When a Quadlet unit file already exists for an orchestrator-managed
backend, sync its on-disk bytes against what the current renderer
produces. write_if_changed makes this idempotent — when bytes match,
no IO; when they differ (post-deploy of a renderer change), the file
is rewritten and systemctl --user daemon-reload runs once.
We deliberately do NOT restart the .service when the file changes:
running containers keep their current config until the operator
restarts them. That's the right tradeoff — file updates are cheap and
non-destructive; service restarts are the SIGKILL cascade we're
trying to eliminate.
Why this matters: pre-this-commit, every renderer change required a
fresh package.install RPC per app to take effect. Observed live on
.228 2026-05-02 — the TimeoutStartSec=600 fix shipped in code but
existing units stayed on the old format because nothing triggered a
re-render. Combined with state.json being empty (so the reconciler's
auto-install path didn't fire either), the fix was invisible until
manual unit deletion.
Companions (UI_APP_IDS) are skipped — companion.rs renders those units
with a different shape; syncing here would clobber them.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Bug surfaced live on .228 2026-05-02 — every backend Quadlet unit
(lnd, electrumx, fedimint, btcpay-server, mempool-api, bitcoin-knots)
hit systemd's default 90s start timeout because Notify=healthy makes
systemctl wait for the first green health probe, but
HealthInterval=30s × HealthRetries=3 = 90s minimum even on a healthy
service. Race: timeout fires the moment the third probe MIGHT succeed.
Result was three different post-states (inactive+running, failed+missing,
inactive+stopped) depending on whether systemd's ExecStopPost ran
podman rm before the orchestrator's adoption logic re-grabbed the
container.
Fix: when health is set, render TimeoutStartSec=600 (10 minutes) into
[Service]. Long enough for slow-starting backends (electrumx index
replay, lnd wallet unlock) without being so long that a truly stuck
unit hangs forever. Companions stay unchanged (no health → no override,
default 90s applies).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two bugs surfaced by the first real-node validation of Phase 3.2-3.4
on .228 (2026-05-02), both caught before flipping the default.
Bug 1 — translate_health_check double-prefixed http://. Manifests in
the wild carry the scheme inside the endpoint string ("http://localhost:8175"),
and we were prepending another http:// unconditionally. Result on .228:
every backend HealthCmd read `curl -fsS -m 5 http://http://localhost...`,
every probe failed, fedimint hit a 14-restart loop. Now we accept either
form and skip appending hc.path when the endpoint already carries one.
Regression test asserts no double-prefix and that an in-endpoint path
is honoured.
Bug 2 — Phase 3.3 migration ran for UI companions (bitcoin-ui /
electrs-ui / lnd-ui) that have shipped via Quadlet since v1.7.41.
Migration tore down the running companion + raced companion.rs render,
producing "Phase 3.3: re-install archy-bitcoin-ui via Quadlet" reconcile
errors and leaving archy-bitcoin-ui down. Companions now short-circuit
out of migrate_to_quadlet_if_needed before any IO. Also: when try_exists
returns Err for an unrelated reason (permissions, EIO), we now skip
migration instead of treating "I can't tell" as "go ahead and migrate" —
migrating on top of a possibly-existing unit is destructive.
What this does not fix yet:
* the orchestrator's reconciler iterating every manifest in
/opt/archipelago/apps/, not just installed apps. Pre-existing
behavior (also affects the legacy path) — separate scope.
* fedimint /data UID mismatch surfaced when Quadlet started fedimint
fresh. Likely orthogonal — defer.
* no rollback when install_via_quadlet fails after a remove_container.
Tracked as Phase 3.3.1 — defer.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds an env-var lever for Phase 3.2's use_quadlet_backends flag so the
20× harness can flip the path on per-node without a config.json edit
(which would require an archipelago.service restart — and that triggers
FM3 cgroup cascade until Phase 3.5 ships, so we can't ask anyone to
reconfigure live nodes that way today).
Truthy parsing centralised in `parse_truthy_env` (1, true, yes, on —
case-insensitive, whitespace-trimmed). Anything else is false. The
helper is unit-tested so future env-var flags can reuse the same shape.
Also adds a default-off regression test for use_quadlet_backends so
flipping the default ahead of the 20× verification fires immediately.
TESTING.md documents the Environment= snippet for the systemd drop-in
so the next operator can flip the flag on a debug node without
re-deriving the recipe.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
A six-test bats suite that validates what install_via_quadlet (Phase 3.2)
is supposed to leave behind:
* `.container` unit on disk in $XDG_CONFIG_HOME/containers/systemd/
with [Container] / [Service] / [Install] sections, Image= present,
and Restart=on-failure (the backend invariant — companions use Always)
* Phase 3.4 cross-check: any unit with HealthCmd= must also emit
Notify=healthy, otherwise systemctl start won't gate on health
* `systemctl --user is-active` returns 0 for the .service
* podman shows the container running
* the container's cgroup is under user.slice/, NOT under
archipelago.service — the kernel-level proof that FM3 cgroup
cascade SIGKILL is structurally fixed for this container
Auto-skips on every test when no backend Quadlet units exist (today's
default state, use_quadlet_backends=false) — so the suite is a no-op
on current fleet boxes and turns into a hard regression gate the
moment anyone flips the flag and reinstalls.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
QuadletUnit gains an optional HealthSpec; from_manifest translates the
manifest's health_check (tcp/http/cmd) into a HealthCmd= directive and
emits Notify=healthy alongside it. systemctl start <unit>.service then
blocks until the container's first green probe — eliminating the
"container up but RPC not ready" race the orchestrator currently papers
over with post-start polling.
Translation policy:
* tcp, endpoint "host:port" -> nc -z host port
* http, endpoint "host:port", path -> curl -fsS -m 5 http://endpoint<path>
* cmd, endpoint "<shell command>" -> verbatim
* unknown type / malformed endpoint -> None (skip Notify=healthy rather
than emit a HealthCmd that hangs the unit start forever)
Companion units leave health: None and remain byte-identical to before
this PR — the renderer only emits the Health* / Notify= block when set.
+4 quadlet unit tests (19 total). Dropped a never-used test setter that
was generating a dead_code warning.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When use_quadlet_backends flips from off → on, existing fleet boxes
have backend containers parented under archipelago.service's cgroup
(the bad shape that triggers FM3 cascade SIGKILL on every archipelago
restart). ensure_running now notices and corrects this:
* If there's already a `<name>.container` unit on disk → no-op
(subsequent reconcile ticks take this fast path).
* Else if a podman container with that name exists → it's a pre-3.3
artifact. Stop+remove it (volumes survive — bind mounts are not
touched by `podman rm`), then write the Quadlet unit, daemon-reload,
and start the new managed service.
* Else → fall through to install_fresh, which already routes through
install_via_quadlet when the flag is on.
The migration is idempotent and self-healing: if a fleet box is
half-migrated (unit on disk but no service active, or service active
but stale unit), the next reconcile tick converges. Bitcoin chain
data, lnd wallet state, and electrumx index all live on host bind
mounts and are unaffected by the container-record swap.
Volume safety audited per backend in `uses_orchestrator_install_flow`
allowlist — every entry mounts its data dir as a host bind mount.
Default still off. To migrate a node:
/etc/archipelago/config.toml: use_quadlet_backends = true
followed by `systemctl restart archipelago` — the next reconcile tick
walks every managed app and migrates each in turn.
Tests: 624 passing, 0 cargo warnings.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
prod_orchestrator::install_fresh now branches on the new
Config::use_quadlet_backends flag (default false):
* off (today's production behavior) — unchanged: runtime.create_container
+ start_container, container parented under archipelago.service's
cgroup, FM3 cascade SIGKILL on every archipelago restart.
* on — install_via_quadlet renders the manifest as a Quadlet unit via
QuadletUnit::from_manifest, writes it atomically into
~/.config/containers/systemd/, calls daemon-reload, and starts the
generated <name>.service. Container ends up under user.slice — no
more cgroup parented under archipelago, so archipelago restarts
don't touch the container's lifetime.
Default off so this commit is structurally safe to ship: nothing
changes at runtime until an operator opts in. Flip the default once
tests/lifecycle/run-20x.sh has gone green against the new path on
.228 + .198 (the v1.7.52 release gate).
Plumbing:
* config.rs — `use_quadlet_backends: bool` w/ Default false
* prod_orchestrator.rs — flag stored on the struct, threaded through
new(), with set_use_quadlet_backends(bool) test setter
* prod_orchestrator.rs — install_via_quadlet helper
* dropped the Phase-3.1 #[allow(dead_code)] markers on from_manifest /
parse_memory_mib / RestartPolicy::OnFailure now that the call path
exists; if a future revert removes the wiring, the warnings come back.
Tests: 624 passing, cargo check clean (0 warnings). Existing companion
behavior unaffected — render_skips_backend_directives_when_default
still passes byte-equal to before quadlet.rs grew the new fields.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The QuadletUnit struct now covers everything a backend manifest needs
(ports, environment, devices, add_hosts, entrypoint+command, read-only
root, no_new_privileges, cpu_quota, restart policy choice). Adds
QuadletUnit::from_manifest(&AppManifest, name) that translates a parsed
manifest into a unit, plus parse_memory_mib for "1g"/"512m"/raw-MiB
forms. The renderer skips empty/false directives so existing companion
units render byte-identically — no behavior change for shipping
companions; the backend renderer is dead code until Phase 3.2 wires it
into the orchestrator.
Eight new unit tests cover:
* parse_memory_mib forms (1024, 512m, 2g, garbage)
* shell_join quoting (whitespace, embedded quotes)
* RestartPolicy → systemd string mapping
* render emits backend directives when set
* render skips them when defaulted (companion regression gate)
* from_manifest happy path on a bitcoin-knots-shaped manifest
* from_manifest read-only volume detection
* from_manifest tmpfs filtering
* end-to-end manifest → render bytes assertion
Tests: 615 → 624 (+9 net; one pre-existing parse_memory_mib path was
implicitly covered before but is now explicit). Cargo warnings: 0.
`from_manifest`, `parse_memory_mib`, and `RestartPolicy::OnFailure` are
marked allow(dead_code) with explicit references to Phase 3.2 — if
3.2 doesn't wire them, the dead-code warning resurfaces.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Brings L1 (RPC API) + L3 (lifecycle survival) parity coverage to the
three multi-app stacks that were previously only touched by
required-stack.bats. Combined with bitcoin-knots / lnd / electrumx
already shipping, the six core apps now have dedicated bats files.
Each suite is shaped like the existing single-container suites
(bitcoin-knots / lnd / electrumx) and gates every assertion on the
backing container actually being present, so a node without the stack
installed gets clean skip messages instead of false fails.
* btcpay.bats — 9 tests, including stack-wide presence and a
"supporting containers don't cascade-restart" guard
* fedimint.bats — 8 tests, single container
* mempool.bats — 9 tests, mixed legacy + orchestrator-managed stack;
reuses the :8999 mempool-api probe from required-stack for parity
Total bats now: 88 (was 53 → +35).
TESTING.md matrix advances 23 → 50 of 110 cells.
UI URL coverage for these three apps already lives in
ui-coverage.bats, so this PR doesn't duplicate proxy-path probes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Single source of truth for "where are we, where are we going" on the
v1.7.52 container excellence work. Replaces ad-hoc tracking in chat.
Sections:
* Test layers L0..L6 with toolchain + per-iteration latency
* Per-app × per-state coverage matrix (23 of 110 cells today; goal 110)
* Layer-by-layer status (L0+L1+L2 ●; L3 ◐; L4..L6 ○)
* Run commands (single suite / full suite / 20×)
* LoC budget — -270 committed, ~1,616 more possible if Phase 3 ships
* Performance KPIs (TBD — measure first, target second)
* Release gates — 8 boxes that must tick before v1.7.52 ships
The file lives in-repo so PR diffs to it answer "what did this commit
improve?". If you can't tick the box, the change isn't ready.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Closes the coverage gap where existing bats suites would report green
on a node whose dashboard tiles 502 because the proxy upstream is dead.
First pass against .198 caught real prod issues immediately:
/app/lnd/ → 502 (lnd container exited)
/app/mempool/ → 502 (mempool container exited)
/app/fedimint/ → 502 (fedimint container exited)
while existing tests reported only "container is up: false" with no
404/502 distinction.
* lib/ui-probes.bash — sourced helper. probe_https_200,
probe_app_url (skip-if-container-down else assert-200),
probe_dashboard_shell (asserts the Vue SPA HTML, not nginx default —
catches the layout regression from feedback_release_tarball_layout.md),
probe_dashboard_catalog (asserts /catalog.json non-empty).
* bats/ui-coverage.bats — 9 @test cases covering the dashboard +
bitcoin-ui :8334 + the seven HTTPS_PROXY_PATHS most users hit
(lnd, electrumx, mempool, fedimint, btcpay, filebrowser).
URL list mirrors HTTPS_PROXY_PATHS in
neode-ui/src/views/appSession/appSessionConfig.ts. Divergence between
the two is the exact bug class we're guarding against.
Loops clean under run-20x.sh. Container-state oracle is via local
podman inspect, so the suite must run on the archy host (same as
companion-survives-archipelago-restart.bats).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Extracted the heal_podman_state cleanup list as a module-level
HEAL_RUNTIME_SUBDIRS const so a unit test can structurally enforce
the invariant: the list must contain "containers" + "libpod" but
must NOT contain "podman" (which holds systemd's podman.sock
listener and was the bug fixed in commit bb421803).
If anyone re-adds "podman" — accidentally, by reverting, or by
copy-paste from old plan memory — this test fires before we ship,
not on the next deploy when it nukes the orchestrator's HTTP path.
Total tests: 614 → 615.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Sister suite to companion-survives-archipelago-restart.bats. That one
tests the same property for UI companions, which already ship via
Quadlet (commit 6e716f68) and so already pass.
This new suite tests the property for backend containers (bitcoin-knots
/ bitcoin-core / lnd / electrumx). Until v1.7.52 Phase 3 ships these
under Quadlet too, the suite is EXPECTED TO FAIL on fleet boxes — it's
the executable definition of "FM3 fixed".
Observed live on .198 on 2026-05-01: `sudo systemctl stop archipelago`
killed every container in archipelago.service's cgroup. The dedicated
"backends survive archipelago restart" test catches exactly that, and
also verifies the SAME container instance survives (compares pre/post
.Id), so an orchestrator that recreates a fresh container after the
SIGKILL doesn't read as pass.
Three @test cases:
* destructive gate (skip-marker for the suite)
* baseline: at least one backend installed + running
* backends survive: same .Id pre + post archipelago restart
Don't gate releases on this passing until Phase 3 lands; before then
treat it as a "expected to fail / shows progress" indicator.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Same shape as bitcoin-knots.bats and lnd.bats so the 20× release-gate
exercises electrumx through the same state matrix it uses for the other
two core apps. electrumx previously had a single TCP-port check inside
required-stack.bats; this adds destructive + cascade-destructive tiers.
10 @test cases:
* read-only: presence, valid state, TCP port (50001) reachable, no
orphan containers beyond {electrumx, archy-electrs-ui}
* destructive: stop, start, restart, TCP port recovers within 120s of
cold restart (longer than bitcoind because electrumx replays its
index against bitcoind on start)
* cascade: uninstall, reinstall (240s timeout for index rebuild)
With this suite, the three single-container core apps (bitcoin-knots,
lnd, electrumx) now have parity coverage. Multi-container stacks
(btcpay, mempool, fedimint) come next.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Mirrors bitcoin-knots.bats so the 20× release-gate run exercises lnd
through the same state matrix. lnd previously had only a single
read-only check inside required-stack.bats; this adds the destructive
and cascade-destructive tiers that match what we already test for
bitcoin-knots.
10 @test cases:
* read-only: presence, valid state, lncli getinfo, no orphan containers
* destructive (ARCHY_ALLOW_DESTRUCTIVE=1): stop, start, restart,
RPC recovers within 90s of cold restart (longer than bitcoind
because the wallet has to unlock first)
* cascade (ARCHY_ALLOW_CASCADE_DESTRUCTIVE=1): uninstall, reinstall
Reuses the same lncli invocation as required-stack.bats so divergence
shows up clearly if either test breaks.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Phase 4 of the v1.7.52 container excellence plan: a release-gate harness
that loops the bats suite N times in a row, with teardown between
iterations, and reports a pass/fail tally.
* setup-teardown.sh — clears /tmp/archy-rpc-session-* between runs so
iteration N+1 doesn't reuse a logged-out cookie from iteration N.
Idempotent; safe to run anytime. Designed to grow as we add suites
that leave other transient state.
* run-20x.sh — wraps run.sh in a loop of ARCHY_ITERATIONS (default 20).
Tracks per-iteration pass/fail with wall-clock timing, prints a
results block, exits non-zero on any failure. Honors ARCHY_FAIL_FAST
for short-circuit during dev.
Suggested release-gate command:
ARCHY_PASSWORD=password123 ARCHY_ALLOW_DESTRUCTIVE=1 \
tests/lifecycle/run-20x.sh
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Observed live on .198: heal_podman_state was removing
$XDG_RUNTIME_DIR/podman/ alongside containers/ and libpod/. That dir
holds the systemd-bound podman.sock — the listener systemd creates for
socket-activated podman.service. Removing it broke every libpod HTTP
call from the orchestrator until `systemctl --user restart
podman.socket` ran. Far worse than any wedge it was trying to repair.
Drop podman/ from the cleanup list. The runtime state we actually want
to clean for FM6 (bolt_state.db drift) lives in containers/ and
libpod/ only.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
cargo check was showing five real warnings, all genuinely dead:
* container/mod.rs — re-exports compute_container_name, AdoptionReport,
ReconcileAction, ReconcileReport were unused outside
prod_orchestrator. Drop from the pub use line.
* prod_orchestrator — with_runtime + insert_manifest_for_test only exist
for the test module in the same file. Mark them
#[cfg(test)] so they don't appear in release builds.
* async_lifecycle — remove_package_entry has no callers; doc claims
"used for install-failure cleanup" but nothing
cleans up. Delete (10 lines).
* registry.rs — `use tracing::{debug, info};` had no consumers.
* fips.rs — unused-assignment chain on last_status. The poll
loop always sets it on every break path, so the
initial `None` and the unwrap_or_else fallback
were both dead. Refactored to `let after = loop
{ ...; break s; };`.
cargo check is now clean. cargo test --workspace --bins: 614 passed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Closes FM6 (podman bolt_state.db / runtime drift) — observed live on
.198 today: bitcoind was running for several minutes, but podman's
state DB reported the container as Exited. The reconciler then tried
to "restart" it, racing the still-bound port 8332 and failing in a
loop.
heal_podman_state() runs as the last bootstrap stage, BEFORE the
orchestrator's reconcile loop ticks. It probes `podman info` with a
5s timeout; on failure it removes the runtime-state dirs under
$XDG_RUNTIME_DIR and re-probes. Persistent storage under
~/.local/share/containers/storage/ is never touched, so containers
re-discover from manifests on next call.
Cleanup never includes `podman system reset` or `system renumber` —
those are destructive and must stay operator-only.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
DependencyResolver had zero call sites in prod or tests outside the
module itself. The actual install-time dependency check lives in
install.rs::detect_running_deps + check_install_deps; this DAG-walk
solver was never wired up. -268 LoC.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
If bitcoin-core was installed but never started (e.g. port 8332 already
bound by bitcoin-knots), the container sticks in `created` state forever.
The old conflict check refused EVERY future bitcoin install — including
re-install of the running variant — leaving no UI path to recovery.
Now the check distinguishes states:
- missing → no conflict, continue
- running → real conflict, refuse install
- created/exited/configured/... → stuck; auto-remove and continue
Volumes are untouched; only the dead container record goes away.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Bitcoin containers were exiting in ms after start because the orchestrator
install path skipped the credential-materialisation step the legacy path
did. resolve_secret_env then failed to read
/var/lib/archipelago/secrets/bitcoin-rpc-password, the container started
with no password, and bitcoind crashed before logs were useful.
Two changes:
1. install.rs — call bitcoin_rpc_credentials() for bitcoin/bitcoin-core/
bitcoin-knots before any install branch runs. The function generates +
persists on first call (OnceCell-cached), so this is idempotent.
2. manifest.rs::resolve_secret_env — return ManifestError::Invalid when a
resolved secret trims to empty, instead of silently producing
`KEY=` env vars that crash auth.
Adds a unit test for the empty-secret rejection.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Phase 3a of the install path consolidation. Two coupled changes:
1. install.rs handle_package_install: gate the legacy "container exists →
adopt + return" probe on !orchestrator_managed. Apps the orchestrator
knows about (bitcoin-knots, bitcoin-core, lnd, electrumx, fedimint,
filebrowser, btcpay-server stack apps, mempool stack apps, plus the
companion UIs that just moved to Quadlet) skip the legacy probe and
fall straight into the orchestrator branch.
The legacy adopt block was returning success on a bare `podman start`
exit-0 — even when the process inside the container crashed seconds
later. That's the .228 "running but unreachable" failure mode. The
orchestrator's ensure_running honors the manifest's health check and
pre-start hooks (e.g. re-renders bitcoin-ui's nginx.conf if the RPC
password rotated), so this is a behavioral upgrade, not just a
refactor.
2. ProdContainerOrchestrator::install: make idempotent. Previously it
blindly called install_fresh which would fail on `podman create` if
the container name already existed. Now it delegates to ensure_running:
- Container Running + healthy → no-op (refresh hooks, restart if
config rewritten)
- Container Stopped/Exited → start (with hook refresh)
- Container missing → install_fresh
- Container in wedged state (Created/Paused/Unknown) → force-recreate
Without this, change #1 would regress every "container already exists"
case for the 18 orchestrator-managed app IDs. With it, install becomes
the single source of truth for "make app X be in the desired state."
Tests: 654 passed across the workspace (614 unit + 37 orchestration + 3
rpc), 0 failures. The 20 prod_orchestrator tests cover the install /
ensure_running / reconcile paths the new install delegates through.
Net delta: install.rs grows by ~30 lines (gating wrapper + comments),
prod_orchestrator.rs grows by ~30 lines (idempotent install body). Both
are temporary — the larger deletions (~1700 lines) come once every app
has been verified through the orchestrator path in subsequent phases.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Companion UI containers (archy-bitcoin-ui, archy-lnd-ui,
archy-electrs-ui) used to be launched as fire-and-forget tokio::spawn
blocks from install.rs. If archipelago crashed mid-spawn or the
container's cgroup was reaped, companions vanished from podman ps -a
and only a manual rm/run could bring them back (the .228 incident).
Now each companion is rendered as a Quadlet .container unit under
~/.config/containers/systemd/, daemon-reloaded, and started via
systemctl --user. systemd owns supervision from that point on:
- archipelago can crash, restart, or be uninstalled without touching
any companion.
- Quadlet's Restart=always + RestartSec=10 handles container exits.
- A 30s reconcile tick in boot_reconciler enumerates expected
companion units and re-installs any whose unit file or service
vanished — defense-in-depth against external tampering.
New module layout:
- container/quadlet.rs: pure unit renderer + atomic write_if_changed
+ systemctl helpers (daemon_reload_user / enable_now / disable_remove
/ is_active). 6 unit tests, no I/O in the renderer.
- container/companion.rs: per-app companion specs, install/remove/
reconcile, image presence (build local first, fall back to insecure
registry only via image_uses_insecure_registry whitelist). 2 tests.
install.rs handle_package_install now ends with a single call to
companion::install_for(package_id), replacing 287 lines of spawn-and-
hope shellouts plus a ~120-line nginx auth-injector helper that worked
around per-node RPC password baking. The helper is gone too — the
pre-start hook renders the per-node nginx.conf to /var/lib/archipelago/
bitcoin-ui/nginx.conf and the Quadlet unit bind-mounts it read-only.
runtime.rs handle_package_uninstall now disables companions before
the container rm loop. Otherwise systemd's Restart=always would
respawn each companion within ~10s of removal.
Tests: 53 container tests pass, including 6 quadlet renderer tests
(host network, bridge network, capability set, atomic write idempotence)
and 2 companion specs (per-app companion lookup, build_unit shape).
boot_reconciler tests gain a #[cfg(test)] without_companion_stage()
flag so the paused-clock fixtures don't race the real systemctl I/O.
A bats regression test (companion-survives-archipelago-restart.bats,
gated on ARCHY_ALLOW_DESTRUCTIVE=1) asserts the .228 failure mode
cannot recur: every installed companion has a unit file, services
stay active across systemctl --user restart archipelago, and a
deleted unit file is recreated within one reconcile tick.
Net delta: +941 / -363, but the +941 is mostly tests (~440 lines)
and the new declarative layer; the imperative tokio::spawn block and
its nginx-auth helper are gone, removing two failure classes
(orphan companions on archipelago crash, and post-start exec races
under tightly-confined cgroups) that previously needed manual SSH
recovery.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three small, focused tightenings:
- core/container/src/podman_client.rs: drop the legacy Hetzner
23.182.128.160:3000 mirror from image_uses_insecure_registry().
It was decommissioned in v1.7.x and is stripped from active
registry config at load time; leaving it in the bypass list let
a stale config still skip TLS. Replace the inline match with a
named INSECURE_REGISTRY_HOSTS slice so future entries are one
line. Test now also pins the spoofing-immune semantics
("evil.example/146.59.87.168:3000/x" must NOT match).
- core/archipelago/src/api/rpc/package/config.rs: split bitcoin
from lnd in get_app_capabilities(). bitcoind never opens raw
sockets — drop CAP_NET_RAW from bitcoin/bitcoin-core/bitcoin-knots.
lnd/fedimint/fedimint-gateway keep it because they enumerate
network interfaces during cert generation.
- core/archipelago/src/bootstrap.rs: tighten_secrets_dir()
enforces 0700 on /var/lib/archipelago/secrets and 0600 on every
file inside on each startup. The dir-mode is the load-bearing
isolation boundary against rootless container escapes (their UID
maps to >=100000, can't traverse uid=1000/0700). The per-file
sweep is defense-in-depth against any installer that wrote 0644.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Snapshots the in-flight hardening work so subsequent reconcile/Quadlet
phases land on a clean before/after diff.
Changes:
- core/container/src/podman_client.rs: image_uses_insecure_registry()
whitelist for the OVH (146.59.87.168:3000) and legacy Hetzner
(23.182.128.160:3000) HTTP mirrors; podman_network_settings() lifts
custom networks into the Networks map so containers can join them.
- core/archipelago/src/container/prod_orchestrator.rs:
ensure_container_network() creates per-manifest networks on demand;
apply_data_uid() now goes through host_sudo for mkdir -p + chown so
bind-mount roots get created and chowned without password prompts.
- core/archipelago/src/api/rpc/package/{install,update,stacks}.rs:
podman pull adds --tls-verify=false only for whitelisted registries.
- core/archipelago/src/bootstrap.rs: removes stale dev-mode systemd
override on startup (live nodes carried it from old installers).
- core/archipelago/src/config.rs: ignore ARCHIPELAGO_DEV_MODE in prod
binaries — it had been silently rerouting volumes to /tmp.
- apps/bitcoin-{core,knots}/manifest.yml: locate bitcoind at runtime
so image-layout differences don't break entrypoint.
- scripts/app-catalog-image-smoke-test.py: production catalog/image
smoke test that probes a target node before users click Install.
- .gitignore: cover .codex, .pnpm-store, __pycache__, *.bak.
Removes filebrowser.rs.bak and two stale catalog.json.bak files
(verified identical to live counterparts).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Hotfix: archipelago.service ExecStartPre now mkdirs /run/containers and
/var/lib/containers before the unit's mount-namespace setup tries to bind
them. Without this, fresh nodes that don't have /run/containers (e.g.
nodes provisioned without a prior podman session) fail at the namespace
step with:
Failed to set up mount namespacing: /run/containers: No such file or directory
Failed at step NAMESPACE spawning /bin/bash: No such file or directory
Existing nodes don't pick up systemd unit changes via OTA — they need a
one-time `systemctl edit archipelago` adding the same mkdir. ISO installs
from this version forward have the fix baked in.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Sync-perf tuning for bitcoin/bitcoin-core/bitcoin-knots/electrumx.
- Drop the --cpus=2 cap on bitcoin/electrumx variants. Script verification
is parallelizable; the cap halved IBD speed on 4-8 core machines.
- Bump bitcoin --memory 4g→8g so dbcache=4096 has headroom for mempool +
connection buffers + I/O. 4g was OOM-prone during heavy IBD.
- Bump electrumx --memory 1g→2g + add CACHE_MB=2048 + MAX_SEND=10MB.
- bitcoin-core CLI args gain -dbcache=4096 -par=0 -maxconnections=125.
- bitcoin-knots manifest matched (1024MB pruned / 4096MB full + par=0).
Future v2: host-RAM-aware dbcache scaling.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Resilience-validated release. Three full sweeps of the new resilience
harness against .228 confirm no shipstoppers.
Big user-visible:
- Bitcoin RPC auth durably correct via host-rendered nginx.conf bind-mount,
replaces fragile post-start exec that failed under restricted-cap rootless
podman ("crun: write cgroup.procs: Permission denied")
- Multi-container stack installs (indeedhub, immich, btcpay, mempool) now
emit phase events at every boundary so the progress bar advances
- Apps no longer vanish from the dashboard mid-install (absent-scanner skips
packages in transitional states)
- Indeedhub fresh installs work end-to-end (was 8500+ restart loop): five
missing env vars (DATABASE_PORT, QUEUE_HOST, QUEUE_PORT,
S3_PRIVATE_BUCKET_NAME, AES_MASTER_SECRET) added to install code
- Tailscale install fixed: --entrypoint string was being passed as a single
shell-line arg; switched to custom_args array
- Catalog cleaned of broken entries (dwn, endurain, ollama removed; nextcloud
restored on docker.io)
- Bitcoin Core update path uses correct image (was looking for nonexistent
lfg2025/bitcoin:28.4)
- ISO installs now allocate swap on the encrypted data partition
Infra:
- New resilience harness (scripts/resilience/) — black-box state-machine
tester, every app × every transition. Run before each release.
Sweep #3 final: PASS 107 / FAIL 12 / SKIP 14. The 12 fails are 1 cosmetic
(homeassistant trusted_hosts), 8 harness/timing false-positives, and 3
non-shipstopper tracked items. Down from 23 in baseline sweep #1.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
self-update.sh previously rebuilt only the backend binary and Vue
frontend. The custom UI containers (archy-bitcoin-ui, archy-lnd-ui,
archy-electrs-ui) were left untouched forever. That meant any change to
docker/<ui>/{Dockerfile, nginx.conf, index.html, ...} never reached a
running node through OTA; it required a manual SSH + rebuild. This is
exactly why the lnd-ui port fix didnt reach .228 in v1.7.43-alpha.
Add a sync-and-rebuild stage:
1. Hash each docker/<ui>/ tree (content-only, path-stable via
`cd && find` so src and dst compare equal when identical).
2. rsync changed trees to /opt/archipelago/docker/<ui>/.
3. For each changed UI: rebuild image as the archipelago user
(rootless podman), then stop+remove+recreate the container using
the canonical spec from scripts/container-specs.sh. Port mappings,
caps, memory, and security opts all come from the spec, so the
runtime cant drift from the tree.
Also install first-boot-containers.sh into /opt/archipelago/scripts/ so
a later reconciler run or reboot picks up current orchestration logic.
