laantungir/fips - fips - Gitea on NOSTR

laantungir/fips

Fork 0

mirror of https://github.com/jmcorgan/fips.git synced 2026-09-14 00:45:08 +00:00

Commit Graph

Author	SHA1	Message	Date
Johnathan Corgan	6a564e26ac	Prepare the v0.5.0 release content Everything the release needs except the version number, which stays at 0.5.0-dev until the tag. The changelog entry covers only the work that is new on this line. The point release's forty-six entries arrived under their own heading with the forward merge and are left alone; the twenty that remained are regrouped by topic and eight more added for changes no entry covered. Three of those eight matter to someone upgrading. Five root modules and four re-exports left the public library surface and Node::connections narrowed, none of it recorded anywhere; the entry names what to use instead and distinguishes the removed connection-phase enum from the Noise type of the same name, which is a different type that still exists. Tracing targets moved, so an existing RUST_LOG filter stops matching rather than erroring. And the handshake resend interval key no longer governs the first resend, which is now a constant, though it still governs later ones. Seven more entries cover the work that landed after the first content pass was written: the experimental native datagram API, the fipsctl probe diagnostic, per-instance transport addressing, the app-owned UDP socket seam, and the connect, disconnect and path-MTU fixes. The four bug fixes among them all reach the deployed line, so the release notes no longer claim this release carries exactly one fix for a shipped bug; it carries four. There is no security section, because after the split every security entry belongs to the point release. The release notes say so plainly rather than leaving a reader upgrading across both releases to conclude this one carries no security work. The notes are organized by audience, since the release spans OpenWrt routers, embedders, FreeBSD, and the existing platforms, and a single list serves none of them. The native datagram API is given a section of its own rather than folded into the embedding seam: it is a client-facing API rather than a way to host a node, and its one rule with no Berkeley-socket counterpart, that the v1 wire carries no half-close, needs to be somewhere a client author will read it. FreeBSD is advertised as supported on x86_64 only, stated wherever the platform appears. Android is advertised as an embedding seam and not as a supported platform: a compile-gated library surface with no artifact and no host application guide. The configuration table rename is carried through every shipped file that taught the old spelling: nine documentation files, the OpenWrt sample config and a test generator, twenty-two sites in all. Guides written this same cycle were among them, which is how the omission was found. The documentation that arrived with the native API was checked for the same omission and was already clean. The compatibility tests keep the old spelling deliberately, since they exist to test the fold. The changelog section is the fold of master's [Unreleased], not a snapshot of it. An earlier version of this commit took a copy that then drifted, so each section ended up holding a bullet the other did not and re-folding them would have picked a winner silently. Both causes were fixed on master instead — the NixOS module had never been recorded there, and the pre-release batch of fixes was new — so [Unreleased] is a strict superset and this is a copy rather than a merge. [0.5.0] carries all forty-six bullets byte for byte, [Unreleased] is empty, and [0.4.2] is untouched, checked by hashing it against master's copy. The BLE work landed after the content pass and gets one summary entry in the changelog and one section in the release notes rather than nine bullets: the ble_available gate replacing target_os = "linux", packet-boundary recovery for stream-oriented backends, peer recognition by node identity instead of a rotating link address, the L2CAP PSM moving into the backend seam and onto the advertisement, the embedder-supplied Android radio, bounded probe retry, and inbound handshakes moved off the accept loop. The two release-notes copies no longer share their link paths. Relative links resolve from one directory only, so the seven written for docs/releases/ all 404ed from the root copy. The root copy now uses paths from the repository root and the versioned copy keeps the ../ form; both sets were resolved against the tree. The same two links are broken the same way in the v0.4.0 through v0.4.2 notes, left as shipped history. The contributor tallies are re-derived against maint..HEAD rather than adjusted: twenty commits from outside the project and 171 from me, with Arjen at fifteen and fr34aky at two. An earlier count of twelve and 138 was carried from a measurement taken three days before this content was written, and the BLE branch widened the gap after it. Arjen's NixOS flake module, the UDP sin6_scope_id fix and most of the BLE rework were uncredited, as was fr34aky's L2CAP PSM seam. They want one last re-derive at tag time if anything lands before the tag. A sweep of all 99 tracked markdown files against the tree corrected fifty-three of them. Four told the reader to run a build.sh that does not exist; the only harness builder is testing/scripts/build.sh. The BLE build prerequisites were described as optional on the strength of a probe that build.rs does not perform, and bluez was named a build prerequisite when libdbus-sys asks only for libdbus-1-dev and pkg-config and bluez is the runtime daemon. Link cost is the primary sort key in next-hop ranking, not reserved for future use; Ethernet runs on macOS as well as Linux; the BLE MTU is the L2CAP CoC MTU rather than a negotiated ATT_MTU; effective Ethernet MTU is 1497; the LAN discovery subsystem is src/mdns and eight citations still named a src/discovery that never existed here. The connectivity states in three tutorials were invented, and their jq filters matched nothing including healthy peers. One command filtered on a literal fd97: address prefix, which only the first byte of fixes, so it returned empty for all but one reader in 256 and every later step using the variable failed silently. transports.tor.advertise_on_nostr was undocumented despite being validated against node.rendezvous.nostr.enabled. The transport design document gains the BLE section it never had, written from the source: the backend cascade and its compile_error tripwire, the platform gate, the PSM advertisement wire layout and the byte budget that forces a 16-bit service-data key, and the probe and admission bounds. Three source files carried the same class of staleness and are corrected with the documentation: the OpenWrt ipk usage line and Makefile error text both named a packaging/openwrt that does not exist, and chaos.sh parsed --subnet without listing it. Folded in with the content commit, having been prepared alongside it: The three GitHub Action pins that had gone stale. Every third-party action is pinned to a commit SHA, nothing reports that a pin has aged, and re-resolving all ten against their tags found dorny/test-reporter@v2, taiki-e/install-action@v2 and vmactions/freebsd-vm@v1 had moved. The three install-action@nextest references stay unpinned, since that action reads the tool to install from the ref name. check-action-pins.sh passes at 75 references and all nine workflow files parse. The lockfile refresh, which is the mutating half of the dependency sweep. Thirty-six packages move to their latest semver-compatible versions and every one is transitive; nothing declared in Cargo.toml changes version. No advisory forces any of them. It was taken before the validation battery, because a gate run against a lockfile that later moves proves nothing about what ships. The sha2 0.10 to 0.11, hkdf 0.12 to 0.13 and bech32 0.11 to 0.12 majors, three of the four deferred at v0.4.0 for change surface rather than security. All three land with no source change. sha2 and hkdf must move together, since both depend on digest 0.11, and neither changes an algorithm. That matters because the chaining-key KDF in the Noise handshake is built on Hkdf::<Sha256>, where an output change would be