fix(nix): require swarm domain only when a hyperhive service is enabled

The swarm.domain assertion in hive-network.nix fired on every host that
imported the module, so a host that enables nothing failed eval. It now
fires only when one of the hyperhive service switches is on (every
deploy.*.enable that runs something, gateway, gateway.dns, network,
otel, snapshotStore). The requirement itself is unchanged: any host that
runs a hyperhive service still needs swarm.domain.

The core-toggle module-eval suite gains a case: a missing swarm.domain is
refused on a hive and on a swarm-service-only host, and a host enabling
nothing passes every assertion.

Closes #4887

docs/swarm/README.md

@@ -310,8 +310,9 @@ services.hyperhive = {
 <!-- vale write-good.Passive = NO -->
 
 Swarm options are identical on every host in the swarm, so `swarm.domain` is
-**required** on every host; `hiveName` is required on a host that runs a
-hive, the secret store or the homeserver. Eval fails with a hint naming
+**required** on every host that runs any hyperhive service; a host that
+imports the module and enables nothing evaluates without it. `hiveName` is
+required on a host that runs a hive, the secret store or the homeserver. Eval fails with a hint naming
 each. Neither defaults, because a guessed value here is a wrong hostname
 that evaluates cleanly and deploys.
 

nix/host-modules/hive-network.nix

@@ -27,6 +27,38 @@ let
   hostCount = pow2 (32 - cfg.bridgePrefixLength);
   # Mask off host bits to get the network base address.
   networkBase = builtins.bitAnd (ipToInt cfg.bridgeIp) (4294967295 - hostCount + 1);
+
+  # Every switch that runs a hyperhive service on this host. Left out:
+  # `github.enable` (on by default, runs nothing by itself), the gateway's
+  # and matrix's sub-features, and `allSwarmServices`/`singleHostSwarm`,
+  # which only take effect through the switches listed here.
+  anyServiceEnabled =
+    let
+      h = config.services.hyperhive;
+      d = h.deploy;
+    in
+    lib.any lib.id [
+      d.hive-controller.enable
+      d.swarm-controller.enable
+      d.swarm-ui.enable
+      d.forgejo.enable
+      d.forgejo.ci.enable
+      d.matrix.enable
+      d.authelia.enable
+      d.bao.enable
+      d.nats.enable
+      d.swarm-otel.enable
+      d.grafana.enable
+      d.victoriametrics.enable
+      d.victorialogs.enable
+      d.swarm-secret-publisher.enable
+      d.wireguard.enable
+      h.gateway.enable
+      h.gateway.dns.enable
+      h.network.enable
+      h.otel.enable
+      h.snapshotStore.enable
+    ];
 in
 {
   # Hive-internal network — the host-side bridge every container in a
@@ -187,25 +219,26 @@ in
       ];
     })
 
-    # Un-gated: the swarm's options are identical on every host, so a host
-    # running no hive and no swarm service needs the value too. The
-    # `.invalid` fallbacks described above keep this message reachable.
-    {
+    # Gated on any hyperhive service, not only the hive: the swarm's options
+    # are identical on every host, so a host running only a swarm service
+    # needs the value too. The `.invalid` fallbacks described above keep
+    # this message reachable.
+    (lib.mkIf anyServiceEnabled {
       assertions = [
         {
           assertion = config.services.hyperhive.swarm.domain != null;
           message = ''
             hyperhive requires services.hyperhive.swarm.domain to be set
-            on every host, to the same value on every host of the swarm —
-            the DNS domain of the swarm, of which each hive occupies one
-            sub-domain. A guessed value would be a wrong hostname that
-            evaluates cleanly and deploys. Set it
+            on every host that runs a hyperhive service, to the same value
+            on every host of the swarm — the DNS domain of the swarm, of
+            which each hive occupies one sub-domain. A guessed value would
+            be a wrong hostname that evaluates cleanly and deploys. Set it
             (`services.hyperhive.swarm.domain = "example.com";`) — with
             `hiveName` it also derives `services.hyperhive.domain` for you.
           '';
         }
       ];
-    }
+    })
 
     # The bridge itself, up only where something hangs off it. Private
     # netns is still the only container mode.

nix/host-modules/hyperhive.nix

@@ -117,8 +117,10 @@ in
       `services.hyperhive.hiveName`, list the hives by name, and no
       hive in the swarm states an address at all.
 
-      **Required** on every host that imports this module, with the
-      same value on every host of the swarm, and deliberately not
+      **Required** on every host that runs a hyperhive service (a
+      `services.hyperhive.deploy.*` service, the gateway, the bridge,
+      `otel` or the snapshot store), with the same value on every host
+      of the swarm, and deliberately not
       defaulted. A guessed swarm domain is a wrong hostname that
       evaluates cleanly and deploys, which is worse than an eval
       failure telling an operator to write down the one address their

nix/module-eval/core-toggle.nix

@@ -503,6 +503,28 @@ let
         }
         && !(refusedHiveName { deploy.hive-controller.enable = false; });
     }
+    {
+      # Refused wherever any hyperhive service runs, the hive or a swarm
+      # service alone. The host enabling nothing is the control that it is
+      # gated at all, and has to pass every assertion: it is what an
+      # operator gets from importing the module.
+      name = "a missing swarm.domain is refused on a hive and on a swarm-service host, and a host enabling nothing evaluates";
+      ok =
+        let
+          noDomain = extra: hive ({ swarm.domain = null; } // extra);
+          refusedSwarmDomain =
+            extra:
+            lib.any (
+              a: !a.assertion && lib.hasInfix "services.hyperhive.swarm.domain to be set" a.message
+            ) (noDomain extra).assertions;
+        in
+        refusedSwarmDomain { }
+        && refusedSwarmDomain {
+          deploy.hive-controller.enable = false;
+          deploy.nats.enable = true;
+        }
+        && lib.all (a: a.assertion) (noDomain { deploy.hive-controller.enable = false; }).assertions;
+    }
   ];
 in
 runGroup "core-toggle" cases