Commit Graph

23 Commits

Author SHA1 Message Date
admin 636c51e542 R-193: the recovery screen — unlocking, and only unlocking (v0.200.0)
A customer whose machine was rebuilt had everything needed to get their data
back and no way to find out: the only route was a command line. This is the
screen that closes that.

IT UNLOCKS, AND ONLY UNLOCKS (operator ruling). It explains, takes the recovery
code, opens the repository and shows what is in there — apps, dates, sizes. It
restores nothing: restore is already per-app and lives in the backups area, and
a screen that unlocks and then offers to overwrite is two decisions wearing one
button.

ONE CORE, TWO CALLERS. RecoverInstallCore is split out of RecoverAndInstall; the
CLI wrapper keeps its exit codes and printed lines byte-identical, and the
handler drives the same function. Two implementations of the one operation that
can permanently lose a customer's data would drift, and only one would be
tested. Asserted from source on both sides by AST.

THREE WAYS OUT, none a dismiss button: recover; 'most nem' (the full page stops
interrupting, the backups-area entry point stays PERMANENTLY, bound to the offer
and never to the postpone flag); and 'I do not want the old data' — confirmed
TWICE and reaching the SHIPPED move-aside, which sets aside and never deletes.

THE CODE IS HANDLED NO MORE LOOSELY THAN ON THE COMMAND LINE: POST body only,
never logged, never persisted, never echoed, cleared on every path, no-store,
autocomplete off. No lockout — the code is a ten-word phrase, and locking a
customer out of their own data for a typo is worse than anything it prevents.

TWO DEFECTS THE TESTS CAUGHT, both fixed: an UNCLAIMED (legacy-open) box would
have been shown the page, because RequireAuth passes such a box through; and the
inventory nil-dereferenced when no off-site target was configured, which is
exactly the pristine rebuilt shape.
2026-08-05 12:45:48 +02:00
admin 3f048e042b R-101 + F-DIAG: the restore dialog names the last SUCCESSFUL copy (v0.182.0)
Tier2LastRun is the attempt clock and was rendered as 'Legutóbbi másolat' in the restore
confirm dialog. New LastSuccess + SuccessTracked anchor; tier2Update makes the three
rebuild sites safe by construction. F-DIAG: six distinct causes, target-aware redaction.
2026-07-28 16:36:47 +02:00
admin 2487681396 style: gofmt normalization — no logic changes
gofmt -w across the controller tree (46 files) so gofmt -l is empty — disarms the
formatting landmine where a targeted edit + accidental gofmt -w swept ~46 unrelated
files. Pure formatting: whitespace + gofmt's optional-semicolon removal in reflowed
inline closures. One doc comment reworded ('' -> 'the empty string') to avoid gofmt's
Go-1.19 doc-comment typographic substitition ('' -> curly quote) muddying its meaning.
No build/vet/test behavior change.
2026-07-25 07:37:02 +02:00
admin 15206314ab v0.165.0: Indítópult megosztása — guest launcher via capability URL (+ optional password, QR)
Mint a 160-bit capability URL (/s/<token>) serving a standalone read-only guest
launcher: same tiles, opens apps in new tabs, no account, no admin session.
Information only, zero control — every privilege stays behind each app's own auth.

- /s/ pre-auth pass-through (after the claim gate) + session-CSRF exemption; guest
  password POST carries its own pre-auth HMAC CSRF.
- Constant-time token match; empty stored token = disabled = byte-identical mux 404.
- Optional per-share password: separate bcrypt hash + own attempt map; signed cookie
  = HMAC(token|passwordHash) keyed with web.session_secret, so rotate/change invalidates.
- Guest labels ride the v0.164.0 ruling; never expose internal state vocabulary.
- Token redacted in logs (/s/<redacted>); never in CHANGELOG/REPORT/CONTEXT.
- Admin modal: copy-link, QR (go-qrcode), set/clear password, rotate, disable.
- Tests: Groups A-G (14) + 3 red-proofs verified red.
2026-07-24 12:08:43 +02:00
admin 987e915bf2 Indítópult launcher page + universal app placeholder icon (v0.163.0)
New /launcher page: a grid of large tappable tiles, one per openable deployed
app (subdomain presence is the single openability criterion, shared with the
Megnyitás button via the extracted Server.subdomainMap helper). Colored tiles
(deterministic slug color or .felhom.yml brand_color), white glyph/monogram
fallback, target=_blank links for operational apps, greyed unclickable tiles for
stopped ones. First sidebar item; / stays the Vezérlőpult.

Universal app placeholder: new AppPlaceholderSVG served at
/static/app-placeholder.svg, now the default FallbackIcon on app_list_row so a
logo-less app shows a placeholder instead of visibility:hidden. Brand mark is
never an app placeholder.