a wire break rather than a compile error; no known-answer vectors exist for that path, so the wire-compatibility gate is what covers it. secp256k1 0.31 is deliberately absent, since nostr's own requirement would leave two copies of the ECC library in the tree. The README support matrix, rebuilt as one feature table broken out by Linux variety. A single Linux column hid that Debian, Ubuntu, Arch and NixOS are one glibc build differing in packaging, that OpenWrt is musl and drops BLE, and that Android is not a daemon platform. Transport rows sort by how many platforms carry them. A Native API row reads its platform set from the cfg gates. The installer row becomes a package format row naming the artifact, and only the .deb is exercised per release. Four changelog and release-note gaps the BLE re-walk found: a Bluetooth LE bullet stranded inside the released 0.4.2 section, a missing Fixed entry for the scan and probe loop counting a pool-refused connection as an established link, the unnamed embedder call that installs an application-owned radio, and the fact that stopping the transport now stops scanning as well as advertising. Three release-document gaps found walking the unsurveyed commits: the UDP reuse-flag fix stated in the direction opposite to the one it was made, with the silent second-daemon bind it prevents left unsaid; the corrected native-API socket paragraph carried into both release-note copies, which still named SOCK_SEQPACKET on FreeBSD and two kernels where three are handled; and the coordinate-cache hardening, which shipped with no text anywhere despite adding four operator-visible status fields. That last entry states plainly that the checks are mitigations and not a closure, since the coordinate is still not authenticated. Also folded in, the documentation pass that followed the content commit: A stage-pipeline diagram for the probe, embedded in the fipsctl reference under the five-stage list. It draws the five stages left to right with each stage's failure reasons below it, and the bypass that skips both lookup stages when the coordinates are cached or the target is a direct peer. Its branches come from the probe state machine rather than from the report, so the path stage is drawn as the one failure that does not stop the probe. A rewrite of the README's "What FIPS does" section. It now opens with what a machine running FIPS gets, rather than with the two deployment modes, and gives the self-organizing and permissionless property its own paragraph since it holds for both modes. A regrouping of the README's feature list into the mesh, getting traffic onto it, and running a node, with a bullet added for the native datagram API, which had none despite sitting in the support matrix. The Quick start now leads with the released packages rather than a source build. It also fixes a real defect: the package enables fips.service and fips-dns.service and starts neither on a fresh install, so .fips name resolution was silently dead until the next reboot and neither page said to start the service. A rewrite of the release notes. They opened with seven subsections of upgrade caveats and reached the first feature two hundred lines in; they now open with a summary of the release and elaborate below it in the same order. Android is stated as supported through an embedded crate rather than as a standalone daemon, consistently across all three documents. The OpenWrt pair is corrected: it is 802.11s between routers with FIPS supplying encryption, authentication and routing, plus a convention of an open !FIPS SSID a client joins over WiFi, not meshing over a router's own radios. The probe's path output is described as the least-common-ancestor walk, which is the worst-case fallback route rather than the route a packet takes. Detail that did not change what a reader does was cut from the notes and kept in the changelog.	2026-08-30 10:42:59 +00:00
Johnathan Corgan	1ec62e5119	Build the native datagram API on macOS, over SOCK_DGRAM macOS does not implement SOCK_SEQPACKET for AF_UNIX, so the listener was gated to Linux and FreeBSD and a Mac got no API at all. It now uses SOCK_DGRAM there, which macOS does implement and which keeps the message boundaries the API's contract with its clients rests on. Both kernels were measured rather than reasoned about, and the Linux answer alone refuted the replacement the source had proposed. On Linux 6.8 a connected SOCK_DGRAM pair reports a closed peer not at all: revents stays empty and recv returns EAGAIN, which is exactly what an idle socket with a live peer does. SOCK_SEQPACKET on the same kernel sets POLLHUP and returns a zero-byte read, which is what the receive path keyed on. Darwin does report the close, with ECONNRESET, errno 54, and does not set POLLHUP. So the receive path treats ECONNRESET as end of file alongside the existing POLLHUP rule. One rule accepting either signal is correct on both kernels, where a rule split by platform would be silently wrong on whichever one it guessed at. EAGAIN is deliberately not in that company: it means the socket is empty and the peer alive, so it stays an error and the caller waits again. The measurements are asserted rather than only written down, so a kernel that gains or loses the signal reds a test and reopens the question instead of leaving a stale comment behind. Each carries its result in the assertion message, since a passing test prints nothing and a negative result would otherwise be as uninformative as no result: the close probe reports the poll return, the whole revents bitmask broken out by flag, the recv result and the errno, which is enough to write the real rule without another round trip. A portable test walks datagram sizes upward, because Darwin bounds a unix-domain datagram with the net.local.dgram.maxdgram sysctl, whose default is small and which is a system tunable rather than something this process controls, while the API advertises 1362 bytes to its clients. Every test recv in the seqpacket suite is bounded in time. This is not tidying. Simulating the Darwin configuration on a kernel that does not report the close, the end-of-file test hung for over ten minutes rather than failing, and a hang is not a red: it would have wedged the macOS runner with no diagnostic instead of naming the assertion. The same simulation now fails by name in five seconds. Three things in the native tree compiled on one platform only, all of them in code that had never been built for Darwin before. suseconds_t is i64 on Linux and i32 there, so the timeval microseconds field is a cast, matching the tv_sec line above it; it cannot truncate, because subsec_micros is below 1_000_000 by construction, and the cast is the only form that compiles on both, since From does not exist for the narrower width and try_from is a clippy error on the wider one. MSG_CMSG_CLOEXEC does not exist on Apple, so the recvmsg flags are chosen per platform and each received descriptor is marked close-on-exec with fcntl where there is no flag to pass; a failure to set it is reported rather than ignored, since the descriptor is live either way and the caller must not be told the receive was clean. socketpair takes SOCK_CLOEXEC in its type argument on Linux and FreeBSD and rejects it on macOS, so Darwin sets FD_CLOEXEC with a second fcntl. Both windows between a call and its fcntl are stated in the code rather than closed, since the daemon spawns no child on this path, and a test asserts both halves of a pair are close-on-exec on every platform, because the failure is a silent descriptor leak into a child and nothing else would report it. Every libc item the native tree uses was then checked against the crate's own Apple definitions rather than from memory, and those two constants are the only ones absent. The close difference turned out to be unhandled in six further places, and the whole native API agrees on it now. Darwin reports a closed AF_UNIX SOCK_DGRAM peer as ECONNRESET, and a later send on the disconnected survivor as EDESTADDRREQ, where Linux SOCK_SEQPACKET gives EPIPE on a write and a zero-byte read plus POLLHUP on a read. Each site below promised one of those spellings and saw another. - The client's recv and send passed ECONNRESET through, so a closed daemon half surfaced as errno 