Codebase-derived research for Phase 1 (FED-01..06): identifies an
unlocked concurrent read-modify-write race on federation/nodes.json
as the likely root cause of nodes reappearing after removal, flags
two stale CONCERNS.md claims already fixed on main (01cbec27), maps
remaining mesh demo-parity gaps (contacts-list/save, reaction/edit/
delete stubs) beyond the already-shipped attachment-send parity fix
(c2ce71c6), and scopes the greenfield Lightning-URI/channel-open
surface needed for FED-05.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Demo attach flow failed with 'Method not found: mesh.send-content-inline'
and force-opened the transport chooser modal real nodes don't show.
- mesh.transport-advice now mirrors the daemon's size-based tier logic
(typed_messages.rs): chooser only in the fits-both 1-2.3KB band
- implement mesh.send-content-inline / send-content / fetch-content and
POST /api/blob; bytes live in the per-visitor session store
- sent texts + attachments persist in mesh.messages so refresh-after-send
shows them, same as a real node
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Research findings and a concrete split of work: web-side items for this
repo (pre-warm camera, torch toggle, continuous focus, keep-stream-
alive) and a native handover list for the companion dev (pre-warmed
CameraX + ML Kit, QR-only format, 720p keep-latest analysis, torch,
zoom nudge, haptic dismiss) with acceptance criteria and how to
measure. Quick win landed now: live scan runs at 10 scans/sec when the
platform has a native BarcodeDetector, keeping 4/s only for the
JS-worker fallback.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Framework-pt report: a paid invoice stalled the UI with no success
shown, and lightning/total balances disappeared until it settled.
Three compounding causes, three fixes:
- Backend payinvoice's synchronous wait drops 120s → 8s. Fast payments
(the majority) still settle in one round trip; slow multi-hop routes
return pending + payment_hash quickly and the caller's 3s poll takes
over — instead of the modal freezing for up to two minutes.
- payLightningInvoice gains an onPending hook: SendBitcoinModal and the
scan modal now flip to a visible "Settling…" success pane the moment
the payment goes pending (safe to close), and the ongoing poll
upgrades it to Paid — or replaces it with LND's real failure.
- One slow lnd.getinfo poll (5s budget) flipped the Home wallet card to
"disconnected", hiding balances the user already knew. Three
consecutive failures are now required (~30s) before the card gives up;
last-known balances keep rendering throughout.
rpc-client tests 75/75.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
"Another update operation is already running" surfaced as a scary
failure while the update was in fact applying fine (OptiPlex, v1.7.118
rollout). The apply path now joins the in-flight install — same
overlay, same wait-for-new-version polling — and a concurrent download
attempt shows a calm in-progress note (EN+ES strings added). The
backend's tarball extractions run under ionice -c3 nice -n10 so a
200MB update can't starve podman/status calls into multi-minute
timeouts on small disks.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- ~8s into a session a second "freshly plugged" RNode appears on
/dev/ttyACM0, so the global mesh setup modal (and its flash step)
demos itself shortly after opening the Mesh page. Fixed plugged_at
means "Not now" sticks for the whole browser session.
- Transport pills (LoRa/FIPS/Tor) now offered for every demo peer
except mountain-node, which stays radio-only for contrast.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- mesh.transport-advice: peer 1 federated (LoRa/FIPS/Tor pills in the
image modal), others radio-only.
- Scripted Flash LoRa job: flash-list-firmware + flash-device +
flash-status advancing download → erase → write → done over ~35s
with live log tail and percent, plus cancel.
- Demo messages carry per-message transport (meshcore/reticulum/fips/
tor variety for the pills + animated route modal) and
sender_pubkey/sender_seq so reactions/replies work.
- Chat-action acks: send-reaction/reply/read-receipt, edit/delete/
forward, send-channel, mesh.refresh, reboot-radio.
- Services classification demo: self-deployed "podsteadr" stack — main
app launchable, its MediaMTX backend (ui:null) files under Services
with no Launch button, mirroring ui_detection's verdicts.
All verified against the running mock: 11/11 runtime checks.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Unread badges came back on every visit (framework showed a phantom "2"
with nothing new): unreadCounts was memory-only, so each page load
replayed the entire message history as "new". Seen-state now persists
as a per-contact highest-seen-message-id watermark in localStorage
(ids are backend-monotonic across restarts); first run after this
ships seeds the watermark from history so nobody gets a wall of stale
badges. Opening a chat advances and persists the watermark for all
twins of the merged conversation.
