
serviceConfig = {
Type = "oneshot";
RemainAfterExit = false;
# Pin the journal identity (else it's the `script` store-path wrapper).
SyslogIdentifier = "forge-avatar-sync";
};
path = [
pkgs.curl
pkgs.coreutils
pkgs.jq
];
script = ''
FORGE_URL=${lib.escapeShellArg config.hyperhive.forge.url}
# $HYPERHIVE_STATE_DIR is set system-wide by the meta flake
# (systemd.globalEnvironment) to `/agents/<name>/state`.
TOKEN_FILE="$HYPERHIVE_STATE_DIR/forge-token"
if [ ! -f "$TOKEN_FILE" ]; then
echo "forge-avatar-sync: no forge-token found; skipping"
exit 0
fi
TOKEN=$(cat "$TOKEN_FILE")
IMAGE=$(base64 -w 0 < ${iconPng})
# Forgejo POST /user/avatar expects {"image":"<base64>"} — just the
# raw base64 string, NOT a data URI (data:image/png;base64,...).
# Use jq to build the payload so the large base64 value is safely quoted.
PAYLOAD=$(jq -n --arg img "$IMAGE" '{image:$img}')
RESP=$(curl -sf --max-time 10 \
-X POST "$FORGE_URL/api/v1/user/avatar" \
-H "Authorization: token $TOKEN" \
-H "Content-Type: application/json" \
-d "$PAYLOAD" \
-w "\n%{http_code}" 2>/dev/null || true)
CODE=$(printf '%s' "$RESP" | tail -1)
if [ "$CODE" = "204" ] || [ "$CODE" = "200" ]; then
echo "forge-avatar-sync: avatar uploaded (HTTP $CODE)"
else
echo "forge-avatar-sync: upload returned HTTP $CODE skipping (non-fatal)"
fi
'';
};
systemd.services.forge-avatar-sync =
lib.mkIf (config.hyperhive.icon != null && config.hyperhive.forge.url != null)
{
description = "sync agent icon to Forgejo user avatar (best-effort)";
wantedBy = [ "multi-user.target" ];
after = [ "tea-login.service" ];
serviceConfig = {
Type = "oneshot";
RemainAfterExit = false;
# Pin the journal identity (else it's the `script` store-path wrapper).
SyslogIdentifier = "forge-avatar-sync";
};
path = [
pkgs.curl
pkgs.coreutils
pkgs.jq
];
script = ''
FORGE_URL=${lib.escapeShellArg config.hyperhive.forge.url}
# $HYPERHIVE_STATE_DIR is set system-wide by the meta flake
# (systemd.globalEnvironment) to `/agents/<name>/state`.
TOKEN_FILE="$HYPERHIVE_STATE_DIR/forge-token"
if [ ! -f "$TOKEN_FILE" ]; then
echo "forge-avatar-sync: no forge-token found; skipping"
exit 0
fi
TOKEN=$(cat "$TOKEN_FILE")
IMAGE=$(base64 -w 0 < ${iconPng})
# Forgejo POST /user/avatar expects {"image":"<base64>"} — just the
# raw base64 string, NOT a data URI (data:image/png;base64,...).
# Use jq to build the payload so the large base64 value is safely quoted.
PAYLOAD=$(jq -n --arg img "$IMAGE" '{image:$img}')
RESP=$(curl -sf --max-time 10 \
-X POST "$FORGE_URL/api/v1/user/avatar" \
-H "Authorization: token $TOKEN" \
-H "Content-Type: application/json" \
-d "$PAYLOAD" \
-w "\n%{http_code}" 2>/dev/null || true)
CODE=$(printf '%s' "$RESP" | tail -1)
if [ "$CODE" = "204" ] || [ "$CODE" = "200" ]; then
echo "forge-avatar-sync: avatar uploaded (HTTP $CODE)"
else
echo "forge-avatar-sync: upload returned HTTP $CODE skipping (non-fatal)"
fi
'';
};
};
}