From 72d9422a7a09e6f4d231ccae28997c6aea9f3f70 Mon Sep 17 00:00:00 2001 From: atlas Date: Mon, 22 Jun 2026 18:49:09 +0200 Subject: [PATCH] feat(#1897): auto-fill the hive domain in peer-config (drop required --domain) Per operator review: hivectl shouldn't make the operator retype this hive's own domain. Add a HostRequest::HiveDomain admin-socket query (c0re answers from HYPERHIVE_HIVE_DOMAIN, already in its service env) and a domain field on HostResponse. hivectl peer-config now resolves the domain as: --domain override (kept for offline/scripted use), else the daemon query; errors with a clear hint when neither resolves. wg init resolves it best-effort too, so it prints the hand-over peer-config block without --domain (skipped, not failed, when the daemon is unreachable). Regenerated docs/tools/hivectl-cli.md. --- docs/tools/hivectl-cli.md | 8 ++--- hive-c0re/src/bin/hivectl.rs | 67 ++++++++++++++++++++++++++++-------- hive-c0re/src/server.rs | 8 +++++ hive-sh4re/src/lib.rs | 25 ++++++++++++++ 4 files changed, 89 insertions(+), 19 deletions(-) diff --git a/docs/tools/hivectl-cli.md b/docs/tools/hivectl-cli.md index ee93a381..d4653290 100644 --- a/docs/tools/hivectl-cli.md +++ b/docs/tools/hivectl-cli.md @@ -50,7 +50,7 @@ Sibling to the `hive-c0re` daemon binary. Covers host-side admin operations that * `gateway` — Gateway htpasswd user management. Add, remove, or list users in an htpasswd file used by the gateway's HTTP Basic auth (`services.hyperhive.gateway.auth`). Credentials are stored as `BCrypt` hashes — no extra service or PAM required * `agents` — Agent container management. Requires the hive-c0re daemon to be running (connects to the host admin socket) * `wg` — WireGuard inter-hive mesh setup helpers (`services.hyperhive.swarm`) -* `peer-config` — Generate the federation peer-config block for THIS hive — the nix a peer operator pastes into their `services.hyperhive.swarm.peers` to trust + reach this hive. Emits `caCert` (+ a `cp` line for the cert) when this hive serves a self-signed CA, the WireGuard public key when the mesh key exists, and the `wireguard{Address,Endpoint}` you pass. Pure output — reads local state (the TLS CA cert, the wg key), never mutates. `wg init` calls this at the end when given a `--domain`, so a fresh mesh setup prints the hand-over block too +* `peer-config` — Generate the federation peer-config block for THIS hive — the nix a peer operator pastes into their `services.hyperhive.swarm.peers` to trust + reach this hive. Emits `caCert` (+ a `cp` line for the cert) when this hive serves a self-signed CA, the WireGuard public key when the mesh key exists, and the `wireguard{Address,Endpoint}` you pass. Reads local state (the TLS CA cert, the wg key) + asks the daemon for this hive's domain; never mutates. `wg init` calls this at the end, so a fresh mesh setup prints the hand-over block too * `choom` — Open an interactive Claude session inside an agent container * `stop` — Stop containers hive-wide in one operator action. Bare `hivectl stop` stops **everything** — all sub-agents plus the ci, forge, gateway, and matrix infra containers. Narrow it with scope flags: `--agents` (all sub-agents), `--ci` / `--forge` / `--gateway` / `--matrix` (named infra), and `--agent ` (repeatable) for specific sub-agents. Flags are additive (e.g. `--agents --matrix`). Requires the hive-c0re daemon (connects to the host admin socket). hive-c0re itself is never stopped — it services the request * `start` — Start containers hive-wide — the inverse of `hivectl stop`. Bare `hivectl start` starts everything back up; the same scope flags as `stop` narrow it (`--agents`, `--ci`, `--forge`, `--gateway`, `--matrix`, `--agent `). Requires the hive-c0re daemon @@ -349,13 +349,13 @@ Show the live mesh interface state (`wg show wg-hive`). Requires the mesh to be ## `hivectl peer-config` -Generate the federation peer-config block for THIS hive — the nix a peer operator pastes into their `services.hyperhive.swarm.peers` to trust + reach this hive. Emits `caCert` (+ a `cp` line for the cert) when this hive serves a self-signed CA, the WireGuard public key when the mesh key exists, and the `wireguard{Address,Endpoint}` you pass. Pure output — reads local state (the TLS CA cert, the wg key), never mutates. `wg init` calls this at the end when given a `--domain`, so a fresh mesh setup prints the hand-over block too +Generate the federation peer-config block for THIS hive — the nix a peer operator pastes into their `services.hyperhive.swarm.peers` to trust + reach this hive. Emits `caCert` (+ a `cp` line for the cert) when this hive serves a self-signed CA, the WireGuard public key when the mesh key exists, and the `wireguard{Address,Endpoint}` you pass. Reads local state (the TLS CA cert, the wg key) + asks the daemon for this hive's domain; never mutates. `wg init` calls this at the end, so a fresh mesh setup prints the hand-over block too -**Usage:** `hivectl peer-config [OPTIONS] --domain ` +**Usage:** `hivectl peer-config [OPTIONS]` ###### **Options:** -* `--domain ` — This hive's DNS domain — the `swarm.peers` attrset key the peer declares. Required: hivectl has no other source for it +* `--domain ` — Override this hive's DNS domain (the `swarm.peers` attrset key the peer declares). Omit to auto-fill from the running daemon (`services.hyperhive.domain`); pass it only when the daemon is down or you're scripting offline * `--wg-address ` — This hive's WireGuard mesh address (e.g. `10.42.0.1/32`), emitted as `wireguardAddress`. Omit when not running the mesh * `--wg-endpoint ` — This hive's public WireGuard endpoint (`host:port`), emitted as `wireguardEndpoint`. Omit when peers dial in / no mesh diff --git a/hive-c0re/src/bin/hivectl.rs b/hive-c0re/src/bin/hivectl.rs index f3027e50..73677da0 100644 --- a/hive-c0re/src/bin/hivectl.rs +++ b/hive-c0re/src/bin/hivectl.rs @@ -94,14 +94,16 @@ enum Cmd { /// to trust + reach this hive. Emits `caCert` (+ a `cp` line for the /// cert) when this hive serves a self-signed CA, the WireGuard public /// key when the mesh key exists, and the `wireguard{Address,Endpoint}` - /// you pass. Pure output — reads local state (the TLS CA cert, the wg - /// key), never mutates. `wg init` calls this at the end when given a - /// `--domain`, so a fresh mesh setup prints the hand-over block too. + /// you pass. Reads local state (the TLS CA cert, the wg key) + asks the + /// daemon for this hive's domain; never mutates. `wg init` calls this + /// at the end, so a fresh mesh setup prints the hand-over block too. PeerConfig { - /// This hive's DNS domain — the `swarm.peers` attrset key the peer - /// declares. Required: hivectl has no other source for it. + /// Override this hive's DNS domain (the `swarm.peers` attrset key + /// the peer declares). Omit to auto-fill from the running daemon + /// (`services.hyperhive.domain`); pass it only when the daemon is + /// down or you're scripting offline. #[arg(long)] - domain: String, + domain: Option, /// This hive's WireGuard mesh address (e.g. `10.42.0.1/32`), /// emitted as `wireguardAddress`. Omit when not running the mesh. #[arg(long)] @@ -590,7 +592,9 @@ async fn main() -> Result<()> { AgentsCmd::RestartAll => agents_restart_all(&socket).await, }, Cmd::Wg { cmd } => match cmd { - WgCmd::Init { address, domain } => wg_init(address.as_deref(), domain.as_deref()), + WgCmd::Init { address, domain } => { + wg_init(&socket, address.as_deref(), domain.as_deref()).await + } WgCmd::Peer { domain, pubkey, @@ -607,6 +611,7 @@ async fn main() -> Result<()> { wg_address, wg_endpoint, } => { + let domain = resolve_hive_domain(&socket, domain).await?; peer_config(&domain, wg_address.as_deref(), wg_endpoint.as_deref()); Ok(()) } @@ -655,11 +660,38 @@ const WG_INTERFACE: &str = "wg-hive"; /// = ACME / operator cert (trusted by the default CA bundle, no `caCert`). const HIVE_TLS_CA_PATH: &str = "/var/lib/hive-tls/ca.pem"; +/// Best-effort query for this hive's domain from the running daemon +/// (`HostRequest::HiveDomain`, which reads `HYPERHIVE_HIVE_DOMAIN` from +/// c0re's service env). `None` when the daemon is unreachable or the +/// domain is unset — callers decide whether that's fatal. +async fn query_hive_domain(socket: &Path) -> Option { + hive_c0re::client::request(socket, hive_sh4re::HostRequest::HiveDomain) + .await + .ok() + .and_then(|r| r.domain) +} + +/// Resolve this hive's domain for snippet generation: the explicit +/// `--domain` override if given, else the daemon. Errors with a clear +/// hint when neither is available, so `peer-config` never silently emits +/// a wrong key. +async fn resolve_hive_domain(socket: &Path, over: Option) -> Result { + if let Some(d) = over { + return Ok(d); + } + query_hive_domain(socket).await.context( + "could not determine this hive's domain from the daemon — is hive-c0re running \ + and `services.hyperhive.domain` set? pass --domain to override", + ) +} + /// `wg init` — generate (if absent) the hive's WireGuard key, print