54 against the EPIPE the documentation promises. The translation is in one function in seqpacket rather than at each call site, since only this one condition has two spellings. - accept propagated the same errno instead of its documented EPIPE. The listener's read reports a closed peer as the empty chunk both of its callers already read as the far end going away, which leaves accept's contract true on both platforms without either caller knowing which it is on. - why() classified a failed hand-off by BrokenPipe alone, so every ordinary macOS listener close was counted under the counter an operator reads to find a client that stopped reading. It recognises all three errnos now, with a test over each. - A full client buffer ended a flow's only writer. On Linux that never arrives, because the send reports EAGAIN and waits for the client to drain; Darwin has no sender-side queue to wait on and reports ENOBUFS on the send itself. Returning left the registration, the port and the reader alive while every later inbound datagram was counted as a full queue for the rest of the flow's life, and a client that resumed reading never recovered. The datagram is dropped instead, which is what a datagram API does when the far end cannot take it. - The flow pair was never sized, and the two kernels charge a queued message to different ends: Linux to the sender's SO_SNDBUF, BSD to the receiver's so_rcv. Sizing only the sender, as the listener pair does, left the flow pair bounded on Darwin by a system default small enough that a batch held for an arriving client could not fit, and the whole flow was destroyed before its client ever saw it. Both halves are sized now. - peer_hung_up polled with an empty events field, on the rule that POLLHUP is reported whether or not it is requested. That holds on Linux, where it was measured, and not on Darwin, where a poll requesting nothing registers no filter. Nothing observable depended on it, because ECONNRESET arrives first and both callers act on it earlier. The cost was elsewhere: three assertions written as tripwires for a change in Darwin's behaviour could not fail there, which is a guard that executes and proves nothing. Requesting POLLIN fixes the function and the guards together. One difference is not an errno at all, and reading the kernel source rather than a manual page is what found it. Darwin's unp_disconnect sets SS_CANTRCVMORE and runs soisdisconnected on both ends for SOCK_STREAM. For SOCK_DGRAM it removes the reflink, clears SS_ISCONNECTED and stops: no sorwakeup, no socantrcvmore, no soisdisconnected. The closing peer deposits ECONNRESET in the survivor's so_error and wakes no knote. The registration is edge-triggered and was made while the socket was healthy, so nothing re-evaluates it, and recv awaited readiness before its syscall, which left the ECONNRESET arm sitting behind an await that never returns. A client closing its descriptor left the daemon's reader parked for ever, and the flow's port and registry entry held for the node's lifetime. recv reads before it waits now, because the latched error is visible to a syscall and only to a syscall, so the attempt that precedes the wait is what sees a close that has already happened. A close can also land while the task is parked, which no first attempt can catch, so on Darwin the wait is bounded and the syscall retried; the error is latched until a read consumes it, so the bound sets how long a dead flow holds its port rather than deciding whether the close is seen at all. On Linux this is one extra recv returning EAGAIN before the wait and changes nothing else, and everywhere else the readiness is authoritative and the wait stays unbounded. The three tests this predicted are the three that had failed: end of file on a closed client half, a listener's port unbound on close, and one flow's port freed while its connection stays open. One test asserted a delivery detail rather than the rule it exists to guard. a_descriptor_lands_on_the_last_complete_line_of_the_read_that_carried_it asserted that a plain write and the sendmsg following it arrive in one recvmsg. Linux coalesces them, so the read returns both lines and the descriptor together; Darwin stops a stream read at the ancillary boundary, so the plain line arrives by itself and the descriptor-bearing line comes on the next read. The rule the module rests on is unaffected, and Darwin satisfies it more easily than Linux, because the read it arrives on holds nothing later. The test fills until both lines are queued and asserts the rule instead of the number of reads it took. The client compiled in /run/fips/api.sock on every platform, and macOS has no /run for that path to be in. The daemon never had this problem: it resolves its socket at startup by looking for a directory, and its macOS branch lands on /var/run/fips. The constant is conditional the same way now, so a client that is told nothing looks where a packaged daemon on its own platform actually is. The reference documentation described that branch as FreeBSD-only and describes both. Windows stays excluded and cannot be included: it has no SCM_RIGHTS, so there is no way to pass a descriptor to another process at all, which is the whole mechanism rather than a detail of it. The platform statements in the source and in the shipped documentation all named Linux and FreeBSD and name macOS now, including the configuration reference, the security reference, the how-to and the walkthrough. The how-to also states how far the testing goes, because the person who would meet the gap first is the one enabling the API on a Mac. The end-to-end suite drives a client container against a node container over a shared volume, which is a Linux arrangement, so the socket lifecycle, the descriptor hand-off across a process boundary and the reclaiming of a port when a client exits are covered on macOS by unit tests rather than by anything that runs a daemon and a client as two real processes. That is a gap in testing and not a known defect, and it is a coverage gap rather than a discharged risk. The same place names the socket-type difference, since a reader who knows the descriptor is SOCK_DGRAM there can make sense of a close arriving as a different errno than the Linux documentation elsewhere describes. The changelog entry for the API is revised rather than followed by a second one: it now names the socket type each platform uses and the two end-of-file signals the receive path accepts. The entry describes what the release ships rather than the order the commits landed in.	2026-08-21 05:48:39 +00:00
Johnathan Corgan	3a789370b9	Add an experimental native datagram API addressed by public key A client process opens a flow to a peer's public key on a chosen port and sends and receives datagrams on a file descriptor the daemon hands it. No IPv6 emulation, no TUN device, no DNS: a datagram travels from key to key. The feature is off by default and is not a stable interface. The wire needs no change and gets none. Every FSP data packet has carried a port pair inside its AEAD envelope since v0.2.0, and port 256 is simply the IPv6 shim. What was missing was a way for a program to ask for a port of its own and be handed the traffic. Addressing is the part worth reading twice, because the obvious design is wrong. The x-only public key is the address. An npub is that key written in bech32, so converting between them is a local encoding rather than a lookup or a name service. The 16-byte node address that travels on the wire is the first half of a SHA-256 of the key: it is a truncated hash, it does not invert, and it appears nowhere a client can see. An earlier iteration of this work reported a peer by that hash and could supply a key only sometimes, which is what treating a wire identifier as an identity produces. An accepted flow therefore always knows its peer. The key is captured where the peer is authenticated rather than looked up when a report is rendered: every inbound datagram passes one call site inside a handler that refuses anything whose session is not established, and the responder has already rejected the session unless the claimed address derives from the key it proved. Reaching for the identity cache instead gives a best-effort answer from a structure that evicts. A