Idempotent: if no UI tree changed since the last update, the whole stage
is a no-op beyond the hash compare. Verified end-to-end on .228 with a
synthetic change to lnd-ui: detection, sync, build, recreate, and HTTP
200 on both the direct container port and the host-nginx /app/lnd/
proxy.
The LND UI container was unreachable on .228 after the v1.7.43-alpha
deploy because three sources of truth disagreed on which port nginx
listens on inside the container:
- docker/lnd-ui/nginx.conf listen 8081
- docker/lnd-ui/Dockerfile EXPOSE 8080
- apps/lnd-ui/manifest.yml host networking, ports: []
- scripts/first-boot-containers.sh -p 8081:8080
- scripts/deploy-to-target.sh -p 8081:80 (de-facto)
- scripts/deploy-tailscale.sh -p 8081:80
- scripts/container-specs.sh SPEC_PORTS=8081:80
Result: podman published host 8081 to container port 80, but no one was
listening on 80 inside, so connections were reset. Canonicalize on
container:80 with host:8081 publish, matching the three deploy paths
already in agreement.
Changes:
- docker/lnd-ui/nginx.conf: listen 8081 -> listen 80
- docker/lnd-ui/Dockerfile: EXPOSE 8080 -> EXPOSE 80
- apps/lnd-ui/manifest.yml: replace host-network (never true) with
bridge networking and explicit 8081:80 port mapping, correcting a
documentation-vs-reality mismatch
- scripts/first-boot-containers.sh: -p 8081:8080 -> -p 8081:80, and
fix the internal-port comment
Verified on .228 after rebuild: curl http://127.0.0.1:8081/ returns HTTP
200 and the /app/lnd/ host-nginx proxy resolves cleanly.
Releases no longer ship as bootable ISOs. Archipelago updates are
distributed as the backend binary plus a frontend tarball referenced by
releases/manifest.json. Nodes OTA-update via scripts/self-update.sh.
Filebrowser and AIUI remain bundled inside the frontend tarball and
deployed atomically, verified present in v1.7.43-alpha release artifact
(189 AIUI files, filebrowser-client bundle).
Archived under image-recipe/_archived/ (resurrectable if ISO distribution
is reintroduced):
- build-auto-installer-iso.sh
- build-unbundled-iso.sh
- test-iso-qemu.sh
- scripts/convert-iso-to-disk.sh
- BUILD-ISO-STATUS.md, ISO-BUILD-CHECKLIST.md
- branding/isohdpfx.bin
- .gitea/workflows/build-iso-dev.yml
Updated release process docs to drop ISO references:
- scripts/create-release.sh (next-steps text)
- docs/BETA-RELEASE-CHECKLIST.md
- docs/hotfix-process.md
- README.md
Every OTA self-update and every ISO capture was implicitly relying on
/opt/archipelago/web-ui/aiui/ already being present on disk. Any node that
had its web-ui directory atomically swapped (for example by a manual
deployment shipping only neode-ui dist output) lost aiui entirely and the
AI Assistant tab fell through to the "needs to be enabled" placeholder.
self-update.sh: drop the rsync --exclude aiui preservation trick and
instead stage demo/aiui into the freshly-built dist tree before rsync.
demo/aiui in the repo is now the source of truth; every update overwrites
the on-disk copy with a matching version rather than carrying forward
whatever stale bundle happened to survive.
build-auto-installer-iso.sh: prepend demo/aiui to the AIUI search list so
ISO builds from a fresh repo clone pick it up automatically, without
requiring a side-checkout of the AIUI project or a live dev server.
This matches create-release-manifest.sh which already bakes demo/aiui
into the release tarball (lines 86-89).
Four production-code fixes merit user-visible mention: the transport
chunking data-corruption fix (real user-affecting bug for multi-chunk
mesh payloads), the avatar u16 overflow panic (backend crash on certain
seeds), the outbox TTL boundary, and the image-versions parser hardening.
Several tests had drifted from the current production behavior:
- identity_manager: create() already auto-provisions a Nostr key, so the
explicit create_nostr_key() call failed with "already exists". Rewrite
the test to assert on record.nostr_npub from create() directly.
- mesh/protocol: test_build_app_start read the app name from frame[4..]
but the v2 layout is [0:marker][1-2:len][3:cmd][4:version][5..:name].
test_identity_broadcast_roundtrip expected input DID = output DID but
the v2 decoder derives DID from the ed25519 pubkey, so the roundtrip
compares against did_key_from_pubkey_hex(&pub) now.
- mesh/bitcoin_relay: test_build_block_header_announcement asserted
sig.is_some(), but the builder intentionally emits an unsigned envelope
to fit the 160-byte LoRa limit; assert sig.is_none(). Also widen
placeholder hashes to the required 64 hex chars (32 bytes).
- update: load_mirrors() now merges default mirrors post-migration, so
the roundtrip test must assert the custom mirror survives alongside
the defaults rather than strict equality.
- wallet/cashu: test_proof_c_as_pubkey used hex that is not on the curve;
replace with the secp256k1 generator point G so parsing succeeds.
- fips: test_status_reports_no_key_pre_onboarding asserted npub.is_none(),
which fails on dev boxes where the fips daemon is already running. Keep
the !key_present assertion and drop the npub one.
Credentials tests created a fresh tempdir and immediately invoked
encrypt/decrypt, but load_encryption_key reads <dir>/identity/node_key
which did not exist, so every test failed with "node key not found".
Add a test_dir_with_node_key() helper that writes a deterministic 32-byte
key and switch all 8 call sites to it.
SessionStore::new() reads /var/lib/archipelago/sessions.json, which on
any node with an active dashboard contains live sessions that pollute
test state and cause intermittent failures. Introduce a cfg(test) only
new_for_tests(PathBuf) constructor and switch the test suite to it so
tests always start from a clean tempdir.
The parser retained any key ending in _IMAGE, so a harmless-looking
variable like NOT_AN_IMAGE="something" would be treated as a pinned
container image. Add a value-shape check: the value must contain both
a registry separator (/) and a tag separator (:) to qualify.
is_expired used age > ttl_secs, so a message with ttl_secs=0 whose age
rounded to 0 seconds was considered live forever. Switch to >= so the
zero-TTL boundary expires on the first check, matching the intuitive
meaning of TTL and the behavior the tests assert.
hue_color and accent_color computed (seed as u16) * 360, which overflows
u16 when seed >= 182 — debug builds panicked, release wrapped silently.
Widen to u32 before the multiplication.
This also unblocks several identity_manager tests that constructed avatars
through master_node_svg and were aborting on the panic.
encode_chunked() split the payload into shards first, then overwrote
the first 4 bytes of shard 0 with a u32 length header, then re-ran
Reed-Solomon to regenerate parity over the now-corrupted shards. The
decoder correctly read the length header and trimmed `[4..4+len]`
from the reconstructed buffer, but those first 4 bytes had already
been destroyed on the encode side, so every chunked mesh payload
lost its first 4 bytes.
Restructure: reserve 4 bytes for the length header up front, build
a single contiguous [len][data][pad] buffer, then split into shards.
Parity is computed over the correct shards on the first pass, no
double-encode needed.
Update test_chunk_roundtrip_medium: 500 bytes + 4-byte header = 504
bytes, which is 5 data shards (ceil(504/124)), not 4. The old test
assertion was wrong all along and masked the corruption bug because
it only checked the roundtripped bytes, which is exactly what we
need to verify. New assertion is correct.
Verified: all 7 transport::chunking tests pass.
The backend runs as `archipelago` and calls `install_log()` to append
audit lines to the install log on every install / update / remove /
start / stop / restart. Target path was /var/log/archipelago-container-installs.log,
which does not exist and cannot be created by the service because
/var/log/ is root-owned. OpenOptions errors were silently swallowed,
so the log was never written on any node.
Ship a tmpfiles.d rule that pre-creates /var/log/archipelago/ and
container-installs.log with archipelago:archipelago ownership. Move
the const path to match, keeping logs inside the directory logrotate
already rotates (image-recipe/configs/logrotate.conf). Install the
rule from both the ISO build and self-update, and apply it
immediately on self-update so existing nodes get a working log
without needing a reboot.
Verified on .228: file created, backend user can write, backend
binary rebuilt with new const.
Document that OTA updates now refresh the reconcile helper scripts,
closing the deploy gap that kept fixes to those scripts from
reaching existing nodes.
The OTA self-update path only refreshed image-versions.sh, leaving
reconcile-containers.sh and container-specs.sh frozen at whatever
version was baked into the ISO that originally provisioned the
node. Any fix to those scripts (notably the --create-missing flag
and the DISK_GB detection fix shipped this round) never reached
existing nodes, and on .228 both scripts were outright missing
because the node predated their inclusion in the ISO recipe.
Install all three helper scripts to /opt/archipelago/scripts/ on
every self-update run. Also preserve the legacy copy of
image-versions.sh at /opt/archipelago/image-versions.sh for any
older backend binaries still looking there first.
The update flow removes the old container before starting the new
one. If the update fails after removal, the rollback path tries
`podman start <name>` first, then falls back to reconcile. But
reconcile without --create-missing treats the now-absent container
as an optional one that the install flow will (re)create later,
and skips it. Result: container stays destroyed until someone
notices and runs reconcile manually.
Add --create-missing to the rollback reconcile invocation so the
fallback actually rebuilds the container from its canonical spec.
Fixes the failure mode observed on .228 where a bitcoin-knots
update left the node with no bitcoin-knots container at all.
Add two user-facing release notes for fixes shipped this round:
- Full-archive Bitcoin nodes no longer silently get pruned on reconcile
because the disk-size check was reading the OS partition.
- Failed updates can now recover via reconcile --create-missing instead
of leaving a destroyed container behind.
The reconcile spec for bitcoin-knots auto-enables prune=550 when
DISK_GB < 1000. DISK_GB was measured via `df /`, which on every
archy install reports the ~30 GB OS partition because user data
lives on a separate encrypted /var/lib/archipelago volume.
Result: every archy node with a 2 TB data drive was silently being
configured as a pruned node, and any bitcoin-knots container
recreated by reconcile would delete its historical blocks down to
the 550 MB prune window on next start.
Observed on .228 (2 TB box): blocks dir went from 384 GB to 926 MB
after a reconcile-triggered restart. Historical archive unrecoverable
without full re-IBD from genesis.
Fix: check /var/lib/archipelago first (where bitcoin data actually
lives). Fall back to / only on first-boot before the data partition
is mounted.
Context: when package update fails after remove-old-container but
before reconcile-recreate, the rollback path in update.rs tries to
restart the old container by name. If the container is already gone
(removed in step 3 of the update), rollback fails silently and the
node is left with no live container for that app but on-disk data
still intact. This is exactly the state .228 ended up in after the
reconcile-script-missing bug killed bitcoin-knots and lnd.
Reconcile was designed to only repair existing containers for
optional apps (SPEC_OPTIONAL=true): it skips "not installed" entries
on the assumption that the install RPC creates them. That safety
check is correct for normal operation but blocks recovery when an
optional-marked container has been destroyed by a failed update.
Fix: add --create-missing flag that overrides the SPEC_OPTIONAL skip.
When set, reconcile treats absent containers exactly the same as
broken containers — it creates them from the canonical spec using
the existing on-disk data directory. Narrow-scope override; the
default behaviour is unchanged.
Updated --help to document all four flags.
Verified on .228: after the failed bitcoin-core update took out both
bitcoin-knots and lnd, running reconcile --container=bitcoin-knots
--create-missing --force (as the archipelago user, not root —
podman is rootless) brought bitcoin-knots back using the pruned
chainstate at /var/lib/archipelago/bitcoin. Repeated for lnd. All
containers now running; electrumx reconnecting; UIs recovering.
Does NOT fix the underlying update-flow rollback hole (rollback
should be able to re-create a container from spec, not just restart
by name). That is a separate commit — this flag is the manual
recovery tool plus the primitive the improved rollback will call.
- AccountInfoSection.vue: append 5th bullet to v1.7.43-alpha entry
explaining that update-available badges and version comparisons
work again now that the pinned-image catalog is found at the
correct deployed path.
- docs/MARKETPLACE-QA.md: new tracker for the upcoming app-by-app
install walk on .228. Documents the per-app fix workflow, the
four layers we might need to fix at (app recipe, registry image,
backend orchestrator, frontend), status-key table for tracking
each catalog entry, and the release-notes policy for the walk.
- docs/RESUME.md: refresh with a9908597 commit, updated binary md5
on .228, and split Immediate Next Step into Phase 1 (browser
verification) and Phase 2 (marketplace walk) with a pointer to
the new tracker.
The Rust search path listed /opt/archipelago/image-versions.sh and
scripts/image-versions.sh (repo-relative for dev), but the image
recipe deploys the file to /opt/archipelago/scripts/image-versions.sh.
Production nodes therefore silently failed every lookup: find_file
returned None, load_image_versions returned an empty HashMap, and
both pinned_image_for_app and pinned_images_for_stack returned no
matches.
Symptom on deployed nodes: every container scan emitted
"image-versions.sh not found in any search path" at DEBUG level, and
the version-comparison logic in docker_packages.rs plus the
update-check logic in api/rpc/package/update.rs silently degraded to
no-op — users would not see update-available badges and upgrade RPCs
could not resolve pinned targets.
Fix: put the canonical deployed path first in PATHS, keep the older
/opt/archipelago/image-versions.sh as a fallback for not-yet-updated
nodes, and retain scripts/image-versions.sh as the dev-repo-relative
fallback. Verified on .228: backend now logs "Parsed 57 image
versions from /opt/archipelago/scripts/image-versions.sh" on scan.
Pre-existing test_parse_image_versions failure in this module is
unrelated (the NOT_AN_IMAGE assertion was broken before this change
because the parser's _IMAGE-suffix retain keeps it). Leaving that for
the general cargo-test cleanup pass.
Consolidated single-file snapshot of plan + progress for a fresh
OpenCode session to pick up the install UX polish work:
- Where we are: v1.7.43-alpha shipped, 5 commits on main, deployed
to .228, browser verification in progress.
- Immediate next step: await user's verification results from
https://192.168.1.228/ browser checklist.
- Working layout: SSHFS mount, ssh archy / archy228, deploy recipes.
- Architecture patterns: async-spawn lifecycle, phase-based install
progress, scanner kick, .23 auto-purge migration.
- Backlog: Vaultwarden exit-on-start, install log perms, 22 stale
cargo test failures, historical changelog entries left intact.
- User preferences: "best long-term first", one-by-one, no push,
Bitcoin-only, conventional commits.
Complements STATUS.md (which remains the engineering log) with a
tighter resume-the-work narrative focused on the current round.
Adds a new top section to STATUS.md covering v1.7.43-alpha:
- Round 3: phase-based install progress bar
- Round 4: post-install scanner kick for instant Launch button
- Round 5: .23 VPS retirement, .168 promoted to Server 1
- Config migration: auto-purge .23 from saved registry/mirror JSONs
- Changelog: new v1.7.43-alpha entry in AccountInfoSection
All 5 commits, deployment md5, verification notes, and git remote
cleanup captured. Round 2 rollback command still valid for the full
stack since backups predate every round in this session.
Four release-note bullets describing the user-visible changes shipped
in this round:
- async-spawn install/update/uninstall (UI no longer freezes)
- phase-based install progress bar (Preparing through Finalizing)
- scanner kick post-install (Launch button appears immediately)
- .23 Hetzner VPS retired, .168 OVH promoted to Server 1 with
auto-purge migration for existing nodes
Matches the tone of existing changelog entries: what changed from the
operator's perspective, not internal implementation detail.
load_registries + load_mirrors normally only ADD missing defaults to
the persisted JSON — explicit removals stick. After retiring the .23
Hetzner VPS we need the opposite: existing nodes have .23 baked into
their saved configs and would spend seconds per install/update timing
out against a dead host until the operator manually removes it via
the Settings UI.
Add a targeted one-time migration in both loaders: if any saved entry
has 23.182.128.160 in its URL, drop it on load and rewrite the file.
This is an exception to the usual "explicit removals stick" rule —
the user never chose to add this mirror, it was a default.
Narrow-scope migration (one hardcoded IP match, no schema version)
because the cost/benefit of a general migration system isn't worth
it for a single decommissioned host. Future retirements can follow
the same pattern.
The Hetzner VPS at 23.182.128.160 was decommissioned. Replace it
everywhere with the OVH VPS at 146.59.87.168, which was previously
the tertiary mirror.
- update.rs: drop DEFAULT_TERTIARY_MIRROR_URL, promote .168 into
the secondary slot as "Server 1 (OVH)"; tx1138 becomes Server 2.
Default mirror list shrinks from 3 to 2.
- container/registry.rs: default RegistryConfig drops .23, promotes
.168 to Server 1 / priority 0, tx1138 stays Server 2 / priority 10.
- api/rpc/package/config.rs: trusted-registry allowlist swaps .23
for .168.
- api/handler/mod.rs: app-catalog fallback URL uses .168.
- neode-ui/views/marketplace/marketplaceData.ts: REGISTRY uses .168.
- scripts/image-versions.sh: ARCHY_REGISTRY_FALLBACK uses .168.
- image-recipe/build-auto-installer-iso.sh: installer ISO registries
use .168 (both podman registries.conf and backend registries.json).
Tests updated to assert on the new 2-entry default lists (registry +
mirror). URL-parser fixture tests in update.rs retain .23 strings —
they exercise string-parsing logic, not mirror policy.
Git remotes: dropped `gitea-vps` and the .23 push URL on the `origin`
multi-push alias (not part of this commit — pure working-copy change).
After install completes, the async-spawn wrapper wrote state=Running
but the skeletal install-time manifest (interfaces: None) persisted
until the next scheduled 60s scan. The frontend saw state=running but
hasUI=false and hid the Launch button for up to a full minute.
Add a shared Notify/watch pair between RpcHandler and the scan loop:
- scan_kick (Notify): scan loop selects! between the 60s interval
and this notify, running immediately on either.
- scan_tick (watch<u64>): scan loop bumps the counter after each
completed scan so callers can await completion.
Install and update success paths now call kick_scanner_and_wait before
flipping to Running. The scan merges via merge_preserving_transitional
(state stays Installing/Updating, manifest refreshed from live podman
with interfaces.main.ui populated from real port bindings). 2s timeout
falls back to pre-fix behavior on slow podman — no regression.
Podman emits zero parseable progress when stderr is piped (no TTY), so
the old byte-counter regex never matched in real installs. Users saw
0% for the whole pull, then a jump to 95%, then silence through
create-container, health-check, and post-install hooks.
Replace with 7 explicit lifecycle phases wired through install.rs and
update.rs: Preparing (5%), PullingImage (20%), CreatingContainer (70%),
StartingContainer (80%), WaitingHealthy (88%), PostInstall (95%),
Done (100%). Each maps to a fixed UI progress and status message.
Frontend PHASE_INFO mapper in stores/server.ts prioritizes phase when
present, falls back to byte-counter for legacy. A Math.max forward-only
guard ensures the bar never regresses. Deleted the duplicate watcher
in Discover.vue that was fighting the store's watcher with stale byte
logic. Added shimmer CSS on the fill (with prefers-reduced-motion
opt-out) so the bar looks alive during long phases.
create_installing_entry hardcoded /assets/img/app-icons/<id>.png for
every new install. About half the app icons ship as .svg or .webp
(lnd.svg, vaultwarden.webp, bitcoin-knots.webp, mempool.webp), so the
browser 404s on the wrong extension and renders the default broken-image
glyph for the 10-30s window before the scanner refreshes with real
manifest data.
Send empty icon. The frontend's icon computed in AppCard.vue falls
through to curatedMap which has correct extensions for bundled apps,
and handleImageError still guards any remaining misses with a
placeholder SVG.
With the backend flipped to async-spawn, install/uninstall/update return
immediately with a { status, package_id } envelope. Client timeouts of
45m/11m were a leftover from synchronous handlers and masked real RPC
failures.
Drop all install/uninstall/update RPC timeouts to 15s. Progress and
terminal state still arrive through the live state stream — the RPC
only needs to confirm the spawn was accepted.
Return-type annotations updated in rpc-client.ts and stores/server.ts.
Five direct rpcClient.call sites across Marketplace.vue, Discover.vue,
and MarketplaceAppDetails.vue updated with the shorter timeout.
Extend the async-spawn treatment previously shipped for Stop/Start/Restart
to the three remaining long-running lifecycle RPCs. Each wrapper validates
params, rejects duplicate in-flight ops, flips state to the transitional
variant (Installing/Removing/Updating), then spawns the existing inner
handler on tokio. RPC returns immediately with { status, package_id }; the
spawn task owns the terminal state write.
Install and update success arms explicitly set state=Running. The scan
loop merge (merge_preserving_transitional) refuses to overwrite
transitional states, so the spawn task must write the terminal state.
Uninstall's inner handler removes the entry entirely, so no explicit
terminal write is needed there.
Dispatcher and handler now thread self as Arc<Self> / &Arc<Self> so
spawned tasks can hold their own Arc without extra field cloning.
Transient install entry uses empty icon string. Hardcoding
/assets/img/app-icons/<id>.png 404s for apps that ship .svg or .webp
assets, which produces a broken-image flicker until the scanner refreshes
with manifest data. Empty string causes the frontend's icon computed to
fall through to the curated map, which has correct extensions.
Removed the inner "already updating" guard in update.rs — the wrapper
now owns duplicate-op detection for all three operations.
Records the four landed commits, the .228 deploy (binary + frontend
paths, backups, md5), the manual LND Stop verification, and the
rollback incantation. Leaves the older "NEXT SESSION" design block
in place as historical reference with a note that it's stale.
Adds a follow-ups list: chaos matrix is now unblocked, bundled-app
RPCs are still sync (deprecate or mirror-async?), transitional_since
is in-memory only, and there are 22 pre-existing test failures in
unrelated modules that should get their own cleanup pass.
The app card and details view previously used a pair of Start/Stop
buttons whose labels were driven off isAppLoading(), a client-side
"I just clicked the button" flag. When the backend's graceful stop
took longer than the RPC round-trip (up to 600s on bitcoin-core),
the flag cleared while the container was still shutting down, the
UI flipped back to "Running" as soon as the next 10s scan saw the
still-alive container, and the user had no indication the stop was
still in flight.
Now that the backend flips PackageState to Stopping / Starting /
Restarting / Installing / Updating / Removing for the duration of
each lifecycle operation and the scan loop preserves those states,
the UI can drive its label off the container state itself. A single
full-width primary button replaces the Start/Stop pair. Its label,
color, and disabled state come from getAppVisualState(), which
collapses resting states (exited/created/paused/installed) into
"stopped" and passes transitional states through untouched.
Changes:
- container-client.ts: widen ContainerStatus.state union to include
the six transitional variants plus "installed". Add
restartContainer() calling the new container-restart RPC.
- stores/container.ts: add getAppVisualState() computed and the
restartContainer() action.
- ContainerApps.vue: single primary button (Start / Stop / Starting
/ Stopping / Restarting etc.) plus a separate circular Restart
button visible only when running. Critically, handleStartApp and
handleStopApp now route through store.startContainer and
stopContainer (which call container-start / container-stop, the
async RPCs) instead of the legacy synchronous bundled-app-start /
bundled-app-stop path. Transitional-state polling widened from
just "created" to the full set of transitional variants.
- ContainerAppDetails.vue: same single-button pattern, Restart
button now calls container-restart instead of the old
stop-sleep-start sequence, added 2s polling interval for
transitional states.
- components/ContainerStatus.vue: widen state prop to match the
shared union, render transitional labels with a trailing ellipsis
and a yellow dot.
No new tests — this is presentation logic. Manual verification on
.228 will confirm the end-to-end async path: click Stop on LND,
button becomes "Stopping" in under a second, stays that way for
roughly 5 minutes, then flips to "Start" with a grey dot. The UI
must never revert to "Running" mid-stop.
The 30s package scan loop used to blindly overwrite every package
entry from podman inspect. While a user-initiated Stop / Start /
Restart was in flight, the RPC spawn task would flip the state to
Stopping / Starting / Restarting, the next scan would see podman
still reporting "running" (for the duration of the graceful stop,
up to 600s for bitcoin-core), and clobber the transitional state
back to Running. The dashboard would then flip Running -> Stopping
-> Running -> Stopped, making it look like the stop had silently
failed until it eventually completed.
The merge loop now treats transitional variants (Stopping, Starting,
Restarting, Installing, Updating, Removing, and the three backup
variants) as owned by the RPC spawn task. For those variants,
merge_preserving_transitional keeps the existing state while still
taking live observability fields (health, exit_code, installed,
lan_address, manifest, static_files, available_update) from the
fresh scan so the UI continues to see live health readings.
Adds an escape hatch via a per-scan transitional_since side table:
if a package has been in a transitional state for more than 1200s
(2x the longest graceful stop at 600s on bitcoin-core), the scan
loop assumes the spawn task died without cleanup and overrides with
podman's live state. Prevents a crashed background task from wedging
a package in Stopping forever.
Three unit tests cover the merge rule, the observability passthrough,
and the transitional-variant classifier.
RPC handlers no longer block on podman operations. container-stop on
bitcoin-core used to hold the connection for up to 600s while the UI
showed a frozen spinner; it now returns in under a second with
{status: stopping} after flipping the package state to Stopping and
broadcasting over WebSocket. Same treatment for container-start and
the new container-restart route.
Widens container-list state mapping to emit the transitional variants
(stopping, starting, restarting, installing, updating, removing,
installed, and the backup states) instead of collapsing them to
"unknown". Keeps the mapping in sync with the UI ContainerStatus.state
union so the dashboard can render the right transitional label.
Mirrors the treatment in package/runtime.rs for package.start,
package.stop, and package.restart. The body of each handler is lifted
into pure do_package_* helpers that the background task runs; state
flipping is bracketed around the spawn with revert on error. The
pre-existing post-start exit-check verification and restart stop+start
fallback run inside the spawned task, not the RPC body.
Adds container-restart route to the dispatcher. mark_user_stopped
continues to run BEFORE the spawn, preserving the ordering contract
with the crash recovery layer at runtime.rs:145-148.
Introduces a new RPC-layer helper that bridges the synchronous
ContainerOrchestrator trait with RPC handlers that must return in <1s.
The helper flips the package state to a transitional variant
(Stopping / Starting / Restarting) in the StateManager so WebSocket
clients see the live label immediately, then tokio::spawns the
actual orchestrator call. On success it writes the final state; on
error it reverts to the pre-transition state and logs via
install_log().
The ContainerOrchestrator trait stays synchronous so the reconciler,
boot flow, unit tests, and chaos harness keep deterministic
behaviour. Async only lives in the RPC layer.
Not wired to any handler yet — Commit 2 consumes this helper.
Widens install_log visibility from pub(super) to
pub(in crate::api::rpc) so the new sibling module can reach it.
Dedicated section covering the file-ops-via-mount + git/cargo-via-ssh
split that makes this dev setup work. Includes:
- Exact running mount command (pulled from ps)
- macFUSE + sshfs-mac brew install path
- Health check + recovery sequence for when mount hangs (it will)
- Full which-path-for-which-operation table
- Don't-do list (cargo from mount, rsync without AppleDouble exclude, etc)
- Cache caveat and inode-sharing note between mount and SSH views
No code change.
Captures full design for the next session:
- Full bug sequence (5.5min blocking RPC + 30s scan clobbering transitional state)
- 4-commit implementation order with exact file:line targets
- Single-button UI spec with full label table
- Verification gates including manual LND stop test on .228
- Architectural decision: spawn lives in RPC layer, orchestrator trait stays sync
No code change yet; next session implements.
The previous code computed HOST_GATEWAY from `ip route show default` to
work around an alleged podman 4.3.x limitation. Two problems:
1. The comment was wrong. Podman 4.4+ supports --add-host=host-gateway
natively, and we ship 5.4.2.
2. More critically, `ip route show default` returns the LAN router
(e.g. 192.168.1.254) — the gateway to the internet, not the gateway
to the host. Every container configured with DAEMON_URL or
--bitcoind.rpchost=host.containers.internal was therefore dialing
the WiFi router instead of the host machine, silently failing.
Symptoms this caused on .228:
- LND crash-looped with "dial tcp 192.168.1.254:8332: connection refused"
- Dashboard showed no LND connect details or QR
- ElectrumX DAEMON_URL broken; stuck at 2 KB index for days
- Any service reaching bitcoin-core through the `archy-net` bridge
Replace the computed value with the literal string "host-gateway",
which podman translates to the correct in-network gateway at container
start. Also drop the stale HOST_GATEWAY reference in the Tor-bootstrap
branch (it always fell back to TARGET_IP anyway). Verified on .228:
after recreating bitcoin-core/electrumx/lnd with the new flag, LND
reached the chain backend, ElectrumX resumed indexing, and the
dashboard /lnd-connect-info endpoint succeeded.
LND's admin.macaroon is owned by a rootless-podman subordinate UID
(typically 100000) with mode 640. The archipelago server runs as UID
1000 and cannot read the file directly, which caused every dashboard
LND RPC (getinfo, connect-info, export-channel-backup) and lnd_client
to fail with "Failed to read LND admin macaroon".
Add a read_lnd_admin_macaroon() helper that first tries a direct read
(for operators who have relaxed permissions) then falls back to
`sudo -n cat`, mirroring the pattern already used for Tor hidden
service hostnames in handle_lnd_connect_info. Centralise the canonical
macaroon path as LND_ADMIN_MACAROON_PATH and route all four callers
through the helper.
Verified on .228: GET /lnd-connect-info now returns 200 with cert,
macaroon, and tor_onion fields. Dashboard QR/connect-string UI
unblocked.
Logs Step 9 acceptance evidence, the two bugs caught and fixed during
the hot-swap (parse_memory_limit IEC suffix bug in 732df1b8 and
cgroup Delegate in ba83f9bc), and outlines the Step 10 plan for .116.
Adds Delegate=memory pids cpu io to the archipelago.service unit.
Context: the service runs as User=archipelago under system.slice with
rootless podman. When podman creates transient libpod-*.scope units for
containers under user.slice, systemd needs the caller to hold
CAP_SYS_ADMIN on the target cgroup subtree \u2014 which happens iff
Delegate= lists the controllers we want to set. Without Delegate, any
future code path that goes through the podman CLI (runtime.rs) instead
of the libpod HTTP API (podman_client.rs) would hit MemoryMax
rejections that have exactly the same symptom as the bug I just fixed
in parse_memory_limit but with a completely different root cause.
Belt-and-braces: current production path uses PodmanClient and was
fixed in the preceding commit. But the DockerRuntime CLI path in
runtime.rs:262-268 (cmd.arg("--memory")) is still reachable via
AutoRuntime fallback on hosts without podman, and future rust
orchestrator code may legitimately need cgroup delegation. This
directive is no-op harmful on hosts that already delegate upstream
(systemd gracefully handles duplicate/nested delegation).
The libpod HTTP API path (PodmanClient::create_container) ran manifest
memory_limit values like "128Mi" through parse_memory_limit which
lowercased+trim_end_matches("m"), leaving "128i" which parse::<f64>()
rejected. The resulting None became 0 via .unwrap_or(0), and podman
serialised that into the OCI config as memory.limit:0. At container
start time systemd then rejected MemoryMax=0 with "Value specified in
MemoryMax is out of range".
Silently wrong for every manifest in apps/ that uses Kubernetes-style
suffixes (all of them). Became visible on .228 when Step 9 first
exercised the ProdContainerOrchestrator path for bitcoin-ui and lnd-ui
installs \u2014 the old first-boot-containers.sh bash script used podman
run --memory 128m directly, which podman-the-CLI parses correctly, so
the bug never surfaced before.
