fix(#946): drop hive-priv self-bind fallback (require socket activation) + clarify child-state rw is intentional

This commit is contained in:
damocles 2026-06-08 19:38:40 +02:00 committed by mara
commit 58b5434466
3 changed files with 40 additions and 36 deletions

View file

@ -1103,6 +1103,15 @@ const HOST_SHARED_ROOT: &str = "/var/lib/hyperhive/shared";
/// `binds`. All three are RW so the parent can read/write state and
/// submit config-change requests. Creates missing host-side directories
/// so nspawn doesn't refuse to start; missing dirs are non-fatal.
/// Mount a child agent's `state`, `harness`, and `config` dirs into the
/// parent, all read-write. The RW on `state` is deliberate (not just a
/// read mount): a parent manages its children, which includes writing
/// into a child's state for recovery (e.g. seeding notes / clearing a
/// stuck sentinel) as well as reading it. `config` is RW because the
/// parent authors proposed config changes for the child (the approval
/// flow commits into the child's config repo). `harness` is RW for the
/// same management reasons. Per-child isolation still holds: a child
/// only ever has its *own* dirs bind-mounted, never a sibling's.
fn bind_child_agent_dirs(child: &str, binds: &mut Vec<BindMount>) {
let state_dir = format!("{HOST_AGENTS_ROOT}/{child}/state");
let harness_dir = format!("{HOST_AGENTS_ROOT}/{child}/harness");
@ -1221,7 +1230,8 @@ async fn set_nspawn_flags(
// Topology-driven child mounts: every direct child of this agent gets
// its state, harness, and config dirs bind-mounted RW so the parent
// can read state and manage config.
// can read AND write child state (recovery) and manage its config.
// See `bind_child_agent_dirs` for why state is RW, not read-only.
let direct_children = crate::topology::children_of(agent_name);
for child in &direct_children {
bind_child_agent_dirs(child, &mut binds);

View file

@ -57,9 +57,14 @@ async fn main() -> Result<()> {
}
fn socket_listener() -> Result<UnixListener> {
use std::os::unix::fs::PermissionsExt as _;
// Socket activation: systemd passes the socket as fd 3 when
// LISTEN_FDS >= 1 and LISTEN_PID matches our pid.
// hive-priv is ALWAYS socket-activated: systemd's `hive-priv.socket`
// unit binds `/run/hive/priv.sock` (SocketGroup=hive-core, mode 0660)
// and passes it as fd 3 via LISTEN_FDS. We require that — there is
// intentionally no self-bind fallback, so dev and prod take the exact
// same path. (The old fallback re-bound the socket itself as root's
// primary group, never `hive-core`, so a hive-core client couldn't
// connect the way the socket unit's grant intends; dropping it removes
// that dev/prod divergence.)
let listen_fds: Option<i32> = std::env::var("LISTEN_FDS")
.ok()
.and_then(|s| s.parse().ok());
@ -67,35 +72,27 @@ fn socket_listener() -> Result<UnixListener> {
.ok()
.and_then(|s| s.parse().ok());
if let (Some(n), Some(p)) = (listen_fds, listen_pid)
&& n >= 1
&& p == std::process::id()
{
// SAFETY: systemd has passed us a ready UnixListener on fd 3.
let std_listener = unsafe {
use std::os::unix::io::FromRawFd;
std::os::unix::net::UnixListener::from_raw_fd(3)
};
std_listener
.set_nonblocking(true)
.context("set socket non-blocking")?;
let listener =
tokio::net::UnixListener::from_std(std_listener).context("wrap systemd socket")?;
tracing::info!("using systemd-activated socket");
return Ok(listener);
let activated =
matches!(listen_fds, Some(n) if n >= 1) && listen_pid == Some(std::process::id());
if !activated {
bail!(
"hive-priv requires systemd socket activation (expected LISTEN_FDS>=1 + \
LISTEN_PID=<self> for {PRIV_SOCK}); run it via the hive-priv.socket unit, \
not directly"
);
}
// Fallback: bind the socket ourselves.
let path = Path::new(PRIV_SOCK);
if let Some(parent) = path.parent() {
std::fs::create_dir_all(parent).with_context(|| format!("create {}", parent.display()))?;
}
let _ = std::fs::remove_file(path);
let listener = UnixListener::bind(path).with_context(|| format!("bind {PRIV_SOCK}"))?;
// Mode 0660: only the hive-core group can connect.
std::fs::set_permissions(path, std::fs::Permissions::from_mode(0o660))
.context("chmod priv.sock")?;
tracing::info!(path = PRIV_SOCK, "bound priv socket");
// SAFETY: systemd has passed us a ready UnixListener on fd 3.
let std_listener = unsafe {
use std::os::unix::io::FromRawFd;
std::os::unix::net::UnixListener::from_raw_fd(3)
};
std_listener
.set_nonblocking(true)
.context("set socket non-blocking")?;
let listener =
tokio::net::UnixListener::from_std(std_listener).context("wrap systemd socket")?;
tracing::info!("using systemd-activated socket");
Ok(listener)
}

View file

@ -894,8 +894,6 @@ in
# Why each entry is needed:
# /etc/nixos-containers — writes <container>.conf (bind mounts,
# network isolation, nspawn flags)
# /run/hive — fallback socket bind if LISTEN_FDS is
# absent (normal path: socket-activated)
# /run/hive-agent — chown/chmod per-agent socket directories
# /run/systemd — container@ unit drop-ins (resource limits)
# + machinectl / systemd-machined state
@ -914,7 +912,6 @@ in
ProtectSystem = "strict";
ReadWritePaths = [
"/etc/nixos-containers"
"/run/hive"
"/run/hive-agent"
"/run/systemd"
"/run/lock"