listener is a descriptor. The daemon writes one message per arrival to it, carrying the new flow's descriptor and the peer's address, so poll, select and epoll work on a listener and accepting is a recvmsg. That is what lets the API be used from a program that already has an event loop, which a command-and-reply listener could not support: an arrival could not be waited on beside anything else. There is no accept command and no reject command. Refusing a flow is closing the descriptor you were handed. The Rust surface mirrors std::net. FipsStream::connect, FipsListener::bind, incoming, accept, io::Result and an errno mapping rather than a bespoke error type. An address is given as an npub, as a key, or as a pair, through one parameter, the way ToSocketAddrs takes several spellings of one thing. Each type holds its descriptor and copies of what setup told it and nothing else, so a stream that outlives its setup connection is not representable. set_nonblocking, AsFd and the four deadline methods carry the names and signatures std::net uses for the same jobs. They were asked for by a user integrating the API with tokio: AsyncFd requires a non-blocking descriptor, and anything receiving from a peer needs a bounded wait. AsFd is the better of the two descriptor accessors, because the borrow cannot outlive the value that owns the descriptor, so a reactor cannot hold a registration for a descriptor that has since been closed and its number reused by the next open. The non-blocking flag is read, modified and written back rather than assigned, since the flag word carries more than that one bit and a caller may have set O_ASYNC. A zero timeout is refused with EINVAL, because the kernel reads a zero timeval as "wait for ever", which inverts what a caller passing zero means; std::net refuses it for the same reason. The two directions are separate options and stay that way. FipsListener gets no timeout methods, matching TcpListener: bounding an accept is set_nonblocking plus the caller's own poll, which the reactor how-to builds. A flow taken from accept is blocking whatever the listener was set to, because the two are separate sockets and the daemon hands over a fresh one. One rule has no counterpart in Berkeley sockets and a client author must know it: the v1 wire carries no half-close, so nothing peer-driven ever closes a flow. A server written to read until the flow ends waits for a signal that cannot arrive, holding a thread and a flow per peer until its process exits. A program decides its own termination, and the example serves one datagram per flow. The tests reach a live daemon rather than a stand-in. Every public item had a unit test against a hand-written stand-in with canned replies, and the five entry points a program actually calls first, connect, connect_from, connect_at, bind and the SOCKET constant, had no coverage of any kind, because the tests that appear to cover them build a Wire over a socket pair and hand it to the private open and hold, so nothing ever resolved a socket path or mapped its errors. examples/native-surface.rs walks all thirty-eight items against a running daemon and reports the number of assertions it made. The count is read from the recorder rather than written as a literal, and the harness asserts the exit status, the completion marker and the count together, so deleting an assertion fails the check rather than quietly shrinking it. Watchdogs turn a hang into a named failure, which several of the walked behaviours would otherwise produce. The shared Docker image is built once for every integration leg, so the new binary is staged at all ten places the existing one is, the interop builder included, which gets a stub because those images exercise the wire between daemon versions and older refs do not carry the example. The platform gating was tested rather than reasoned about: flipping all eleven gates so the native API is excluded leaves the crate compiling clean across the workspace, every target and the profiling feature. The shipped docs tree gains what only the LaTeX manual under design/ had, which is not published with the daemon. A reference entry covers the whole surface: addressing and the port tiers, the Berkeley mapping, every method on FipsAddr, FipsStream, FipsListener and Incoming, the errno table, the ceilings, the four places data disappears with nothing reported, the line protocol and the command reference. The errno table gives names rather than numbers, since the client maps each name onto the libc constant for the platform it was built for and the supported platforms disagree on the numbers. A tutorial side trip stands up two throwaway nodes on one machine, peered over loopback UDP with no TUN and no DNS, then writes a listening program and a connecting program against them; it needs neither the public mesh nor root, because the native path is the one that does not go through the IPv6 adapter. The obligations a client in another language carries are a how-to of their own, since they are a task rather than a description: reading the setup connection with recvmsg, associating a descriptor with the last complete line, telling an empty datagram from a close, and six others. Serving many peers from one poll loop is another, with the whole program, because the straightforward listener spawns a thread per flow and that is wrong at the node's ceiling of 256. The drop causes are a table mapping each of the seven texts DropReason::as_str produces to the counter it increments, with drop_oversize called out as the ninth counter that is not in the table. What a daemon restart costs is a section of its own: every flow and listener ends, descriptors do not survive, there is no resumption, and datagrams sent but not yet forwarded are lost through a window nothing bounds. A stack comparison diagram places the interface against the stack a reader already knows: the same application over HTTP, TLS, TCP, IP and Ethernet on one side, and over its own format, FSP, FMP and a FIPS transport on the other, aligned so each row is one concern. The two columns are not alternatives and are not drawn as such. An unmodified IPv6 program's packets reach fips0, and the adapter hands each one to FSP as a payload, so the left stack runs inside the right one; the left column ends at a fork, eth0 for the ordinary internet and fips0 for the mesh, and an arrow leaves fips0 and runs back up into FSP's input. The row where TCP would be is empty on purpose and names Reliable Object Delivery, which is where that capability is expected to land. ROD is a v2 capability, the box is dashed because none of it exists yet, and the design entry says the part a reader needs most: nothing on the surface anticipates it, so a program written today should assume it does not exist. Both endpoints carry a scheme and a worked port, https://<npub>.fips:443 and fips://<npub>:443, with a footnote saying the two ports are not the same kind of thing, a TCP port inside the tunnel on the left and an FSP port on the right. The fips:// form is a coinage: nothing in the tree parses it, nothing registers the scheme, and the API takes a key and a port as separate arguments rather than a URL. The diagram also says where the right column stops, since FIPS over UDP still rides IP and Ethernet beneath. It appears in fips-concepts.md and fips-ipv6-adapter.md, which were making its argument in prose without a picture, and deliberately not in fips-architecture.md, which already carries the OSI mapping and makes the same point about the transport row. The gateway's control socket moves onto the same bind policy this API uses, which is the one change here that touches deployed behaviour: fips-gateway now tightens /run/fips to 0750. That is unreachable under the packaged deployment, where fips.service has already created the directory at that mode, and reachable for a source build or a container that starts the gateway alone. One changelog entry under Added, describing the released state: what a client opens and reads, the addressing and why the node address is not it, the listener being a descriptor, the std::net shape of the Rust surface, and the one rule Berkeley sockets have no counterpart for. It says in as many words that the wire is unchanged.	2026-08-21 05:48:23 +00:00