its -/// public key + the nix snippet to enable the mesh. When `domain` is set, -/// also prints the `peer-config` block peers paste to federate with this -/// hive (so a fresh setup is one command). -fn wg_init(address: Option<&str>, domain: Option<&str>) -> Result<()> { +/// public key + the nix snippet to enable the mesh, then (best-effort) +/// the `peer-config` block peers paste to federate with this hive, so a +/// fresh setup is one command. The domain comes from `--domain` or the +/// daemon; if neither resolves, the peer block is skipped (init still +/// succeeds — its core job is enabling the mesh locally). +async fn wg_init(socket: &Path, address: Option<&str>, domain: Option<&str>) -> Result<()> { use std::os::unix::fs::PermissionsExt as _; let key_path = Path::new(WG_KEY_PATH); if key_path.exists() { @@ -703,11 +735,16 @@ fn wg_init(address: Option<&str>, domain: Option<&str>) -> Result<()> { println!(" # listenPort = 51820; # default"); println!(" }};"); - // When the operator names this hive's domain, also print the block a - // peer pastes to federate with us (CA + this mesh key) — one-stop setup. - if let Some(d) = domain { + // Also print the block a peer pastes to federate with us (CA + this + // mesh key) — one-stop setup. Domain from --domain or the daemon; + // best-effort, so init still succeeds when neither resolves. + let resolved = match domain { + Some(d) => Some(d.to_owned()), + None => query_hive_domain(socket).await, + }; + if let Some(d) = resolved { println!(); - peer_config(d, address, None); + peer_config(&d, address, None); } Ok(()) } diff --git a/hive-c0re/src/server.rs b/hive-c0re/src/server.rs index 8fc56bc5..fb7d05b2 100644 --- a/hive-c0re/src/server.rs +++ b/hive-c0re/src/server.rs @@ -114,6 +114,12 @@ async fn dispatch(req: &HostRequest, coord: Arc) -> HostResponse { } HostRequest::Rebuild { name } => handle_rebuild(&coord, name).await?, HostRequest::List => HostResponse::list(lifecycle::list().await?), + // The hive domain is injected into c0re's service env by + // hive-c0re.nix (`HYPERHIVE_HIVE_DOMAIN`); surface it so the + // operator CLI can fill in this hive's own identity. + HostRequest::HiveDomain => { + HostResponse::hive_domain(std::env::var("HYPERHIVE_HIVE_DOMAIN").ok()) + } HostRequest::Pending => HostResponse::pending(coord.approvals.pending()?), HostRequest::Approve { id } => { actions::approve(coord.clone(), *id).await?; @@ -210,6 +216,7 @@ async fn handle_restart_all() -> Result { error: Some(errors.join("; ")), agents: Some(ok_agents), approvals: None, + domain: None, }) } } @@ -368,6 +375,7 @@ fn finish_lifecycle(ok_items: Vec, errors: &[String]) -> HostResponse { error: Some(errors.join("; ")), agents: Some(ok_items), approvals: None, + domain: None, } } } diff --git a/hive-sh4re/src/lib.rs b/hive-sh4re/src/lib.rs index b49f1963..ed121f15 100644 --- a/hive-sh4re/src/lib.rs +++ b/hive-sh4re/src/lib.rs @@ -44,6 +44,11 @@ pub enum HostRequest { Rebuild { name: String }, /// List managed containers. List, + /// Report this hive's canonical DNS domain + /// (`services.hyperhive.domain`), or `None` when unset. Lets the + /// operator CLI fill in the hive's own identity (e.g. the federation + /// peer-config block) without the operator retyping it. + HiveDomain, /// List pending approval requests. Pending, /// Approve a pending request by id; the action runs immediately. @@ -131,6 +136,10 @@ pub struct HostResponse { pub agents: Option>, #[serde(default, skip_serializing_if = "Option::is_none")] pub approvals: Option>, + /// This hive's canonical DNS domain — `HiveDomain` result. `None` + /// when the domain is unset (no `services.hyperhive.domain`). + #[serde(default, skip_serializing_if = "Option::is_none")] + pub domain: Option, } /// One row in the approval queue. `commit_ref` is overloaded per @@ -218,6 +227,7 @@ impl HostResponse { error: None, agents: None, approvals: None, + domain: None, } } @@ -228,6 +238,7 @@ impl HostResponse { error: Some(message.into()), agents: None, approvals: None, + domain: None, } } @@ -238,6 +249,7 @@ impl HostResponse { error: None, agents: Some(agents), approvals: None, + domain: None, } } @@ -248,6 +260,19 @@ impl HostResponse { error: None, agents: None, approvals: Some(approvals), + domain: None, + } + } + + /// `HiveDomain` result — this hive's canonical domain (or `None`). + #[must_use] + pub fn hive_domain(domain: Option) -> Self { + Self { + ok: true, + error: None, + agents: None, + approvals: None, + domain, } } }