Two parts:
- parse_memory_limit now recognises Ki/Mi/Gi/Ti (IEC binary, what k8s
and our manifests use), kB/MB/GB/TB (SI decimal), k/K/m/M/g/G/t/T
(docker shorthand, treated as IEC binary for backwards compat), and
bare byte integers. Filters out zero/negative results.
- create_container omits the memory/cpu fields entirely when the
manifest has no limit or parsing fails, rather than emitting 0. The
libpod API treats absent as unlimited; 0 is "set MemoryMax=0" which
systemd rightly rejects. Defence in depth against the next weird
suffix someone puts in a manifest.
Six regression tests in the new tests module cover IEC, SI, shorthand,
raw bytes, invalid input (empty/garbage/0/negative), and whitespace.
BootReconciler (in-process, 30s interval, spawned from main.rs as of
Step 6 commit 48f08aa3) fully replaces the timer-driven bash
reconciliation path. Delete the systemd unit + timer and their
ISO-builder touchpoints.
Removed:
- image-recipe/configs/archipelago-reconcile.service
- image-recipe/configs/archipelago-reconcile.timer
- image-recipe/build-auto-installer-iso.sh L412-413 (COPY unit+timer)
- image-recipe/build-auto-installer-iso.sh L449 (systemctl enable)
- image-recipe/build-auto-installer-iso.sh L542-543 (cp to WORK_DIR)
Kept (intentionally):
- scripts/reconcile-containers.sh
- scripts/container-specs.sh
Reason: core/archipelago/src/api/rpc/package/update.rs still invokes
reconcile-containers.sh at two sites (OTA update + rollback paths).
Porting those call sites to ContainerOrchestrator::upgrade() requires
manifests for every container update.rs might touch — that scope
belongs in Step 8b. Until then the script stays on disk, just no
longer runs on a periodic timer.
No Rust code changes. cargo check -p archipelago clean, 6 pre-existing
warnings. Skipped full ISO rebuild validation per user decision —
edits are 5 textual deletions with zero behavioral ambiguity; Step 9
live hot-swap on .228 will catch any regression.
Discovered during Step 8 execution that first-boot-containers.sh
creates 30+ containers with per-container logic (wallet loads, DB
init, rpcauth derivations, post-create health waits) and does
substantial non-container setup (secret gen, rootless-podman subuid
chowns, Tor hostnames, WireGuard, firewall, nostr-relay). Only 3 of
the 30+ containers have manifests today (the UIs from Step 7).
Deleting the bash in a single step bricks first-boot on fresh
installs. Split into:
- 8a: delete reconcile-containers.sh + container-specs.sh + reconcile
systemd unit + timer. BootReconciler fully covers these. Safe,
atomic, no manifest porting required.
- 8b: port remaining ~25 containers into apps/<id>/manifest.yml. One
manifest per commit, validated against current bash behavior.
Multi-day scope.
- 8c: rename first-boot-containers.sh -> first-boot-setup.sh, strip
container ops, keep secret/dir/Tor/WG/firewall setup. Final
one-way door, requires 8b complete.
Replaces the first-boot-containers.sh sed/envsubst approach with a
Rust-native render step bound into the ContainerOrchestrator lifecycle.
- New container::bitcoin_ui module: embeds the nginx.conf template via
include_str!, reads the plaintext RPC password from
/var/lib/archipelago/secrets/bitcoin-rpc-password, substitutes
{{BITCOIN_RPC_AUTH}} with base64(archipelago:<password>), and atomic-
writes (tmp + rename) to /var/lib/archipelago/bitcoin-ui/nginx.conf.
Idempotent: byte-compares before writing so unchanged input is a
no-op (no inode churn, no restart cascade).
- ProdContainerOrchestrator gains run_pre_start_hooks(app_id) returning
HookOutcome::{Rewritten, Unchanged}. Fires in install_fresh before
create_container, and in ensure_running: on Running + Rewritten
triggers a restart; on Stopped re-renders then starts.
- bitcoin-ui Dockerfile no longer COPYs a default.conf; the file now
arrives via runtime bind-mount of the rendered config. If the bind-
mount is ever missing, nginx starts with no site configured and
returns 404 everywhere — safe failure vs. serving upstream RPC with
a stale Authorization header.
- apps/{bitcoin,electrs,lnd}-ui/manifest.yml land as first-class
manifests. bitcoin-ui declares the bind-mount target and a dependency
on bitcoin-core; electrs-ui and lnd-ui declare their own deps and
health checks.
- 8 new unit tests on the render fn (idempotency, rotation, trimming,
missing/empty secret, template invariants) plus an integration test
asserting install(bitcoin-ui) actually lands a substituted nginx.conf
on disk via the hook. 39/39 container:: tests pass
(test_parse_image_versions pre-existing failure unchanged, out of
scope).
Step 6 of the rust-orchestrator migration. Construct the container
orchestrator once in main.rs, call load_manifests + adopt_existing
immediately after Config::load, log the adoption report, and spawn
BootReconciler::run_forever with the 30s default interval. Thread the
orchestrator through Server::new -> ApiHandler::new -> RpcHandler::new
so the reconciler and RPC layer share one instance.
Wire a tokio::sync::Notify through the SIGTERM/SIGINT shutdown path so
the reconciler exits cleanly alongside the server drain. Uses notify_one
so the signal stores a permit if the reconciler is mid reconcile_all
when the signal fires.
Delete the commented-out run_boot_reconciliation block in main.rs that
documented the prior bash-script approach being unsafe on unbundled
installs — the new reconciler is manifest-driven and only touches apps
present in /opt/archipelago/apps, fixing that concern.
cargo check -p archipelago clean (6 pre-existing dead-code warnings on
trait methods not yet exercised until Step 9 hot-swap). Container test
suite 43/44 pass; the one failure (container::image_versions::
test_parse_image_versions) is pre-existing and unrelated.
Step 5 of the rust-orchestrator migration. New file boot_reconciler.rs holds a
small Tokio task that calls ProdContainerOrchestrator::reconcile_all() on a
30-second cadence (answered design Q3).
* BootReconciler::new(orch, interval, shutdown) — shutdown is an Arc<Notify>
so callers can trigger a graceful exit without pulling in tokio-util.
* run_forever(self) — does one reconcile immediately, then loops on
tokio::select! { sleep_until | shutdown.notified() }. Shutdown interrupts
the sleep but never an in-flight reconcile_all call.
* Per-pass outcomes are logged at debug/warn; failures never propagate out
because reconcile_all already absorbs per-app errors into ReconcileReport.
Four tokio::test(start_paused = true) tests verify the loop cadence against a
CountingRuntime test double:
* initial_pass_fires_immediately — first reconcile runs with no delay
* second_pass_fires_after_interval — second pass fires after exactly
interval elapses in paused-clock time
* shutdown_terminates_loop — notify_one() lets run_forever return
* failure_in_one_pass_does_not_stop_loop — the loop keeps ticking even when
the first pass had to install a missing container
Not wired into main.rs yet — that is Step 6. Re-exported from container::mod
as BootReconciler + RECONCILER_DEFAULT_INTERVAL for the wire-up step.
Records acceptance evidence for Steps 1-4 (container tests 21/21 pass, build
clean with expected unused-method warnings) and queues the BootReconciler
implementation for Step 5.
The laptop mounts ~/Projects/archy over SSHFS and macOS finder / Spotlight
sidecars write ._<name> resource-fork files alongside every edit. They are
noise; keep them out of git.
Step 4 of the rust-orchestrator migration. Unifies the container lifecycle
surface behind a single trait so the RPC layer stops caring whether it is
talking to the dev or prod orchestrator.
* New trait core/archipelago/src/container/traits.rs: ContainerOrchestrator
with install / start / stop / restart / remove / upgrade / status / list /
logs / health, all keyed by app_id. Every method is async_trait-based.
* ProdContainerOrchestrator: the lifecycle methods are moved from inherent
impl into the trait impl (avoids name-shadowing recursion). Adoption and
reconcile remain inherent since only main.rs / BootReconciler call them.
* DevContainerOrchestrator: new trait impl that forwards to the existing
Dev-named methods, applying the dev container-name + port-offset rules
internally. New load_manifest_for() helper resolves app_id to
<data_dir>/apps/<app_id>/manifest.yml so trait-level install(app_id)
works in dev too. install_container(manifest, path) stays inherent for
the manifest-path RPC shape.
* RpcHandler now holds Option<Arc<dyn ContainerOrchestrator>> and, when in
dev mode, a separate Option<Arc<DevContainerOrchestrator>> for the
manifest_path install RPC. In prod mode RpcHandler::new() constructs a
ProdContainerOrchestrator and calls load_manifests() at startup.
* All seven container-* RPC guards no longer say dev mode required.
container-install still requires dev mode because its manifest_path
argument has no prod meaning; every other container RPC now works in both
modes via the trait.
BOOT STILL DOES NOT USE THIS. main.rs wire-up (Step 6) and BootReconciler
(Step 5) come next. Until then the prod orchestrator is constructed but nothing
populates /opt/archipelago/apps so it has zero manifests to manage, matching
the pre-Step-4 behaviour.
Verification: cargo build -p archipelago clean (11 expected unused method
warnings for methods not yet wired from main.rs). cargo test -p archipelago:
all 21 container::* tests pass (16 prod_orchestrator + 5 others). 24 other
test failures are pre-existing and unrelated (identity_manager / session /
wallet / mesh / credentials — all independently flaky on file-backed state).
Step 3 of the rust-orchestrator-migration. New file prod_orchestrator.rs (999 LOC)
implements the full public surface that will replace scripts/first-boot-containers.sh:
* install / start / stop / restart / remove / upgrade / status / list / logs / health
* adopt_existing: read-only scan that claims containers matching our manifests by
name, without recreating — preserves the v1.7.42 fixture on .116.
* reconcile_all: level-triggered, per-app failures collected rather than aborting.
* install_fresh: build-or-pull (Step 2 trait methods), relative build contexts
resolved against the manifest directory.
Naming rule (answered design Q1): UI app IDs (bitcoin-ui/electrs-ui/lnd-ui) get the
archy- prefix; backends keep their bare ID. An explicit extensions.container_name
always wins. Codified in compute_container_name() with unit tests for all three tiers.
Concurrency (answered design Q4): per-app tokio::sync::Mutex<()> created lazily,
protecting every mutating op against the reconciler loop. Acquiring the per-app
lock only needs a read lock on the map, so independent apps do not serialize.
16 tests: 3 sync naming rule tests + 13 tokio async tests covering install (pull,
build-absent, build-present, relative-context), reconcile (noop/exited/missing/
mixed-failure), adopt-by-name, upgrade sequence ordering, list filtering, health
state mapping, and unknown-app-id rejection. All pass.
Not wired into main.rs yet — that is Step 6. Crate builds clean with expected
unused warnings for the new re-exports.
Adds two methods to ContainerRuntime so the upcoming ProdContainerOrchestrator
can inspect local image storage and build images from BuildConfig:
- image_exists(image_ref) -> Result<bool>: local-storage check only, does
not consult registries. Distinguishes exit 0 (present) from exit 1
(absent) from other failures (environment error).
- build_image(&BuildConfig) -> Result<()>: shells out to podman/docker
build with -t, -f, deterministically-sorted --build-arg pairs, and the
context path last.
Implemented on all three runtimes:
- PodmanRuntime: new podman_cli helper shells out alongside the existing
HTTP API calls (build and image inspect are awkward over the HTTP API)
- DockerRuntime: native docker CLI, same exit-code semantics
- AutoRuntime: delegates to the selected inner runtime
Argv construction extracted into pure build_args_for_podman helper so it
can be unit-tested without a real podman. 4 new tests cover minimal args,
custom Dockerfile path, deterministic build-arg sorting (guards against
HashMap iteration non-determinism), and context-is-last (positional arg
placement is load-bearing for podman build).
Step 2 of docs/rust-orchestrator-migration.md. 25/25 tests pass.
ContainerConfig.image is now Option<String>, mutually exclusive with a new
optional ContainerConfig.build: Option<BuildConfig>. Exactly one of image
or build must be present, enforced in AppManifest::validate.
Adds ResolvedSource enum (Pull | Build) and ContainerConfig::resolve +
::image_ref helpers so the orchestrator can treat pull and build uniformly.
All 26 existing pull-only manifests continue to parse unchanged
(covered by existing_pull_only_manifests_still_parse test).
Call sites updated: podman_client, runtime::DockerRuntime, dev_orchestrator.
Dev orchestrator errors out cleanly on Build sources until Step 2 lands
build_image support on the runtime trait.
Step 1 of docs/rust-orchestrator-migration.md. 10 new unit tests, all pass.
Also includes: docs/rust-orchestrator-migration.md (design spec) and
docs/STATUS.md resume section for the next session.
Closes failure mode adjacent to FM3 (docs/bulletproof-containers.md): on
a syncing pruned node, bitcoind's RPC thread blocks for 5-10s during block
validation. The old 10s client-side timeout was rejecting roughly 30% of
UI calls even though the node was perfectly healthy. 20x stress test on
the live .116 node (caught in IBD catch-up at block 797k) used to drop
10 of 20 calls; now drops 0 of 20.
What changed:
- core/archipelago/src/api/rpc/bitcoin.rs: bitcoin_rpc_call now retries up
to 3 times with 500ms and 1500ms backoffs between attempts. Only
transient transport errors (timeout, connect refused, send/recv IO)
trigger retry. A well-formed bitcoind error response is surfaced
immediately - real RPC bugs are never masked.
- Per-attempt hard deadline (tokio::time::timeout, 15s) layered on top
of reqwest's own timeout, so DNS starvation or TLS wedging can't
steal the entire retry budget.
- handle_bitcoin_getinfo client builder gained a 3s connect_timeout
so a dead bitcoind is fast-failed inside the first attempt instead
of eating the whole 15s.
- Retry policy extracted into a RetryConfig struct so tests can dial
down timeouts to ~100ms per attempt. Production defaults live in
RetryConfig::production().
Not changed (tracked as follow-up):
- mesh/mod.rs bitcoin_rpc_getblockcount and related helpers use the
same 10s-timeout pattern. Not migrated to the new wrapper in this
release; scheduled for v1.7.43 alongside the render_bitcoin_conf
work.
- lnd/info.rs and electrs_status have similar 10s/15s timeouts but
different failure profiles - audit first, migrate only the ones
that actually exhibit the bug.
Tests: 6 new unit tests under api::rpc::bitcoin::tests, all passing.
Uses an in-process hyper server (already a transitive dep) to simulate
bitcoind responses; no new crates required.
- happy_path_first_attempt: no retry when first attempt succeeds
- retries_on_timeout_then_succeeds: first attempt times out, second
succeeds, returns OK (uses a short-timeout RetryConfig so the test
runs in <1s instead of 15s)
- retries_exhausted_on_persistent_connect_refused: all attempts fail
against a closed port, error bubbles up, elapsed time confirms
backoffs actually ran
- does_not_retry_on_rpc_level_error: bitcoind-returned error body is
surfaced immediately, no retry
- does_not_retry_parse_errors: non-JSON response (e.g. 503 with html
body) is NOT retried - guards against the tempting "retry all
non-2xx" mistake that would mask real bitcoind misconfig
- retry_budget_invariants: asserts total wall-time ceiling stays
under 60s so a bumped constant can't silently hang a UI call
forever
Validated live on .116: 20/20 bitcoin.getinfo calls succeed during IBD
catch-up (chain at block 797419 -> 797464), vs ~40% baseline under the
old 10s timeout. Worst-case latency was 48.9s during peak validation;
happy-path latency (cached result) remains 28-77ms.
Closes failure mode FM5 from docs/bulletproof-containers.md: the v1.7.38 +
v1.7.39 rollouts left every affected node on an unreachable UI (nginx 500)
with no recovery path short of SSH. This release adds a self-check
guardrail to the update flow.
What changed:
- apply_update() writes a pending-verify marker with old+new version and
a 150s deadline immediately before scheduling the service restart.
- verify_pending_update() runs from main.rs startup. If the marker is
present and within its freshness window, the new binary waits 15s for
nginx + backend to settle, then probes https://127.0.0.1/ every 5s for
up to 90s (self-signed certs accepted).
- On any probe success within the window, the marker is cleared and
nothing else happens.
- On window-exhaust, the new binary:
1. Moves the broken /opt/archipelago/web-ui to web-ui.failed.<ts>
(quarantined, not deleted, so we can post-mortem).
2. Restores web-ui.bak on top of web-ui.
3. Calls rollback_update() to restore the previous binary.
4. Updates state.current_version to reflect the rollback.
5. systemctl --no-block restart archipelago so the OLD binary boots.
- Markers older than 10 minutes are treated as stale and cleared without
probing, so a crashed-during-startup marker from weeks ago cannot
spontaneously roll back a healthy node on a later reboot.
- rollback_update() binary copy now goes through host_sudo instead of
tokio::fs::copy, so it escapes the service's ProtectSystem=strict
mount namespace. Without this, the rollback silently failed with
EROFS on /usr/local/bin and orphaned the rollback - the exact
opposite of what auto-rollback is for.
Tests: 4 new unit tests in update::tests covering marker round-trip,
absent-marker noop, no-panic on verify_pending_update with nothing to
verify, and an invariant assert that the 90s probe window stays below
the 600s stale threshold. All passing.
Side fix: scripts/create-release-manifest.sh was dying with exit 141
(SIGPIPE from tar tvzf pipe head pipe awk) under set -euo pipefail.
Replaced with a single awk NR==1 that doesn't short-circuit the upstream
pipe, so the release-build flow is idempotent again.
v1.7.38 and v1.7.39 both shipped with `./` inside the frontend tarball marked
drwx------ (700). Tar extraction preserves archive perms, so every node that
pulled the OTA landed with /opt/archipelago/web-ui at 700, nginx (www-data)
returned 500 "permission denied" on every page, and the browser showed
"Internal Server Error nginx". .116 hit this on both v1.7.38 and v1.7.39
rollouts. The v1.7.39 runtime self-heal in main.rs was the wrong layer —
systemd's ReadOnlyPaths namespace made /opt/archipelago read-only from inside
the archipelago service, so chmod from there returned EROFS.
Root cause: create-release-manifest.sh used mktemp -d (700 default umask) for
staging, then tar preserved that 700 in the archive's root entry.
Fix the archive itself:
- chmod 755 staging dir + `find -type d -exec chmod 755` + `-type f chmod 644`
before tar, so the on-disk entries are correct.
- tar --owner=0 --group=0 --mode='u=rwX,go=rX' to normalize archive perms
belt-and-braces in case file-mode drift ever reappears.
- Post-tar verify: `tar tvzf | head -1` must show drwxr-xr-x at root, or
the release script aborts before the manifest is even generated.
Binary unchanged semantically — the main.rs self-heal stays in as a last-
resort belt (can't hurt on nodes whose FS isn't namespace-isolated), and the
update.rs in-extractor chmod stays in so v1.7.40-onwards extractors are
double-safe. The authoritative fix is the archive.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
v1.7.38 shipped with an OTA bug: the tar-extracted staging dir inherited 700
perms and nginx (www-data) returned 500/403 on every request after the swap.
.116 hit this on rollout; had to chmod by hand to recover.
- update.rs: after extraction, explicitly chmod 755 dirs + 644 files on the
new staging dir before the mv into place, so nginx can stat/serve them.
- main.rs: self-heal on startup — if /opt/archipelago/web-ui is not
world-readable, run `sudo chmod -R u=rwX,go=rX` to repair. This is what
rescues nodes upgrading from v1.7.37/v1.7.38, since their extractor
(running on the old binary) doesn't have the chmod fix yet — the new
binary's first boot fixes the mess before nginx serves a single request.
Everything v1.7.38 shipped is still in this release:
- auth.rs auto-heals is_onboarding_complete() from setup_complete +
password_hash so nodes don't bounce back to /onboarding/intro after
browser clear / reboot / update
- useOnboarding tri-state: backend-unreachable no longer defaults to intro
- login sounds gated by isFirstInstallPhase() — silent after onboarding,
typing sounds unaffected
- FIPS app / Nostr Relay / Nostr VPN / Routstr / Penpot removed from
catalog + frontend + Rust + docker + icons; 15 image versions deleted
from tx1138, .168, gitea-local
- AIUI baked into release tarball via demo/aiui/
- prebuild hook syncs app-catalog/catalog.json → public/catalog.json
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- auth.rs now infers onboarding-complete from setup_complete + password_hash so
nodes stop bouncing users through the intro wizard after browser clear / update
/ reboot; the flag self-heals to disk on next check
- frontend: "backend uncertain" no longer defaults to /onboarding/intro —
useOnboarding returns null + callers poll / retry instead of flashing the wizard
- login sounds (synthwave, welcome voice, pop, whoosh, oomph) gated by
isFirstInstallPhase(); typing sounds unaffected
- removed FIPS app, Nostr Relay, Nostr VPN, Routstr, Penpot from catalog,
frontend config, Rust AppMetadata + install dispatch + install_penpot_stack;
docker/fips-ui + docker/nostr-vpn-ui + apps/penpot dirs and 5 icons deleted;
15 image versions deleted from tx1138, .168, gitea-local registries (.160
Gitea was 502 at release time — follow-up)
- AIUI baked into frontend release tarball via demo/aiui/; deploy-to-target
falls back to demo/aiui/ when the AIUI sibling checkout is missing
- prebuild hook syncs app-catalog/catalog.json → public/catalog.json so the
two copies can no longer drift (was the source of the "apps still visible"
bug — public/ had stale data)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Install flow
- api/rpc/package/install.rs: always append the literal image URL as a
last-resort pull candidate in do_pull_image, so images not carried by
any configured mirror (docker.io/bitcoin/bitcoin:28.4) still install
instead of masquerading as a generic pull failure across every mirror.
- api/rpc/package/install.rs: write_bitcoin_conf now skips on any stat
error, not just "file exists". Once bitcoin-knots' first-boot chowns
/var/lib/archipelago/bitcoin into the container's user namespace (700
perms, UID 100100/100101), the archipelago daemon can't even traverse
in — try_exists returns Err which unwrap_or(false) treated as "not
present" and drove a doomed write. Now errors out of the directory
traversal are treated as "conf already owned by container user" and
the write is skipped. Mirrors the lnd.conf pattern.
- api/rpc/package/install.rs: drop the hardcoded `prune=550` from the
conf default. Operators with multi-TB drives shouldn't be silently
pruned; users who want a pruned node can set it in bitcoin.conf
themselves. Full archive is the only honest default.
- api/rpc/package/config.rs: bitcoin-core now passes explicit
-server/-rpcbind/-rpcallowip/-rpcport/-printtoconsole/-datadir CLI
args. Vanilla bitcoin/bitcoin:28.4 has no entrypoint wrapper and
reads conf + argv only; without these the RPC listens on 127.0.0.1
inside the container and rootlessport can't reach it, so the
bitcoin-ui companion gets 502 on every /bitcoin-rpc/ call.
Bitcoin Knots keeps its own entrypoint-driven defaults.
- container/docker_packages.rs: split bitcoin-core out of the shared
AppMetadata arm. bitcoin-core now surfaces as "Bitcoin Core" with
bitcoin-core.svg and a Reference-implementation description; the
bitcoin + bitcoin-knots ids keep the Knots branding. Fixes the home
card showing "Bitcoin Knots" for a Core install.
Bitcoin node UI (docker/bitcoin-ui)
- index.html: impl name/tagline/logo now dynamic. applyImplBranding()
reads subversion from getnetworkinfo — /Satoshi:X/Knots:Y/ resolves
to Bitcoin Knots, plain /Satoshi:X/ resolves to Bitcoin Core. Both
get their own icon and subtitle. Settings modal replaced its
hardcoded Regtest/txindex=1/port-18443 placeholders with live values
from getblockchaininfo + getindexinfo + getzmqnotifications.
- index.html: new Storage info card (Full Archive · X GB /
Pruned · X GB from blockchainInfo.pruned + size_on_disk) visible on
the main dashboard, same level as Network. Settings modal mirrors it
with the prune height when applicable.
- Dockerfile + assets/: bitcoin-core.svg, bitcoin-knots.webp, and the
bg-network.jpg used by the dashboard are now COPY'd into the image
under /usr/share/nginx/html/assets. Previously the <img src> pointed
at paths that 404'd into the SPA fallback and the onerror handler
hid the broken logo silently.
Frontend
- appSession/appSessionConfig.ts: add bitcoin-core to APP_PORTS (8334),
HTTPS_PROXY_PATHS (/app/bitcoin-ui/), and APP_TITLES (Bitcoin Core).
Without these the AppSessionFrame showed "No URL found for
bitcoin-core" and the home/app-list title fell through to the raw id.
- settings/AccountInfoSection.vue: backfill What's New entries for
v1.7.31 through v1.7.37 that had been missed in earlier cuts.
Release plumbing
- releases/v1.7.37-alpha/: binary + frontend tarball.
- releases/manifest.json: v1.7.37-alpha, sha256/size refreshed.
- Cargo.toml / package.json: version bumps.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The trusted-registry allowlist in api/rpc/package/config.rs splits the
image on '/' and matches the first segment against a fixed set (docker.io,
ghcr.io, git.tx1138.com, 23.182.128.160:3000, ghcr.io, localhost). A bare
'bitcoin/bitcoin:28.4' splits to registry="bitcoin" which isn't on the
list, so the install RPC was returning 'Invalid Docker image format'.
Live catalogs on .160 and gitea-local already hotfixed directly; these
static copies keep ISO builds and the final hardcoded fallback in sync.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- neode-ui/public/assets/img/app-icons/bitcoin-core.svg (NEW): 256×256
Umbrel community Bitcoin icon sourced from getumbrel.github.io/
umbrel-apps-gallery/bitcoin/icon.svg. Referenced by the static
catalog, the curated fallback, and the upstream lfg2025/app-catalog
entry so every surface shows the same image.
- app-catalog/catalog.json + neode-ui/public/catalog.json: add
bitcoin-core (v28.4) entry pointing at bitcoin/bitcoin:28.4. Same
entry pushed to the lfg2025/app-catalog repo on .160 and the local
gitea mirror so nodes see it without needing a full archipelago
update. Sovereignty Stack entry added to FEATURED_DEFINITIONS with
a description that frames it as a Knots alternative, not a rival.
- core/archipelago/src/api/handler/mod.rs: handle_app_catalog_proxy
is now instance-scoped (&self) and derives its upstream list from
load_registries — each active container registry contributes one
`<scheme>://<reg.url>/app-catalog/raw/branch/main/catalog.json` URL
in priority order (scheme follows tls_verify). When the operator
switches mirrors in Settings, the App Store now follows. Falls back
to the legacy hardcoded .160/tx1138 pair only when registry config
can't be loaded, so the App Store still renders on nodes that
haven't persisted one yet.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- core/archipelago/src/bootstrap.rs (NEW): embed scripts/container-doctor.sh
and image-recipe/configs/archipelago-doctor.{service,timer} via
include_str! and sync to disk + enable the timer on every archipelago
startup. Idempotent (content-hash compare), dev-box symlink guard keeps
the git checkout untouched, best-effort (warn-only on failure) so
bootstrap never blocks server readiness. Wired in main.rs as a
background tokio task.
- scripts/container-doctor.sh: add fix_rootless_netns_egress(). Detects
when the rootless-netns has lost its pasta tap (container-to-container
still works but outbound DNS/TCP fails) via an nsenter probe into
aardvark-dns; with a two-probe 10s debounce to rule out transients and
a host-precheck that bails out if the host itself is offline. When the
rootless-netns is truly broken, does a graceful podman stop --all /
start --all so pasta + aardvark-dns rebuild the netns from scratch.
Bitcoin-knots and every other outbound container recover in one cycle.
- core/archipelago/src/update.rs: host_sudo → pub(crate) so bootstrap.rs
can reuse the existing systemd-run escape hatch.
- apps/bitcoin-core/manifest.yml: bump app version 24.0.0 → 28.4.0 and
image bitcoin/bitcoin:24.0 → bitcoin/bitcoin:28.4. Resources aligned
with the real container-specs.sh large-disk tune (4 GiB memory cap,
cpu_limit: 0 so bitcoind can run -par=auto across every core).
- neode-ui/src/views/apps/AppCard.vue + Apps.vue: add an Update button
+ Updating spinner to every app card that has available-update set.
Wires through serverStore.updatePackage(id) — the same RPC the detail
view already calls. common.update / common.updating i18n keys added in
en.json and es.json.
- core/archipelago/src/identity_manager.rs: add create_from_signing_key()
that mirrors an existing Ed25519 key as a manager-level identity with
a deterministic id (`node-<pubkey16>`). Idempotent across restarts,
gets the hex-SVG master avatar.
- core/archipelago/src/server.rs: the auto-create path on first boot now
mirrors the node's own signing_key (seed-derived on onboarded installs)
as a "Node" identity instead of generating a random "Default" keypair.
Once this ships, the DID on the Web5 DID Status card (via node.did
RPC), the Node entry on the Identities page (via identity.list), and
the DID used for peer-to-peer connects (via server_info.pubkey) all
resolve to the same seed-derived pubkey.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- useOnboarding.ts: when the backend gives a definitive answer
(true/false, not a null retry failure), re-seed the
neode_onboarding_complete localStorage flag accordingly. Fixes the
case where a user clears site data on an already-onboarded node —
OnboardingWrapper's useVideoBackground computed reads localStorage
synchronously, so without this re-seed the intro video would fire
again on /login even though RootRedirect correctly sent them
straight to /login.
- OnboardingWrapper.vue: login background now rotates through
bg-intro-1..6 on each /login mount, with the current index
persisted to localStorage (neode_login_bg_idx) so subsequent
logouts advance rather than repeat the same image.
- Dashboard.vue: subsequent-login branch drops the 1.2s showZoomIn
entirely. Only the first dashboard entry after onboarding plays
the full zoom + glitch reveal; every re-login now just fades in
with the welcome typing (~300ms).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- useOnboarding.ts: prefer the backend over localStorage when checking
onboarding completion. The old order (localStorage first) meant any
browser that had ever onboarded a node would treat every new fresh
node as already-onboarded and skip the wizard, dumping the user
straight at the inline set-password form. Backend is now authoritative;
localStorage stays as the offline fallback.
- OnboardingWrapper.vue: skip the intro video on `/login` once
`neode_onboarding_complete` is set. Returning logged-out users now
get the static lock-screen background + glitch overlay instead of
replaying the full intro on every logout.
- RootRedirect.vue: when the health check fails, only show the full
BootScreen if the node was never onboarded. For already-onboarded
nodes (i.e. an OTA-update blip), keep the spinner and poll the
health endpoint every 2s for up to 60s before falling back to the
boot screen. Fixes the "fake boot loader" / "server starting up"
screens flashing on every successful update.
- loginTransition store: new `justCompletedOnboarding` flag distinct
from `justLoggedIn`. Set true only by the inline setup-password
flow (handleSetup). Dashboard.vue branches on it: full glitch+zoom
reveal for the post-onboarding entry, quick zoom + welcome typing
on every other login (no triple glitch flashes, ~1.2s vs 8s).