Author

SHA1

Message

Date

Johnathan Corgan

6a564e26ac

Prepare the v0.5.0 release content

Everything the release needs except the version number, which stays at
0.5.0-dev until the tag.

The changelog entry covers only the work that is new on this line. The
point release's forty-six entries arrived under their own heading with the
forward merge and are left alone; the twenty that remained are regrouped by
topic and eight more added for changes no entry covered. Three of those
eight matter to someone upgrading. Five root modules and four re-exports
left the public library surface and Node::connections narrowed, none of it
recorded anywhere; the entry names what to use instead and distinguishes
the removed connection-phase enum from the Noise type of the same name,
which is a different type that still exists. Tracing targets moved, so an
existing RUST_LOG filter stops matching rather than erroring. And the
handshake resend interval key no longer governs the first resend, which is
now a constant, though it still governs later ones.

Seven more entries cover the work that landed after the first content pass
was written: the experimental native datagram API, the fipsctl probe
diagnostic, per-instance transport addressing, the app-owned UDP socket
seam, and the connect, disconnect and path-MTU fixes. The four bug fixes
among them all reach the deployed line, so the release notes no longer
claim this release carries exactly one fix for a shipped bug; it carries
four.

There is no security section, because after the split every security entry
belongs to the point release. The release notes say so plainly rather than
leaving a reader upgrading across both releases to conclude this one
carries no security work.

The notes are organized by audience, since the release spans OpenWrt
routers, embedders, FreeBSD, and the existing platforms, and a single list
serves none of them. The native datagram API is given a section of its own
rather than folded into the embedding seam: it is a client-facing API
rather than a way to host a node, and its one rule with no Berkeley-socket
counterpart, that the v1 wire carries no half-close, needs to be somewhere
a client author will read it. FreeBSD is advertised as supported on x86_64
only, stated wherever the platform appears. Android is advertised as an
embedding seam and not as a supported platform: a compile-gated library
surface with no artifact and no host application guide.

The configuration table rename is carried through every shipped file that
taught the old spelling: nine documentation files, the OpenWrt sample
config and a test generator, twenty-two sites in all. Guides written this
same cycle were among them, which is how the omission was found. The
documentation that arrived with the native API was checked for the same
omission and was already clean. The compatibility tests keep the old
spelling deliberately, since they exist to test the fold.

The changelog section is the fold of master's [Unreleased], not a snapshot
of it. An earlier version of this commit took a copy that then drifted, so
each section ended up holding a bullet the other did not and re-folding
them would have picked a winner silently. Both causes were fixed on master
instead — the NixOS module had never been recorded there, and the
pre-release batch of fixes was new — so [Unreleased] is a strict superset
and this is a copy rather than a merge. [0.5.0] carries all forty-six
bullets byte for byte, [Unreleased] is empty, and [0.4.2] is untouched,
checked by hashing it against master's copy.

The BLE work landed after the content pass and gets one summary entry in
the changelog and one section in the release notes rather than nine
bullets: the ble_available gate replacing target_os = "linux",
packet-boundary recovery for stream-oriented backends, peer recognition by
node identity instead of a rotating link address, the L2CAP PSM moving
into the backend seam and onto the advertisement, the embedder-supplied
Android radio, bounded probe retry, and inbound handshakes moved off the
accept loop.

The two release-notes copies no longer share their link paths. Relative
links resolve from one directory only, so the seven written for
docs/releases/ all 404ed from the root copy. The root copy now uses paths
from the repository root and the versioned copy keeps the ../ form; both
sets were resolved against the tree. The same two links are broken the
same way in the v0.4.0 through v0.4.2 notes, left as shipped history.

The contributor tallies are re-derived against maint..HEAD rather than
adjusted: twenty commits from outside the project and 171 from me, with
Arjen at fifteen and fr34aky at two. An earlier count of twelve and 138
was carried from a measurement taken three days before this content was
written, and the BLE branch widened the gap after it. Arjen's NixOS flake
module, the UDP sin6_scope_id fix and most of the BLE rework were
uncredited, as was fr34aky's L2CAP PSM seam. They want one last re-derive
at tag time if anything lands before the tag.

A sweep of all 99 tracked markdown files against the tree corrected
fifty-three of them. Four told the reader to run a build.sh that does not
exist; the only harness builder is testing/scripts/build.sh. The BLE build
prerequisites were described as optional on the strength of a probe that
build.rs does not perform, and bluez was named a build prerequisite when
libdbus-sys asks only for libdbus-1-dev and pkg-config and bluez is the
runtime daemon. Link cost is the primary sort key in next-hop ranking, not
reserved for future use; Ethernet runs on macOS as well as Linux; the BLE
MTU is the L2CAP CoC MTU rather than a negotiated ATT_MTU; effective
Ethernet MTU is 1497; the LAN discovery subsystem is src/mdns and eight
citations still named a src/discovery that never existed here. The
connectivity states in three tutorials were invented, and their jq filters
matched nothing including healthy peers. One command filtered on a literal
fd97: address prefix, which only the first byte of fixes, so it returned
empty for all but one reader in 256 and every later step using the
variable failed silently. transports.tor.advertise_on_nostr was
undocumented despite being validated against node.rendezvous.nostr.enabled.

The transport design document gains the BLE section it never had, written
from the source: the backend cascade and its compile_error tripwire, the
platform gate, the PSM advertisement wire layout and the byte budget that
forces a 16-bit service-data key, and the probe and admission bounds.

Three source files carried the same class of staleness and are corrected
with the documentation: the OpenWrt ipk usage line and Makefile error text
both named a packaging/openwrt that does not exist, and chaos.sh parsed
--subnet without listing it.

Folded in with the content commit, having been prepared alongside it:

The three GitHub Action pins that had gone stale. Every third-party
action is pinned to a commit SHA, nothing reports that a pin has aged,
and re-resolving all ten against their tags found dorny/test-reporter@v2,
taiki-e/install-action@v2 and vmactions/freebsd-vm@v1 had moved. The
three install-action@nextest references stay unpinned, since that action
reads the tool to install from the ref name. check-action-pins.sh passes
at 75 references and all nine workflow files parse.

The lockfile refresh, which is the mutating half of the dependency sweep.
Thirty-six packages move to their latest semver-compatible versions and
every one is transitive; nothing declared in Cargo.toml changes version.
No advisory forces any of them. It was taken before the validation
battery, because a gate run against a lockfile that later moves proves
nothing about what ships.

The sha2 0.10 to 0.11, hkdf 0.12 to 0.13 and bech32 0.11 to 0.12 majors,
three of the four deferred at v0.4.0 for change surface rather than
security. All three land with no source change. sha2 and hkdf must move
together, since both depend on digest 0.11, and neither changes an
algorithm. That matters because the chaining-key KDF in the Noise
handshake is built on Hkdf::<Sha256>, where an output change would be a
wire break rather than a compile error; no known-answer vectors exist for
that path, so the wire-compatibility gate is what covers it. secp256k1
0.31 is deliberately absent, since nostr's own requirement would leave
two copies of the ECC library in the tree.

The README support matrix, rebuilt as one feature table broken out by
Linux variety. A single Linux column hid that Debian, Ubuntu, Arch and
NixOS are one glibc build differing in packaging, that OpenWrt is musl
and drops BLE, and that Android is not a daemon platform. Transport rows
sort by how many platforms carry them. A Native API row reads its
platform set from the cfg gates. The installer row becomes a package
format row naming the artifact, and only the .deb is exercised per
release.