The hop-route modal the user asked for is now reachable from a visible
per-message "⋯" button (the transport pill remains clickable too), has
a fallback title/branch for messages that predate transport tracking,
and animates: endpoints and link reveal in sequence, a pulse travels
the link, and relay dots blink in order — all disabled under
prefers-reduced-motion.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
v1.7.117 broke Reticulum mesh on OTA-only fleet nodes two ways: the
update swaps only the backend binary and frontend tarball, so nodes
kept a stale archy-reticulum-daemon whose argparse exits on the new
--enable-transport flag (mesh session died on every spawn — confirmed
on framework-pt), and they never had archy-rnodeconf at all, so the
in-app Flash LoRa flow failed with a bare "No such file or directory".
Four-part fix:
- The Rust supervisor probes `daemon --help` and only passes
--enable-transport when the daemon advertises it; unsupported daemons
run edge-only exactly as pre-1.7.117 (tested against stub daemons
both ways + missing-binary fail-safe).
- Both PyInstaller tools ride the frontend tarball's runtime payload
(radio-tools/) and bootstrap.rs promotes them to /usr/local/bin on
startup when bytes differ — the first OTA path that ever updates
them. create-release.sh now rebuilds them every release and the
manifest script hard-fails if they're missing.
- The ISO bundles archy-rnodeconf alongside the daemon (it never did).
- Flash LoRa reports "tool not installed — update the node" instead of
the bare spawn error when rnodeconf is absent everywhere.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The Reticulum daemon gains --enable-transport, which writes
enable_transport = yes into the RNS config it regenerates on every
start, and the Rust supervisor always passes it. Archy nodes now relay
RNS traffic and rebroadcast announces, so archy nodes (and Sideband/
NomadNet peers) beyond direct RF range discover and reach each other
through any archy node in between — edge-only operation left every
node limited to its own radio horizon. RNS's per-interface airtime
caps bound the extra announce overhead on LoRa.
Verified: config generation with the flag on/off, daemon --selftest
green with transport enabled, mesh test module 116/116.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Rewrites the prepared 1.7.117 section to include the mesh flash flow,
first-class Reticulum fixes, radio-first routing, mesh chat polish, the
transactions-modal phone fixes, services-vs-apps classification, the
lightning slow-payment fix, cached-resource page loads, load-shedding,
FIPS uptime hardening, and companion 0.5.25. What's New modal block
regenerated from the new bullets (sync-whats-new --check passes).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A published port no longer implies a web UI. The package scanner used to
synthesize interfaces.main.ui="true" for any container with a port or
onion address, so headless backends — including self-deployed compose
stacks like podsteadr — showed up as launchable apps. New ui_detection
module decides instead: a manifest interfaces declaration (catalog
overlay first, disk second) is definitive; undeclared apps get a short
HTTP probe of the launch port (HTML page, redirect, or browser auth
wall = UI; JSON APIs, raw TCP, dead ports = service), with cached
verdicts and probes gated on running containers. Frontend canLaunch
now refuses curated services outright and only treats a bare runtime
address as launchable for curated known apps.
Works identically for manifest apps and containers deployed by hand
outside the orchestrator. ui_detection tests 6/6, frontend suite
696/696.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Chips become sticky inside the modal's scroll region with a dimmed
blurred band, so the rail filter stays reachable while rows scroll
underneath. Verified in headless chromium: pinned at scroll-region top
after deep scroll, touch swipes starting on the chips still scroll.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Replaces the external flasher.meshcore.co.uk link with a button that
opens the global device-setup modal directly at its flash step, via a
new manual entry point in the mesh store (flashFlowPath). Manual opens
target the connected radio (else the first detected stick), skip the
read-only probe — the port is held by the live session and a second tty
opener corrupts it; the backend flash job stops the listener itself —
and close the modal instead of stepping back to the detection screen.