- vite.config.ts: bump assets cache from `assets-cache-v2` to
`assets-cache-v3` so service workers running the previous bundle
invalidate their cache and pick up the new UI cleanly.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- HOTFIX: v1.7.31-alpha's frontend tarball was packaged with a
`neode-ui/` top-level directory instead of the flat layout v1.7.30
and earlier used. Nodes that applied v1.7.31 ended up with
`/opt/archipelago/web-ui/neode-ui/index.html` instead of
`/opt/archipelago/web-ui/index.html`, and nginx returned 403/500.
v1.7.32's tarball is built with `tar -C web/dist/neode-ui .` so
files land directly at web-ui root. Broken nodes auto-heal on this
update (web-ui dir is replaced).
- transport/lan.rs: add Drop impl that calls ServiceDaemon::shutdown()
on the mdns_sd daemon. Without this the OS thread it spawns, plus
the blocking `receiver.recv()` task, keep the tokio runtime alive
past SIGTERM — long enough for systemd's TimeoutStopSec to SIGKILL
the service and mark it Failed. Was visible on every update:
"shut down cleanly" logged, then 15s later systemd forcibly kills.
- main.rs: after logging "Archipelago shut down cleanly", call
`std::process::exit(0)` explicitly. Belt-and-suspenders against
any future non-daemon thread creeping in (reqwest resolver pool,
etc.) and causing the same SIGKILL regression.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Backend: install.rs registry reachability probe now strips the
`host[:port]/namespace` suffix before appending `/v2/` (the Docker
V2 API lives at the host root, not under the namespace) and accepts
HTTP 405 in addition to 200/401 as "registry daemon alive". This
fixes false "unreachable" reports on the Test button for Gitea and
other registries that protect their /v2/ endpoint.
- Backend: stacks.rs install_indeedhub_stack now force-removes any
leftover indeedhub-* containers and indeedhub-net before creating
the stack. A partial install (or the old first-boot stub racing the
installer) used to leave containers around that blocked re-install
with "name already in use". Re-running the App Store install now
self-heals.
- Backend: registry.rs load_registries auto-merges any default
registry URLs missing from the saved config (appended with priority
max+10+i, persisted). Lets new default mirrors (e.g. Server 3 OVH)
roll out to existing nodes without manual config edits. Explicit
removals still stick — URLs absent from disk AND absent from
defaults stay gone.
- Backend: update.rs adds DEFAULT_TERTIARY_MIRROR_URL at
http://146.59.87.168:3000/ (Server 3 OVH) to default_mirrors, with
the same auto-merge-on-load behavior as registries. Test updated
for 3-mirror default (.160, tx1138, .168).
- Scripts: dropped the first-boot IndeedHub stub (~38 lines in
first-boot-containers.sh §8b). It predated the proper stack
installer, raced it, and was the main source of the name-conflict
mess the stacks.rs cleanup above now also guards against.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Backend: unified pull-progress streaming across primary AND fallback
registries. Earlier code only streamed for the primary attempt; if it
failed fast (VPS 404, etc.) the UI froze at 0% until the fallback
finished. The waterfall now uses a single shared helper that streams
podman stderr through update_install_progress for every URL tried.
- Backend: PackageDataEntry gains uninstall_stage, set at each phase of
handle_package_uninstall ("Stopping containers (i/total)",
"Cleaning up volumes", "Removing app data"). State flips to Removing
during the pipeline.
- Frontend: MarketplaceAppCard renders the live progress bar with byte
counts during installs, matching the System Update download bar style.
- Frontend: AppCard renders the live uninstall stage label per app.
Modal closes immediately on confirm so concurrent uninstalls each
show their own progress on their own card.
- Cleanup: removed dead helpers (image_candidates, rewrite_for_primary,
primary_image_url, pull_from_registries_with_skip) made unused by
the install.rs refactor.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- New Settings → App registries page (/dashboard/settings/registries)
that mirrors the update-mirrors experience: list of configured
registries, test reachability, set primary, add/remove. New
registry.set-primary RPC; existing registry.{list,add,remove,test}
reused.
- Default RegistryConfig flipped: VPS (23.182.128.160:3000/lfg2025) is
now Server 1 (primary), tx1138 is Server 2 (fallback).
- Install pipeline now rewrites the first pull to the primary registry
URL before attempting it. Before this, installs always hit whichever
registry the image was hardcoded to, so changing the primary didn't
actually affect where images came from. On failure, the existing
fallback walk skips the primary (already tried) and walks the rest.
- App catalog proxy UPSTREAMS order flipped so the catalog follows the
same VPS-first rule.
- Reboot overlay: animated "a" logo now sits in the center of the ring
(matches the screensaver composition). Extracted the logo-wrapper
pattern inline.
7/7 registry tests pass.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- New reboot progress overlay: full-screen black with the screensaver's
pulsing ring, rebooting → reconnecting → back-online → stalled stages,
elapsed counter, auto-reload on health-check success, manual reload
button at 3 min stall. Mirrors the existing update overlay.
- Ring extracted from Screensaver.vue into a reusable ScreensaverRing
component so the reboot overlay reuses the same animation.
- default_mirrors() now puts the VPS as Server 1 (primary) and tx1138 as
Server 2 — new nodes fetch manifests from VPS first; existing nodes
keep whatever mirror order they've customized.
- What's New entry prepended for v1.7.28-alpha.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- New "Served by {mirror}" line on the System Update page so operators can see
which mirror actually served the available manifest (vs. which is configured
primary). Backend threads the served URL through UpdateState.manifest_mirror.
- New update.test-mirror RPC + per-row lightning-bolt button that pings a
mirror and renders reachable/latency or error inline under the URL.
- UI polish on the mirrors section: Set Primary, Remove, and the new Test
action are compact icon buttons; add-mirror form moved into a dialog.
- "What's New" block prepended for v1.7.27-alpha.
21/21 update module tests pass. vue-tsc + vite build clean.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds a multi-mirror manifest fetch. `check_for_updates` walks a
configurable list (data_dir/update-mirrors.json) in priority order
and falls through to the next mirror on any HTTP / parse / timeout
failure. Two defaults bake in: Server 1 (git.tx1138.com) and Server 2
(23.182.128.160:3000).
Critical fix: after parsing a manifest, rewrite every component's
`download_url` so its origin matches the manifest URL we fetched.
Before this, the manifest hard-coded absolute URLs pointing at one
specific server — so even when a node fetched the manifest from a
faster mirror, the actual 200MB download went back to the slow
original. Now the faster mirror wins end-to-end.
New RPCs: update.list-mirrors, update.add-mirror, update.remove-mirror,
update.set-primary-mirror. New UI section on the System Update page
for operator management. 5 new unit tests for origin parsing and
manifest rewriting (21/21 green).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Re-adds the TCP transport (`0.0.0.0:8443`) to the rendered fips.yaml
alongside UDP. Upstream factory default enables both; we had
inadvertently narrowed to UDP-only when the yaml rewriter was last
touched, which left nodes unable to reach fips.v0l.io (the public
anchor only answers on TCP right now) or talk across networks that
block UDP.
Backend startup now compares the installed yaml against the current
rendered schema and restarts whichever fips unit is active when they
differ — so OTA-upgrading nodes pick up the new transport without
anyone having to click Reconnect.
Dropped the earlier plan to auto-add federated peers as seed anchors:
invites don't carry a FIPS-reachable IP:port, and once TCP reconnects
the public mesh, federated peers become npub-routable without needing
a seed entry.
Seed Anchors modal cleanup: replaced malformed header icon with a
three-arc broadcast glyph, and the close button now matches the
What's New modal (embedded in the card header, same icon + hover
style) instead of the earlier floating off-design placeholder.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The npm run build step in the release ritual had been silently failing
for roughly seven releases. vue-tsc died with EACCES on a root-owned
node_modules/.tmp, exited non-zero, and my `tail -5` of the build
output happened to only show vite's precache summary — which makes
vite look successful even when the typecheck that precedes it failed.
The resulting archipelago-frontend-*.tar.gz files were rebuilds from
whatever content happened to live in web/dist/neode-ui/ at the moment
(files left over from v1.7.9, owned root:root from an earlier sudo'd
operation, unchanged since).
Fixed by chowning both paths back to the archipelago user and
rebuilding. Every published frontend tarball from v1.7.17 through
v1.7.23 therefore shipped the same frozen UI; v1.7.24 is the first
release in that stretch whose frontend actually matches its backend.
Recorded the build-verification rule as a persistent feedback memory
(feedback_frontend_build_verify.md) — future ships must grep the
packaged tarball for the new version string before push.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds a gear button next to the FIPS Mesh card's status pill that
opens a Teleport-ed modal containing FipsSeedAnchorsCard. The card
was landed on disk in v1.7.21 but never wired into a UI entry point
per the entry-point convention, so users couldn't access the
Add/Remove/Apply controls at all. One gear click now opens them.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- fips::service::active_unit() picks whichever fips unit is running
(archipelago-fips.service vs upstream fips.service) so
handle_fips_restart and handle_fips_reconnect don't silently no-op
on hosts where the archipelago-managed unit was never created.
- peer_connectivity_summary(anchor_candidates) replaces the old
identity-cache check. anchor_connected is now true when at least
one authenticated peer's npub matches the public anchor OR any
entry in seed-anchors.json, which matches what the user actually
cares about ("am I in the mesh?") rather than what the card used
to claim ("is this one specific public anchor reachable?").
- FipsStatus::query takes data_dir now (so it can read seed-anchors)
rather than identity_dir. All call-sites updated.
- handle_fips_reconnect re-pushes seed anchors after restart so the
new daemon gets dialed without waiting for the 5-min apply loop.
- FipsNetworkCard label drops "(fips.v0l.io)" — misleading now that
multiple anchors may be configured.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds a local seed-anchor list at <data_dir>/seed-anchors.json. Each
entry is {npub, address, transport, label}. On archipelago startup
and every 5 minutes the list is pushed into the running fips daemon
via `fipsctl connect <npub> <addr> <transport>`, so a cluster can
anchor itself independently of the global fips.v0l.io. A flaky or
unreachable public anchor no longer strands a fresh install.
New RPCs:
- fips.list-seed-anchors
- fips.add-seed-anchor (validates npub1… + host:port)
- fips.remove-seed-anchor
- fips.apply-seed-anchors (on-demand re-dial)
New standalone UI card at views/server/FipsSeedAnchorsCard.vue. Not
wired into Home.vue / Server.vue — operator places it per the
entry-point convention.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The 3AM auto-update path called std::process::exit(0) immediately
after apply_update returned. apply_update had already spawned a 2s-
delayed systemctl restart, but exit(0) killed the runtime before that
spawned task could run — and the unit's Restart=on-failure does not
trigger on a clean exit 0, so the service stayed dead until someone
SSH'd in and started it manually (.253 hit this today).
Scheduler now returns from the task without killing the process;
apply_update's existing restart path (same one the UI's Install
Update button uses) brings the new version up cleanly.
Also hardens the ISO CI: the AIUI inclusion step now falls back to
extracting from the newest release tarball if the runner's cached
/opt/archipelago/web-ui/aiui path is missing, so a reprovisioned
runner can't silently ship a frontend tarball without AIUI. The ISO
build step also sanity-checks the binary exists before invoking the
builder.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
load_state now drops any stored available_update whenever the running
binary version differs from what's on disk — the old migration only
cleared it when the stale entry happened to match the new version, so
skipping releases (e.g. sideloading 1.7.16 → 1.7.18 without 1.7.17)
left a pointer to an intermediate version as the "update available",
which the UI then offered as a downgrade prompt.
check_for_updates also uses a numeric version comparator so a stale or
cached manifest with an older version can't offer itself as an
update, and 1.7.10 correctly outranks 1.7.9 past the single-digit
patch boundary.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Flip transitively-discovered federation peers to Trusted instead of
Observer. Hints are already only ingested from peers we trust and only
peers we trust are re-exported via build_local_state, so the chain of
trust is already vetted end-to-end — making the user promote each
newcomer by hand was friction with no security win.
Backend:
- federation/sync.rs: merge_transitive_peers now inserts TrustLevel::Trusted
(doc comment updated to explain the transitive-trust rationale)
- update.rs: info! log at download start (version, components, total_bytes,
staging path), cancel (staging wiped?, marker cleared?), and apply (backup
path) so journalctl reveals where a stuck update actually is
Frontend:
- SystemUpdate What's New block gets a v1.7.18-alpha entry
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Federation join flow now notifies the inviter with the joiner's name and
immediately bumps state so the Federation UI reloads without a manual
Sync click. Accepting an invite that points back at the local node is
rejected up front (DID/pubkey/onion match). After a peer joins, we spawn
a transitive sync that pulls the new peer's federated peer hints so all
nodes in the federation learn about each other as Observer entries.
Federation.vue polls every 5s while mounted.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
download_update
Each component download is now resumable via HTTP Range requests
(Range: bytes=N-) and retried up to 6 times with exponential
backoff (5/15/30/60/120/180s). On a dropped connection the next
attempt picks up at the last written byte offset instead of
restarting at zero. Streams via reqwest::Response::chunk() to the
staging file so a 160 MB frontend tarball doesn't sit in RAM. SHA
is verified over the complete file at the end of each component;
mismatch nukes the staged file and restarts from scratch.
Real download progress counters
New AtomicU64 globals DOWNLOAD_BYTES/DOWNLOAD_TOTAL are updated
from the chunk loop. update.status exposes them as
download_progress.{bytes_downloaded, total_bytes, active}. The
SystemUpdate.vue progress bar now polls update.status every
second instead of incrementing a fake random counter — and
crucially, if the user navigates away and back, the component
picks up the in-progress download from the backend atomics
immediately.
Update-check retries
handle_update_check now retries the manifest fetch up to 3 times
with a 5s gap if the first try hits a transport error, so a
momentary gitea hiccup doesn't make a node report "up to date"
when there actually is a new release. Tight 10s connect timeout
per attempt keeps the total bounded.
Artefacts:
archipelago 1070c87f…c081c162b 40584792
archipelago-frontend-1.7.15-alpha.tar.gz 8e630eba…63fd43f 162078068
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Install UX
SystemUpdate.vue now shows a full-screen overlay after apply: the
BitcoinFaceAscii logo, a target-version label, an indeterminate
progress stripe (solid orange; solid green on ready), and an
elapsed-time readout. Polls /health every 1.5s and auto-reloads
once the backend reports the new version. 3-min stall → "Reload
now" button. Download UI also shows a spinner + "Finishing
download — verifying checksum…" while the fake bar sits at 95%.
FIPS reconnect — for real this time
New fips.reconnect RPC does stop → start → wait 20s → re-poll →
classify. Classification buckets: connected / daemon_down /
no_seed_key / no_outbound_udp_or_anchor_down / peers_but_no_anchor,
each with a plain-language hint surfaced verbatim by the Reconnect
button. The real reason nodes like .198/.253 couldn't reach the
anchor: identity::write_fips_key_from_seed was writing fips_key.pub
as a bech32 npub TEXT file, but upstream fips expects 32 raw
bytes. The daemon silently authenticated with garbage. Fix:
PublicKey::to_bytes() → raw 32 bytes, and new
fips::config::normalize_pub_file migrates legacy files by decoding
the npub and rewriting in place. fips.reconnect also re-installs
the config + healed keys to /etc/fips before restarting.
AIUI preservation + restore
apply_update was wiping /opt/archipelago/web-ui/aiui because the
Vue build doesn't include it — every OTA lost the Claude sidebar.
The preserve block now copies aiui/ + archipelago-companion.apk
from the old web-ui into the staging dir before the swap, and
prefers new-tar versions if present. To restore it on the three
nodes that already lost it (.116/.198/.253), this release bundles
the 85 MB aiui build into the frontend tarball. Frontend component
size is now ~155 MB.
Download / install timeouts
Backend download client timeout 1800s → 3600s (1 h). Larger
tarball + slow gitea raw throughput put us above the old cap.
Frontend update.download rpc timeout 30 min → 65 min to match.
package.install rpc timeout 15 min → 45 min — IndeedHub pulls
6 images and was timing out mid-install.
UI nit
"Rollback to Previous" → "Rollback Available".
App-catalog proxy already landed in v1.7.13.
Artefacts:
archipelago 725e18e6…3c525e6 40462288
archipelago-frontend-1.7.14-alpha.tar.gz c35284be…ff2c16 162077052 (+aiui)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The Discover / Marketplace page fetched the app catalog directly from
git.tx1138.com/lfg2025/app-catalog/raw/.../catalog.json in the
browser. Two blockers hit the fleet simultaneously: (1) tx1138's
Gitea doesn't emit Access-Control-Allow-Origin so the HTTPS fetch
got CORS-blocked; (2) the HTTP IP-port fallback
(http://23.182.128.160:3000/...) falls outside the node's
`connect-src` CSP. Users saw the hardcoded fallback instead of the
live catalog.
Backend: new authenticated GET /api/app-catalog handler uses reqwest
to pull catalog.json server-side (15s timeout) and returns it with
application/json + 1h Cache-Control. Tries the HTTPS URL first,
HTTP IP-port second.
Frontend: curatedApps.ts now calls /api/app-catalog (same-origin,
no CORS/CSP) with credentials included so the session cookie
authenticates the proxy. Baked /catalog.json stays as the last
resort.
Artefacts:
archipelago 0aaf7262…b979f22c 40371192
archipelago-frontend-1.7.13-alpha.tar.gz 27505811…efc6f4142 76982505
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Version-only bump. Sits above v1.7.11-alpha which user has verified
runs the full Install Update pipeline end-to-end (check → download
→ install → auto-restart). Freshly-installed nodes from the 1.7.11
ISO will see 1.7.12 as their first OTA target.
Frontend tarball byte-identical to v1.7.11 (same sha).
Artefacts:
archipelago 247f65c2…54f40df9 40385472
archipelago-frontend-1.7.12-alpha.tar.gz 0644a436…54f58 76983846 (reused)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Version-only bump. Frontend tarball byte-identical to v1.7.10. First
OTA-testable release where the running backend (v1.7.10) has the
host_sudo/systemd-run apply fix — clicking Install Update should
walk through check → download → install → auto-restart with no
manual intervention.
Artefacts:
archipelago cf003f62…65465f 40378752
archipelago-frontend-1.7.11-alpha.tar.gz 0644a436…54f58 76983846 (reused)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
THE apply fix
archipelago.service uses ProtectSystem=strict, so /opt and /usr are
read-only inside the service's mount namespace. sudo inherits that
namespace — every sudo mkdir/mv/chown from apply_update was hitting
EROFS even as root. Every prior "Failed to apply update" was a
symptom of this. New `host_sudo()` helper wraps every filesystem
call in `sudo systemd-run --wait --collect --pipe -- <cmd>`, which
spawns a transient unit with systemd's default (no ProtectSystem)
protections — the command runs in the host namespace and can touch
/opt/archipelago + /usr/local/bin normally.
FIPS cascade (#2)
Home.vue and Server.vue both carry a FIPS row that previously only
looked at {installed, service_active, key_present}. Now they also
read anchor_connected + authenticated_peer_count and mirror the
full FIPS card: green "Active · N peers" when healthy, orange "No
anchor" when the DHT bootstrap has failed.
Profile paste URL fallback (#4)
Web5Identities.vue list + editor previously had `@error="display:none"`
on the <img>, which hid the tag without re-rendering the fallback —
a broken pasted URL showed up blank. Replaced with reactive
pictureLoadFailed / listPictureFailed flags plus a watcher that
resets on URL change. Broken URL now falls back to the initial (or
identicon for seed-derived identities).
Small-upload data URL (#3)
Uploaded profile pictures ≤ 64 KB are now inlined as
`data:image/png;base64,...` into profile.picture on the client
before calling update-profile. That kind-0 event is fetchable by
any Nostr client — no Tor needed. Larger uploads fall back to the
onion-rooted public_url with a hint telling the user to paste a
public https:// URL for broader visibility.
Deferred: #1 FIPS Reconnect "actually fixes" — the current Reconnect
calls fips.restart which clears the daemon state, but when the
anchor is truly unreachable (UDP 8668 blocked by network/ISP), no
amount of restart can help. A richer diagnostic is out of scope for
this bundle.
Artefacts:
archipelago 4a77c704…82aa6f8 40379696
archipelago-frontend-1.7.10-alpha.tar.gz 0644a436…54f58 76983846
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Version-only bump. First release where .116/.198/.253 (running v1.7.8
with the mv-based apply) should walk through Check → Download →
Install → auto-restart cleanly via UI, no sideload intervention.
Artefacts:
archipelago 1ec7383d…301629 40378536
archipelago-frontend-1.7.9-alpha.tar.gz 4fb79664…0172e9 76984615 (reused)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
apply_update's binary swap called `sudo install -m 0755 src
/usr/local/bin/archipelago`. install opens the destination for write
with O_TRUNC; the kernel returns ETXTBSY (exit 1) when the path is a
currently-running executable, which it always is during apply because
apply_update is called by the archipelago RPC handler — running as
archipelago itself. Every previous "Failed to apply update" was this
one root cause; the manual sideload path only worked because we
stopped the service first.
rename() doesn't modify the file it replaces — it repoints the path
at a new inode while the old inode stays alive for any process that
has it mapped. `mv` uses rename(). Switched to `sudo mv` (with prior
chmod+chown on the staging file) so the swap is atomic and tolerant
of the running binary.
Frontend tarball byte-identical to v1.7.7-alpha; only the binary
version string changes.
Artefacts:
archipelago 2753daec…48094d 40377648
archipelago-frontend-1.7.8-alpha.tar.gz 4fb79664…0172e9 76984615 (reused)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Pure version bump. No code changes. First release shipped with the
reinforced apply_update (timestamped staging + all-mv) and frontend
with 95% progress cap — this OTA should walk through cleanly from
.116/.198/.253 without any sideload intervention.
Artefacts:
archipelago e3f1740d…006025 40373392
archipelago-frontend-1.7.7-alpha.tar.gz 4fb79664…0172e9 76984615 (reused)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
apply_update frontend swap
Transient EROFS on .198 (filesystem hiccup — root FS mounts with
errors=remount-ro so a fleeting glitch can bounce /opt to RO for a
moment) caught the pre-cleanup `rm -rf web-ui.new web-ui.bak` mid-
stride and aborted the apply. Rewrote the swap to use a timestamped
staging dir (web-ui.new.<ms>) and a timestamped old-copy path so
nothing needs to be rm'd before the extract. After the new tree is
mv'd into place, the previous rollback copy is rotated aside with a
.<ms> suffix (best-effort) and this apply's old copy becomes the new
web-ui.bak. If the final mv fails, the staged old is restored so
nginx keeps serving.
handle_update_check manifest override
handle_update_check takes the git path whenever ~/archy/.git exists.
On the dev box (.116) that meant the Pull & Rebuild button was
always the only option even though the manifest-path OTA was
already wired via ARCHIPELAGO_UPDATE_URL. Now: if that env var is
set, we skip the git detection entirely and use the manifest path.
The regular fleet (no env var, no repo) hits the manifest branch
naturally; beta dev nodes (repo + no env var) still get Pull &
Rebuild; dev nodes with the env var explicitly set can finally test
the manifest OTA end-to-end.
Artefacts:
archipelago 356e78cc…91a6dd 40372288
archipelago-frontend-1.7.6-alpha.tar.gz 4fb79664…0172e9 76984615 (reused)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Trivial version-only bump. No code changes; binary differs only in its
embedded CARGO_PKG_VERSION string. Frontend tarball is byte-identical
to v1.7.4-alpha's (same sha), copied under the new filename to satisfy
the manifest component naming.
This exists so the fleet nodes (.116/.198/.253), all now running
v1.7.4-alpha with the fixed apply_update tar flow, can exercise the
full OTA pipeline from the UI: Check → Download (30-min timeout) →
Install (sudo install binary + sudo tar to web-ui.new + atomic swap) →
auto-restart (systemctl --no-block) → sidebar updates → state sync.
Artefacts:
archipelago 7422a695…a1a2a6 40362432
archipelago-frontend-1.7.5-alpha.tar.gz 4fb79664…0172e9 76984615 (reused)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
apply_update was extracting the frontend tarball with
`tar -xzf -C /opt/archipelago`, but the tar contents are the *inside*
of web-ui/ (root entries are ./test-aiui.html, ./assets/, etc.). So
the files landed directly in /opt/archipelago instead of under web-ui/,
and tar bailed on nginx-owned paths mid-extraction. First end-to-end
OTA test (.198) found it: "tar: ./assets/SystemUpdate-…js: Cannot
open: No such file or directory".
Now extracts into web-ui.new, chowns, then atomically swaps: move
existing web-ui → web-ui.bak, then web-ui.new → web-ui. Same pattern
as the manual sideload that's been working.
Frontend: SystemUpdate.vue fake download progress was capped at "<90"
with a Math.random()*15 increment — the last tick could push to
~104.99%. Capped at 95% with a smaller step so it stops at 95 and the
real RPC completion jumps it to 100.
Artefacts:
archipelago a14ad7e4…2a2be3 40361984
archipelago-frontend-1.7.4-alpha.tar.gz 4fb79664…0172e9 76984615
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Sidebar version
detect_build_version() no longer reads /opt/archipelago/build-info.txt
first. That file was written by the ISO installer at flash time and
never rewritten by OTA or sideload, so after any binary swap the
sidebar kept advertising whatever the ISO shipped with. Now just
returns env!("CARGO_PKG_VERSION") unconditionally — always matches the
running binary.
FIPS card
The two-column grid in FipsNetworkCard.vue placed version/npub boxes
side-by-side on mobile but the anchor-status panel forced col-span-2,
creating an unbalanced empty column at every desktop width. Anchor
status moves to its own full-width row below the grid. When the
anchor is not reached, a "Reconnect" button appears next to the
status line; it calls fips.restart (45s timeout), waits 5s for the
daemon to come back, then reloads fips.status. Surfaces whether the
restart actually recovered the anchor in a status flash.
Profile picture render
Uploaded profile pictures are stored with an onion-rooted URL so
external Nostr clients can fetch them. The local browser isn't
Tor-routed though, so the <img src> silently 404'd and the UI fell
back to showing initials. Added a displayableUrl() helper on
Web5Identities.vue that rewrites http://<onion>/blob/<cid>[?...] to
same-origin /blob/<cid> for rendering, while the stored URL keeps
its onion prefix so publishing to Nostr still works for external
viewers. Pass-through for data: URLs and already-relative paths.
Identity row title
The identity list header now renders profile.display_name (when set)
and keeps identity.name as a muted parenthetical. Before, only the
internal name was shown and a user who'd customised their Nostr
display_name saw a mismatch between their own UI and what peers
rendered.
Artefacts:
archipelago 99184b95…22dc1b 40350664
archipelago-frontend-1.7.3-alpha.tar.gz 7b933cf4…74a8bc 76987031
Changelog layman-style per the saved feedback.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three user-visible fixes shipped together.
1. update.apply permission-denied
apply_update() was doing fs::copy into /usr/local/bin/archipelago and
tar xzf into /opt/archipelago as the archipelago user — both root-owned.
The backup step succeeded (it wrote to data_dir) but the swap failed
with a silent permission denied, wrapped as "Failed to apply archipelago".
Now uses `sudo install -m 0755` for the binary and `sudo tar -xzf` for
the frontend, plus a post-apply `sudo systemctl --no-block restart
archipelago` scheduled 2s after the RPC reply so the UI sees success.
2. Apply → Install label
en/es locale strings: applyUpdate / applyTitle / applyNow changed from
"Apply" to "Install". Matches the user's mental model and distinguishes
the user-facing verb from the internal apply_update() function.
3. Identity avatar backfill
Identities created before df83163f had profile=None on disk and so
rendered as initials. load_record() now synthesizes an IdentityProfile
with a default picture (identicon for regular identities, the hex node
SVG for derivation_index=0) when profile is missing. The synthetic
profile lives only in the returned record; the file stays untouched so
a later explicit Save persists whatever the user actually chose.
Artefacts:
archipelago 70e5444e…67c589 40381960
archipelago-frontend-1.7.2-alpha.tar.gz 806b027b…358a824 76983699
Changelog rewritten layman-style per saved feedback.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Trivial bump on top of df83163f. No code changes — this exists purely so
the fleet nodes now sitting on 1.7.0-alpha have a real target to exercise
the OTA pipeline against: check → download → apply → restart → state
reconciliation. The binary content differs only in the embedded
CARGO_PKG_VERSION string.
Frontend tarball reused from v1.7.0-alpha (same bytes, copied to a new
filename to match the manifest component name convention).
Artefacts:
archipelago 7f7981bd…56eef0 40391760
archipelago-frontend-1.7.1-alpha.tar.gz (dup of 1.7.0) dc3b63af…e9a8370 76984288
Manifest changelog is a single plain-language sentence explaining that
this is the test release — per the saved feedback about keeping
fleet-facing strings readable.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Follow-up to 1fb71b4b on the same v1.7.0-alpha line.
Identity avatars
• New module `avatar.rs` generates two deterministic SVG styles keyed
off the pubkey: a 5×5 mirrored identicon for sub-identities and a
hexagonal-network motif for the master (seed index 0) identity.
Both returned as base64 data URLs, so a fresh identity has a
recognisable picture before the user uploads anything.
• `IdentityManager::create()` and `create_from_seed()` populate
`profile.picture` on creation. Index 0 gets the node SVG; all
other seed-derived + ad-hoc identities get the identicon.
Blob store — public flag for profile assets
• `BlobMeta.public` (default false) added; `BlobStore::put()` takes
a `public: bool`. Missing in legacy meta files = false.
• `POST /api/blob` now stores uploads with public=true and returns
`public_url` alongside `self_test_url`. public_url is
`http://<node-onion>/blob/<cid>` (no cap) if Tor has published the
archipelago hidden service, else falls back to the local path.
• `GET /blob/<cid>` bypasses the HMAC capability check when the
requested blob is flagged public — external Nostr clients fetching
a kind-0 `picture` URL can't hold a cap.
• Mesh callers (content_ref attachments, dispatch rehydration) pin
public=false explicitly so nothing leaks out of the mesh path.
Profile editor UX
• Collapsed Save + Save & Publish into one button — the Save action
now persists locally AND publishes the kind-0 metadata event in
one step. Uploads store `public_url` into `profile.picture` /
`profile.banner` so the published URL is reachable by external
clients.
Update client — the 15-second cliff
• Frontend `rpcClient.call` for `update.download` now has an
explicit 30-minute timeout (was falling back to the default 15 s).
`update.apply` gets 5 min, `update.git-apply` gets 15 min. Matches
what the backend is actually willing to wait for.
• Backend `load_state()` reconciles `state.current_version` with
`CARGO_PKG_VERSION` on every start. Sideloaded or reflashed nodes
were stuck advertising the old version even with a new binary in
place, which kept re-offering the same release as an update.
Manifest changelog rewritten for fleet readers per the saved feedback
(no function names, no file paths). Artefacts refreshed:
binary 12f838c5…5ba82d 40381864
frontend dc3b63af…e9a8370 76984288
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Follow-up to 56d4875b, same v1.7.0-alpha shipping band.
Backend download timeout bumped from 300s to 1800s (update.rs) with an
explicit 30s connect timeout. git.tx1138.com raw-file throughput can sit
around 70–80 KB/s, which meant OTA downloads were timing out at ~55%
through the 40 MB binary even though the SHA would have matched on a
full pull. 30 min gives ample headroom for the worst LAN-to-VPS link we
actually hit.
Frontend: SystemUpdate.vue now formats downloadPercent with toFixed(2)
via a new computed, so the progress card shows "45.23%" instead of
"45.270894%". Cosmetic only; the underlying ref still tracks raw floats.
Manifest changelog rewritten in user-facing language per the saved
feedback — no file paths, function names, or "root cause" phrasing.