Four changelog and release-note gaps the BLE re-walk found: a Bluetooth
LE bullet stranded inside the released 0.4.2 section, a missing Fixed
entry for the scan and probe loop counting a pool-refused connection as
an established link, the unnamed embedder call that installs an
application-owned radio, and the fact that stopping the transport now
stops scanning as well as advertising.

Three release-document gaps found walking the unsurveyed commits: the UDP
reuse-flag fix stated in the direction opposite to the one it was made,
with the silent second-daemon bind it prevents left unsaid; the corrected
native-API socket paragraph carried into both release-note copies, which
still named SOCK_SEQPACKET on FreeBSD and two kernels where three are
handled; and the coordinate-cache hardening, which shipped with no text
anywhere despite adding four operator-visible status fields. That last
entry states plainly that the checks are mitigations and not a closure,
since the coordinate is still not authenticated.

Also folded in, the documentation pass that followed the content commit:

A stage-pipeline diagram for the probe, embedded in the fipsctl
reference under the five-stage list. It draws the five stages left to
right with each stage's failure reasons below it, and the bypass that
skips both lookup stages when the coordinates are cached or the target
is a direct peer. Its branches come from the probe state machine rather
than from the report, so the path stage is drawn as the one failure that
does not stop the probe.

A rewrite of the README's "What FIPS does" section. It now opens with
what a machine running FIPS gets, rather than with the two deployment
modes, and gives the self-organizing and permissionless property its own
paragraph since it holds for both modes.

A regrouping of the README's feature list into the mesh, getting traffic
onto it, and running a node, with a bullet added for the native datagram
API, which had none despite sitting in the support matrix. The Quick
start now leads with the released packages rather than a source build.
It also fixes a real defect: the package enables fips.service and
fips-dns.service and starts neither on a fresh install, so .fips name
resolution was silently dead until the next reboot and neither page said
to start the service.

A rewrite of the release notes. They opened with seven subsections of
upgrade caveats and reached the first feature two hundred lines in; they
now open with a summary of the release and elaborate below it in the
same order. Android is stated as supported through an embedded crate
rather than as a standalone daemon, consistently across all three
documents. The OpenWrt pair is corrected: it is 802.11s between routers
with FIPS supplying encryption, authentication and routing, plus a
convention of an open !FIPS SSID a client joins over WiFi, not meshing
over a router's own radios. The probe's path output is described as the
least-common-ancestor walk, which is the worst-case fallback route
rather than the route a packet takes. Detail that did not change what a
reader does was cut from the notes and kept in the changelog.

2026-08-30 10:42:59 +00:00

Johnathan Corgan

1ec62e5119

Build the native datagram API on macOS, over SOCK_DGRAM

macOS does not implement SOCK_SEQPACKET for AF_UNIX, so the listener was
gated to Linux and FreeBSD and a Mac got no API at all. It now uses
SOCK_DGRAM there, which macOS does implement and which keeps the message
boundaries the API's contract with its clients rests on.

Both kernels were measured rather than reasoned about, and the Linux answer
alone refuted the replacement the source had proposed. On Linux 6.8 a
connected SOCK_DGRAM pair reports a closed peer not at all: revents stays
empty and recv returns EAGAIN, which is exactly what an idle socket with a
live peer does. SOCK_SEQPACKET on the same kernel sets POLLHUP and returns a
zero-byte read, which is what the receive path keyed on. Darwin does report
the close, with ECONNRESET, errno 54, and does not set POLLHUP. So the
receive path treats ECONNRESET as end of file alongside the existing POLLHUP
rule. One rule accepting either signal is correct on both kernels, where a
rule split by platform would be silently wrong on whichever one it guessed
at. EAGAIN is deliberately not in that company: it means the socket is empty
and the peer alive, so it stays an error and the caller waits again.

The measurements are asserted rather than only written down, so a kernel that
gains or loses the signal reds a test and reopens the question instead of
leaving a stale comment behind. Each carries its result in the assertion
message, since a passing test prints nothing and a negative result would
otherwise be as uninformative as no result: the close probe reports the poll
return, the whole revents bitmask broken out by flag, the recv result and the
errno, which is enough to write the real rule without another round trip. A
portable test walks datagram sizes upward, because Darwin bounds a
unix-domain datagram with the net.local.dgram.maxdgram sysctl, whose default
is small and which is a system tunable rather than something this process
controls, while the API advertises 1362 bytes to its clients.

Every test recv in the seqpacket suite is bounded in time. This is not
tidying. Simulating the Darwin configuration on a kernel that does not report
the close, the end-of-file test hung for over ten minutes rather than
failing, and a hang is not a red: it would have wedged the macOS runner with
no diagnostic instead of naming the assertion. The same simulation now fails
by name in five seconds.

Three things in the native tree compiled on one platform only, all of them in
code that had never been built for Darwin before. suseconds_t is i64 on Linux
and i32 there, so the timeval microseconds field is a cast, matching the
tv_sec line above it; it cannot truncate, because subsec_micros is below
1_000_000 by construction, and the cast is the only form that compiles on
both, since From does not exist for the narrower width and try_from is a
clippy error on the wider one. MSG_CMSG_CLOEXEC does not exist on Apple, so
the recvmsg flags are chosen per platform and each received descriptor is
marked close-on-exec with fcntl where there is no flag to pass; a failure to
set it is reported rather than ignored, since the descriptor is live either
way and the caller must not be told the receive was clean. socketpair takes
SOCK_CLOEXEC in its type argument on Linux and FreeBSD and rejects it on
macOS, so Darwin sets FD_CLOEXEC with a second fcntl. Both windows between a
call and its fcntl are stated in the code rather than closed, since the
daemon spawns no child on this path, and a test asserts both halves of a pair
are close-on-exec on every platform, because the failure is a silent
descriptor leak into a child and nothing else would report it. Every libc
item the native tree uses was then checked against the crate's own Apple
definitions rather than from memory, and those two constants are the only
ones absent.

The close difference turned out to be unhandled in six further places, and
the whole native API agrees on it now. Darwin reports a closed AF_UNIX
SOCK_DGRAM peer as ECONNRESET, and a later send on the disconnected survivor
as EDESTADDRREQ, where Linux SOCK_SEQPACKET gives EPIPE on a write and a
zero-byte read plus POLLHUP on a read. Each site below promised one of those
spellings and saw another.