Button is disabled with a hint when no radio is present.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The tx list kept a vestigial overflow-y-auto from before the modal
contract refactor made BaseModal's slot wrapper the scroller. With a
modal open, modal-scroll-locked applies overscroll-behavior:contain to
every .overflow-y-auto inside the overlay, so touch scrolls latched
onto the non-scrollable inner list and could not chain up to the real
scroller — the modal was unscrollable on any touch device. Wheel input
latches onto the scrollable ancestor directly, which is why desktop
never showed it. Verified with headless-chromium touch synthesis at
360x640/320x568: list scrolls to bottom, background stays contained.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Post-merge regression from combining two individually-correct changes:
main's 2026-07-23 fix makes open_preferred_path bail WITHOUT touching the
port when no device_kind is pinned, while the hw-config branch's skip_path
dedup excludes the preferred path from the auto-detect fallback on the
assumption it was already probed this cycle. Together, on a single-radio
node with no pin (the common fleet state), the only candidate was never
probed at all and the mesh never came up — hit live on archi-dev-box
right after deploying merged main.
Fix: when device_kind is None, skip open_preferred_path entirely and go
straight to auto_detect_and_open with skip_path=None. The pinned path
keeps the existing probe-then-skip fallback.
Verified live on archi-dev-box: radio auto-detected, Reticulum daemon
ready, 5 persisted peers loaded. Mesh tests 116 passed / 0 failed.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Image quality modal: 'Send via' pills (LoRa / FIPS / Tor) when the
peer is federation-reachable — mesh.transport-advice now returns
has_fips + last_transport alongside has_tor. Picking FIPS/Tor routes
the image over the content-ref path instead of the radio.
- Attachment modals (transport chooser, image quality, new hop modal)
Teleport to body so the backdrop dims the FULL viewport — rendered
in-place they sat inside a transformed glass panel that trapped
position:fixed to the right chat panel.
- Click a message's transport pill → route modal: radio hops + live
SNR/RSSI quality for LoRa transports, overlay/circuit shape for
FIPS/Tor, delivery + E2E state.
- Reactions move behind a compact 'React ▾' dropdown with a larger
12-emoji palette.
- Unread badges now clear like a normal chat app: opening a contact
clears ALL twins of the merged conversation (badge sums every
contact_id — clearing just the clicked one left it stuck), and only
once the chat has scrolled to the latest messages; scrolled up into
history, new arrivals accumulate until you scroll back down.
- Refresh button shows only the spinner while refreshing (text+spinner
overflowed the fixed button width).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
E2E-verified both directions dev-box<->x250 over real RF (5KB image ~60s):
1. Sender daemon never called link.identify() — receiver's
get_remote_identity() was None, so every arrived transfer carried an
empty source_hash and the Rust side dropped it (now also warns
instead of silently vanishing it).
2. Receiver treated resource.data as bytes, but RNS hands a concluded
Resource's data as a file-like BufferedReader — b64encode raised
TypeError and the transfer was lost even when attributed.
3. Radio twins of merged contacts carry the peer's Archipelago ed25519
key as pubkey_hex, not an RNS hash — prefix lookup could never match
('Unknown Reticulum prefix', observed live). resolve_dest_hash now
falls back to matching the announce-bound arch_pubkey_hex.
4. The daemon RPC socket kept asyncio's default 64KiB line limit; any
attachment >~48KB overflowed it and tore down the whole daemon
connection ('reticulum-daemon is gone'). Raised to 16MiB.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Both sides independently fixed the serial-alias dedup and the ESP32
boot-reset races; kept the branch's defer-to-auto-detect for unpinned
preferred paths (single probe pass per cycle) on top of main's
advert-name threading, Reticulum name propagation and radio-first
routing. Modal keeps main's 'Set Recommended' naming + probe progress
bar alongside the branch's in-app firmware flasher step.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Two halves of the same twin-resolution gap, found live while testing
images between archi-dev-box and archy-x250-dev:
- peer_dest_prefix resolved the given contact row's own pubkey. For the
UI's merged conversation (the federation-synthetic id) that's the
Archipelago ed25519 identity key, NOT a radio routing key — so every
Reticulum resource send (images/files over LoRa) failed with 'Unknown
Reticulum prefix' while the UI showed the message as sent. It now
resolves through the radio twin (same arch identity, radio-range id).
- send_typed_wire sent EVERY federation-synthetic contact over the
federation path (FIPS→Tor), even with the same node one LoRa hop away.
Policy per operator: LoRa first when the payload fits and the radio
twin is reachable, then FIPS, then Tor. Verified live: text to the
merged contact now logs 'Radio-first routing' and lands with
transport=reticulum on the peer.
