Watch
0
0
Fork
You've already forked hyperhive
0

operator admin sync: cap the grant at one send, split out response verdicts

The admin-room command stays in #admins' history, so a grant tuwunel
ignores must not be re-sent every five minutes. The controller now
sends `make-user-admin` at most once per operator per process; an
operator still missing from the room on a later pass is logged as
"make-user-admin for an operator failed to take" and left alone until
the controller restarts. The cap lives in the pure `plan()`.

swarm-matrix-client's `user_exists` and `joined_members` read their
responses through pure `profile_verdict` / `joined_verdict`, the shape
`whoami` / `whoami_verdict` already has, with every branch tested: only
M_NOT_FOUND is "no account", and a 403 or a list without `joined` is an
error rather than an answer.

The bridge's handling after authentication moves into `respond()`,
which takes the config and write lock instead of the whole AppState, so
a test can drive ListOperators through the real request path and see
the 403 without an HTTP client or an authelia.

docs/integrations/matrix.md no longer says @swarm is the only homeserver
admin.
This commit is contained in:
atlas 2026-10-04 15:59:35 +02:00 • committed by mara
commit cf88fbba9d
4 changed files with 261 additions and 42 deletions

View file

@ -150,11 +150,19 @@ async fn handle_request(
Ok(caller) => caller,
Err(resp) => return resp,
};
if let Err(refusal) = permit(
&req,
caller.as_deref(),
&state.config.operator_reader_client_id,
) {
respond(&state.config, &state.write_lock, caller.as_deref(), req).await
}
/// Everything after authentication: the per-request [`permit`] check, then
/// the operation. Takes the config and the write lock rather than
/// [`AppState`], whose HTTP client only [`authorize`] uses.
async fn respond(
config: &Config,
write_lock: &tokio::sync::Mutex<()>,
caller: Option<&str>,
req: BridgeRequest,
) -> Response {
if let Err(refusal) = permit(&req, caller, &config.operator_reader_client_id) {
// The client id is not a credential; naming it is what lets an
// operator see which client asked for something it may not have.
tracing::warn!(client_id = ?caller, request = ?req, "refused a request scoped to another client");
@ -162,10 +170,10 @@ async fn handle_request(
}
let (op, result) = match req {
BridgeRequest::EnsureAgentIdentity { name } => {
("ensure_identity", handle(&state, name).await)
("ensure_identity", handle(config, write_lock, name).await)
}
BridgeRequest::ListAgentIdentities => ("list_identities", list(&state)),
BridgeRequest::ListOperators => ("list_operators", list_operators(&state)),
BridgeRequest::ListAgentIdentities => ("list_identities", list(config)),
BridgeRequest::ListOperators => ("list_operators", list_operators(config)),
};
match result {
Ok(body) => (StatusCode::OK, Json(body)).into_response(),
@ -304,20 +312,23 @@ fn refuse(refusal: Refusal) -> Response {
.into_response()
}
async fn handle(state: &AppState, name: String) -> Result<BridgeResponse> {
async fn handle(
cfg: &Config,
write_lock: &tokio::sync::Mutex<()>,
name: String,
) -> Result<BridgeResponse> {
store::validate_username(&name)?;
// After the charset check and before anything touches the store: a
// reserved name is refused on its own terms, not as a side effect of a
// collision with whatever happens to be in the file today.
store::reject_reserved_name(&name)?;
let cfg = &state.config;
// Held across the whole load → insert → publish sequence, not just the
// write — two concurrent `EnsureAgentIdentity` calls must not both read
// the same on-disk snapshot before either publishes. See the field's
// doc comment on `AppState::write_lock` for why this is reachable in
// practice, not just theoretically.
let _write_guard = state.write_lock.lock().await;
let _write_guard = write_lock.lock().await;
let mut user_store = store::load_store(&cfg.users_file)?;
if user_store.users.contains_key(&name) {
@ -365,8 +376,8 @@ async fn handle(state: &AppState, name: String) -> Result<BridgeResponse> {
/// reader sees either the whole old file or the whole new one — and holding
/// the lock would make a read wait behind a password mint, which is an
/// external process.
fn list(state: &AppState) -> Result<BridgeResponse> {
let user_store = store::load_store(&state.config.users_file)?;
fn list(config: &Config) -> Result<BridgeResponse> {
let user_store = store::load_store(&config.users_file)?;
Ok(BridgeResponse::Agents {
agents: store::agent_names(&user_store)
.into_iter()
@ -395,10 +406,10 @@ fn list(state: &AppState) -> Result<BridgeResponse> {
}
/// The operator group's members. No write lock, for [`list`]'s reason.
fn list_operators(state: &AppState) -> Result<BridgeResponse> {
let user_store = store::load_store(&state.config.users_file)?;
fn list_operators(config: &Config) -> Result<BridgeResponse> {
let user_store = store::load_store(&config.users_file)?;
Ok(BridgeResponse::Operators {
operators: store::group_members(&user_store, &state.config.operator_group),
operators: store::group_members(&user_store, &config.operator_group),
})
}
@ -444,7 +455,9 @@ async fn generate_password(bin: &std::path::Path) -> Result<String> {
#[cfg(test)]
mod tests {
use super::{BridgeRequest, Refusal, StatusCode, classify, introspect::Verdict, permit};
use super::{
BridgeRequest, Config, Refusal, StatusCode, classify, introspect::Verdict, permit, respond,
};
/// Fail-closed, asserted per variant rather than as one negation:
/// splitting blame out of a single denial is exactly the change that
@ -559,4 +572,63 @@ mod tests {
"the token is valid; a 401 would send the caller off to mint a new one"
);
}
/// A config whose users database holds one operator, `mara`. Every field
/// the request path does not read is left empty.
fn config_with_one_operator() -> Config {
let dir = std::env::temp_dir().join(format!(
"swarm-authelia-bridge-respond-{}-{}",
std::process::id(),
std::time::SystemTime::now()
.duration_since(std::time::UNIX_EPOCH)
.unwrap()
.as_nanos()
));
std::fs::create_dir_all(&dir).unwrap();
let users_file = dir.join("users.yml");
std::fs::write(
&users_file,
"users:\n mara:\n displayname: mara\n password: \"$argon2id$x\"\n groups: [admins]\n",
)
.unwrap();
Config {
bind: String::new(),
users_file,
authelia_bin: std::path::PathBuf::new(),
introspection_url: String::new(),
client_id: String::new(),
client_secret: String::new(),
operator_reader_client_id: CONTROLLER.to_owned(),
operator_group: "admins".to_owned(),
}
}
/// The request path, not only [`permit`]: a dispatch that skipped the
/// check would answer the hive below with the operator list.
#[tokio::test]
async fn the_request_path_answers_the_operator_list_to_the_controller_only() {
let config = config_with_one_operator();
let lock = tokio::sync::Mutex::new(());
for caller in [Some("hive-alpha"), None] {
let refused = respond(&config, &lock, caller, BridgeRequest::ListOperators).await;
assert_eq!(refused.status(), StatusCode::FORBIDDEN, "{caller:?}");
}
let answered = respond(
&config,
&lock,
Some(CONTROLLER),
BridgeRequest::ListOperators,
)
.await;
assert_eq!(answered.status(), StatusCode::OK);
let body = axum::body::to_bytes(answered.into_body(), usize::MAX)
.await
.unwrap();
assert_eq!(
serde_json::from_slice::<serde_json::Value>(&body).unwrap(),
serde_json::json!({"status": "operators", "operators": ["mara"]})
);
let _ = std::fs::remove_dir_all(config.users_file.parent().unwrap());
}
}