Artifacts refreshed:
binary d85a71c5…982f4 40360936
frontend 8adcdacf…e687f6 76986852
ISO at image-recipe/results/archipelago-installer-unbundled-x86_64.iso
(Apr 20 09:00) carries both fixes for fresh installs.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Follow-up to 8b7cb002 (no version bump — same v1.7.0-alpha manifest):
* WireGuard peer persistence. Kernel peer state is ephemeral; the add-peer
RPC wrote each peer to data_dir/nostr-vpn/peers/*.json but nothing
re-pushed them on reboot. Result on .198: wg0 came up listening with zero
peers after last night's reboot. Added vpn::restore_wg_peers() — reads
the peers dir, waits up to 30s for wg0 to exist, then replays each via
`archipelago-wg add-peer`. Spawned from main.rs alongside the other
startup tasks.
* Reconcile + filebrowser drift. scripts/container-specs.sh load_spec_
filebrowser now declares SPEC_NETWORK="archy-net" (to match what
first-boot-containers.sh creates) and pins the filebrowser-data volume
+ wget-style healthcheck so the reconciler stops reporting network
drift. Without this, reconcile would kill the healthy first-boot
filebrowser container and recreate it on bridge, breaking the archy-net
DNS name the backend proxies to.
Manifest binary sha/size refreshed:
6c178a76…3582cc, 40361912 bytes.
Rebuilt ISO at image-recipe/results/archipelago-installer-unbundled-x86_64.iso
(Apr 20 07:10) carries both fixes baked in.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two fixes bundled into the OTA:
1. update.download hard-fail on git-path nodes. handle_update_check's git
branch reported update_available=true + update_method="git" but never
populated state.available_update, so update.download returned "No update
available to download" even though the UI showed one. SystemUpdate.vue
now routes update_method=="git" through update.git-apply (pull+rebuild+
restart via self-update.sh); manifest-path nodes keep the download→apply
flow. i18n strings + confirm modal added for the git path.
2. Reconciler creating containers behind the user's back. On fresh
unbundled installs (.198, .253) archy-mempool-db and archy-btcpay-db
materialised ~10 min after first boot because reconcile-containers.sh
walked container-specs.sh's canonical tier list and created any
"missing" container. reset_spec() now defaults SPEC_OPTIONAL="true",
so reconcile is strictly a repair tool — baseline comes from
first-boot-containers.sh (filebrowser on unbundled), everything else
from the install RPC.
Also forces OTA trigger for nodes on 1.6.0-alpha that otherwise saw
"I'm at manifest.version, nothing to do" and skipped the refreshed 1.6
artifacts.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
1. FIPS auto-activate at server startup only fires if fips_key already
exists on disk, which on a fresh install is never true until AFTER
onboarding. By the time the user completes seed-generate/restore,
archipelago has been running for minutes and the startup task has
long since exited. User still had to hit Activate.
Fix: call spawn_post_onboarding_fips_activate() from the tail of
handle_seed_generate and handle_seed_restore — the moment the
fips_key materialises, a detached task runs `fips::config::install`
+ `archipelago-fips.service activate`. Logged only, never blocks
the onboarding RPC.
2. Kiosk health-poll window was 30 × 2s (configs/ copy was 60 × 2s
but unused — the heredoc in build-auto-installer-iso.sh is what
actually lands on disk). On .198's slower hardware archipelago
/health wasn't ready within 60s, so Chromium launched against a
not-yet-running backend → blank window until manual reboot. Bumped
to 150 × 2s (5 min) + TimeoutStartSec=360. .253 was already well
within the window; this protects the slower box too. Standalone
configs/archipelago-kiosk.service updated in lockstep so the two
copies don't drift.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Supersedes the earlier v1.6.0-alpha artifacts from today (which were
identical to v1.5.0-alpha and only existed for the update-flow smoke
test). This drop actually changes behaviour:
- archipelago-fips auto-activates on startup when fips_key exists; no
Activate button needed.
- fips_key on-disk format migrated to bech32 nsec; legacy raw-byte
files from v1.5.0-alpha self-heal when this version reads them.
- fips.yaml schema matches upstream jmcorgan/fips 0.3+.
- VPN status row shows "Not configured" instead of "Starting…" when
wg0 isn't up — no VPN peer added yet is not a failure state.
New SHA256s + sizes in manifest.json. Fleet nodes .116/.228/.253 will
notice within 30 min (periodic update-check). Also lets .198 self-heal
its crashlooping archipelago-fips when it picks up the update.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Problems addressed (all observed on .198):
* fips_key was written as raw 32 bytes; upstream fips daemon reads it
with read_to_string() and bailed with "stream did not contain valid
UTF-8", crashlooping indefinitely.
* Activate button racy: user had to hit it, and it would keep failing
silently because the daemon couldn't parse its own config.
* FIPS schema drift (already fixed in 7d8a5864) put the config write
path behind the same broken "Activate" flow, so the fix alone
didn't help existing nodes.
* Journal was on tmpfs — every reboot wiped install/onboarding history,
making post-hoc debugging impossible.
Changes:
* identity.rs: write fips_key as bech32 nsec + newline. load_fips_keys
now auto-migrates legacy 32-byte files to bech32 the first time it
reads them, so OTA updates from v1.5.0-alpha self-heal without user
action.
* server.rs: post-onboarding auto-activate task runs on every
archipelago startup. If fips_key exists it ensures /etc/fips/fips.yaml
is schema-current and starts archipelago-fips.service. Pre-onboarding
nodes stay quiet (guarded on fips_key_exists).
* ISO build: un-mask archipelago-fips + archipelago-wg + wg-address —
all use ConditionPathExists on their key files, so systemd silently
skips them pre-onboarding (no MOTD [FAILED]). Only nostr-vpn stays
masked (legacy service, superseded by upstream fips).
* Journald made persistent via /var/log/journal + 500M cap, so
install and first-boot logs survive reboots for diagnosis.
After this, a fresh install + onboarding should bring FIPS up automatically
with no user interaction. The UI "Activate" button can stay as an escape
hatch (the RPC is still there) but is no longer on the critical path.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
No functional changes from v1.5.0-alpha — this release exists only to
validate the in-app update pipeline end-to-end (manifest check → staged
download → apply → restart → version bump in UI sidebar).
Dropping just the manifest + artifacts; no manual deploy to the fleet.
.116/.228/.253 should notice within 30 min (periodic update-check
interval) and surface the update in the dashboard.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Anchored regex was too strict — `strings` concatenates adjacent printable
bytes so the version never sits on its own line. The 1.5.0-alpha binary
DOES contain the version but as part of `1.5.0-alpharpcNot Found`. Fixed
by switching to `grep -qF $VERSION`: substring match is safe because the
version string is specific enough that accidental collisions are
vanishingly unlikely.
Caught mid-build today: check rejected the correct local binary, fell
through to container source-build — ISO still produced correctly but
wasted ~10 min on an unnecessary rebuild.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
1. FIPS daemon config schema drifted: upstream jmcorgan/fips now takes
`node.identity.persistent: true` (keys read from config-dir/fips.key)
and `transports.udp.bind_addr: "0.0.0.0:PORT"` instead of
`identity.key_file/pub_file` + `transports.udp.enabled/port`. The
`tor:` transport was dropped entirely; archipelago handles Tor
fallback itself. fips.yaml generated by archipelago::fips::config
now matches the upstream schema, and archipelago-fips.service stops
crashlooping on Activate. Observed on .198: 52 restarts with
"data did not match any variant of untagged enum TransportInstances
at line 7 column 3".
2. ISO backend-binary capture didn't verify that the captured binary
matched the checked-out Cargo.toml version. Today's 14:40 ISO
shipped a stale 1.4.0 binary because `core/target/release/archipelago`
pre-dated the 1.5.0-alpha bump — the build grabbed it via the
first-priority "local release build" path without looking at it.
All four capture sources now go through verify_backend_version()
which greps the binary for the expected version string; mismatches
are skipped so the build falls through to the source-build path.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
On this host (and potentially others with a particular podman/overlay
state), passing the multi-hundred-line stage-2 script via
`debian:trixie bash -c '...'` caused debootstrap to fail at
"Extracting apt... tar failed" on the very first package — no matter
what patch, storage cleanup, or env-reset we tried.
Running the exact same script body via a bind-mounted file
(`bash /installer-env.sh`) succeeds. So: write the body to a temp
file in WORK_DIR, bind-mount it read-only, and have the container
bash execute it from the file. Same behavior, different invocation,
works.
Was blocking every ISO rebuild since ~10:57 local. First successful
build since: 14:40, sha256 41fad2ff…, 2.3GB.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Debian Trixie apt 3.0.3's data.tar has duplicate entries for the same
path (regular file + symlink at e.g. libapt-private.so.0.0), and tar
bails on the second entry with "Cannot create symlink: File exists",
failing debootstrap on the very first package. Patch debootstrap's
tar invocation to use --skip-old-files so the duplicate is ignored.
Was blocking every unbundled ISO rebuild since the Trixie apt bump.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Previous build's STEP 47/50 log showed:
RUN for svc in nostr-vpn archipelago-wg archipelago-wg-address; do
rm -f /etc/systemd/system/.service
ln -sf /dev/null /etc/systemd/system/.service
done
The Dockerfile is generated via <<DOCKERFILE heredoc in the build
script, so unescaped $svc resolved in the outer bash BEFORE Docker
ever saw it, leaving nostr-vpn/wg masks as a hidden `.service` file
with no effect. nostr-vpn still tries to start on boot → [FAILED].
Fixed with \$svc so the literal lands in the Dockerfile for Docker's
shell to expand per iteration.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Fresh install of .198 reported "FIPS has an npub but says inactive".
The debian package writes /etc/fips/fips.pub during install (whence
the npub) but leaves the upstream fips.service disabled. Result:
FipsStatus.service_active = false, dashboard shows "inactive" until
the user hits Activate. Explicit `systemctl enable fips.service`
in the Dockerfile so first boot brings the daemon up immediately.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
1. nostr-vpn still failing despite last mask attempt — confirmed in
the 6th ISO's rootfs.tar: the .service file was present but
not in multi-user.target.wants. Previous `systemctl mask` silently
no-oped because the real file was already there. Fixed properly
with explicit `rm -f` + `ln -sf /dev/null` for nostr-vpn,
archipelago-wg, and archipelago-wg-address — same /dev/null
symlink state that `mask` would produce on a clean install.
2. Kiosk didn't come up on first boot, only on reboot. Extended the
ExecStartPre health-poll from 30s → 120s (unbundled ISO takes
longer to settle on first boot: archipelago initializes state,
pulls FileBrowser, frontend settles), raised TimeoutStartSec to
180s, and added After=systemd-user-sessions.service +
After=network-online.target so X / Chromium aren't racing.
3. /init: line 29: can't create /root/etc/network/interfaces error
on installer boot — debootstrap --variant=minbase omits ifupdown
so the target has no /etc/network/ directory, and live-boot's
init tries to seed it. Non-fatal but noisy. Added ifupdown +
isc-dhcp-client to the debootstrap --include list.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
5th ISO attempt died in rustables's build.rs (which uses bindgen to
wrap libnftnl) with "couldn't find any valid shared libraries
matching: libclang". bindgen requires libclang.so at build time
to parse C headers. rustables also needs libnftnl-dev + libmnl-dev
for the actual wrappers.
Added to the fips-builder stage apt install line.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Third time's the charm. The upstream fips Cargo.toml puts fips-gateway
behind features.gateway = ["dep:rustables"], so the previous two
attempts (--bins, --workspace --bins) never produced the binary —
only the default feature set was compiled. cargo deb --no-build then
panics looking for the missing binary.
Inspected /tmp/fips-investigate (fresh clone of upstream main on
2026-04-19) to confirm — the feature flag is the gate, not a
workspace layout issue.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Plain `cargo build --release --bins` only built the root crate's
binary targets. fips-gateway is a workspace member, so we need
--workspace to pull every member's bins. Without it cargo deb
--no-build panics looking for target/release/fips-gateway.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Consumes the new authenticated_peer_count + anchor_connected fields
from fips.status. Shows a cyan dot + "connected" when fips.v0l.io is
in the identity cache (DHT routing to unknown npubs will work), or
an orange "not reached" with a one-line explainer that federation
and messaging will fall back to Tor until the anchor reconnects.
Peer count appears on the same row so users see "3 peers" when the
fleet-pair script has been run, or "0 peers" on a fresh install
still waiting for the anchor handshake.
Block only renders when service_active — pre-onboarding the FIPS
package is masked so there's nothing meaningful to report.
Covers the UI half of task #20. Multi-anchor defaulting is still
open (need real anchor addresses beyond fips.v0l.io).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two new fields on the /rpc fips.status payload:
- authenticated_peer_count: how many FIPS peers the daemon has an
authenticated session to right now. 0 means isolated / not on
the mesh; >0 means traffic to any known npub can DHT-route.
- anchor_connected: true when the public anchor (fips.v0l.io,
npub1zv58cn7…) is present in the daemon's identity cache. The
anchor bootstraps DHT routing for general-case deployments, so
this is the best single-value indicator the UI can show for
"will federation traffic over FIPS work between previously-
unknown peers?"
Implementation: fips::service::peer_connectivity_summary shells
out to `sudo -n fipsctl show peers` + `... show identity-cache`
(archipelago user already has NOPASSWD:ALL per the ISO sudoers
and live fleet nodes, confirmed). Failure returns (0, false) so
the UI degrades to "unknown" state without crashing.
Only queried when service_active — pre-onboarding / daemon-down
nodes skip the fipsctl call entirely.
UI side (FipsNetworkCard) consumes the full status JSON, so the
two new fields are available via existing prop plumbing; visual
treatment can come later.
Also fixes ISO build (commit 3e04456c wasn't sufficient): the
Dockerfile needs `cargo build --release --bins` — upstream FIPS
added a `fips-gateway` binary target, and plain `cargo build
--release` only builds the default bin list, which caused
`cargo deb --no-build` to fail hunting for the missing binary.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Until now federation.sync-state only fired on (a) user clicking Sync
in the UI or (b) server-name push. That meant own_fips_npub,
last_transport, peer state updates — all the things v1.5 added for
auto-upgrade from Tor to FIPS — didn't propagate until the user
poked the button.
Fix: spawn a background task in server.rs that runs
federation::sync_with_peer for every Trusted peer every 30 minutes.
First run is 60s after boot (let onboarding settle) and peers are
staggered 5s apart to not hammer Tor's SOCKS proxy with concurrent
connects.
The sync path already prefers FIPS (via PeerRequest), so once peers
have learned each other's fips_npub (now automatic thanks to the
own_fips_npub broadcast in state snapshots), subsequent periodic
syncs route over FIPS — transport badge cycles from 'tor' to 'fips'
on its own without user action.
Covers task #30.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
rust:1-slim-bookworm doesn't include dbus/ssl dev headers, and
jmcorgan/fips upstream started linking against libdbus-sys + openssl
at some recent commit. Observed during the 2026-04-19 v1.5.0-alpha
rebuild: libdbus-sys's build.rs panics when pkg-config can't find
dbus-1.pc, which kills the whole cargo build → the whole ISO build
→ ships an ISO without FIPS installed.
Also mask nostr-vpn.service + archipelago-wg*.service in the rootfs
Dockerfile: these have WantedBy=multi-user.target so systemd pulls
them into the default boot target, but their EnvironmentFile + an
ExecStartPre guard cause them to [FAILED] in the boot MOTD on every
fresh install until onboarding writes their env files. Masking
keeps the startup clean; the onboarding / install RPC handlers
unmask + start them when prerequisites exist (same model as
archipelago-fips).
Bonus discovery from same diag: the default build was silently
reusing a stale rootfs cache from Apr 12 — before the FIPS
integration landed. So the v1.5.0-alpha ISO I shipped had no FIPS
package at all. Rebuild pass with --rebuild forces fresh rootfs.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Previously the profile editor only accepted external URLs for
picture/banner — typing in a URL works, but anyone without their
own image host couldn't use an avatar at all. Now there's an
"Upload" button next to each field that pushes the selected file
to /api/blob and pastes the returned capability-signed local URL
(`/blob/<cid>?cap=…&exp=…&peer=…`) straight into the form field.
- Two new refs: avatarUploading / bannerUploading so each button
shows "Uploading…" independently.
- uploadAsset(ev, 'picture' | 'banner') wraps the POST, validates
HTTP 200 + presence of self_test_url, surfaces failures in the
existing profileError banner.
- File input is re-cleared on completion so the user can pick the
same file again without refreshing.
- Live preview in the <img> at the top of the editor updates
immediately because profileForm[field] is reactive.
Image persists through Save & Publish via the existing
identity.update-profile + identity.publish-profile (both now
multi-relay). The image URL is still local-only — external nostr
clients won't resolve it until we integrate a public image host
(noted in task #29).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Pairs with the backend change that broadcasts kind:0 to every
enabled relay. Web5Identities.vue's publishProfile() now reads the
richer response ({accepted, rejected, relays_attempted, published})
and shows one of three states:
- all relays accepted → "Published to all N relays (event_id…)"
- partial → "Published to X/N relays" plus a warning with the first
relay's rejection reason
- zero → "Published to 0/N relays — check Manage Relays" (error)
User can now tell at a glance whether their profile actually made
it to the wider nostr network or only the local relay.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Previously handle_identity_publish_profile defaulted to a single
hard-coded relay (ws://localhost:18081) so the user's kind:0 profile
event only ever landed on the local relay — hence "Manage Relays
shows N connected, but profile edits don't propagate" from testing.
Fix — two-layer change:
- identity_manager::publish_profile now takes `&[String]` relays
instead of one URL. Adds each relay to the nostr-sdk client,
gives 15s for handshakes, publishes, then surfaces per-relay
accept/reject in a new ProfilePublishOutcome struct so the UI
can show WHICH relays accepted vs. rejected and WHY.
- RPC handle_identity_publish_profile no longer defaults to the
local relay: pulls the ENABLED list from nostr_relays::list_relays
(the same table that powers Manage Relays) and publishes to every
entry. Accepts an optional `relays: [...]` override for tests.
- At-least-one-accept guarantee: if every relay rejects, the call
errors instead of silently reporting published=true. User gets a
real error message listing the failures.
- Response shape: `{event_id, accepted: [urls], rejected: [[url,
reason]], relays_attempted: N, published: bool}` so the UI can
show a useful status block after clicking Publish.
relay_url_matches is tolerant of trailing-slash / case differences
since nostr-sdk canonicalises URLs internally.
Covers the publishing half of task #29; avatar/banner upload UI is
still open.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Before: Alice sent /network.send-request to Bob, Bob accepted via
/network.accept-request and gained Alice in his peers list, but Alice
was never notified — her pending row sat there and she had to
manually add Bob separately. User complaint: "it's strange you have
to do it both ways."
Fix — the accept now fires a best-effort connection_accepted message
back to the requester:
- handle_network_accept_request: after writing the local peer record,
assembles a `{type: "connection_accepted", request_id, from_did,
from_onion, from_pubkey}` JSON, signs + encrypts + POSTs it to the
requester via node_message::send_to_peer. Uses PeerRequest internally
so it prefers FIPS and falls back to Tor.
- handle_node_message: parses incoming plaintext as JSON; on a match
for type=connection_accepted, auto-adds the sender to peers.json
(the existing self-pubkey guard in add_peer still applies) and
short-circuits the normal store_received path so the acceptance
doesn't also land as a chat message in Alice's inbox.
Offline handling: if Alice is offline when Bob accepts, the notify
warns and the local accept still succeeds. Alice will receive any
subsequent message from Bob normally; future iteration could
retry on reconnect.
Federation-invite flow (federation.accept-invite → notify_join) was
already bidirectional; this closes the gap for the peer flow.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
need an archipelago restart
Previously the server checked `fips0` once at startup; if the
interface wasn't up (pre-onboarding, or post-onboarding before the
user clicked Activate FIPS), the peer listener never bound and stayed
unreachable until the next archipelago restart.
Replaced with a `peer_late_bind_loop` background task: polls every
30s for an fd00::/8 address on `fips0` and binds the listener the
moment one appears. First tick fires immediately so the hot path —
fips0 already up at startup — is still zero-cost. Cancellation
cascades through the same `tokio::sync::watch` channel the main
listener uses.
Side effects:
- main.rs no longer computes peer_addr eagerly; dropped the unused
param from serve_with_shutdown.
- FipsTransport::is_available already caches the service probe so
the 30s poll doesn't thrash systemctl.
Covers task #21. Unblocks the first-boot + onboarding flow for
fresh ISO installs on .253.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Pre-v1.4 federation pairs (who exchanged invites before fips_npub
was part of the invite code) had no path to learn each other's FIPS
npub — they'd stay Tor-only forever even after upgrading. Fix:
every state snapshot now carries the sender's own_fips_npub, and
update_node_state refreshes the stored fips_npub on the receiver
side whenever it differs.
- NodeStateSnapshot.own_fips_npub (serde default for back-compat).
- build_local_state takes own_fips_npub alongside the other
single-value fields.
- handle_federation_get_state populates own_fips_npub from
identity::fips_npub, with a fallback to the upstream daemon's
/etc/fips/fips.pub for legacy nodes that never materialised a
seed-derived key.
- storage::update_node_state now writes fips_npub into the
FederatedNode when a new value arrives and trims whitespace
before comparing, so key rotations also flow through.
- Test fixtures (storage + transport/delta + sync) updated for the
new field; existing tests pass.
Net effect: on the next sync, .116 and .228 learn each other's
fips_npub (currently null from the old invite) and subsequent
federation calls route FIPS-first automatically.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Observed on .228: /var/lib/archipelago/peers.json contained an entry
matching the node's own node_key.pub pubkey. It had been added
2026-03-02 and stuck around forever since add_peer() only dedupes by
pubkey — nothing stops a pubkey that happens to be ours.
How it probably got there: somewhere in the auto-add paths
(node-message receive, mesh federation bridge, invite back-and-forth)
a message we'd sent was fed back and the receiver-side add used the
echoed from_pubkey without realising it was us. Doesn't matter which
path — the guard belongs in storage.
add_peer now short-circuits when the candidate pubkey matches
data_dir/identity/node_key.pub. Helper is_own_pubkey best-effort:
unreadable identity → returns false so normal peers aren't blocked.
Also manually purged the one stray entry on .228 (1 removed, 2 real
peers remain). Future deploys include this guard so the phantom can't
come back.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Real 413 root cause on .116 and .228 turned out not to be the body-size
limit — their /etc/nginx/sites-enabled/archipelago was a stale regular
FILE, not a symlink to sites-available, so every nginx update since
someone froze the active config had been invisible to running nginx.
The /api/blob location, added at some point after that freeze, didn't
exist in sites-enabled, so every attachment upload hit nginx's default
1m client_max_body_size and returned 413 regardless of attachment
size.
Deploy now re-creates the symlink on every run: if sites-enabled is a
regular file or missing, we replace it with a symlink to
sites-available. Idempotent if it's already correct.
Also applied the fix live on all 4 fleet nodes — /api/blob now
responds 401 (session-auth required, as designed) instead of 413 on
2MB+ test uploads.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Was seeing "upload failed: 413" on mesh attachment sends between
federated nodes — a ~7MB image becomes ~10MB base64 in the
typed_envelope wire and hit the 10m client_max_body_size on
/archipelago/, /content/, and /dwn/. Bumped those six locations
(two per server block, regular + HTTPS) to 256m so modern
attachments/blobs don't trip the proxy. /rpc/ stays at 1m —
internal JSON-RPC calls are small and don't need the headroom.
Applied to all 4 fleet nodes live; ISO source config updated so
fresh installs get the same limits.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Versioning was drifting on three axes — fixed all of them:
1. Cargo.toml → 1.5.0-alpha (was 1.5.0). User wants `-alpha` suffix
on every pre-stable release; this is the current state of main.
2. neode-ui/package.json was still 1.3.5 — brought in line.
3. /opt/archipelago/build-info.txt was stale on .198 (1.3.4) and
.253 (1.3.5), absent on .116/.228. That file OVERRIDES the
binary's CARGO_PKG_VERSION for the UI sidebar, which is why
.198/.253 kept showing old versions even with fresh binaries.
scripts/deploy-to-target.sh now writes build-info.txt on every
deploy, reading the version straight from Cargo.toml — so the
sidebar can never drift from the binary again.
Release artifacts + manifest:
- releases/v1.5.0-alpha/archipelago (40M, sha in manifest)
- releases/v1.5.0-alpha/archipelago-frontend-1.5.0-alpha.tar.gz (51M)
- releases/manifest.json bumped with full 7-line changelog covering
FIPS-first routing, Settings toggle, transitive federation, cancel
button, transport badges, peer listener, and the build-info fix.
- scripts/check-release-manifest.sh — new pre-publish guard. Refuses
to pass if: Cargo.toml ≠ manifest version, changelog is empty
(release notes are mandatory), or any component's sha256/size
doesn't match the file on disk. Run locally or from CI.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Every federated node card now shows a colored badge indicating how
archipelago actually reached the peer on the most recent successful
call — FIPS / TOR / LAN / MESH — not a prediction based on available
addresses. The badge is hidden when we've never reached the peer.
Backend:
- Cargo.toml: 1.4.0 → 1.5.0 (visible in the sidebar health endpoint).
- FederatedNode gains last_transport + last_transport_at (serde
default for back-compat with v1.4 nodes.json files).
- federation::storage::record_peer_transport(did, onion, transport)
— writes both fields plus last_seen after each successful peer
call. Matches by DID first, falls back to onion.
- federation::sync::sync_with_peer now calls record_peer_transport
immediately after a successful PeerRequest return, so the badge
on the sync'ing peer's card reflects the transport the call
actually rode (fips vs tor).
Frontend:
- types.ts FederatedNode gains last_transport / last_transport_at
(union-typed to the four known kinds).
- NodeList.vue: new transportBadge(node) returns {label, cls, title}
tuned per transport. Hidden when last_transport is absent so we
never lie. Tooltip shows "Last reached via <x> · <time ago>" so
stale data is self-evident. Removed the predictive icon from the
transport store — badge is now 100% ground-truth.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Previously the Pending Peer Requests panel only had Approve/Reject for
inbound rows; outbound rows in the 'sent' state had no action and
would sit there until the target explicitly approved or rejected. Now
you can Cancel an outbound request — the local row is dropped and a
PeerCancel nostr DM is sent so the target's inbound row also
disappears.
Backend:
- HandshakeMessage::PeerCancel {reason: Option<String>} variant.
- nostr_handshake::send_peer_cancel() mirrors send_peer_reject.
- handshake.poll handler dispatches inbound PeerCancel: finds the
matching inbound pending row (same from_nostr_pubkey, state=Pending)
and deletes it. Reply shape gains `cancelled_inbound: [id]`.
- federation::pending::delete() — hard-remove (set_state only
transitions; we don't want 'Cancelled' ghosts in the audit trail).
- federation.cancel-request RPC: outbound+Sent only, default
notify=true (cancelling silently is a footgun), best-effort DM
(relay failure doesn't block local deletion). Wired in dispatcher.
Frontend:
- PendingRequestsPanel.vue: Cancel button appears only on
outbound+sent rows. Emits 'cancel' event with request id.
- Federation.vue: cancelPending(id) handler calls
rpcClient.federationCancelRequest and reloads the list.
- rpcClient.federationCancelRequest(id, reason?, notify=true).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- TransportPrefsCard.vue: import from '@/api/rpc-client' (not
'@/api/rpc') so vue-tsc resolves the module during build.
- scripts/fleet-fips-unpair.sh: companion to the fleet-pair script —
rewrites each node's fips.yaml to anchor-only (fips.v0l.io) so we
can prove the general-case deployment works without the LAN
fast-path. Prints per-node peer counts + DHT AAAA resolution for
every cross-node pair after the change.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When Alice syncs state with a Trusted peer Bob, she now learns about
Bob's other Trusted peers and auto-adds them as Observers on her side
— so Carol's fips_npub is known locally and subsequent federation
traffic to Carol can route directly over FIPS without a separate
invite round-trip.
- NodeStateSnapshot gains a `federated_peers: Vec<FederationPeerHint>`
field (serde default for backward compat with v1.4 snapshots).
- FederationPeerHint is a minimal projection: did, pubkey, onion,
name, fips_npub — excludes per-receiver fields (trust_level,
added_at, last_seen, last_state).
- build_local_state takes the local federation list and includes only
Trusted peers. Observer/Untrusted peers are NOT re-exported — a
node shouldn't launder other people's federation through its own
authority.
- sync_with_peer merges the received hints via merge_transitive_peers
when the source is Trusted: existing entries get fips_npub
refreshed if missing; unknown DIDs are added at Observer trust
(never auto-promoted to Trusted).
- Bounded to 1 hop: merged Observer entries do NOT get re-exported in
the local node's own snapshots. So Bob → Alice learns Carol, but
Alice's snapshots to Dave do not include Carol.
- Tests: round-trip + filter-non-trusted-from-snapshot coverage.
- Storage + delta test fixtures updated for the new field.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
scripts/fleet-fips-pair.sh writes a deterministic /etc/fips/fips.yaml
on each of our 4 dev fleet nodes (.116/.198/.228/.253), listing the
other three as static FIPS peers over their LAN IPs (UDP 2121 / TCP
8443). Also flips `node.identity.persistent: true` so the npub stays
stable across restarts — without this the daemon rolls a new keypair
on every restart and federation invites that carried the previous
npub go stale.
The script is NOT the general deployment mechanism:
- Every archipelago install already ships fips.v0l.io as an anchor
peer, so any node can DHT-route to any npub that has ever announced
on the public mesh.
- Federation invites (v1.4+) carry the peer's fips_npub, so accepting
an invite is enough for crate::fips::dial::peer_base_url(npub) to
reach the peer through the anchor network.
- This script is a LAN fast-path optimization so intra-fleet traffic
stays on the wire instead of bouncing through fips.v0l.io.
Usage:
scripts/fleet-fips-pair.sh # apply to all nodes
scripts/fleet-fips-pair.sh --verify # print current peer state
Verified: all 4 fleet nodes now report 3 authenticated peers each
(their 3 fleet siblings), plus fips.v0l.io in the identity cache.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds a user-configurable toggle for how each peer-to-peer service
reaches federated peers. Three options per service:
- Auto (default) — FIPS preferred, Tor fallback (current behavior).
- FIPS only — fail rather than fall through to Tor.
- Tor only — explicit opt-in to onion anonymity for that service.
Services covered (matching the UI rows):
- Federation — state sync, invites, peer notifications
- Peers — address/DID rotation broadcasts
- Peer Files — content catalog download/browse/preview
- Messaging — archipelago channel + mesh bridge
- Mesh File Sharing — content_ref blob fetches
Implementation:
- settings::transport — persisted struct + process-wide OnceLock handle
(so deep call sites don't need data_dir threaded through signatures).
On-disk file: <data_dir>/settings/transport_preferences.json; missing
or corrupt → defaults (Auto everywhere).
- settings::transport::init() called from main.rs after config load.
- fips::dial::PeerRequest gains a .service(kind) builder; send_* checks
the preference before choosing a transport. FIPS-only fails loudly
when FIPS is unavailable (so users who pick it know when something
falls back).
- Every FIPS-first migration site tags its PeerRequest with the
matching PeerService so the toggle actually applies.
- transport.preferences + transport.set-preference RPCs added; wired
into the dispatcher.