Also restyles the mesh-chat attachment download controls: the pre-fetch
button was a bare .btn that squished to text width in the narrow mobile
bubble; now a full-width glass pill with a download icon and fetch
spinner, and the on-image overlay swaps the emoji glyph for a crisp SVG
in a properly-sized glass circle.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
tests/mesh/run-mesh-tests.sh: cheap-first layers — 116 Rust mesh unit
tests, the daemon --selftest (now also asserting the announce app_data
wire contract and the set_name verb), and opt-in live assertions against
a running node (radio connected, named, mesh.refresh/broadcast, no
ARCHY-blob peer names). Verified green on archi-dev-box and archy-x250-dev.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Refresh button: real handler — calls the new mesh.refresh (radio
re-query) plus contacts/federation/outbox re-reads, disabled with a
spinner while running (was an unawaited cache repaint with no feedback
that skipped half the list's data sources).
- Broadcast button: success ('Sent ✓') and failure states with the error
in the tooltip; failures no longer vanish as unhandled rejections.
- The store's 5s status poll no longer wipes the error banner each tick.
- Contacts/aliases, federation nodes and the outbox badge refresh every
~30s (were mount-only and went permanently stale).
- Peer-list empty state keys on the merged list, so federation rows and
the channel rows still render with no radio attached.
- Device setup modal: 'Set Recommended' naming, probe progress bar with
stage labels instead of an anonymous spinner.
- Device panel: name save clears properly (empty = fall back to server
name) and the confirmation reflects the new live apply.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Root causes found and fixed after live debugging on archi-dev-box (all
verified on real RNode hardware, archi-dev-box <-> archy-x250-dev E2E):
- probe_rnode raced the board's own boot: opening the port pulses DTR/RTS
through USB-UART bridges (CP2102/Heltec V3), the ESP32 power-cycles and
spends ~2.5-3s in boot ROM, and the KISS DETECT written 300ms after open
landed in the void — so an RNode could NEVER connect on these boards.
Now: immediate probe (fast path), then drain-until-quiet boot settle and
a second DETECT with a fresh response window.
- MeshService::configure() only restarted the listener on enable/disable —
device_kind/device_path/advert_name/RF-param changes were silent no-ops
until a full process restart (the setup modal's apply/keep-as-is did
nothing). Material config changes now bounce the listener; the open
sequence races the shutdown signal so stop() no longer burns the full
15s timeout mid-probe; mesh.configure applies in the background instead
of stalling every status poll behind the service write-lock.
- The mesh name was write-only: config.advert_name had no reader,
server.set-name never reached the mesh service, and Reticulum's
set_advert_name was a no-op (daemon display name fixed at spawn, and the
ARCHY:2 announce blob REPLACED the LXMF display name — every archy node
was anonymous on RNS). Now: advert_name > server name precedence feeds
the session, renames restart it live, the daemon gets --display-name at
spawn plus a set_name RPC verb, and announces carry the LXMF-standard
msgpack name with the identity blob appended as an extra list element
stock clients (Sideband/NomadNet) ignore.
- Dead reticulum-daemon was invisible for up to 30min (RX-stall watchdog):
child exit / RPC-EOF now fails try_recv_frame so the session reconnects.
- Setup modal re-trigger loop: plugged_at used the tty node's mtime, which
bumps on every open — each probe invalidated the dismissal key. Use
btime/ctime (only change on real plugs).
- ARCHY:2 identity adverts (re-emitted every 60s over Reticulum) stomped
the federation twin's real name with a synthetic Archy-… placeholder and
nulled its position; blob-only announces no longer assert a name, blob
strings can never become display names, and stale blob names are healed
at peers.json load.
- mesh.broadcast on Meshtastic sent heartbeat+time only (no identity);
SendAdvert now also fires a want_response NodeInfo broadcast.
- New mesh.refresh RPC: actively re-queries the radio contact table (the
UI Refresh button previously only re-read server caches).
- Reticulum peers now track last_advert (announce time) and mark existing
peers reachable on inbound traffic.