- The client's recv and send passed ECONNRESET through, so a closed daemon
  half surfaced as errno 54 against the EPIPE the documentation promises.
  The translation is in one function in seqpacket rather than at each call
  site, since only this one condition has two spellings.
- accept propagated the same errno instead of its documented EPIPE. The
  listener's read reports a closed peer as the empty chunk both of its
  callers already read as the far end going away, which leaves accept's
  contract true on both platforms without either caller knowing which it is
  on.
- why() classified a failed hand-off by BrokenPipe alone, so every ordinary
  macOS listener close was counted under the counter an operator reads to
  find a client that stopped reading. It recognises all three errnos now,
  with a test over each.
- A full client buffer ended a flow's only writer. On Linux that never
  arrives, because the send reports EAGAIN and waits for the client to drain;
  Darwin has no sender-side queue to wait on and reports ENOBUFS on the send
  itself. Returning left the registration, the port and the reader alive
  while every later inbound datagram was counted as a full queue for the rest
  of the flow's life, and a client that resumed reading never recovered. The
  datagram is dropped instead, which is what a datagram API does when the far
  end cannot take it.
- The flow pair was never sized, and the two kernels charge a queued message
  to different ends: Linux to the sender's SO_SNDBUF, BSD to the receiver's
  so_rcv. Sizing only the sender, as the listener pair does, left the flow
  pair bounded on Darwin by a system default small enough that a batch held
  for an arriving client could not fit, and the whole flow was destroyed
  before its client ever saw it. Both halves are sized now.
- peer_hung_up polled with an empty events field, on the rule that POLLHUP is
  reported whether or not it is requested. That holds on Linux, where it was
  measured, and not on Darwin, where a poll requesting nothing registers no
  filter. Nothing observable depended on it, because ECONNRESET arrives first
  and both callers act on it earlier. The cost was elsewhere: three
  assertions written as tripwires for a change in Darwin's behaviour could
  not fail there, which is a guard that executes and proves nothing.
  Requesting POLLIN fixes the function and the guards together.

One difference is not an errno at all, and reading the kernel source rather
than a manual page is what found it. Darwin's unp_disconnect sets
SS_CANTRCVMORE and runs soisdisconnected on both ends for SOCK_STREAM. For
SOCK_DGRAM it removes the reflink, clears SS_ISCONNECTED and stops: no
sorwakeup, no socantrcvmore, no soisdisconnected. The closing peer deposits
ECONNRESET in the survivor's so_error and wakes no knote. The registration is
edge-triggered and was made while the socket was healthy, so nothing
re-evaluates it, and recv awaited readiness before its syscall, which left
the ECONNRESET arm sitting behind an await that never returns. A client
closing its descriptor left the daemon's reader parked for ever, and the
flow's port and registry entry held for the node's lifetime. recv reads
before it waits now, because the latched error is visible to a syscall and
only to a syscall, so the attempt that precedes the wait is what sees a close
that has already happened. A close can also land while the task is parked,
which no first attempt can catch, so on Darwin the wait is bounded and the
syscall retried; the error is latched until a read consumes it, so the bound
sets how long a dead flow holds its port rather than deciding whether the
close is seen at all. On Linux this is one extra recv returning EAGAIN before
the wait and changes nothing else, and everywhere else the readiness is
authoritative and the wait stays unbounded. The three tests this predicted
are the three that had failed: end of file on a closed client half, a
listener's port unbound on close, and one flow's port freed while its
connection stays open.

One test asserted a delivery detail rather than the rule it exists to guard.
a_descriptor_lands_on_the_last_complete_line_of_the_read_that_carried_it
asserted that a plain write and the sendmsg following it arrive in one
recvmsg. Linux coalesces them, so the read returns both lines and the
descriptor together; Darwin stops a stream read at the ancillary boundary, so
the plain line arrives by itself and the descriptor-bearing line comes on the
next read. The rule the module rests on is unaffected, and Darwin satisfies
it more easily than Linux, because the read it arrives on holds nothing
later. The test fills until both lines are queued and asserts the rule
instead of the number of reads it took.

The client compiled in /run/fips/api.sock on every platform, and macOS has no
/run for that path to be in. The daemon never had this problem: it resolves
its socket at startup by looking for a directory, and its macOS branch lands
on /var/run/fips. The constant is conditional the same way now, so a client
that is told nothing looks where a packaged daemon on its own platform
actually is. The reference documentation described that branch as
FreeBSD-only and describes both.

Windows stays excluded and cannot be included: it has no SCM_RIGHTS, so there
is no way to pass a descriptor to another process at all, which is the whole
mechanism rather than a detail of it.

The platform statements in the source and in the shipped documentation all
named Linux and FreeBSD and name macOS now, including the configuration
reference, the security reference, the how-to and the walkthrough. The how-to
also states how far the testing goes, because the person who would meet the
gap first is the one enabling the API on a Mac. The end-to-end suite drives a
client container against a node container over a shared volume, which is a
Linux arrangement, so the socket lifecycle, the descriptor hand-off across a
process boundary and the reclaiming of a port when a client exits are covered
on macOS by unit tests rather than by anything that runs a daemon and a
client as two real processes. That is a gap in testing and not a known
defect, and it is a coverage gap rather than a discharged risk. The same
place names the socket-type difference, since a reader who knows the
descriptor is SOCK_DGRAM there can make sense of a close arriving as a
different errno than the Linux documentation elsewhere describes.

The changelog entry for the API is revised rather than followed by a second
one: it now names the socket type each platform uses and the two
end-of-file signals the receive path accepts. The entry describes what the
release ships rather than the order the commits landed in.

2026-08-21 05:48:39 +00:00

Johnathan Corgan

3a789370b9

Add an experimental native datagram API addressed by public key

A client process opens a flow to a peer's public key on a chosen port and
sends and receives datagrams on a file descriptor the daemon hands it. No
IPv6 emulation, no TUN device, no DNS: a datagram travels from key to key.
The feature is off by default and is not a stable interface.

The wire needs no change and gets none. Every FSP data packet has carried a
port pair inside its AEAD envelope since v0.2.0, and port 256 is simply the
IPv6 shim. What was missing was a way for a program to ask for a port of its
own and be handed the traffic.

Addressing is the part worth reading twice, because the obvious design is
wrong. The x-only public key is the address. An npub is that key written in
bech32, so converting between them is a local encoding rather than a lookup
or a name service. The 16-byte node address that travels on the wire is the
first half of a SHA-256 of the key: it is a truncated hash, it does not
invert, and it appears nowhere a client can see. An earlier iteration of this
work reported a peer by that hash and could supply a key only sometimes,
which is what treating a wire identifier as an identity produces.

An accepted flow therefore always knows its peer. The key is captured where
the peer is authenticated rather than looked up when a report is rendered:
every inbound datagram passes one call site inside a handler that refuses
anything whose session is not established, and the responder has already
rejected the session unless the claimed address derives from the key it
proved. Reaching for the identity cache instead gives a best-effort answer
from a structure that evicts.