- neode-ui/src/views/settings/TransportPrefsCard.vue — standalone card
with a 5-row Auto/FIPS/Tor tri-state. Not wired into Settings.vue —
the user places components themselves (see feedback_ui_entry_points).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Migrates the remaining Tor-direct peer call sites to PeerRequest so
FIPS is the default when the peer is federated and running the daemon:
- node_message::send_to_peer / check_peer_reachable: gain a
fips_npub parameter. Error messages updated to reference both
transports.
- Callers (api/rpc/network.rs, api/rpc/peers.rs, server health
loop): look up fips_npub from federation storage by onion and
pass it.
- mesh::send_typed_wire_via_federation: the spawned background POST
for the /archipelago/mesh-typed endpoint now uses PeerRequest with
federation-resolved fips_npub. Signature domain unchanged.
- api/rpc/mesh/typed_messages.rs fetch_blob_from_peer: blob URL
rebuilt as (base_url, path_with_query) so PeerRequest can append
the query string after swapping the host. Cap/exp/peer
parameters are still signed over the content ref itself, so
transport choice is invisible to the signature.
- network/dwn_sync.rs sync_with_peers: per-peer fips_npub lookup
before sync_single_peer; health/pull/push each dial through
PeerRequest, so any DWN peer known to federation gets FIPS.
Left Tor-only on purpose:
- api/rpc/identity/handlers.rs handle_identity_resolve_peer_onion —
resolving TO a DID, no anchor yet.
- content.browse / preview calls to non-federated peers fall
through to Tor naturally inside PeerRequest (no fips_npub → skip
FIPS branch).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
All four content-over-peer handlers prefer FIPS when the peer is in
our federation and has advertised a FIPS npub; fall back to Tor
otherwise (unknown peers, FIPS daemon down, transient failure).
- content.handle_content_download_peer / _paid: DID-authenticated
fetch, payment token header threaded through both transports.
- content.handle_content_browse_peer / _preview: no DID header by
design (anonymous browse) — still benefits from FIPS when the
peer happens to be federated.
- federation::fips_npub_for_onion: storage helper that looks up a
peer's FIPS npub from the federation nodes file given their onion
address. Suffix-tolerant (`abc` matches `abc.onion`).
Preserves the Tor-only path for truly unknown peers: PeerRequest
returns Err from the Tor branch instead of silently succeeding,
matching the previous behavior when the peer was unreachable.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Every federation peer-to-peer call now prefers FIPS (direct ULA dial
over `fips0`, ~LAN latency) and falls back to Tor only on network
failure. Per-method ed25519 signatures are preserved on both
transports so authenticity doesn't change.
- fips::dial::PeerRequest — fluent builder that owns transport
selection. Returns the Response plus the TransportKind that carried
it, so handlers can log or expose which path was used.
- fips::dial::is_service_active — free-standing async probe used by
migration sites (the transport::fips::is_available cache is keyed
to a `&self`, not usable from static contexts).
- federation/sync.rs: sync_with_peer + deploy_to_peer drop the
hand-rolled reqwest::Proxy dance, call PeerRequest instead.
- federation/invites.rs: notify_join takes the remote's fips_npub
(already parsed out of the invite code since v1.4) and dials over
FIPS when available. The "peer-joined" signature domain is
unchanged.
- api/rpc/federation/handlers.rs: DID rotation broadcast loops over
federated peers through PeerRequest; the per-peer result payload
gains a `transport` field so the UI can surface mesh vs. onion.
- api/rpc/tor/mod.rs: onion-address-change propagation is now the
most useful FIPS-first call — fips_npub is stable across onion
rotation, so peers get the new address even when the old onion
is already dead.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Wires the FIPS transport end-to-end so peer-to-peer calls can reach
other nodes over the mesh without going through Tor:
- fips::dial — raw RFC 1035 DNS client (zero new deps) that queries the
FIPS daemon's local resolver at 127.0.0.1:5354 for `<npub>.fips` AAAA
records. Exposes peer_base_url(npub) → "http://[fd9d:…]:5679" plus a
reqwest client factory for call-site migrations.
- fips::iface — parses /proc/net/if_inet6 to find the ULA address on
`fips0`. Runs under the archipelago service user without extra caps.
- FipsTransport::is_available() — live probe of archipelago-fips and
upstream fips.service via `systemctl is-active`, cached 10s so the
send hot path doesn't thrash DBus.
- FipsTransport::send() — resolve npub, POST TransportMessage JSON to
the peer's /transport/inbox. Today /transport/inbox isn't wired on
the receive side, so call-site migrations use dial::peer_base_url
directly against the already-signed endpoints (/rpc/v1,
/archipelago/node-message, /content/*). The inbox handler lands as
part of the Settings/transport work.
- server::serve_with_shutdown — takes an optional peer_addr and spawns
a second listener bound specifically to the fips0 ULA on port 5679.
The peer listener applies is_peer_allowed_path() — a whitelist of
endpoints that already do per-request signature auth — and returns
404 for everything else. Shutdown cascades to both listeners via a
watch channel; 5s drain window preserved.
- main.rs — if fips0 has a ULA at startup, pass the peer SocketAddr to
serve_with_shutdown; otherwise run the main listener only.
Security: the peer listener is bound to the fips0 ULA directly, not
wildcard, so it's unreachable from WAN IPv6. The path whitelist limits
exposure to endpoints whose handlers verify ed25519 signatures or
federation DID headers server-side.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Nodes without a seed-derived FIPS key (legacy deploys, fresh pre-onboarding
installs) were reporting "Awaiting seed" in the dashboard even when the
upstream fips.service was running — status.npub was None unless
/data/identity/fips_key.pub existed.
- fips/service.rs: new read_upstream_npub() reads /etc/fips/fips.pub
(bech32 text or raw 32 bytes) from the debian package.
- fips/mod.rs: FipsStatus::current() prefers the seed-derived npub,
falls back to the upstream key. service_active is now TRUE if either
archipelago-fips.service OR upstream fips.service is active; adds
upstream_service_state to the status payload.
- fips/update.rs: resolve the upstream default branch from the GitHub
repo API (jmcorgan/fips is on `master`, not `main`) instead of
hardcoding — future repo rename just works.
- network/router.rs + api/rpc/router.rs: diagnostics gain wifi_ssid from
`nmcli -t device` so the Network card can show the connected SSID.
- UI: Home.vue adds a FIPS row to the Local Network card; Server.vue
mounts the new FipsNetworkCard and shows SSID + FIPS Mesh rows;
HomeNetworkCard.vue removed (superseded by the inline rows).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Bakes the FIPS (Free Internetworking Peering System) mesh daemon into
the node stack, supervised by archipelago alongside Tor. Runs as a
system service, identity derives from the same BIP-39 master seed, and
user-triggered updates track upstream main.
Identity
seed.rs: new HKDF label archipelago/fips/secp256k1/v1 → dedicated
secp256k1 key, distinct from the Nostr-node key for crypto isolation
but still seed-recoverable
identity.rs: writes fips_key[.pub] to /data/identity on onboarding,
chmod 0600; fips_key_exists / load_fips_keys / fips_npub accessors
Transport
TransportKind::Fips=3 inserted between LAN and Tor (Tor bumps to 4)
→ router prefers FIPS over Tor for all peer traffic
PeerRecord gains fips_npub + last_fips fields (serde(default) for
backward-compat with older nodes)
transport/fips.rs: NodeTransport stub, reports unavailable until the
daemon is live so router falls through to Tor cleanly
Federation invites
FederatedNode and FederationInvite carry optional fips_npub
create_invite / accept_invite / peer-joined callback thread it end
to end; signature domain deliberately unchanged — FIPS Noise does
its own session auth, so the unsigned hint only affects path
selection
crate::fips
config.rs: renders /etc/fips/fips.yaml and sudo-installs key material
service.rs: systemctl status/activate/restart/mask wrappers
update.rs: GitHub API check against upstream main; apply stubbed
until per-commit .deb artefact source is decided
RPC + dashboard
fips.status / fips.check-update / fips.apply-update / fips.install /
fips.restart registered in dispatcher
HomeNetworkCard.vue shipped standalone (unmounted — place in Home.vue
when ready); shows state pill, version, FIPS npub, update button,
activate button when key is present but service is down
ISO + systemd
archipelago-fips.service: conditional on key presence, masked by
default — backend unmasks after onboarding writes the key
build-auto-installer-iso.sh: multi-stage Dockerfile builds the FIPS
.deb from jmcorgan/fips main (fail-loud), COPYs it into rootfs, apt
installs it so trixie resolves deps; unit copied + masked
Version bump: 1.3.5 → 1.4.0
Tests: 33 new/updated passing (seed, identity, transport, federation,
fips module, transport::fips).
Known gaps: fips.apply-update returns a clear stub error until
upstream publishes per-commit .deb artefacts; HomeNetworkCard is not
mounted in Home.vue by default.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Pure formatter output — no semantic changes. Sweeping these into their
own commit so the FIPS integration diff that follows stays scoped to
the actual feature.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The CI workflow calls `test-iso-qemu.sh "$ISO" 120`. The old arg parser
had a `case *) ISO=...` fallthrough that silently let the second
positional `120` overwrite ISO, so QEMU went looking for a file literally
named "120". That's the "failed step" the user was seeing on recent ISO
runs — the rest of the job succeeded because the QEMU step has
`continue-on-error: true`.
Changes:
- Treat `--timeout=N` or a bare numeric first-match as a CI timeout in
seconds; the original ISO path still wins the positional.
- When a timeout is set, force `--nographic` (CI has no DISPLAY anyway)
and wrap the QEMU invocation in coreutils' `timeout` so the script
always returns instead of hanging.
- After termination (or timeout), grep the serial log for well-known
systemd/live-boot markers. Pass if the kernel reached userspace, fail
if no marker appeared within the window — useful signal rather than
the previous "did the VM shut itself off" proxy.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The .github/workflows/ci.yml Rust job runs cargo fmt --check, clippy
with -D warnings, and tests. All three were failing. This commit:
- Applies rustfmt across the tree (the bulk of the diff — untouched
since the last toolchain bump, so a wide sweep was unavoidable).
- Fixes the correctness-level clippy errors:
container/bitcoin_simulator.rs wildcard-in-or-pattern
container/manifest.rs from_str rename to parse (reserved name)
container/podman_client.rs .get(0) -> .first()
container/runtime.rs manual += collapse
archipelago/src/constants.rs doc-comment → module-doc
api/rpc/package/install.rs stray /// comment above a non-item
container/docker_packages.rs redundant field init
streaming/advertisement.rs missing Metric import in tests
tests/orchestration_tests.rs `vec!` in non-Vec contexts
mesh/listener/dispatch.rs unused store_plain_message import
api/rpc/tor/mod.rs and mesh/steganography.rs: push-after-new → vec!
- Quiets wide legacy surfaces with crate-level allows in main.rs for
stylistic lints (too_many_arguments, type_complexity, doc indent,
enum variant prefix, wildcard-in-or, assertions-on-constants,
drop_non_drop, unused_io_amount, ptr_arg) — these fired in dozens
of places with no correctness payoff and have been churning every
toolchain bump.
- Tags intentional-dead-code helpers: wallet/ and streaming/ modules
are WIP, mesh::send_chunked_payload and DM_V1_MARKER are kept for
rollback compatibility, vpn::get_nostr_vpn_status is surface-area
for a not-yet-landed RPC.
cargo fmt --check, cargo clippy --all-targets --all-features
-- -D warnings, and cargo test --all-features now all pass locally.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Root cause of the "every bubble shows twice" complaint after the prior
dedup fix: the frontend was firing mesh.send twice per user action. A
held/repeating Enter key on the input fires a keydown per repeat, and
handleSendMessage didn't guard on mesh.sending, so both calls queued
through the store's sendQueue and both executed against the same
contact_id (backend logs show two mesh.send RPCs 13ms apart, same text).
That's why sender and receiver both saw doubles — the envelope actually
was transmitted twice.
Mesh.vue: handleSendMessage now early-returns if mesh.sending or
sendingArch is already set. Send button replaces the `...` placeholder
with a proper spinning ring (`.mesh-send-spinner`) so the held-Enter case
stops looking like the app is ignoring the user.
mesh/mod.rs: send_typed_wire_via_federation no longer blocks on the Tor
POST. Sent MeshMessage is recorded synchronously (UI bubble appears
instantly); the HTTP goes in tokio::spawn. Tor circuit setup was the
1–5s lag the user was seeing on every send to a federation peer. Delivery
failure still shows as `delivered: false` via the read-receipt path.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two mesh fixes bundled so the deploy lands them together:
Doubled messages (radio + federation): dedup at store_message now runs
a third cross-transport check keyed on (sender_seq, plaintext, 120s).
The existing (sender_pubkey, sender_seq) match missed the common case
where the same envelope arrives via LoRa radio (sender_pubkey looked
up from the firmware key) and again via Tor federation (sender_pubkey
= archipelago ed25519), because the two lookups disagree. The new
cross-transport match closes that gap without loosening legacy paths.
Stale contacts after clear-all: meshcore's on-device contact table is
persistent and reads back into peers on the next refresh_contacts, so
the previous "nuclear" clear wiped app state for a few seconds before
the old rows reappeared. New persistent `radio_contact_blocklist`
(mesh-ignored-radio-contacts.json) captures the pubkeys present at
clear-time; `refresh_contacts` filters them on read and the filter
survives restart. Federation-synthetic peers are excluded from the
snapshot so the list rebuilds normally on the next gossip.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Clear All now deletes messages.json, mesh-contacts.json, sessions.json,
mesh-outbox.json and clears shared secrets for a truly clean slate.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Mesh adverts now use the node's configured server name (e.g. "ThinkPad",
"Arch Dev") instead of DID key fragments ("Archy-z6MkmkSB")
- Added mesh.clear-all RPC to reset peers, messages, contacts, and history
- Added "Clear All" button in Mesh UI peers panel
- Both glibc and musl builds verified
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add deploy_secondary() function for deploying to multiple LAN nodes
- --both now deploys to .198 and .253 (previously .198 only)
- Fleet deploy updated for 3 LAN nodes
- Mesh DM fixes: protocol frame format, DM-via-channel routing
- Federation pending requests, discover modal
- VPN status UI improvements
- Image versions and container specs updates
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Bundles the Phase 2b/3/4/5 work that accumulated across prior sessions
and the new attachment chunking router from this session. Everything
ships in one shot so the full mesh surface stays coherent on-wire.
Telegram primitives (variants 13–18, 20–22):
- Reply / Reaction / ReadReceipt / Forward / Edit / Delete
- Presence heartbeat + last-seen tracking
- ChannelInvite + ContactCard payload types
- MessageKey (sender_pubkey, sender_seq) as cross-transport identity
- Action menu, reply banner, edit banner, tombstones, (edited) marker
- Debounced auto-read-receipts on scroll + message arrival
Activated prototypes (Phase 4):
- PsbtHash send RPC
- Contacts CRUD (in-memory alias/notes/pinned/blocked)
- Outbox 📤 badge, rotate-prekeys button
- Chunked send fallback (MCIIXXTT framing) as auto-failover inside
send_typed_wire when a typed wire exceeds the LoRa per-frame budget
Unified inbox (Phase 1):
- conversations.list + conversations.messages RPCs (UI collapse deferred)
Attachment transport router (new this session):
- ContentInline variant 23 + ContentInlinePayload carrying file bytes
directly in the envelope for small files with no Tor path
- mesh.send-content-inline RPC — mirrors to local BlobStore, rides
send_typed_wire which auto-chunks over MCIIXXTT framing (~2.3 KB cap)
- mesh.transport-advice RPC as single source of truth for tier
decisions: auto-mesh / choose / tor-only / impossible
- Receive arm writes inline bytes to local BlobStore so the existing
content_ref card renderer handles both transports uniformly
- MeshState.blob_store field + order-independent propagation from
RpcHandler::set_blob_store / set_mesh_service
- Frontend handleAttachFile calls advice first, branches into silent
auto-send, transport-chooser modal, Tor-only path, or red error
- Transport modal with 📡 mesh / 🧅 Tor options + ETA + disabled
state when peer has no Tor reachability
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
MeshCore firmware frame for cmd 0x03 is
`[cmd][txt_type][channel][timestamp_le32][text]`, not `[cmd][channel][text]`.
Missing txt_type + timestamp caused every channel broadcast to come
back with ERR_UNSUPPORTED, which broke the DM-via-channel path
entirely (nothing was reaching the radio). Bring the frame into
spec — verified against meshcore-dev/MeshCore docs/companion_protocol.md.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Firmware rejected send_channel_text(1, ...) with "Unsupported command"
because channel 1 isn't configured on the device. Revert to channel 0
for the DM wrapper — the 0xD1 marker + dest_prefix header still
disambiguates DMs from plain public-channel text. Also revert
Mesh.vue publicChannel back to index 0 so user-typed broadcasts
target the same (only) working channel.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Meshcore direct unicast silently drops between our two Archy nodes
(firmware reports flood sends with resp_code=6 but nothing arrives).
Wrap DMs as channel-1 broadcasts with a [0xD1][dest_prefix(6)][inner]
header; receivers filter by prefix and dispatch the inner payload
through the existing typed/base64/chunk ladder. Shrink chunk body to
125B so the wrapper still fits the 160B LoRa budget. Auto-heal
routing: CMD_RESET_PATH (0x0D) any type-1 contact with path_len=0 on
refresh so floods take over. send_text now returns the firmware's
flood/direct mode flag for diagnostics.
Disable the 120s presence heartbeat broadcaster — its CBOR payload
was being re-echoed as plaintext by the shared repeater, spamming
every visible node with garbled "Archy-…: av�…fstatusfonline…"
messages on channel 0. mesh.broadcast-presence RPC stays registered
but no longer transmits. Re-enable only once presence moves off the
shared broadcast path.
Also: MeshState.cmd_tx behind RwLock so stop()→start() cycles don't
fail with "command channel already consumed"; MeshService.send_cmd
helper; drop_message_by_id for control envelopes that shouldn't
appear as Sent bubbles; self_advert_name reflected into MeshStatus
after set; path_len/flags parsed out of RESP_CONTACT.
Frontend: unified inbox merges mesh peers with federation nodes by
DID/pubkey/name; hide presence/read_receipt/edit/channel_invite/
contact_card from chat stream; publicChannel index → 1 to match the
new DM-via-channel routing.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* Replaces click-anywhere-on-bubble with a tiny ⋯ trigger in the meta row
that fades in on hover (always visible on touch devices). Outside-click
closes the menu, bubble gets a `menu-open` class so the trigger stays lit.
* Action menu gains Forward (any message) + Edit + Delete (own messages
only, delete is red). Reaction spinner + reply preview upgraded to handle
typed targets (attachment/invoice/location/alert) via summarizeForPreview.
* Pending-edit banner with ✎ icon mirrors the reply banner; Send flushes as
mesh.edit-message when pendingEdit is set.
* Forwarded bubbles render "↪ Forwarded from {orig_name}" header; tombstone
+ (edited) markers; pending-reply close button upsized (28px, red hover).
* Scroll + message-arrival watcher fires a debounced 400ms read receipt
with per-peer seq dedup so we never double-ack.
* Chat header: ⟲ rotate-prekeys button next to the shield badge; 📤 outbox
count when mesh.outbox reports queued messages. Blob-store test widget
removed and chat list now sorts by most-recent message timestamp.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds every remaining wire variant and RPC needed to finish the Telegram-quality
mesh plan in a single pass:
* Variants 15 ReadReceipt, 16 Forward, 17 Edit, 18 Delete, 20 Presence,
21 ChannelInvite; plus MeshMessageType::ContactCard(22) cleanup (was
enum-only, now wired through from_u8/label/from_label).
* MessageType::from_label() as the inverse of label() — used by the Forward
path to re-encode a stored typed body back through its original variant.
* RPCs: mesh.send-psbt (variant 3 was previously enum-only),
mesh.send-read-receipt, mesh.forward-message, mesh.edit-message,
mesh.delete-message, mesh.broadcast-presence, mesh.presence-list,
mesh.contacts-list, mesh.contacts-save, mesh.contacts-block,
mesh.send-channel-invite, conversations.list, conversations.messages.
* MeshState gains presence (pubkey → status+timestamps) and contacts
(pubkey → ContactEntry{alias,notes,pinned,blocked}) in-memory stores.
* MeshService gains find_message_by_id (Forward lookup), apply_local_edit /
apply_local_delete (optimistic local echo), and send_chunked_payload — an
MC-framed base64 splitter that fires as a fallback inside send_typed_wire
when wire > MAX_MESSAGE_LEN and no federation path is known. Reuses the
existing receive-side reassembly in listener/decode.rs.
* Receive dispatch arms for PsbtHash, Presence, ChannelInvite, ReadReceipt
(rolls forward `delivered` flag on own-Sent ≤ seq for that peer), Forward,
Edit, Delete. Edit/Delete guard against cross-peer tampering by matching
the target MessageKey pubkey against the sender's advertised pubkey_hex.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Mesh peer pubkeys (LoRa advert ed25519) differ from federation node
pubkeys (archipelago identity), so matching on pubkey always missed
and attachments >160B had no transport. Match on master DID instead;
also accept an explicit peer_onion override from the frontend, which
resolves the peer by display name against federation.list-nodes.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
mesh.send-content was failing with "Message too large for LoRa: 624
bytes (max 160)" because a single ContentRef envelope (cid + onion +
cap_token + thumb) dwarfs a LoRa frame. Add a federation Tor fallback:
- New POST /archipelago/mesh-typed endpoint accepts
{from_pubkey, typed_envelope_b64, signature}, verifies ed25519 over
the raw wire bytes, and injects the decoded envelope into MeshState
via a new MeshService::inject_typed_from_federation helper. This
shares the same dispatch match as LoRa receives via a new pub(crate)
handle_typed_envelope_direct extracted from handle_typed_message.
- MeshService::send_typed_wire_via_federation POSTs the signed wire to
a peer's onion over TOR_SOCKS_PROXY and records a local Sent record.
- handle_mesh_send_content looks up the peer's onion in federation
storage and routes via federation when available, falling back to
LoRa only when no federation presence is known (still fails on
oversized — chunking is Phase 4).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Tap a bubble to open an action menu with Reply + 6 quick reactions.
Reply stashes the target MessageKey and flips the Send button to
"Reply" mode, routing through mesh.send-reply. Reactions call
mesh.send-reaction immediately and render as chips under the target
bubble, collapsed per emoji with a count and self-highlight. Reaction
messages are filtered out of the main chat stream so they don't create
standalone bubbles. Reply bubbles show a "↳ quoted snippet" header
when the target is still in the local window.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Per-target outbound seq counter on MeshState allocates a monotonic seq
before each typed envelope is encoded; send_typed_wire +
send_channel_typed_wire record it (alongside our own pubkey_hex) on the
Sent MeshMessage so the local store carries the same MessageKey the
receiver will see. TypedEnvelope.with_seq lets the RPC layer stamp the
seq AFTER signing (signature covers t/v/ts only).
New MessageKey struct pairs sender_pubkey+sender_seq as the stable
cross-transport identity. Adds variants 13 Reply and 14 Reaction with
ReplyPayload {target, text} and ReactionPayload {target, emoji}, plus
mesh.send-reply / mesh.send-reaction RPCs and receive-side dispatch
arms that store the payload json for the UI to index.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
App.vue listens for postMessage({type:'share-to-mesh',cid,...}) from
marketplace app iframes, stashes the payload in sessionStorage, and
routes to /mesh. Mesh.vue reads the stash on mount (and on a synthetic
'archipelago:share-to-mesh' event when already on the view), showing a
pending-attachment banner in the compose area. Send becomes Share and
flushes the CID via mesh.send-content with the input text as caption.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Marketplace app iframes (Penpot, Gitea, IndeedHub, ...) can POST a file
to /api/share-to-mesh and postMessage the returned CID to the parent
window. The endpoint mirrors /api/blob's body format but adds CORS for
the requesting app origin (any port on host_ip) so proxied apps can
reach it with credentials:'include'. Session cookie is still the primary
auth; the origin check is a sanity guard.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Compose row gains a 📎 attach button that uploads the file via /api/blob
and calls mesh.send-content for the selected peer. Received content_ref
bubbles render as a caption+filename card with either an inline image
preview or a Download button that calls mesh.fetch-content and swaps in
the returned local_url.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds attachment sharing over the mesh: a ContentRef envelope (variant 19)
carries the blob CID, size, mime, optional thumb/caption, and a per-peer
HMAC capability URL so the recipient fetches the full blob out-of-band via
`GET {sender_onion}/blob/{cid}?cap=..&exp=..&peer=..`. BlobStore is shared
from ApiHandler into RpcHandler so mesh.send-content and mesh.fetch-content
(reqwest via TOR_SOCKS_PROXY) hit the same store and cap_key.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Plumbs the BlobStore from blobs.rs into ApiHandler. The HMAC capability
key is derived from the node's Ed25519 signing key via a domain-separated
SHA-256 — rotating the identity rotates every outstanding cap (intentional
so a replaced node cannot honour old tokens).
New routes (added to nginx config in both server blocks):
- POST /api/blob — session-authenticated raw upload, returns
{cid, size, mime, filename, self_test_url}. The self_test_url is a
pre-signed cap pointing at the local node so the UI can verify the
round-trip without needing a peer pubkey.
- GET /blob/<cid>?cap=<hex>&exp=<epoch>&peer=<pubkey> — peer-facing,
HMAC-verified in constant time, expiry-checked, then streams bytes.
Mesh.vue gets a minimal "Attachment test (blob store)" section: file
picker → upload → cid display → "Verify round-trip" and "Open in new
tab" buttons. This validates Phase 3a end-to-end before we layer the
ContentRef typed envelope variant on top.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds core/archipelago/src/blobs.rs: a SHA-256 content-addressed store
that writes bytes to ${data_dir}/blobs/<cid> with a sibling <cid>.meta
JSON file (mime, filename, size, created_at, optional tiny thumbnail).
BlobStore::put is idempotent, max 64 MiB per blob, and issues HMAC-SHA256
capability tokens scoped to (cid, peer_pubkey_hex, expiry_epoch). Tokens
are verified in constant time and rejected on expiry. This is the
foundation piece for the mesh ContentRef typed envelope — the /blob/<cid>
HTTP route and ContentRef variant will land in a follow-up increment
once the HMAC key is plumbed from node identity.
No consumer yet, so the module compiles with dead_code warnings; these
will clear when the HTTP handler and ApiHandler state wiring land next.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds sender_pubkey + sender_seq fields to MeshMessage so received
messages carry a stable cross-transport identity: (sender_pubkey,
sender_seq) pair. This is the foundation for the upcoming reply,
reaction, edit, and read-receipt variants — they need to target a
message by an ID that is meaningful on every node, not just locally.
Receive-side population lives in dispatch.rs::store_typed_message,
which now looks up the peer's pubkey_hex and copies envelope.seq from
the decoded TypedEnvelope. Sent-side population will land when we
plumb a per-node monotonic seq counter through the RPC layer.
Also adds mesh.debug-dump: a full in-memory state snapshot returning
peers, messages, status, shared-secret peer ids, encrypt_relay flag,
and stego mode — intended for smoke tests and bug investigation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
IndeedHub, Bitcoin, and Penpot installs routinely exceed the default
RPC timeout on first pull. Bump package.install specifically to
900s so the frontend doesn't drop the request while the backend is
still downloading images.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds renderers for tx_relay, tx_relay_response, tx_confirmation,
lightning_relay, and lightning_relay_response message types so these
appear as rich cards in the chat stream. sendArchMessage now looks up
our own onion via getTorAddress and skips federation peers that match,
preventing the duplicate "echoed back to self" message we were seeing
on single-node test federations. Empty-federation error message is
also clearer.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adds message_type + typed_payload (JSON) to MeshMessage so the UI can
render invoice/alert/coordinate/tx/lightning messages as structured
cards in both directions instead of showing raw wire bytes on the
Sent side. RPC handlers now route through send_typed_wire /
send_channel_typed_wire which transmit the binary envelope directly
(no utf8_lossy corruption) and record a rich Sent MeshMessage.
Also: store_message deduplicates echo-back doubles (20-msg lookback,
30s window), from_name is plumbed through the federation Incoming
path, and peer_dest_prefix / send_raw_payload are factored out of
send_message.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Swapped all registry references: image-versions.sh, marketplaceData.ts,
curatedApps.ts, catalog.json. git.tx1138.com is now fallback only.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Registry fallback now only tries DIFFERENT registries (skips original
that already failed). 120s timeout per fallback attempt. WireGuard
keys generated on unbundled first-boot. Gitea ROOT_URL uses port 3001.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Gitea image pushed to Archipelago registry. PROXY_APPS stays empty
per user preference - direct port only. Gitea config uses
INSTALL_LOCK + dark theme.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Installs all 7 containers (postgres, redis, minio, relay, api,
ffmpeg, frontend) on indeedhub-net with proper env vars and volumes.
Fixes pull timeout to cover stderr reader. Catalog registry set to
23.182.128.160:3000.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Timeout now wraps stderr reader + wait (was only wrapping wait, so
hung pulls were never killed). 23.182.128.160:3000 is now primary
registry since git.tx1138.com is unreachable.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Previous timeout used ExitStatus::default() which is success on Linux,
so the fallback never triggered. Now properly kills process, awaits
exit, and forces fallback path on timeout.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Image pulls now timeout after 60s and fall through to dynamic registry
fallback instead of hanging forever when primary is unreachable.
Gitea external port corrected to 3001. WireGuard key generation
added to first-boot for fresh installs.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
1. registries.conf includes docker.io search + fallback 23.182.128.160
2. First-boot pull_with_fallback() tries primary then fallback registry
3. FileBrowser created with noauth config on persistent volume
4. Backend dynamic registries.json pre-created in ISO
5. Filebrowser password secret created for token flow
Fixes: apps stuck at 0% download, filebrowser not working, dynamic
catalog not loading on fresh installs.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
App catalog served from Gitea repos (app-catalog) with 35 apps.
Nodes fetch catalog dynamically — new apps appear without frontend
rebuild. Test app added and removed to verify pipeline.
Gitea manifest updated with internal_port/nginx_proxy for iframe.
Updated catalog.json, nginx configs, app session configs.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Configurable registry list persisted to config/registries.json.
Image pulls try all registries in priority order — if primary fails,
fallback registries are attempted automatically. RPC endpoints:
registry.list, registry.add, registry.remove, registry.test.
Replaces hardcoded fallback logic with extensible registry system.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Added gitea to PROXY_APPS so it always routes through /app/gitea/
nginx proxy (same origin as parent page). Fixes X-Frame-Options
SAMEORIGIN rejection when loading via direct port.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
When primary registry (git.tx1138.com) fails, image pull automatically
retries from Gitea registry at 23.182.128.160:3000. Tags pulled image
with original name so install continues seamlessly. Gitea added as
external app in app session config.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Videos and audio now stream directly via URL with auth token query
param instead of downloading entire file into a JS blob. Fixes
playback of large videos (170MB+ was timing out). Images still use
blob URLs. streamUrl() added to filebrowser client and cloud store.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Cloud subpages (Music, Photos, etc.) now show bg-cloud.jpg instead
of falling through to bg-home.jpg.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Video thumbnail in card is pointer-events-none so clicks pass through
to the play handler. Better error messages when preview fetch fails.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Cookie was scoped to /app/filebrowser but Cloud page reads it from
/dashboard/cloud — cookie was invisible. Changed to path=/ so the
auth token is accessible from any page for fetchBlobUrl calls.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Video fills entire viewport with no padding/border-radius. Double-click
toggles native fullscreen. Reduced padding for all media types.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Fedimint can use a remote Bitcoin RPC (e.g., over Tailscale or Tor).