New Metadata.BrandColor; new funcmap tileColor/initial. 10 new test functions +
4 red-proofs. No agent coupling; MinAgent unchanged.
2026-07-24 09:16:28 +02:00
admin 5fdd2039fd catalog: the lifecycle implementation itself (fixes the previous commit)
The previous commit landed only the new test/badge files: a 'git stash' used
to compare REUSE.md ref-check output silently dropped the staged index, so
every modification to an existing file was left behind and that commit does
not build. This adds the metadata field, the predicates, the fail-closed
deploy gate, the catalog filter, the funcmap entries, the template edits and
the docs that those tests exercise.
2026-07-21 16:20:24 +02:00
admin 285dd1032f feat(v0.156.0): dead-primary alerting (R-51) + boot desired-state reconciliation (R-52)
R-51: aggregateState's mixed branch returned StateRunning ("partial"), so a stack whose
MAIN container was dead behind live helpers alerted on nothing — immich-server sat Exited
for 18 h, 100 % unreachable, no banner and no app_start_failed (audit F4). New
StateDegraded: a DOWN member whose docker restart policy is always/unless-stopped is a
fault (degraded, a down state); no/on-failure is a finished one-shot and stays benign; an
unreadable policy fails CLOSED. The unhealthy/restarting/paused/unknown exclusions are
byte-identical — folding unhealthy into down is the flapping fix-3 avoided.

R-52: new internal/bootrecon — one bounded start-once sweep at startup (2 attempts, 30 s
apart) for apps an interrupted boot left behind, inside the 90 s boot grace so a success
is silent and a failure still alerts. A zero-container stack is NEVER touched: the UI's
Stop is compose down, so a deliberate stop survives a reboot.

Both features carry a production-path wiring test (the v0.154.0 / v0.91.0 inert-seam
class). The main() assertion is an AST walk, not strings.Contains — the substring version
passed its own red-proof, because a commented-out call still contains the string.

Red-proofs run and restored: mix branch reverted -> "running" on the immich fixture;
boot hook commented out -> wiring test fails; zero-container gate dropped -> the
user-stopped app gets started.

