- scripts/build-iso-release.sh: single gated command from signed release
to tested ISO (preflight version/signature parity, release gate harness,
strict catalog drift, full Rust suite, artifact version checks, build,
mount smoke, best-effort QEMU boot)
- scripts/iso-smoke-test.sh: standalone mount-level ISO verification incl.
stale-binary version assertion inside the payload
- .gitea/workflows/build-iso.yml: resurrected ISO CI as dispatch-only job
calling the gated orchestrator (old one was stranded in _archived/)
- tests/release/run.sh: catalog drift now runs --strict (was silently
always-pass)
- test-iso-qemu.sh: headless mode used -append without -kernel, which
QEMU rejects; use -display none so -serial file: capture works
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
docker login and buildkit both refused the HTTP registry. Host daemon on vps2
now lists it under insecure-registries (SIGHUP reload, no downtime); the
workflow configures buildkit with http=true for the registry host.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Companion QR overlay: demo builds encode the demo's own origin
(/packages/archipelago-companion.apk ships in the web image) instead of
the vps2 raw URL.
- Dockerfile.web/.backend: build-time scrub replaces every remaining
occurrence of the release-server address with registry.demo.internal and
fails the build if any survives.
- Mock backend update-mirror list, sideload help text, and changelog prose
no longer name the server address.
- Copy demo-images workflow into .gitea/workflows/ — Gitea ignores
.github/workflows when .gitea/workflows exists, so the CI never ran.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
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
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>
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>
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>
- 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>
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>
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>
- 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>
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>
- 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>
- 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>
- 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>
- 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>
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>
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>
The act_runner on .228 cannot git-fetch from git.tx1138.com via the
actions/checkout action (auth/network issue). Without continue-on-error
the build dies before the ~/archy rsync fallback can run. Restore it
so the fallback works. The red cross on checkout is cosmetic — the
fallback step provides the correct code.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The continue-on-error flag causes the checkout step to always show a
red cross in Gitea UI even on success. Removed it since the rsync
fallback is now conditional and ~/archy is up to date. Increased
timeout from 3 to 5 minutes for slow LAN fetches.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The rsync step was unconditionally overwriting the git checkout with
~/archy (which had diverged commit history), causing every CI build to
use wrong code. Now only falls back to rsync if checkout didn't produce
a valid workspace. Also removed --delete to prevent destroying checkout
files, and updated verification checks.
Root cause of CI build #373 using stale code.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Container stability:
- Merge scan results instead of full replacement (prevents UI flapping)
- Absence threshold: 3 consecutive missed scans before removing from state
- container-list RPC uses cached scanner state for consistency
- Increased Podman API timeout 30s → 60s (scanner + health monitor)
- Keep crashed containers visible as "exited" instead of podman rm -f
- Resolve host-gateway IP via ip route (podman 4.3.x compatibility)
ISO build fixes:
- AIUI web app inclusion: searches 5 paths + CI step to copy from build server
- Claude API proxy: systemctl enable with symlink fallback
- AIUI nginx: try_files =404 (was /aiui/index.html redirect loop)
- Build version set to 1.3.0
Container fixes:
- lnd-ui: nginx listens on 8080 (was 80, Permission denied in rootless)
- first-boot: image-versions.sh sourced from correct path with validation
- first-boot: host-gateway resolved to actual gateway IP
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The backend's post-install hooks create archy-bitcoin-ui, archy-lnd-ui,
archy-electrs-ui containers but with only NET_BIND_SERVICE cap. Nginx
inside these containers crashes on chown in rootless podman.
Added --user=0:0, CHOWN, DAC_OVERRIDE, SETUID, SETGID caps to match
the first-boot-containers.sh pattern. Also fixed manifest publish
Python error (git log fails in rsync'd workspace with no .git).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
actions/checkout fetches from Gitea via WAN which is unreliable (times out
on large repos). Added fast LAN fallback that syncs from ~/archy which is
kept current via rsync from dev machine. Includes verification step to
confirm changes are present before building.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- update.rs: extract frontend .tar.gz archives during apply (was TODO/skip)
- update.rs: back up current frontend before extraction, set binary perms
- server.rs: periodic scan reads update_state.json, sets status_info.updated
flag and broadcasts via WebSocket so frontend gets notified automatically
- build-iso-dev.yml: publish binary + frontend archive + manifest.json with
SHA256 hashes to /Builds/releases/v{version}/ after each build
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The workflow was workflow_dispatch ONLY — pushes never triggered builds.
Every ISO was built from whatever commit was current when someone
manually triggered the workflow from Gitea UI.
Changes:
- Add on.push.branches: [main] trigger
- Set clean: true on checkout to prevent stale cached code
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Critical:
- fix: container installs fail with "statfs: no such file or directory"
Root cause: NoNewPrivileges=yes in systemd blocks sudo inside backend.