Dependency check now logs info instead of blocking installation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
MediaLightbox: full glassmorphic redesign with dark backdrop, smooth
transitions, proper video/audio/image support. FileBrowser: noauth
config on persistent volume. Deploy script: fixed sed quoting.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Cashu ecash protocol (BDHKE blind signatures, cashuA token format,
mint HTTP client) replacing the stub wallet. TollGate-inspired streaming
data payment system with step-based pricing (bytes/time/requests),
session management with incremental top-ups, usage metering, and
Nostr kind 10021 service advertisements.
13 new streaming.* RPC endpoints. Content server now verifies real
Cashu tokens. Profits tracking includes streaming revenue.
Frontend: GlobalAudioPlayer (persistent bottom bar across all pages),
video lightbox with full controls, audio in MediaLightbox, free file
previews (no blur), paid 10% audio/video previews, separated play
vs download buttons in PeerFiles.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Prevents duplicate resolver directive error when both
nginx-archipelago.conf and external-app-proxies.conf are loaded
at http context level.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Convert botfights from external link to real container app on port 9100.
Add manifest, update marketplace/discover/kiosk/session configs, switch
registry URLs to git.tx1138.com.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add git.tx1138.com to trusted registries (replaces old 80.71.235.15)
- Add botfights app config: port 9100, data volume, JWT_SECRET auto-gen, fight loop
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Apps are now installed exclusively via the Marketplace UI.
The deploy script handles code sync, backend/frontend builds,
and service restarts only. The legacy container creation code
is wrapped in `if false` to preserve git history.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- ISO builder: run npm ci before npm run build to prevent stale UI artifacts
- Unbundled ISO: clean container-images dir to prevent bundled tars leaking
- WireGuard: use After=network.target instead of network-online.target for
faster wg0 startup on install
- VPN status: check actual nvpn0 interface instead of config tunnel_ip to
prevent NostrVPN from showing standalone WireGuard IP
- ContainerApps: filter out not-installed bundled apps (fixes Bitcoin Knots
appearing on clean unbundled installs)
- Kiosk: persist kiosk mode to localStorage before /kiosk redirect so
App.vue can skip remote relay (fixes input doubling with companion app)
- IndeedHub: fix port mapping and X-Forwarded-Prefix passthrough
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- IndeedHub container port changed from 7777 to 7778 (7777 used by nostr-relay)
- Nginx proxy updated to route to 7778
- Backend config.rs port mapping updated
- Podman registries.conf switched to v2 format (fixes mixed v1/v2 error)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Registry migration to git.tx1138.com/lfg2025, version bump for
release testing across nodes.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
All hardcoded references to the old IP-based registry replaced across
Rust backend, Vue frontend, shell scripts, Dockerfiles, CI, and docs.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- VPN status: don't show WG IP as NostrVPN IP when tunnel not up
- VPN section polls every 15s so IP updates after pairing
- NostrVPN shows "Pair a device" when service active but no tunnel
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Unbundled ISO: first-boot only creates FileBrowser (marker file .unbundled)
Users install apps from Marketplace — no more bitcoin/mempool on clean install
- VPN status: read tunnel IP from config file (instant) instead of nvpn status (22s)
- Kiosk: App.vue skips remote relay on /kiosk path (prevents duplicate input)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
VPS runner was sniping jobs and failing instantly (no build env).
Changed runs-on from ubuntu-latest to iso-builder label, which only
the ThinkPad runner has registered.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
act_runner runs non-interactive shells where nvm isn't loaded.
Cargo steps already source .cargo/env but frontend steps were missing
the equivalent nvm.sh sourcing, causing "npm: command not found".
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Linux-only option that mirrors ISO install exactly: builds backend
(release), frontend (with typecheck), syncs all configs, and restarts
all system services (Tor, WireGuard, NostrVPN, nginx, backend).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
ISO boot failed in emergency mode because:
- fsck.ext4 binary missing (no e2fsprogs in rootfs)
- LUKS data volume never opened (no cryptsetup-initramfs in initramfs)
Both packages were in the installer debootstrap but not the target rootfs
Dockerfile. The initramfs regeneration at install time now includes LUKS
support since cryptsetup-initramfs is present.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Update all references from Debian 12 (Bookworm) to Debian 13 (Trixie)
- Enable SystemCallArchitectures, RestrictAddressFamilies, RestrictRealtime
in archipelago.service (safe on systemd 256+ which respects NoNewPrivileges=no)
- Update GLIBC compatibility checks from 2.36 to 2.40
- ISO filename, build container, and docs updated throughout
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
ISO build no longer copies netavark from build host (Debian 13/GLIBC 2.41)
which broke container networking on Debian 12 targets. Rootfs already
installs netavark from Debian 12 repos — just configure the backend.
Install RPC now adopts existing containers (from first-boot) instead of
erroring on duplicates. Container scanner extracts real versions from
image tags and detects available updates against pinned versions.
Frontend shows update button with version info when updates are available.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Backend: cache status in RwLock, refresh every 15s via background task.
Eliminates per-request TCP race to ElectrumX that caused volatile errors.
Fix error classification so "Failed to read" is transient, not hard error.
Frontend: keep last-known-good data across failed polls, persist Tor
onion once discovered, adaptive polling (5s active / 30s synced).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Replace aardvark-dns container names with host.containers.internal
for all cross-app connections (LND→Bitcoin, ElectrumX→Bitcoin,
Mempool→ElectrumX, Fedimint→Bitcoin, NBXplorer→Bitcoin P2P+RPC)
- Add BTCPay multi-container stack installer (postgres + nbxplorer +
btcpay-server) with proper secrets, data dir ownership, NOAUTH
- Add Mempool multi-container stack installer (mariadb + mempool-api +
mempool-frontend) with host.containers.internal for RPC
- Immediately remove apps from state on uninstall (no 3-min ghost delay)
- Include archy-bitcoin-ui in bitcoin uninstall container list
- Fix LND UI port 8081 (was 8080, conflicting with LND gRPC)
- Fix ElectrumX UI: proxy /electrs-status to backend, cache-busting
headers, graceful fallback when backend returns HTML
- Add Tor hidden services for ElectrumX and LND in torrc template
- Remove unused detect_bitcoin_container_name() (replaced by
host.containers.internal)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Standalone WireGuard (wg0:51820):
- New archipelago-wg.service creates wg0 independent of NostrVPN
- Keypair generated on first-boot, persisted on LUKS partition
- vpn.create-peer uses wg genkey/pubkey (no nvpn dependency)
- wg-address service depends on archipelago-wg, not nostr-vpn
Networking fixes:
- Remove nos.lol from default relays (requires PoW, events rejected)
- Add Tor hidden service for private relay (port 7777) — NAT'd peers
can reach relay over Tor for NostrVPN signaling
- Fix Tor hostname sync race: wait loop before copying hostname files
- Add tor-hostnames + wireguard dirs to LUKS partition setup
- Include relay in hostname sync loops (setup-tor.sh + first-boot)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace disconnected "Generate Invite" + "Add participant" with a 2-step
wizard: enter phone npub → get invite QR + mesh details. Backend vpn.invite
now accepts optional npub param to add participant in the same call. Modal
shows network ID, node npub, and relay URLs for manual app configuration.
Also includes nostr-vpn service hardening (rate-limit restarts, reset-failed
before enable).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
polkit denies reboot/shutdown for non-root users without a local seat
(e.g. SSH sessions). Since archipelago has NOPASSWD sudo, add shell
aliases so reboot/shutdown/halt/poweroff transparently use sudo.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Two issues on fresh ISO install:
1. nostr-vpn.service was enabled in rootfs but env file doesn't exist
until first-boot generates Nostr identity — crash-loop on boot.
Now only enabled by first-boot-containers.sh after identity exists.
2. LUKS encrypted partition mounts over /var/lib/archipelago/, hiding
the relay config.toml the Dockerfile put there. Now copies relay
config and creates nostr-relay/nostr-vpn dirs on the LUKS partition.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The nvpn daemon config at /var/lib/archipelago/nostr-vpn/ is owned by
root, but the backend runs as archipelago. Direct write silently failed,
so adding a second phone's npub never reached the daemon — service
restarted with stale config. Now falls back to sudo cp for root-owned
paths, and first-boot sets ownership to archipelago.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
podman export creates paths without ./ prefix, but tar tf checks
used ./etc/... which never matched. List once, grep without prefix.
Also fix git commands to use $HOME/archy (workspace has no .git).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Dockerfile RUN steps execute under /bin/sh (dash on Debian), which
doesn't support brace expansion {a,b,c}. The nostr-relay directory
was never created, causing the config copy to fail (build #444).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Runner is Debian 13 (glibc 2.41), ISO rootfs is Debian 12/bookworm
(glibc 2.36). Dynamic binary crashes with GLIBC_2.41 not found.
Musl static build eliminates the dependency entirely.
Also set GRUB_DISTRIBUTOR="Archipelago" so installed system boot
menu says "Archipelago" not "Debian GNU/Linux".
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add nostr-rs-relay as native system service (port 7777) for VPN
signaling. Every node runs its own private relay from first boot.
Update nvpn binary from v0.3.4 to v0.3.7 (fixes mesh event
processing). Add WireGuard helper and address service for peer VPN.
First-boot script configures relay, nvpn identity, relay URLs
(direct + Tor onion), and syncs daemon config.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- VPN card: relay URLs, device management, invite QR, add participant
- Backend: vpn.invite, vpn.add-participant, vpn.peer-config RPCs
- nvpn v0.3.7 system service (fixes event processing bug in v0.3.4)
- First-boot: auto-configure nvpn with node identity and endpoint
- Service: AF_NETLINK for WireGuard, NoNewPrivileges=no for sudo wg
- TASK-50: networking stack reliability from first install
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Revert CI to normal cargo build --release (musl was false positive)
- Add acpid + acpi-support-base to rootfs packages
- Add acpi=force to GRUB and ISOLINUX boot params (installer + installed)
- Fixes "Maybe missing ACPI. Shutdown not powering off" on some hardware
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Install progress bar replaces action buttons (no overlay)
- Hide status badge during install/uninstall
- Uninstall keeps progress state until container disappears from WebSocket
- Uninstall RPC timeout increased to 660s (Bitcoin UTXO flush)
- Installing apps appear in My Apps immediately as placeholders
- Auto-configure Tor hidden service for every app on install
- Widen Tor module visibility for install hooks
- Only clear stale install entries on error status
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Build server (Debian 13) has GLIBC 2.41 but ISO targets Debian 12
(GLIBC 2.36). Switching to x86_64-unknown-linux-musl produces a
fully static binary that runs on any Linux.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
nvpn status command hangs for seconds (connects to relays), causing
the Network page to never finish loading. Read tunnel_ip from the
local config file instead (instant).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Remove ReadWritePaths sandbox (causes namespace error when /run/nostr-vpn
doesn't exist after reboot — /run is tmpfs)
- Detect both 'active' and 'activating' states in VPN status check
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Prevents burst of health checks, scans, and snapshots after slow
podman responses by using MissedTickBehavior::Skip. Bumps container
scan interval from 30s to 60s to reduce DB lock contention.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add vpn.create-peer, vpn.list-peers, vpn.remove-peer RPC methods
- Generate WireGuard config + QR code (SVG) for mobile device connection
- Add "Add Device" modal on Network page with QR scanner support
- Remove old build-iso.yml (replaced by build-iso-dev.yml)
- Remove container-tests.yml (tests run in dev workflow)
- Remove container orchestration tests from dev workflow (redundant)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Build in $HOME/archy to reuse target/ cache across CI runs
- Skip apt-get install when ISO build deps already present
- Cargo tests also use persistent target dir
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Kiosk was redirecting /kiosk → /dashboard, bypassing RootRedirect
and BootScreen entirely. This caused the kiosk to land on Login.vue
showing "server is starting up" in a loop instead of the proper
terminal-style boot progression screen.
Now /kiosk → / → RootRedirect → BootScreen, matching what remote
browsers see.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
nvpn binary writes to $HOME/.config/nvpn. Set HOME to data dir,
create runtime dirs in ExecStartPre, remove overly restrictive
ProtectSystem/ProtectHome that blocked the binary.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Fix vpnStatus type mismatch (provider: string|undefined vs string|null)
- Remove redundant history_dirty assignment in health_monitor.rs
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add system.settings.get/set RPC methods for Claude API key management
- Save key to secrets/claude-api-key, restart claude-api-proxy service
- Home Network card now fetches VPN status via vpn.status RPC
- Shows provider name (nostr-vpn, tailscale) instead of just "Connected"
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
FIPS stays in the marketplace as an installable container app.
NostrVPN is the native system service; FIPS is a separate optional app.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Convert NostrVPN from container app to native systemd service
- Auto-configure VPN with node's Nostr identity after onboarding
- Add nostr-vpn.service with proper capabilities (NET_ADMIN, NET_RAW)
- Remove FIPS from marketplace, container config, nginx, image-versions
(consolidated into NostrVPN — same mesh VPN concept)
- Add AIUI inclusion step to dev CI workflow
- AIUI installed on VPS build server for ISO inclusion
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add NostrVPN as a native systemd service (extracted from container)
- Add VPN status detection for nostr-vpn in backend vpn.rs
- ISO build extracts nvpn binary from container image
- First-boot auto-configures NostrVPN with node's Nostr identity
- Change Claude Auth from login iframe to API key input field
- Remove duplicate ChangePasswordSection from Settings.vue
- FIPS and Routstr remain as installable container apps
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- FIPS container expects FIPS_NSEC/FIPS_NPUB, not FIPS_NOSTR_SECRET
- NostrVPN container doesn't have a 'start' binary — use image default
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Remove MemoryDenyWriteExecute=yes from archipelago.service — ring
(rustls) and secp256k1 (bitcoin/nostr) crypto libraries need
executable memory mappings that this restriction blocks
- Add + prefix to ExecStartPre so mkdir/chown run as root
- Use $HOME/archy instead of /home/archipelago/archy in CI workflows
so builds work on both .228 and VPS CI runners
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace hardcoded /run/user/1000 with $(id -u archipelago) so first-boot
works regardless of the archipelago user's UID.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add three new marketplace apps:
- Routstr (v0.4.3): Decentralized AI inference proxy with Cashu payments
- Nostr VPN (v0.3.4): Mesh VPN with Nostr signaling + WireGuard tunnels
- FIPS (v0.1.0): Self-organizing encrypted mesh network
Includes status UI dashboards for headless apps (nostr-vpn-ui, fips-ui)
with usage instructions, node identity display, and container logs.
Nostr identity injected via env vars for all three apps.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Deploy backend binary + frontend to VPS after successful build
- Fix ISO ownership to use runner's UID instead of hardcoded 1000
- FileBrowser on VPS serves ISOs at :8083
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
bitcoin.conf already has server=1, rpcbind=0.0.0.0, rpcallowip, listen.
Passing them again via command-line causes bitcoin to try binding port
8332 twice → "Address already in use" → container crashes on every start.
Now only passes pruning/txindex args and dbcache via CLI.
Health check uses cookie auth (-datadir) instead of plaintext password.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Critical:
- BUILD_VERSION was hardcoded as "1.3.0-alpha" — now reads from Cargo.toml
This caused ALL ISOs to show v1.3.0 regardless of actual binary version
Kiosk:
- Remove --disable-gpu flags (broke display scaling on some monitors)
- Add --start-fullscreen --window-size for reliable fullscreen
New apps:
- Nostr VPN, FIPS, Routstr, noStrudel, BotFights, NWNN, 484 Kitchen,
Call the Operator, Arch Presentation, Syntropy Institute, T-0
Rust: suppress dead_code and unused_assignments warnings
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
We were editing build-iso.yml but Gitea runs build-iso-dev.yml.
Replaced actions/checkout@v4 with direct git fetch+rsync.
This is the root cause of stale builds all day.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
actions/checkout@v4 uses a broken Gitea-generated token that always
fails. Replaced with direct git fetch+reset on the local repo, then
rsync to workspace. No more stale builds. Verified with version check.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add firmware-linux-nonfree to ISO (fixes missing Realtek NIC firmware)
- Pre-create nbxplorer/Main and btcpay/Main data directories
- Fix fedimint data dir permissions (chmod 775 for non-root container)
- GRUB GFX fallback: gfxpayload=keep + console fallback for incompatible hardware
- Kill stale Chromium before kiosk restart (prevents duplicate processes)
- Suppress Rust warnings: #[allow(dead_code)] on run_boot_reconciliation,
#[allow(unused_assignments)] on history_dirty
- Version bump to 1.3.3
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The actions/checkout@v4 step fails with stale Gitea token but leaves
a cached .git dir, preventing the fallback from triggering. Now we
always rsync from ~/archy/ which is kept up-to-date via git pull.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The actions/checkout fails (Gitea token issue) and falls back to
~/archy local copy. But local copy was stale — builds were missing
fixes. Now: always git pull in local repo before rsync fallback.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Pre-create Documents/Photos/Music/Downloads/Builds dirs for FileBrowser
- Add "Already set up? Log in" link on onboarding intro page
- Prevents users from getting stuck in onboarding loop
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Claude proxy no longer crashes when ANTHROPIC_API_KEY is not set.
Instead serves a 401 with a helpful message telling users to configure
their API key in Settings. Fixes blank AIUI on fresh installs.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Onboarding:
- Persist current step in localStorage — page refresh resumes where user was
- Router afterEach saves step; guard redirects to saved step, not always intro
- Show npub alongside DID on restore success screen
UI fixes:
- Clipboard polyfill for HTTP contexts (fixes Copy DID crash on non-HTTPS)
- AppCard installing overlay shows for pkg.state=installing (survives refresh)
- Hide uninstall button during installation
- Frontend version bumped to 1.3.2
App store:
- OnlyOffice fully removed from marketplace, curated apps, app config
- Replaced with CryptPad references throughout
- Remove OnlyOffice from ISO capture patterns
Container stability:
- UI containers (bitcoin-ui, lnd-ui, electrs-ui) pull from registry first
- Added --cap-add FOWNER for rootless Podman compatibility
- electrs-ui now included in first-boot loop alongside bitcoin-ui and lnd-ui
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
# ISO will be created in: results/archipelago-auto-installer-*.iso
```
## What the ISO Includes
✅ Complete Debian 12 root filesystem
✅ Pre-built Archipelago backend
✅ Pre-built frontend (web UI)
✅ **Prepackaged container images** (Bitcoin Knots, LND, UIs, and other bundled apps), loaded on first boot
✅ Nginx configuration (HTTPS ready)
✅ Auto-installer that:
- Detects internal disk
- Creates partitions (EFI + root)
- Extracts pre-built system
- Installs bootloader
- Reboots to working system
## What Users Need to Do Post-Install
1.**Start apps from the Web UI**– Container images are prepackaged and loaded on first boot. Bitcoin Knots + UI, LND + UI, and other bundled apps are ready to start from the Web UI without manual `podman run`. No need to pull or deploy core containers.
2.**Access Web UI**– Navigate to `http://[server-ip]`
## Testing the ISO
```bash
# Use VirtualBox, QEMU, or real hardware
qemu-system-x86_64 \
-m 4G \
-cdrom results/archipelago-auto-installer-*.iso \
-hda archipelago-test.qcow2 \
-boot d
```
## Important Notes
⚠️ **The auto-installer will ERASE the target disk!**
⚠️ Make sure to test on a non-production machine first
⚠️ Minimum 20GB disk space required (500GB+ recommended for Bitcoin)
## Build from Source (Alternative)
If you want to build everything from scratch instead of capturing the live server:
```bash
BUILD_FROM_SOURCE=1 ./build-auto-installer-iso.sh
```
This will:
- Build backend from Rust source
- Build frontend with `npm run build`
- Create fresh SSL certificates
- Generate default configs
## Troubleshooting
**ISO won't boot:**
- Ensure UEFI mode is enabled
- Try disabling Secure Boot
**Installer hangs:**
- Check the auto-start script fix is applied (see DEPLOYMENT.md)
- Fleet now preserves the last known node list, alerts, and selection locally while telemetry refreshes in the background, so the dashboard no longer blanks on tab switches or update scans.
- Connected nodes and identities now reuse their last loaded data instead of reloading the visible list every time the user revisits the tab.
- The Fleet matrix and detail views now show actual node names and host information instead of raw node id prefixes.
- The network map only redraws when its graph data actually changes, which stops the D3 scene from visually resetting on every refresh tick.
- Mobile federation and system-update actions now stack full width, and the ElectrumX app health check allows a long startup window so slow sync nodes do not restart mid-index.
- Validation passed with `git diff --check`, focused frontend tests, and `npm run type-check`.
## v1.7.85-alpha (2026-06-12)
- ElectrumX now runs with less cache pressure and more memory headroom, reducing the restart loop seen during sync catch-up.
- Portainer is pinned to `2.19.4` instead of `latest`, avoiding schema-drift restarts from surprise image updates.
- LND receive-address creation now asks for a native SegWit address and returns clearer wallet/readiness failures when an address is not available.
- Fleet telemetry now carries server name, hostname, and server URL, and the Fleet dashboard shows those names instead of hashed node ids.
- Trusted federation peers are still auto-added transitively, but the local node no longer imports itself back into the fleet list.
- Validation passed locally for the touched frontend helpers, `git diff --check`, and Rust formatting.
## v1.7.84-alpha (2026-06-11)
- Bitcoin trusted-node relay approvals now generate restricted `txrelay` RPC credentials when needed and restart the active Bitcoin backend so bitcoind loads the new `rpcauth` whitelist.
- Kiosk mode now includes a browser safe-area path for HDMI displays that crop edges, and self-update refreshes kiosk launcher/systemd files so display fixes ship to existing nodes. The experimental X11 scaling safe-area is opt-in to avoid stretching TV output.
- Wi-Fi setup now reports scan errors instead of showing an empty network list, supports retrying scans from the modal, parses escaped `nmcli` SSIDs correctly, and can join open networks without forcing a WPA password.
- Bitcoin Core now matches Bitcoin Knots for restricted relay RPC support, including the txrelay secret injection and transaction broadcast whitelist.
- The restricted Bitcoin relay whitelist now includes `submitpackage` and `gettxout`, covering newer wallet/package-relay broadcast flows without opening wallet/admin RPC.
- The Bitcoin UI companion image is pinned to `1.7.84-alpha` across release metadata and the Quadlet fallback path, avoiding stale `latest` detection during OTA updates.
- Container scanning now uses an RAII in-flight guard so timeout and error paths cannot leave the scanner stuck in a permanently busy state.
- Validation passed with `cargo fmt`, `cargo check -p archipelago`, `git diff --check`, and focused source review of the relay message/approval path.
## v1.7.83-alpha (2026-06-11)
- App launch metadata now derives more consistently from app manifests, with typed launch interfaces and catalog generation updates that keep packaged apps aligned with their runtime ports and launch surfaces.
- Revoked or unsupported app surfaces were removed from the catalog and release path, including OnlyOffice and the unvalidated Saleor surface, so the Marketplace no longer exposes apps that cannot be safely supported in this release.
- The frontend production build now passes strict TypeScript checks after tightening app details, Web5, cloud refresh, and credential test typing.
- Mobile and desktop app surfaces received release polish: improved mobile app layout, safer mesh desktop/tablet scrolling, and the Home system card now routes directly to monitoring.
- Bitcoin UI status rendering now avoids false stale/reconnecting states when fresh block snapshots advance, and guards optional DOM updates so the standalone Bitcoin UI is more resilient.
- Deploy tooling now excludes local Codex scratch output, archived image-build artifacts, and upload screenshots from target syncs, and bounded optional IndeedHub fixups so a stuck Podman helper cannot hold the deploy.
- Validation passed with `npm run type-check`, production `npm run build`, backend `cargo build --release`, catalog/release manifest checks, focused frontend tests, and live `.198` deploy verification through the frontend/service restart phase.
## v1.7.82-alpha (2026-05-22)
- Saleor storefront proxying now forwards `X-Forwarded-Host`, fixing Next.js Server Actions requests that compared the browser origin with the internal `storefront-app:3000` upstream host.
- Saleor storefront media now routes `/thumbnail/` and `/media/` through the same `9011` proxy to the Saleor API, fixing product image optimizer failures caused by `localhost:8000` media URLs.
- The Saleor storefront container receives an explicit internal media origin so rewritten media URLs resolve inside the Podman network without exposing private API ports to browsers.
- Validation passed with `cargo fmt --all --check --manifest-path core/Cargo.toml`, `cargo check -p archipelago --manifest-path core/Cargo.toml`, and live checks on `100.114.134.21` for storefront HTML, static assets, GraphQL, media redirects, and optimized product images.
## v1.7.81-alpha (2026-05-21)
- Saleor storefront installs now use the prebuilt registry image instead of building the Next.js app on-device, avoiding Podman build failures during stack installation.
- Existing Saleor stacks are repaired on adoption by recreating missing storefront containers, forcing the storefront app to bind `0.0.0.0:3000`, and resolving nginx upstreams dynamically after container restarts.
- The shipped Saleor storefront image now includes public assets and omits Vercel-only Speed Insights injection, fixing broken static asset responses and the local `/_vercel/speed-insights/script.js` browser warning.
- Validation passed with `cargo fmt --all --check --manifest-path core/Cargo.toml`, `cargo check -p archipelago --manifest-path core/Cargo.toml`, and live checks on `100.114.134.21` for `9011` storefront, static assets, and proxied GraphQL.
## v1.7.80-alpha (2026-05-21)
- Saleor storefront proxying now falls back to the direct request scheme when no forwarded protocol header is present, fixing direct `http://node:9011` launches that could generate an invalid same-origin GraphQL URL.
- The Saleor storefront release path keeps public proxy support intact by still honoring forwarded HTTPS headers for Nginx Proxy Manager domains while repairing local/direct port launches.
- Validation passed with `cargo fmt --check` and `cargo check` for the Archipelago backend before release staging.
## v1.7.79-alpha (2026-05-20)
- Saleor now installs the official Saleor Storefront as part of the stack, built from the pinned `saleor/storefront` source and served as the customer-facing shop on port `9011`.
- Saleor app launches now open the storefront while the admin dashboard remains available on port `9010` with the generated `admin@example.com` credentials shown in Archipelago.
- Public Nginx Proxy Manager hosts forwarding to the Saleor storefront also expose same-origin `/graphql/`, so public storefront domains can talk to the local Saleor API without mixed-content or private-LAN reachability failures.
- Saleor stack metadata, marketplace descriptions, catalog ports, scanner exclusions, and app-session routing now describe the storefront/dashboard/API split explicitly.
## v1.7.78-alpha (2026-05-20)
- Public Nginx Proxy Manager hosts for Saleor now keep browser GraphQL calls same-origin at `/graphql/` and proxy them to the local API on `8000`, fixing `Failed to fetch` when a public domain such as `noderunner.shop` was loaded from devices that cannot reach the node's private LAN/tailnet API address.
- Saleor's validated stack changes are now release-ready: dashboard origins on port `9010` are explicitly allowed for dashboard/API calls, preserving the working test-node install path for production nodes.
- NetBird launches now stay pinned to the unified dashboard/proxy origin on port `8087` instead of following stale runtime-discovered server URLs on `8086`.
- NetBird's local nginx proxy now routes browser API, OAuth, relay, and WebSocket traffic through `host.containers.internal:8086` instead of a hard-coded rootless Podman gateway IP, and includes the upstream `management.ProxyService` gRPC path.
- The mobile credentials interstitial now keeps credential lists scrollable and action buttons reachable in both My Apps and the mobile app icon grid.
- Android WebView popup windows now hand external popup URLs to the system browser, covering app login/signup flows that open secondary windows.
- Validation passed with `git diff --check`, `cargo check -p archipelago`, and the focused `npm test -- src/views/appSession/__tests__/appSessionConfig.test.ts` suite.
## v1.7.77-alpha (2026-05-20)
- Saleor first-use now exposes generated credentials through Archipelago instead of leaving users at an unexplained dashboard login: App Details shows copyable `admin@example.com` credentials, and My Apps/mobile icon launches show a pre-launch credentials modal.
- Saleor installs now create or repair the `admin@example.com` staff account idempotently after sample data loads, use the correct dashboard mount path, and re-check stack containers after startup so stopped containers are caught.
- NetBird embedded login now uses the upstream-compatible IdP signing-key behavior and sends ID tokens from the dashboard to the management API, fixing the post-signup `Unauthenticated` state while preserving the unified local proxy/logout routes.
- Transient unnamed Podman helper containers created during app install tasks are hidden from My Apps, so generated names like `eager_keldysh` no longer appear as user applications.
- Validation passed with catalog/release JSON checks, `npm run type-check`, and `cargo fmt --all --check --manifest-path core/Cargo.toml`; live checks on `100.114.134.21` confirmed Saleor dashboard/API availability, generated Saleor admin login, NetBird OAuth availability, and NetBird logout redirects.
## v1.7.76-alpha (2026-05-20)
- Saleor installs now use dashboard port `9010`, avoiding the existing Portainer `9000` binding on the test node while keeping API `8000`, Mailpit `8025`, and Jaeger `16686` unchanged.
- Saleor's Valkey cache no longer bind-mounts `/var/lib/archipelago/saleor-cache`, and the dashboard container has the minimal rootless nginx capabilities it needs to chown cache files, bind port 80 inside the container, and drop workers to the nginx user.
- NetBird's browser proxy now sends API, OAuth, relay, WebSocket, and management traffic through the stable host-published server port at `169.254.1.2:8086`, avoiding stale rootless Podman DNS/IPs after `netbird-server` restarts.
- Mobile App Store category chips now stay visible above the tab bar, Discover is available on mobile, and category selection updates the page route/query so the selected category is actually shown.
- Apps that require a real browser tab now open directly from the app icon tap instead of first entering an in-shell app-session route, including BTCPay, Grafana, Home Assistant, Vaultwarden, Nextcloud, Portainer, OnlyOffice, Tailscale, Uptime Kuma, Gitea, and Nginx Proxy Manager.
- Validation passed with catalog JSON checks, `npm run type-check`, `cargo fmt --all --check --manifest-path core/Cargo.toml`, and `cargo check -p archipelago --manifest-path core/Cargo.toml`; live checks on `100.70.96.88` confirmed Saleor dashboard `9010`/API `8000` and NetBird API/OAuth routes survive `netbird-server` restart.
## v1.7.75-alpha (2026-05-19)
- Saleor is now published as a recommended commerce app with catalog metadata, icon, direct app-session launch on port `9000`, scanner metadata, image pins, and a full stack installer for dashboard, API, worker, PostgreSQL, Valkey, Mailpit, and Jaeger.
- Existing NetBird installs are repaired more aggressively by rewriting unified-origin config, recreating the dashboard/proxy containers, restarting the server, preserving data, and handling exact `/api` and `/oauth2` routes plus dashboard logout redirects through the local proxy.
- Desktop dashboard scrolling now hands focus back from the sidebar to the main content when the pointer or wheel moves over the main pane, preventing the sidebar scroll area from trapping wheel input on short screens.
- Validation passed with catalog JSON checks, `npm run type-check`, `cargo fmt --all --check --manifest-path core/Cargo.toml`, and `cargo check -p archipelago --manifest-path core/Cargo.toml` before release.
## v1.7.74-alpha (2026-05-19)
- App-session right panels now re-focus the iframe after load and when the frame area is activated, so wheel/touch scrolling works immediately after switching tabs or selecting an app on shorter screens.