NOTE: controller/cmd/controller/ is matched by .gitignore's `controller` entry, so new
files there need `git add -f` (and ripgrep silently skips main.go without --no-ignore).
2026-07-21 12:27:33 +02:00
admin 4f06149abc Part B: infra stacks carry identity — curated Hungarian metadata for cloudflared/traefik/filebrowser (v0.126.0)
- inframeta.go: static display-only map (name/description/Linked) + infraMeta template func;
  filebrowser is the ONLY Linked stack (files.<domain> Megnyitás on the dashboard row;
  the stacks card already links via Subdomains)
- generic infra icon: embedded /static/infra-logo.svg (Lucide-style server, monochrome)
  wired through the app-row FallbackIcon slot + the stacks-card onerror chain
- dashboard rows + app cards: infra name + description + existing Védett chip
- guarded WRONG outcome: no customer link for cloudflared/traefik — render tests count
  exactly ONE https:// link; red-proven (Linked:true on cloudflared → test FAILS)
2026-07-13 11:31:38 +02:00
admin 3eacb6c326 Part A: shared app_list_row partial — ONE row grammar on four surfaces (v0.126.0)
- templates/app_row.html: app_list_row/app_list_row_end (layout_start/_end idiom) — icon+name
  (+optional secondary) left, caller action block right; compact 44px row; logoURL→PNG→
  optional FallbackIcon→hidden onerror chain
- applied: Távoli mentés toggle list, Visszaállítás restore-to-verify + .fab lists, dashboard
  Telepített alkalmazások (state edge + data-href preserved); Alkalmazások collapsed headers
  ALIGNED (icon+name left, status dot moved right before chevron; expander untouched)
- funcmap: dict + appHref; OffboxAppRow/AppBackupRow gain Slug
- style.css: .app-row family; stack-card/stack-info/stack-logo/stack-name/stack-desc/
  stack-actions rules retired (dashboard rows now shared); .app-backup-row-header matched
  to the shared row height
- scripts/app_row_dedup_gate.py: row markup single-sourced (red-proven: pasted old
  storage-path-item block → exit 1); render tests per surface (app_row_test.go)
- NO behavior change: toggle/download/expander actions byte-identical
2026-07-13 11:26:48 +02:00
admin 4906524c11 fix(backups): off-box LastRun is an RFC3339 string — 500'd the backups page
Found during the D0 click-through: OffboxTarget.LastRun persists as an
RFC3339 string (settings.json), but backups.html passed it to timeAgo
(expects time.Time), so GET /backups returned 500 on any node where an
off-box backup had ever run. Pre-existing since v0.93.0 (NAS Part B),
unrelated to the re-skin. New timeAgoStr funcmap helper parses the
string (falls back to the raw value); also v2-token fix for the offbox
error hint (var(--danger) fallback hex) and the dead stat-running/
stat-stopped classes on the offbox status card.
2026-07-02 14:47:22 +02:00
admin 5dc277f2b1 D0 Part 2: design system v2 — tokens, meter/tag/metarow, funcmap remap, dashboard + stacks
- style.css: navy token palette (--bg-0/1/2, --line, --text-1/2/3, --blue,
  --warn, --crit), single 2px radius, all box-shadows and the bg grid
  overlay removed, fonts via --font-ui/--font-data.
- Components: .meter (3px hairline track, blue nominal fill, neutral
  70/85 ticks, warn/crit flag), .tag (square 2px state chip + dot, pulse
  on progress, reduced-motion respected), .metarow (icon + text, no
  container), .panel/.list/.section-h primitives, boxless .stats with
  hairline dividers, buttons 2px (danger = crit outline).
- funcmap: stateColor -> run/progress/warn/neutral/off (stopped is
  neutral, NOT red — operator-approved exception-color change),
  usageColor/tempColor -> nominal/warn/crit (thresholds unchanged);
  stateLabel Hungarian copy untouched (guarded by test).
- layout.html: sprite nav icons (layout-grid/cloud/shield/cpu/wrench/
  settings), alert banner emoji -> triangle-alert/info icons.
- dashboard.html: meters with disk warn/crit flags (Fogyóban a hely /
  Kritikusan kevés hely), boxless stats (Leállítva 0 muted, >0 amber),
  single-panel stack list with 2px state edges, tags instead of badges,
  icon action buttons.
- stacks.html: state tag + metarow rows; catalog keeps its grid.
- setup minimalCSS retokened to v2 (drops GitHub-dark hexes).
- Tests: §8 truth tables for stateColor/usageColor/tempColor +
  stateLabel byte-identity guard (red-proven vs pre-change funcmap:
  stopped->red and 0->green failed as required).
2026-07-02 14:22:27 +02:00
admin 803ce50578 F5 (dashboard): surface 'route unpublished' for unhealthy/restarting deployed apps
Traefik only publishes a route to a healthy container, so an unhealthy deployed app
returns 404 at its URL though the container runs — previously shown only as 'Nem
egészséges' with no hint the URL is dead. New routeUnpublished() funcmap helper +
a distinct indicator on the dashboard and stacks cards (gated on .Deployed). Tests:
routeUnpublished across all states, real templateFS parses with the funcmap, and the
card guard renders the indicator only for deployed+unhealthy.
2026-06-14 09:56:30 +02:00
admin 0a85b5cd69 fix: add json template function for post-deploy data embedding
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-23 15:29:31 +01:00
admin 563cf07ec8 feat(deploy): async compose-up for instant UI feedback (v0.28.2)
Deploy API now returns immediately after validation + config save.
docker compose up -d runs in a background goroutine so the UI shows
progress during image pulls instead of blocking for 30-60s.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-23 12:08:08 +01:00
admin 6713df2186 v0.15.5: Disaster recovery — Hub-based infra backup, auto-mount, restore UI
Complete DR implementation (TASK2.md Phases 1-4):
- Hub infra-backup push/pull endpoints (controller.yaml, disk layout, stacks)
- Fresh-deployment detection pulls config from Hub, auto-mounts drives by UUID
- Full-page restore UI with drive status, app table, sequential restore
- docker-setup.sh shows DR instructions when customer_id is configured

New files: disk_layout.go, restore_scan.go, restore_app_linux.go,
restore_drives_linux.go, infra_backup.go, infra_pull.go,
handler_restore.go, restore.html

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-19 13:16:46 +01:00
admin 90826dec7a v0.13.0: UI polish fixes (8 improvements)
- Fix 1: Dashboard backup card border (verified already correct)
- Fix 2: Show auto-generated env values on deploy page with copy/reveal
- Fix 3: Temperature value pill for better visibility on dashboard
- Fix 4: Rework dashboard backup section (remove manual trigger, add Tier 2 summary)
- Fix 5: Scope HDD warning banner to dashboard and monitoring pages only
- Fix 6: Move Tárhely section up in monitoring page
- Fix 7: Snapshot table clarity (HOZZÁADOTT header, n/a instead of -)
- Fix 8: Restructure Tároló section into tiered storage view

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-02-18 13:30:21 +01:00
admin d160c6c06d v0.12.4 — 15 bug fixes (CRITICAL/HIGH/MEDIUM)
CRITICAL:
- C1: SetAppBackupBulk data loss + nil map panic (settings.go)
- C2: UpdateStackConfig nil Env map panic (deploy.go)
- C3: ValidateDump missing scanner.Err() check (dbdump.go)