A listener is a descriptor. The daemon writes one message per arrival to it,
carrying the new flow's descriptor and the peer's address, so poll, select
and epoll work on a listener and accepting is a recvmsg. That is what lets
the API be used from a program that already has an event loop, which a
command-and-reply listener could not support: an arrival could not be waited
on beside anything else. There is no accept command and no reject command.
Refusing a flow is closing the descriptor you were handed.

The Rust surface mirrors std::net. FipsStream::connect, FipsListener::bind,
incoming, accept, io::Result and an errno mapping rather than a bespoke
error type. An address is given as an npub, as a key, or as a pair, through
one parameter, the way ToSocketAddrs takes several spellings of one thing.
Each type holds its descriptor and copies of what setup told it and nothing
else, so a stream that outlives its setup connection is not representable.

set_nonblocking, AsFd and the four deadline methods carry the names and
signatures std::net uses for the same jobs. They were asked for by a user
integrating the API with tokio: AsyncFd requires a non-blocking descriptor,
and anything receiving from a peer needs a bounded wait. AsFd is the better
of the two descriptor accessors, because the borrow cannot outlive the value
that owns the descriptor, so a reactor cannot hold a registration for a
descriptor that has since been closed and its number reused by the next
open. The non-blocking flag is read, modified and written back rather than
assigned, since the flag word carries more than that one bit and a caller may
have set O_ASYNC. A zero timeout is refused with EINVAL, because the kernel
reads a zero timeval as "wait for ever", which inverts what a caller passing
zero means; std::net refuses it for the same reason. The two directions are
separate options and stay that way. FipsListener gets no timeout methods,
matching TcpListener: bounding an accept is set_nonblocking plus the caller's
own poll, which the reactor how-to builds. A flow taken from accept is
blocking whatever the listener was set to, because the two are separate
sockets and the daemon hands over a fresh one.

One rule has no counterpart in Berkeley sockets and a client author must know
it: the v1 wire carries no half-close, so nothing peer-driven ever closes a
flow. A server written to read until the flow ends waits for a signal that
cannot arrive, holding a thread and a flow per peer until its process exits.
A program decides its own termination, and the example serves one datagram
per flow.

The tests reach a live daemon rather than a stand-in. Every public item had a
unit test against a hand-written stand-in with canned replies, and the five
entry points a program actually calls first, connect, connect_from,
connect_at, bind and the SOCKET constant, had no coverage of any kind,
because the tests that appear to cover them build a Wire over a socket pair
and hand it to the private open and hold, so nothing ever resolved a socket
path or mapped its errors. examples/native-surface.rs walks all thirty-eight
items against a running daemon and reports the number of assertions it made.
The count is read from the recorder rather than written as a literal, and the
harness asserts the exit status, the completion marker and the count
together, so deleting an assertion fails the check rather than quietly
shrinking it. Watchdogs turn a hang into a named failure, which several of
the walked behaviours would otherwise produce. The shared Docker image is
built once for every integration leg, so the new binary is staged at all ten
places the existing one is, the interop builder included, which gets a stub
because those images exercise the wire between daemon versions and older refs
do not carry the example. The platform gating was tested rather than reasoned
about: flipping all eleven gates so the native API is excluded leaves the
crate compiling clean across the workspace, every target and the profiling
feature.

The shipped docs tree gains what only the LaTeX manual under design/ had,
which is not published with the daemon. A reference entry covers the whole
surface: addressing and the port tiers, the Berkeley mapping, every method on
FipsAddr, FipsStream, FipsListener and Incoming, the errno table, the
ceilings, the four places data disappears with nothing reported, the line
protocol and the command reference. The errno table gives names rather than
numbers, since the client maps each name onto the libc constant for the
platform it was built for and the supported platforms disagree on the
numbers. A tutorial side trip stands up two throwaway nodes on one machine,
peered over loopback UDP with no TUN and no DNS, then writes a listening
program and a connecting program against them; it needs neither the public
mesh nor root, because the native path is the one that does not go through
the IPv6 adapter. The obligations a client in another language carries are a
how-to of their own, since they are a task rather than a description:
reading the setup connection with recvmsg, associating a descriptor with the
last complete line, telling an empty datagram from a close, and six others.
Serving many peers from one poll loop is another, with the whole program,
because the straightforward listener spawns a thread per flow and that is
wrong at the node's ceiling of 256. The drop causes are a table mapping each
of the seven texts DropReason::as_str produces to the counter it increments,
with drop_oversize called out as the ninth counter that is not in the table.
What a daemon restart costs is a section of its own: every flow and listener
ends, descriptors do not survive, there is no resumption, and datagrams sent
but not yet forwarded are lost through a window nothing bounds.

A stack comparison diagram places the interface against the stack a reader
already knows: the same application over HTTP, TLS, TCP, IP and Ethernet on
one side, and over its own format, FSP, FMP and a FIPS transport on the
other, aligned so each row is one concern. The two columns are not
alternatives and are not drawn as such. An unmodified IPv6 program's packets
reach fips0, and the adapter hands each one to FSP as a payload, so the left
stack runs inside the right one; the left column ends at a fork, eth0 for the
ordinary internet and fips0 for the mesh, and an arrow leaves fips0 and runs
back up into FSP's input. The row where TCP would be is empty on purpose and
names Reliable Object Delivery, which is where that capability is expected to
land. ROD is a v2 capability, the box is dashed because none of it exists
yet, and the design entry says the part a reader needs most: nothing on the
surface anticipates it, so a program written today should assume it does not
exist. Both endpoints carry a scheme and a worked port,
https://<npub>.fips:443 and fips://<npub>:443, with a footnote saying the two
ports are not the same kind of thing, a TCP port inside the tunnel on the
left and an FSP port on the right. The fips:// form is a coinage: nothing in
the tree parses it, nothing registers the scheme, and the API takes a key and
a port as separate arguments rather than a URL. The diagram also says where
the right column stops, since FIPS over UDP still rides IP and Ethernet
beneath. It appears in fips-concepts.md and fips-ipv6-adapter.md, which were
making its argument in prose without a picture, and deliberately not in
fips-architecture.md, which already carries the OSI mapping and makes the
same point about the transport row.

The gateway's control socket moves onto the same bind policy this API uses,
which is the one change here that touches deployed behaviour: fips-gateway
now tightens /run/fips to 0750. That is unreachable under the packaged
deployment, where fips.service has already created the directory at that
mode, and reachable for a source build or a container that starts the gateway
alone.

One changelog entry under Added, describing the released state: what a
client opens and reads, the addressing and why the node address is not it,
the listener being a descriptor, the std::net shape of the Rust surface,
and the one rule Berkeley sockets have no counterpart for. It says in as
many words that the wire is unchanged.

2026-08-21 05:48:23 +00:00