- NetBird now launches through a unified local origin on port `8087` that proxies the dashboard plus `/oauth2`, `/api`, relay, WebSocket, and gRPC routes to `netbird-server`, fixing the embedded login flow that previously ended in `Unauthenticated` or `404 page not found` after logout.
- Existing NetBird installs are repaired on adopt/start by rewriting `config.yaml`, `dashboard.env`, and the local nginx proxy config, then creating the missing `netbird-dashboard` and `netbird` proxy containers when needed while preserving NetBird data.
- Saleor is still pending and is not included in this release; its registry/installer work remains local until it can be validated separately.
- Validation passed with catalog JSON checks, `npm run type-check`, `cargo fmt --all --check --manifest-path core/Cargo.toml`, and `cargo check -p archipelago --manifest-path core/Cargo.toml`.
## v1.7.73-alpha (2026-05-19)
- Mobile app launches for iframe-blocked apps now open the direct app URL in a new browser tab immediately instead of landing in a broken in-shell webview that requires a second tap.
- Mobile My Apps/Websites tabs now react to route query changes, App Store pages label the mobile view as Discover, mobile filters have safe bottom spacing, and App Store search ignores the current category so searches cover all available apps.
- My Apps search now surfaces matching App Store entries when the app is not installed, making it possible to jump directly from a failed My Apps search to the installable app details.
- NetBird self-host installs now prefer a `100.x` tailnet/CGNAT address for dashboard, management, relay, STUN, and auth redirect origins when one is present; live repair on `100.89.209.89` updated the existing stack from LAN origins to `100.89.209.89` and restored `netbird-server`.
- App-session iframe frames now focus automatically and wrap the iframe in a scroll host so wheel/touch scrolling works in the active right frame without requiring an initial click.
## v1.7.72-alpha (2026-05-19)
- Settings What's New now includes the missing release notes for `v1.7.68-alpha` through `v1.7.71-alpha`, so the modal reflects the current OTA history instead of stopping at `v1.7.67-alpha`.
- The follow-up release carries the NetBird install fix, Gitea icon polish, mobile app-session fallback updates, and rounder app icon masks from `v1.7.71-alpha` with the Settings modal notes included.
- The local Cargo lockfile version metadata is kept in sync with the release bump after the previous release build updated it.
## v1.7.71-alpha (2026-05-19)
- NetBird stack installs now pre-create `/var/lib/archipelago/netbird/data` before binding it into `netbird-server`, fixing the failed install/start path seen on `100.70.96.88` where Podman rejected the missing host directory.
- NetBird start/restart ordering now starts `netbird-server` before the dashboard container so lifecycle actions bring the control plane up before the UI.
- App-session invalid IDs and panel-mode fallbacks now return to `/dashboard/apps`, avoiding the stale `/apps` route that could render a 404.
- Mobile launches for apps that block iframes now stay inside the Archipelago app-session fallback instead of automatically opening an external browser tab.
- Installed Gitea containers now report the packaged Gitea icon, and app icon masks use a rounder radius on mobile grids, app cards, and detail headers.
- Validation passed with `npm run type-check`, focused Vitest app-session/app-grid tests, `cargo fmt --all --check --manifest-path core/Cargo.toml`, and `cargo check -p archipelago --manifest-path core/Cargo.toml`.
## v1.7.70-alpha (2026-05-19)
- NetBird is being corrected from the peer/client daemon image to the self-hosted NetBird control-plane stack with a launchable dashboard on port `8087`, a combined management/signal/relay server on `8086`, and STUN on UDP `3478`.
- App sessions now always launch local apps through direct host ports and carry an explicit dashboard return target, so closing an iframe returns to the launching dashboard screen instead of falling through to browser history or a 404.
- Mobile app launches ignore stale desktop panel state and route into the full app-session webview consistently.
- The desktop sidebar now pins the logo/version at the top and controller/online/mode controls at the bottom, with only the navigation section scrolling on shorter screens.
- Validation passed with catalog JSON checks, `scripts/image-versions.sh` syntax check, `npm run type-check`, `cargo fmt --all --check --manifest-path core/Cargo.toml`, and `cargo check -p archipelago --manifest-path core/Cargo.toml`.
## v1.7.69-alpha (2026-05-19)
- App installs now allow up to 10 minutes for the initial `package.install` RPC to return, matching slow container image pulls and preventing apps from disappearing from My Apps while the backend is still pulling or retrying mirrors.
- Live diagnostics on `100.70.96.88` confirmed the Gitea install did not fail; the primary registry pull timed out after 300 seconds, the fallback mirror succeeded, and Gitea came up healthy on `3001` while the frontend had already timed out at 15 seconds.
- Gitea and other Docker-image app installs now stay visible during slow registry pulls instead of being marked as failed by the browser before backend install progress can complete.
- Gitea is now categorized as a known Data app in My Apps, so a running Gitea container appears with installed apps instead of being filtered into the Websites/Services split.
- NetBird `0.71.2` is now available in the app catalog and fallback marketplace data as a recommended networking app using the official `docker.io/netbirdio/netbird:0.71.2` image.
- NetBird installs get persistent state under `/var/lib/archipelago/netbird`, `NET_ADMIN`/`NET_RAW`, `/dev/net/tun`, `slirp4netns`, image-version pinning, backend metadata, and health checks through `netbird status`.
- The Archipelago terminal now includes `nano` on new disk installs and ISO builds, and self-update installs it on existing nodes if it is missing.
- Validation passed with catalog JSON checks, shell syntax checks, `npm run type-check`, `cargo fmt --all --check --manifest-path core/Cargo.toml`, and `cargo check -p archipelago --manifest-path core/Cargo.toml`.
## v1.7.68-alpha (2026-05-19)
- BTCPay Server now ships on the official `docker.io/btcpayserver/btcpayserver:2.3.9` image, fixing the plugin catalog crash caused by newer plugin dependency version metadata while preserving existing datadirs and Postgres databases.
- BTCPay release and first-boot health checks no longer depend on `curl` inside the container; they use a bash TCP probe that works with the official image out of the box.
- Host nginx now serves Nginx Proxy Manager HTTP-01 challenge files before the Archipelago SPA fallback and is marked as the default HTTP/HTTPS virtual host, so public proxy hosts can issue certificates without hijacking local API traffic.
- Nginx Proxy Manager first-boot, runtime repair, and container-doctor paths now pre-create the ACME webroot, keep bind mounts owned by the rootless Archipelago user, and sync issued public proxy hosts into host nginx vhosts.
- The Nginx Proxy Manager host-nginx sync now skips proxy hosts with missing certificate files and rolls back the generated nginx include if validation fails, preventing a bad certificate path from poisoning later nginx reloads.
- App session close buttons now return to the previous dashboard screen when possible and otherwise fall back to My Apps, avoiding the 404 page after closing an app launched from an invalid or stale history entry.
- System Update confirmation and mirror modals now teleport to the document body with a full-screen overlay, so they cover the whole app instead of only the right-hand dashboard panel.
- Mobile app launches stay inside Archipelago's app-session webview and hide desktop-only new-tab launch affordances, including apps such as Home Assistant that previously looked like they would leave the mobile shell.
- Live recovery on `100.70.96.88` upgraded only the `btcpay-server` container to `docker.io/btcpayserver/btcpayserver:2.3.9`, preserved the existing datadir and Postgres database, and confirmed the container is healthy after a pre-upgrade backup.
- Public validation confirmed `spay.tx1138.com`/`www` redirect to BTCPay login over HTTPS and `sapien.tx1138.com`/`www` serve the L484 page over HTTPS using the issued Let's Encrypt certificates.
## v1.7.67-alpha (2026-05-18)
- Home dashboard status cards now keep the last known good system, VPN, Bitcoin, and FIPS values while route changes or transient RPC failures are in flight, avoiding false "not configured" or "not running" flashes.
- Home, Web5 Monitoring, and the Monitoring page headline cards now share the same live system-stat snapshot for CPU, memory, disk, uptime, and load so the visible numbers agree across the UI.
- Settings What's New is filled through `v1.7.67-alpha`, including the missing historical `v1.7.44-alpha` through `v1.7.66-alpha` entries.
- Bitcoin/Knots/Core shell lifecycle specs now match the Rust app config memory policy: 8 GiB on normal hosts, 4 GiB on low-memory hosts, and pruned Knots uses a larger dbcache on hosts with enough RAM to improve IBD throughput.
- ElectrumX/electrs shell lifecycle specs now match the 4 GiB memory policy used by the Rust app config, reducing drift between first boot, reconcile, and app lifecycle paths.
- Live assessment of `100.70.96.88` identified the current IBD bottlenecks as CPU/thermal/I/O pressure rather than RAM exhaustion, with follow-up work planned for existing-node swap repair, kiosk Chromium CPU reduction, and reconcile failure cleanup.
## v1.7.66-alpha (2026-05-18)
- Nginx Proxy Manager stale-port repair now detects stopped or `Created` Podman records by inspecting `podman ps -a` port metadata, covering records where `podman port nginx-proxy-manager` returns no mapping until start.
- Live recovery on `100.70.96.88` removed only the stale Nginx Proxy Manager container record and recreated it with `8081:81`, `8084:80`, and `8444:443`, preserving `/var/lib/archipelago/nginx-proxy-manager` data.
- Validation confirmed Nginx Proxy Manager recovered as healthy and responds through direct admin port `8081`, host compatibility port `81`, and `/app/nginx-proxy-manager/`.
## v1.7.65-alpha (2026-05-18)
- Orchestrator-backed app starts now run the same pre-start repairs as the legacy Podman path, so Nginx Proxy Manager stale `81:81` container metadata is removed and recreated before the orchestrator tries to start it.
- Live diagnostics on `100.70.96.88` confirmed host nginx is healthy while Nginx Proxy Manager has no listeners on `8081`, `8084`, or `8444`, causing host nginx `502` responses for NPM proxy paths.
## v1.7.64-alpha (2026-05-18)
- Update apply rate limiting is relaxed for authenticated admins from 2 attempts per 10 minutes to 10 attempts per minute, preventing the System Update page from getting stuck behind `429 Too Many Requests` during legitimate OTA retry/troubleshooting flows.
- The corrected backend artifact rebuild protection from `v1.7.63-alpha` remains in place, so this release is built from a fresh Rust backend binary before publishing.
## v1.7.63-alpha (2026-05-18)
- Release automation now rebuilds the Rust backend after bumping the version and before hashing release artifacts, preventing OTA manifests from pointing at a stale backend binary.
- This corrected release carries the Nginx Proxy Manager stale-port repair in an updated backend binary, so nodes running `1.7.61-alpha` can actually receive and execute the fix.
- Validation confirmed the previously published `v1.7.62-alpha` backend artifact still contained `1.7.61-alpha`, explaining why nodes did not advance after applying that update.
## v1.7.62-alpha (2026-05-18)
- Nginx Proxy Manager start and restart now repair stale Podman containers that still publish the admin UI on host port `81`, which conflicts with host nginx on updated nodes.
- The repair recreates only the stale Nginx Proxy Manager container metadata while preserving `/var/lib/archipelago/nginx-proxy-manager` data and using the current `8081:81`, `8084:80`, and `8444:443` mappings.
- Runtime stale-listener cleanup for Nginx Proxy Manager is shared across start and restart paths so rootless port helper leftovers are still cleared before lifecycle retries.
- Validation passed with `cargo fmt --all --check --manifest-path core/Cargo.toml` and `cargo check -p archipelago --manifest-path core/Cargo.toml`.
## v1.7.61-alpha (2026-05-18)
- Multi-container stack installs now keep their app card in the `Installing` state for up to 20 minutes while dependency containers are being pulled and prepared.
- BTCPay Server installs no longer appear to vanish or fail after two minutes while Postgres and NBXplorer are still being created before the primary `btcpay-server` container exists.
- The stale-transition escape hatch remains short for start, stop, restart, update, and removal operations, so genuinely wedged lifecycle actions still recover quickly.
- Live validation on `100.70.96.88` confirmed BTCPay Server completed installation and responds on port `23000` with the expected HTTP redirect.
## v1.7.60-alpha (2026-05-18)
- Meshtastic serial detection now rejects malformed or incomplete handshakes instead of accepting unrelated serial devices as a fallback Meshtastic radio.
- Mesh radio auto-detection now skips known non-mesh serial devices such as Sierra Wireless LTE modems and Zooz/Z-Wave sticks, avoiding interference with production peripherals.
- Meshtastic config sync now sends `want_config_id` with the correct protobuf wire type, fixing radio-side `ignore malformed toradio` errors and allowing node-info/contact ingestion.
- The stable `/dev/mesh-radio` udev rule no longer claims every `ttyACM*` device; it only matches known mesh USB serial adapters and known USB CDC ACM radio vendors.
- Live validation on `100.70.96.88` confirmed Archipelago selects `/dev/ttyUSB0`, identifies the Meshtastic node, and refreshes 103 mesh contacts.
## v1.7.59-alpha (2026-05-17)
- Mobile app launching now keeps known container apps inside Archipelago's app-session flow instead of forcing desktop-only new-tab behavior on phones.
- App sessions on mobile now respect the status-bar safe area so foreground iframe content starts below the device chrome while the fullscreen backdrop remains edge-to-edge.
- Prepackaged website launch buttons now resolve their curated website URLs before website-container fallback logic, restoring launches for the L484 sites and adding the Arch Presentation bookmark.
- Meshtastic contact discovery now drains the radio config stream through completion and retries config sync when the contact cache is empty, so nearby nodes already known by the radio are more likely to appear in Archipelago.
- The Apps page now includes a compact sideload button and modal for installing trusted Docker images with optional title, description, and port mapping metadata.
- Sideloaded app title and description metadata now persist through the backend app-config file so refreshed package scans do not collapse custom apps back to generic IDs.
- Validation passed with `npm test -- appLauncher`, `npm run build`, `cargo check -p archipelago`, and `cargo fmt --all --check`.
## v1.7.58-alpha (2026-05-17)
- Mesh networking now supports Meshtastic radios over the Meshtastic serial API in addition to existing MeshCore Companion USB radios.
- The mesh listener now probes preferred and auto-detected serial paths for both MeshCore and Meshtastic firmware, preserving the existing reconnect loop so unplug/replug and firmware hot-swap behavior stays consistent.
- Meshtastic text packets are translated into the existing Archipelago mesh frame pipeline, so current RPC handlers, transport routing, message storage, typed-message decoding, and UI state continue to work without a separate frontend path.
- Meshtastic node information is surfaced as normal mesh contacts using stable synthetic public keys derived from Meshtastic node numbers, allowing peer refresh and message attribution to reuse existing MeshCore contact handling.
- Outbound Archipelago mesh messages can now be sent through Meshtastic as channel text packets using the same command path used by MeshCore channel broadcasts.
- Device status now reports the detected firmware family as `meshcore` or `meshtastic` from the shared listener abstraction.
- Radio udev rules now include USB CDC ACM serial devices (`ttyACM*`) alongside CP2102, CH340, and FTDI adapters so Meshtastic boards are more likely to appear through the stable `/dev/mesh-radio` symlink.
- Host nginx now serves `/assets/*` hashed frontend chunks as immutable static files with a hard 404 on misses instead of falling back to `index.html`, preventing strict MIME errors when a browser has a stale pre-update HTML shell.
- The SPA HTML shell and service-worker files now revalidate on every load, reducing stale frontend references after OTA updates.
- OTA runtime promotion now installs the bundled `nginx-archipelago.conf` into `/etc/nginx/sites-available/archipelago` and reloads nginx after a successful config test, so frontend cache/fallback fixes reach existing nodes without a manual deploy.
- Local validation passed with `cargo check -p archipelago`; live SSH testing against `100.70.96.88` was not completed because temporary public-key authentication was rejected on the target.
## v1.7.57-alpha (2026-05-17)
- Nginx Proxy Manager now avoids privileged rootless Podman host port `81`, preferring `8081:81` while host nginx keeps a compatibility proxy on `:81` for stale cached launch buttons.
- App installs now allocate ports by checking live host bind availability, falling back to a free high port when preferred ports are already occupied.
- Portainer-created launchable containers are separated into a `Websites` tab and launch through their discovered published host port instead of hard-coded app URLs.
- Internal BuildKit helper containers such as `buildx_buildkit_default` are hidden from the Apps UI.
- Portainer works out of the box on Debian 13/Podman installs by including `catatonit` and by preserving the Podman socket mount as a socket rather than creating it as a directory.
## v1.7.56-alpha (2026-05-15)
- Health notifications now clear when an app is no longer unhealthy, including stale alerts for removed containers such as Portainer.
- Fresh installs now include the full Wi-Fi userspace stack (`wpasupplicant`, `wireless-regdb`, `iw`, `rfkill`, `polkitd`, `pciutils`, and `usbutils`) so NetworkManager can scan and connect with Intel Wi-Fi cards out of the box.
- The installed system now grants the `archipelago` service user explicit NetworkManager PolicyKit access for web-triggered Wi-Fi scans and connection changes.
- Wi-Fi connect now replaces stale/partial NetworkManager profiles and creates an explicit WPA-PSK profile with the supplied password, avoiding no-secret retry failures after a failed attempt.
- Settings password changes now update the Linux/SSH password through non-interactive sudo, so the web password and SSH password stay in sync when the checkbox is enabled.
- Quadlet environment values with spaces or shell metacharacters are quoted consistently, preventing env drift recreate loops for apps like nostr-rs-relay and Grafana.
- Boot/bootstrap reconcile avoids restarting running Bitcoin containers while repairing RPC config, preserving IBD progress on active nodes.
- Exit code 137 is labeled as SIGKILL instead of assuming OOM, avoiding false OOM alerts for orchestrator-managed recreates.
- Container reconcile force-recreates Podman records stuck in `Stopping`, preserving bind-mounted app data while recovering wedged containers automatically.
- Container health reporting is honest for running containers: Archipelago surfaces Podman's actual health state instead of marking every running container healthy.
- Quadlet reconciliation restarts services when stale health gates, port bindings, network aliases, exec commands, or healthchecks drift from the current manifest.
- Bitcoin Knots sync performance improves on fresh installs and updates with 8Gi container memory, a 4Gi dbcache, and full CPU parallelism.
- ElectrumX initial indexing gets more headroom: CPU caps are removed, memory is raised to 4Gi, cache is raised to 3Gi, and oversized sends are allowed for heavier wallet/indexing workloads.
- Mempool/ElectrumX lifecycle qualification respects pruned/non-archival Bitcoin nodes instead of installing a half-running stack with unhealthy dependencies.
- LND wallet/RPC helpers are more tolerant of container-owned files and updated REST port metadata, improving LND lifecycle and wallet-connect flows.
- Marketplace/catalog metadata carries richer container config so remote lifecycle tests install apps using the same settings users get from the UI.
- The app screensaver no longer activates during media-heavy app sessions such as IndeeHub, Jellyfin, Immich, PhotoPrism, and File Browser; apps can also pause/resume it with media playback messages.
- A fresh `1.7.56-alpha` unbundled installer ISO is built from the same primary VPS2 release line for easy download and USB flashing.
## v1.7.55-alpha (2026-05-13)
- Container reconcile now force-recreates Podman records stuck in `Stopping`, preserving bind-mounted app data while recovering wedged containers automatically.
-`.198` is green after the container-layer hardening pass: focused and broad non-destructive lifecycle audits pass, raw Podman health/state sweep is clean, and direct app probes return healthy responses.
- Release-candidate artifacts are staged separately from live update publishing while Gitea artifact hosting is repaired.
## v1.7.54-alpha (2026-05-06)
- Existing installs now self-repair nginx backend proxy locations for `/bitcoin-status` and `/api/app-catalog`, including hosts where `sites-enabled/archipelago` is a copied active file instead of a symlink.
- LND UI is consistently served on `18083` across first boot, Tor config, companion Quadlet reconciliation, OTA runtime payloads, and ISO scripts; stale companion units/images are rewritten instead of only checking service active state.
- OTA frontend tarballs now carry a clean runtime payload with updated scripts, docker UI sources, and canonical nginx config, preventing startup promotion from reintroducing stale host assets.
- Release ISO builds now support the primary HTTP app registry when bundling core images, so unbundled media includes File Browser/Cloud support instead of requiring a post-install Marketplace download.
-`.116` was live-updated with the new backend and runtime scripts; focused non-destructive lifecycle audit passes for Bitcoin Knots, LND, BTCPay, Mempool, and Grafana.
## v1.7.53-alpha (2026-05-05)
- Bitcoin Knots/Core config generation no longer duplicates RPC bind and port settings between `bitcoin.conf` and container command args, fixing `Unable to bind all endpoints for RPC server` startup failures.
- Legacy Bitcoin container healthchecks no longer depend on `bitcoin-cli`, which is absent from current Knots images and can wedge Podman healthcheck runners.
- Update checks now prefer manifest OTA releases over stale git remotes unless `ARCHIPELAGO_GIT_UPDATES` is explicitly enabled, so installed nodes can see published releases from the VPS mirror.
## v1.7.52-alpha (2026-05-05)
- Tailscale now launches the local installed web UI on port `8240` and starts `tailscaled` before `tailscale web`, fixing unreachable installs after container creation.
- Grafana install/start/restart now repairs missing rootless host listeners on port `3000`, matching the existing SearXNG, Uptime Kuma, and Gitea recovery path.
- Debian 13/Trixie ISO and disk-install paths now force security updates from `trixie-security` during image/install creation so rebuilt release media includes patched base packages.
- Broad `.198` lifecycle audit passes with the current qualified app set; known absent blockers remain `electrumx`, `photoprism`, `dwn`, and `ollama`.
## v1.7.49-alpha (2026-04-30)
- Bitcoin Knots/Core UI now reports connection, reconnecting, syncing, and error states from a backend status bridge instead of showing a stale "Unable to connect" message while the node is warming up.
- ElectrumX UI now exposes indexed height, local Bitcoin height, known headers, status, and progress source so indexing/waiting states are readable during long initial sync.
- Added container doctor timer and smoke/lifecycle test coverage for Bitcoin Knots/Core, ElectrumX, Mempool, BTCPay/NBXplorer, and UI surface availability.
- Bitcoin Core and Bitcoin Knots are mutually exclusive variants, with a real Bitcoin Core manifest and corrected install conflict handling.
- IndeeHub now launches only on direct web UI port `7778`; the broken `/app/indeedhub/` path proxy was removed, and port `7777` remains the Nostr relay.
- BTCPay/NBXplorer Postgres environment formatting fixed so installs do not carry malformed connection strings.
## v1.7.48-alpha (2026-04-29)
- archipelago.service no longer fails to start with "Failed to set up mount namespacing: /run/containers: No such file or directory" on nodes where /run/containers wasn't pre-created. ExecStartPre now creates it. Existing nodes need a one-time `systemctl edit archipelago` to add the mkdir; ISO installs from this version forward have the fix baked in.
## v1.7.47-alpha (2026-04-29)
- Bitcoin Knots/Core sync is now significantly faster. The container now uses every available core for script verification (was capped at 2) and has 8GB of memory instead of 4GB so its 4GB UTXO cache has headroom for the mempool and peer connections. Existing nodes pick up the new limits on next install/update; freshly-installed nodes start at full speed.
- ElectrumX initial indexing is faster too. Its CPU cap is removed, container memory is 4GB, and its internal cache is now 3GB (default was 1.2GB).
## v1.7.46-alpha (2026-04-29)
- Health monitor no longer pages "Auto-restart failed" for orphaned containers. After a variant switch (bitcoin-core ↔ bitcoin-knots) the previous variant's container could survive uninstall and the health monitor would try restarting it forever. Now skipped silently with a debug log.
- Apps no longer disappear from My Apps when an install fails. The card stays visible with state=Stopped so the user can retry or uninstall, with the failure reason surfaced via the new install_progress.message field.
- "Downloading…" progress now actually advances during multi-image stack pulls. Was sticking at 20% until all pulls finished; now interpolates 20%→70% based on which image of N has landed.
- Pulled four docker.io images (bitcoin, gitea, nextcloud, valkey) into the lfg2025 registries on OVH and tx1138. Removes a docker.io dependency from first-boot installs.
- Resilience harness improvements: install-fail entries no longer vanish, install/uninstall/probe cells are timing-tolerant (60s retry on ui_probe and auth_probe), dep snapshots no longer leak companion containers into the dependent app's "new containers" set.
## v1.7.45-alpha (2026-04-29)
- Bitcoin RPC auth is durable. The dashboard reliably connects across container restart, image update, and reboot. Was failing on registry-pulled images that shipped a stale baked-in password.
- Multi-container apps show real install progress. IndeedHub (7), BTCPay (4), Mempool (3), Immich (3) — bar advances through Preparing → Pulling → Creating → Done instead of sitting at 0% until the very end.
- Apps no longer disappear from the dashboard mid-install. The container scanner now respects in-flight installs and updates instead of evicting an entry while its containers are still being created.
- IndeedHub installs cleanly on a fresh node. Five missing environment variables fixed; Nostr sign-in works on first install.
- Tailscale install no longer fails with "executable not found". Container command was a malformed shell string; now a proper command array.
- Removed three catalog entries that hung installs for ten minutes (dwn, endurain, ollama — no source images in our registries). Restored Nextcloud, sourced from docker.io.
- Bitcoin Core update path uses the correct image name (was pulling from a non-existent path).
- New ISO installs now allocate swap (sized to RAM, capped at 8GB, on the encrypted data partition). Without swap, container image builds and memory spikes were hitting OOM under load.
## v1.7.44-alpha (2026-04-28)
43de3b73 feat(orchestrator): complete container migration and release hardening
ce39430b feat(self-update): sync and rebuild UI containers on OTA
72dec5aa fix(lnd-ui): align container port across all specs
83aacdf2 chore(release): archive ISO build recipes, tarball-only releases
All notable changes to Archipelago will be documented in this file.
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
1.**Never build Rust on macOS** — deploy script handles cross-compilation via rsync + remote build
2.**Always deploy after changes** — `./scripts/deploy-to-target.sh --live`
3.**Frontend builds to `web/dist/neode-ui/`** — not `neode-ui/dist/`
4.**Container images**: `scripts/image-versions.sh` is the single source of truth. All scripts use `$*_IMAGE` variables, never hardcoded registry paths.
Default: **4848**. To change, edit the ports mapping in `docker-compose.demo.yml`:
```yaml
ports:
- "YOUR_PORT:80"
```
## Chat (Claude AI)
Set `ANTHROPIC_API_KEY` in the Portainer stack environment to enable real AI chat:
1. In the stack editor, add under **Environment variables**:
-`ANTHROPIC_API_KEY` = your Anthropic API key (starts with `sk-ant-api...`)
2. Redeploy the stack
Without this key, chat shows a "not configured" error. The key is passed to the `neode-backend` container which proxies requests to `api.anthropic.com`.
## Dev Mode
`VITE_DEV_MODE=existing` skips setup/onboarding and goes straight to login. For other flows:
**Archipelago** is a bootable personal server OS. Flash it to a USB drive, install on any x86_64 or ARM64 machine, and manage Bitcoin infrastructure, self-hosted apps, and decentralized identity through a glassmorphism web UI.
"description":"Bitcoin documentary streaming platform featuring God Bless Bitcoin and other educational content about Bitcoin, sovereignty, and decentralized technology. Sign in with your Nostr identity.",
"description":"Bot competition arena with 2-player arcade fighting mode. AI bots battle in trivia challenges while humans duke it out with controllers. Built for Bitcoiners.",
"tailscaled --tun=userspace-networking & for i in $(seq 1 30); do [ -S /var/run/tailscale/tailscaled.sock ] && break; sleep 1; done; tailscale web --listen 0.0.0.0:8240 & wait"
]
}
},
{
"id":"portainer",
"title":"Portainer",
"version":"2.19.4",
"description":"Container management web UI for the local Podman socket.",
"notes":"Installed as a two-container stack: netbird dashboard on 8087 and netbird-server control plane on 8086 plus UDP 3478. For production clients, publish a DNS name over HTTPS with gRPC/WebSocket routing."
This will build all apps that have Dockerfiles. Standard apps (bitcoin-core, lnd, etc.) will use their official images, while custom apps (router, did-wallet, web5-dwn) will be built from source.
This will build all apps that have Dockerfiles. Standard apps (bitcoin-core, lnd, etc.) will use their official images, while custom apps (router, did-wallet) will be built from source.
### Build Specific App
```bash
./build.sh router
./build.sh did-wallet
./build.sh web5-dwn
```
## Running Apps via Archipelago
@@ -64,7 +63,6 @@ In development mode, apps are accessible on offset ports:
- **Router**: http://localhost:18084
- **DID Wallet**: http://localhost:18083
- **Web5 DWN**: http://localhost:13000
- **Nostr RS Relay**: http://localhost:18081
- **Strfry**: http://localhost:18082
@@ -72,7 +70,7 @@ See [PORTS.md](./PORTS.md) for complete port mapping.
## Development Workflow
### For Custom Apps (router, did-wallet, web5-dwn)
### For Custom Apps (router, did-wallet)
1. **Make changes** to source code in `apps/<app-id>/src/`
description:Bot competition arena with 2-player arcade fighting mode. AI bots battle in trivia challenges while humans duke it out with controllers. Built for Bitcoiners.
category:community
container:
image:146.59.87.168:3000/lfg2025/botfights:1.1.0
pull_policy:always
dependencies:
- storage:500Mi
resources:
cpu_limit:2
memory_limit:512Mi
disk_limit:500Mi
security:
capabilities:[]
readonly_root:true
no_new_privileges:true
user:1001
seccomp_profile:default
network_policy:bridge
apparmor_profile:default
ports:
- host:9100
container:9100
protocol:tcp # Web UI + API
volumes:
- type:bind
source:botfights-data
target:/app/server/data
- type:tmpfs
target:/tmp
options:[rw,noexec,nosuid,size=64m]
environment:
- NODE_ENV=production
health_check:
type:http
endpoint:http://localhost:9100
path:/api/health
interval:30s
timeout:10s
retries:3
start_period:30s
interfaces:
main:
name:Web UI
description:Bot arena and arcade fighter with controller support
description:Bitcoin documentary streaming platform featuring God Bless Bitcoin and other educational content about Bitcoin, sovereignty, and decentralized technology. Sign in with your Nostr identity.
category:media
category:community
container:
image:git.tx1138.com/lfg2025/indeedhub:latest
image:146.59.87.168:3000/lfg2025/indeedhub:1.0.0
pull_policy:always # Pull from registry; falls back to local build
network:indeedhub-net
dependencies:
- storage:1Gi
@@ -27,9 +28,9 @@ app:
apparmor_profile:default
ports:
- host:7777
container:3000
protocol:tcp # Web UI (Next.js)
- host:7778
container:7777
protocol:tcp # Web UI. Port 7777 on the host is reserved for Nostr relay.
volumes:
- type:tmpfs
@@ -38,6 +39,12 @@ app:
- type:tmpfs
target:/app/.next/cache
options:[rw,noexec,nosuid,size=128m]
- type:tmpfs
target:/run
options:[rw,nosuid,nodev,size=16m]
- type:tmpfs
target:/var/cache/nginx
options:[rw,nosuid,nodev,size=32m]
environment:
- NODE_ENV=production
@@ -57,14 +64,15 @@ app:
name:Web UI
description:Stream Bitcoin documentaries with Nostr identity
"restart_hint": "Archipelago restarts the active Bitcoin backend after generating txrelay credentials so bitcoind loads the restricted rpcauth whitelist.",
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.