- Boot auto-enable no longer force-enables mesh when an operator
explicitly disabled it (only fires when no config file exists).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The two FIPS containers on the Server page were the last network cards
still fetching-on-mount into local refs — every visit was a blank card
until fips.status / fips.list-seed-anchors answered. Both now ride the
cached-resource layer: FipsNetworkCard shares the server.fips-summary
key with the Local Network card's FIPS row (one fetch, never disagree),
seed anchors cache under server.fips-seed-anchors, and mutations
write the RPC's authoritative result straight into the cache. The 15s
status poll skips hidden tabs — revalidate-on-focus covers the return.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The HTTP accept loop parked on acquire_owned() when the connection
budget drained, freezing accept() for every client (the .228
session-flapping / CLOSE-WAIT signature). Permits drained because
half-open clients and hung upstreams held them indefinitely.
- try_acquire_owned + immediate 503-and-close when the budget is
exhausted; the accept loop itself never blocks
- 30s http1_header_read_timeout drops slowloris/half-open clients
- 900s watchdog bounds non-upgraded connections; websocket upgrades
are exempt (legitimately long-lived)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Root cause of the .198 load spiral (2026-07-28): needs_repair() called
ensure_image_present() every 30s tick to render the expected unit, so
under IO pressure the image-existence check timed out, read as "image
missing", and a 900s podman build ran inside the PROBE while the
companion was up — each build pegging the disk that made probes fail.
- needs_repair() is now build/pull-free: unit file present → service
is-active (10s cap; a hung systemctl reads as "assume active", never
as dead) → unit matches one of the three image refs install_one could
have written → context-newer-than-image staleness only when the unit
uses the auto-built :latest.
- Per-companion 10-min repair cooldown after a failed install_one, so a
failing build retries at most every REPAIR_COOLDOWN instead of every
reconcile tick.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Bridge WebSocket patches into the resource layer: a /peer-health/<onion>
patch invalidates that peer's cloud.peer-browse entry and the federation
node list; /package-data patches invalidate the tor-services list.
invalidate() debounces 800ms and refetches only keys with mounted
subscribers, so patch storms cost one revalidation per key; the 30s
staleness reconciliation remains the backstop for unmapped data.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Credentials: identity.list + identity.list-credentials become
useCachedResource entries; explicit reloads still toast on failure.
- OpenWrtGateway: openwrt.get-status caches in the shared store (revisits
paint the last router state instantly); the connect flow's
params/No-router-configured semantics are preserved on top of the entry.
- ContainerApps assessed and left as-is: its Pinia store already persists
across navigation, keeps last data on error, and gates the spinner on
empty — same class as Apps/Marketplace/Fleet.
This closes the B4 rollout list from docs/FIPS-UPTIME-AND-UI-STATE-PLAN.md.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The keeps-data-visible-while-refreshing tests for PeerFiles, Server, and
LightningChannels mounted without Pinia (the converted components now
pull the resources store in setup) — add createPinia to the mounts.
LightningChannelsPanel: refresh the main channel list before the closed
history so the primary entry gets the first response, and null-guard
both fetchers' response shapes.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
network summary (4-RPC allSettled aggregate), fips row, vpn peers,
interfaces, and tor services become useCachedResource entries — revisits
paint instantly, background refreshes keep content on screen. Mutations
write through the cache: DNS apply + the 15s vpn poll patch the network
aggregate via optimistic() instead of refetching all four RPCs; peer
removal filters the cached list. loading/refreshing flags derive from
entry loadState (drops the hand-rolled hasLoaded bookkeeping).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
federation nodes + dwn.status become useCachedResource entries: revisits
paint the node list instantly, `loading` fires on true first-load only
(the old showLoader semantics), the 5s poll refreshes silently like the
old surfaceErrors:false path, and explicit reloads after mutations still
surface failures in the error banner. Replaces the hand-rolled
loadNodesWithOptions SWR.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
lnd.listchannels (+summary) and lnd.closedchannels become separate
useCachedResource entries: reopening the panel paints the last channel
lists instantly and revalidates behind them; a closed-history failure
keeps its last list without touching the main view (same semantics as
the old nested try). Open/close mutations still force a refresh.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
monitoring.current/history/alerts/alert-rules become useCachedResource
entries: revisiting the page paints the last snapshot, chart, and alert
list instantly and the 5s poll revalidates behind them (refreshes dedup
in the store; errors keep last-known values instead of blanking).
Alert-rule toggles and acknowledgements refresh their entries after the
mutation as before.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
