fix(fips): P0 uptime fixes — open peer port 5679, allow /blob+/dwn, fix LAN anchor port, un-deaden direct peering, fast-fail budgets

Phase A1 of docs/FIPS-UPTIME-AND-UI-STATE-PLAN.md — the five changes that
made FIPS fall back to Tor even when a FIPS path existed:

- RC0: the fips.d drop-in now opens PEER_PORT 5679 (was 80+8443 only, so
  every hardened node firewalled peers' FIPS dials; 28k drops on .198)
- RC4: /blob/ and /dwn/ added to the peer-path allowlist — mesh file
  sharing and DWN sync were 404 → 100% Tor by construction
- RC2-G2: lan_fips_anchors dials PUBLISHED_UDP_PORT (2121) instead of the
  dead 8668, with a drift-guard test against the rendered daemon config
- RC2-G1: direct LAN peering actually runs now — mDNS TXT advertises the
  FIPS npub, discovery calls set_fips_npub, and the anchor tick hydrates
  npubs from federation storage for peers on older builds
- RC3: FIPS attempt budget is a hard cap (retry no longer doubles it) and
  the 12 hot call sites get explicit fips_timeout fast-fail so Tor keeps
  its full budget (browse-peer, preview, /blob, DWN, node-message,
  rotation notifies)

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
archipelago
2026-07-27 16:42:44 -04:00
co-authored by Claude Fable 5
parent 94b5374f66
commit eb2fc0f37b
12 changed files with 176 additions and 23 deletions
+39 -3
View File
@@ -433,8 +433,18 @@ impl Server {
),
));
// LAN transport (mDNS discovery)
let mut lan = crate::transport::lan::LanTransport::new(&did, &pubkey_hex, 5678);
// LAN transport (mDNS discovery). Advertise our FIPS npub in
// the TXT record so co-located peers can form a direct FIPS
// link (see `lan_fips_anchors`).
let local_fips_npub = crate::identity::fips_npub(&data_dir.join("identity"))
.await
.unwrap_or(None);
let mut lan = crate::transport::lan::LanTransport::new(
&did,
&pubkey_hex,
5678,
local_fips_npub,
);
match lan.start(registry.clone()) {
Ok(()) => info!("📡 LAN transport (mDNS) started"),
Err(e) => debug!("LAN transport init (non-fatal): {}", e),
@@ -753,12 +763,25 @@ impl Server {
// (often flaky) global anchor's spanning tree to route to each
// other. For every peer the registry knows both a LAN address
// AND a FIPS npub for, dial it on its FIPS UDP transport port
// (8668) at its LAN IP. This is FIPS's own transport over the
// at its LAN IP. This is FIPS's own transport over the
// LAN — NOT Tailscale, NOT the HTTP/LAN messaging port. Pure
// FIPS. `fipsctl connect` is idempotent, so re-applying every
// tick just keeps the direct link warm; unknown/remote peers
// (no LAN address) are left to the anchor as before.
if let Some(reg) = fips_peer_registry.as_ref() {
// Hydrate FIPS npubs into the registry from federation
// storage (did-keyed). Peers discovered before the mDNS
// TXT `fips` key existed — or running builds that don't
// advertise it yet — would otherwise never satisfy the
// `fips_npub` requirement in lan_fips_anchors(), leaving
// direct LAN peering a no-op.
if let Ok(nodes) = crate::federation::load_nodes(&data_dir).await {
for n in &nodes {
if let Some(npub) = n.fips_npub.as_deref() {
reg.set_fips_npub(&n.did, npub).await;
}
}
}
let direct = crate::fips::anchors::lan_fips_anchors(&reg.all_peers().await);
if !direct.is_empty() {
let _ = crate::fips::anchors::apply(&direct).await;
@@ -1241,6 +1264,12 @@ pub fn is_peer_allowed_path(path: &str) -> bool {
)
// Prefix-matched content endpoints (peer file browse + fetch)
|| path.starts_with("/content/")
// Mesh file sharing — blob fetch by CID, signature-gated in the
// handler. Absent from this list it 404'd over FIPS and the feature
// was 100% Tor by construction.
|| path.starts_with("/blob/")
// DWN sync — /dwn/health is step 1 of every sync; same story.
|| path.starts_with("/dwn/")
}
async fn accept_loop(
@@ -1944,10 +1973,17 @@ mod merge_tests {
);
assert!(is_peer_allowed_path("/rpc/v1"));
assert!(is_peer_allowed_path("/health"));
// Mesh blob fetch + DWN sync — both were missing from the allowlist,
// which made them deterministically 404 over FIPS and therefore
// 100% Tor by construction.
assert!(is_peer_allowed_path("/blob/abc123"), "blob fetch by CID");
assert!(is_peer_allowed_path("/dwn/health"), "DWN sync step 1");
// Not on the allow-list → rejected (no broad surface over the mesh).
assert!(!is_peer_allowed_path("/contention"), "must not prefix-leak");
assert!(!is_peer_allowed_path("/"));
assert!(!is_peer_allowed_path("/rpc/v2"));
assert!(!is_peer_allowed_path("/blobber"), "must not prefix-leak");
assert!(!is_peer_allowed_path("/dwnx"), "must not prefix-leak");
}
#[test]