HIGH:
- H1: nextDailyRun DST bug — use time.Date(day+1) not Add(24h)
- H2: Cache Europe/Budapest timezone with sync.Once in scheduler
- H3: settings.save() leaks .tmp file on WriteFile failure
- H4: SetNotificationPrefs nil pointer panic
- H5: appDirSize + getDirSizeBytes ignore Sscanf return value
- H6: getDirSizeBytes has no timeout — add 30s context
- H7: dbdump.go tmpFile not using defer Close
- H8: UpdateCrossDriveStatus misleading comment

MEDIUM:
- M1: Replace custom containsBytes with strings.Contains
- M2: scheduler.Every() validates interval > 0
- M3: executeJob panic recovery now sets LastRun
- M4: logPostStartStatus copies env slice before goroutine
- M5: Cache timezone in web package via getTimezone() sync.Once

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-02-18 07:50:02 +01:00
admin 29be81f1be fix: require_arg for --hdd-path, explicit event in stackAction, title separator, nextPruneLabel Sunday edge case
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-16 16:28:41 +01:00
admin ded0cbb842 v0.6.1: Code review bugfixes — 7 correctness/safety/quality fixes
- Fix http.NotFound(w, nil) → pass actual request in handlers
- Fix dashboard running/stopped counts to match displayed stacks
- Fix Secure cookie blocking HTTP login (dynamic based on request)
- Remove misleading subtle.ConstantTimeCompare in session check
- Fix cleanupSessions goroutine leak (proper ticker + done channel)
- Add http.MaxBytesReader (1MB) to API POST endpoints
- Cache time.LoadLocation("Europe/Budapest") in template funcmap

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-16 14:40:13 +01:00
admin 37ff296a0d v0.4.5: Add dedicated Backup page (Biztonsági mentés)
New /backups page with full backup system visibility:
- Status overview cards (local/remote backup, DB count, repo size)
- Schedule section with next-run times and retention policy
- Database table with type, size, validation (table count), status
- Snapshot history table with per-snapshot stats
- Repository info card with paths, integrity status, remote placeholder
- "Mentés most" button with auto-refresh polling
- Empty state when backup not configured

Backend: SnapshotRecord history (ring buffer), DumpValidation,
ListDumpFiles, ListSnapshots, GetFullStatus, restic check tracking.
Server accepts scheduler for next-run time calculation.

Sidebar nav updated with 3rd item, dashboard backup card title clickable.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-16 07:43:24 +01:00
admin d51e67f199 v0.4.1: app filtering + clickable dashboard stat cards
Add filter bar (Mind/Futó/Leállítva/Telepíthető) to Alkalmazások page with
URL-based filter persistence. Dashboard stat cards are now clickable links
that navigate to the filtered stacks view.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-15 12:19:26 +01:00
admin d32d9fb44b v0.4.0: monitoring & backup — scheduler, CPU/temp metrics, healthchecks, restic backups
Phase 2 (Monitoring & Health):
- Central job scheduler replacing ad-hoc goroutines (internal/scheduler)
- CPU usage collector via /proc/stat background sampling (internal/system/cpu_linux.go)
- Temperature reading from /sys/class/thermal + /host/sys (Docker mount)
- Load average from /proc/loadavg
- Healthchecks.io-compatible HTTP pinger (internal/monitor/pinger.go)
- System health checks: disk, memory, CPU, temp, Docker, protected containers (internal/monitor/healthcheck.go)

Phase 3 (Backups):
- Database auto-discovery via docker ps + docker inspect (internal/backup/dbdump.go)
- Database dumping via docker exec (pg_dump / mariadb-dump) with atomic writes
- Restic backup integration with auto-password generation (internal/backup/restic.go)
- Backup orchestrator: DB dumps + restic snapshots + weekly prune (internal/backup/backup.go)
- Manual backup trigger via dashboard button and POST /api/backup/run

Dashboard UI:
- CPU usage bar with load average display
- Temperature with colored indicator dot
- Backup status card with last run time, DB count, repo stats
- "Mentés most" button for manual backup trigger

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-15 11:17:10 +01:00
admin 57cacd9233 v0.3.0: structural refactor — go:embed templates, server split, domain rename
- Migrate all 7 HTML templates + CSS from Go string constants to individual
  go:embed files in internal/web/templates/ (templates.go: 2150→35 lines)
- Split server.go into auth.go, handlers.go, funcmap.go (server.go: 540→120 lines)
- Rename controller subdomain from dashboard.* to felhom.* in Traefik labels
- Update documentation (CLAUDE.md, README.md, CONTEXT.md)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-15 10:38:06 +01:00