Fix: use std::fs::create_dir_all + podman unshare chown (no sudo needed)
- fix: Tor services.json never written — \$ARCHY_TOR_DIR escaping bug
- fix: kiosk white screen — increase health wait to 60s, add --disable-gpu
Improvements:
- feat: LUKS encryption badge in Server disk stats (backend detects dm-crypt)
- fix: GRUB theme text scaling on 4:3 monitors — explicit fonts, wider menu
- fix: suppress default Debian MOTD (custom profile.d welcome is enough)
- fix: install error messages now show "Failed to pull/start" instead of
generic "Operation failed" (middleware.rs allowlist expanded)
- fix: container-tests CI — source cargo env before running tests
- docs: interactive container architecture diagram (HTML)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- build-iso-dev.yml now triggers on both main and dev-iso
- build-iso.yml (full Debian) is workflow_dispatch only
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Version per build:
- Health endpoint returns "1.2.0-alpha-{git_hash}" using GIT_HASH env
- CI passes git hash to cargo build
FileBrowser auto-login:
- filebrowser-client.ts: include CSRF token + credentials:include
- First-boot: generate random password, store at secrets/filebrowser/
- Set FileBrowser admin password to match after container creation
Nostr relay:
- Use docker.io/scsibug/nostr-rs-relay:0.9.0 (not in our registry)
UID mappings:
- Added electrumx (UID 1000), mysql-mempool, archy-btcpay-db, nextcloud-db
522 tests pass, Rust compiles clean.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add orchestration_tests.rs + mock_podman.rs (container unit tests)
- Add container-tests.yml CI workflow
- Add dev-container-test.sh for local testing
- MASTER_PLAN.md: add TASK-49 (P0) with 6-phase plan
- Login.vue: minor fixes from user testing
- AppCard.vue: enter key handler fix
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The clean:false setting causes checkout to fail when previous runs
leave corrupted workspaces. Default clean behavior ensures fresh
checkout each run.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The orchestration_tests integration test file is not yet committed,
causing CI to fail with "no test target named orchestration_tests".
Gracefully skip if not present.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
CI now runs a headless QEMU boot test after the smoke test:
- Boots ISO with -nographic, captures serial output
- Watches for "Press Enter to start installation" (pass)
- Detects kernel panic or initramfs shell (fail)
- 120 second timeout, runs as continue-on-error
Also: updated iso-debug reference with embedded vs appended EFI
findings from real hardware testing.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Boot fix:
- Ship proven Debian Live MBR (4552) as branding/isohdpfx.bin — the
ISOLINUX package MBR (33ed) doesn't boot on all hardware. This was
the root cause of "machine doesn't pick up the USB".
Branding:
- Custom GRUB background: pixel-art floating island (1024x574)
- Archipelago pixel-art logo for Plymouth boot splash
- GRUB theme: dark background, orange selected item, no broken font refs
- Plymouth theme: script-based with progress bar, LUKS prompt support
- Plymouth + splash added to target rootfs packages
- GRUB theme installed on both installer ISO and target system
- Serial console (ttyS0) added to kernel params for QEMU debugging
CI improvements:
- Smoke test step: mounts ISO, verifies all critical files, checks
initrd has live-boot, confirms boot=live in grub.cfg. Fails build
before copying to Builds if any check fails.
Dev workflow:
- dev-branding.sh: extract ISO, swap branding, repackage, boot in QEMU
(~10 seconds vs 20 min full rebuild)
- generate-grub-background.py: procedural cyberpunk background generator
- generate-plymouth-logo.py: procedural logo generator
- Improved test-iso-qemu.sh: --bios/--nographic flags, serial logging
Build:
- Simplified live-boot install (clean chroot, no complex fallbacks)
- Static branding images preferred, generators as fallback
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Major ISO build overhaul on dev-iso branch:
- Replace ~800MB Debian Live download with debootstrap --variant=minbase
(~150MB installer squashfs built from scratch)
- Custom initramfs with archipelago-mount hook for boot media detection
- Systemd service auto-starts installer (replaces profile.d hack)
- GRUB + ISOLINUX configs written from scratch (no Debian Live dependency)
- EFI boot image built with grub-mkimage (no more MBR extraction)
- Archipelago GRUB theme: dark background, Bitcoin orange accents
- Theme installed on both installer ISO and target system
- Rootfs optimizations: --no-install-recommends, strip docs/man/locales,
remove firmware-misc-nonfree/wget/htop, add explicit font deps
- Separate CI workflow (build-iso-dev.yml) for dev-iso branch
- Includes pre-existing fixes from main (build-iso.yml, middleware, Login)
Target: sub-2GB unbundled ISO (down from 3.9GB)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Add identity.create + server.echo to UNAUTHENTICATED_METHODS
- Clear web/dist before frontend build to prevent stale artifacts
- Add autocomplete attrs to login inputs
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The Build report step was failing the entire job because `du -h` and
`tar tf` on root-owned rootfs.tar returned permission denied. Added
sudo and continue-on-error: true so the report never fails the build.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- CI: configure root podman with insecure registry so FileBrowser
image can be pulled during ISO build
- CI: chmod u+rwX on workspace and act cache to fix cleanup failure
- ISO: auto-login on tty1 (no password prompt on console)
- Frontend: add console.log debug output for onboarding routing,
health checks, and 401 redirects to diagnose session issues
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The checkout action post-cleanup fails on root-owned files in the
workspace, marking the build as failed even though the ISO was built.
Chown the entire act cache dir so cleanup succeeds.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
