Root cause: backend filtered switches by status='online', so a mid-poll-cycle
device (briefly 'unknown') caused the endpoint to return [] — and the frontend
rendered nothing when the array was empty.
- Backend: query ALL switch devices (not just online ones); unreachable devices
already fall through to the Promise.allSettled error path and render with an
error badge, so no data is lost
- Frontend: replace silent `length > 0 &&` guard with a three-way ternary that
shows a "No managed switches found" message when the array is empty, so the
user always sees feedback instead of a blank section
- Frontend: add staleTime: 60_000 to the SNMP query to prevent the hair-trigger
background refetch that raced with the poller on every navigation
Backend: GET /api/system/version-check fetches frontend/package.json from
the main branch on GitHub, caches result in app_settings for 24 hours,
and compares semver against the running version. Fails silently (no banner)
for air-gapped or firewalled deployments.
Frontend: dismissible blue update banner in SecurityBanners (rendered
between TopBar and main content) — only visible when a newer version is
detected. Shows target version and docker compose update command.
Backend: accept target_device_id — connect to target, enable btest
server, run test from source using target credentials, disable server
in finally (even on failure). Manual IP mode preserved as fallback.
Frontend: device picker dropdown (all managed devices except self) as
default mode; toggle to Manual IP for non-managed targets with a warning
that the server must be pre-enabled. Results show target IP in labels.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Frontend: blob error responses (from responseType:'blob' calls) are now
decoded back to text/JSON so the actual backend error message surfaces
in the UI instead of generic "Request failed with status code 500"
- Backend: step-by-step console.log through capture flow so the exact
failure point is visible in server logs
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Packet capture: RouterOS sniffer + SFTP download to PCAP for Wireshark analysis
- Bandwidth test: RouterOS /tool/bandwidth-test with TX/RX Mbps result display
- Both exposed in the Tools tab on device detail pages
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- aggregateRules/aggregateActions: deduplicate coverage per device using a per-device seenKeys set,
preventing a device with duplicate topic|action rules from inflating the coverage count above total
- executePush (add path): skip devices that already have an identical rule/action before pushing to
all devices, preventing duplicate RouterOS logging entries from being created in the first place
- TopBar: remove hardcoded decorative 'live' indicator that had no backend connection logic
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add 'Copy VLANs from Another Switch' — a 3-step modal available on the
VLANs tab of any switch device. Step 1 selects a source switch and which
VLAN IDs to copy (VLAN 1 deselected by default with a management risk
warning). Step 2 lets the user assign interfaces on the target device
per VLAN: clicking a port chip cycles it through None → Tagged → Untagged
→ None, with a live summary and bridge picker per VLAN. Conflict detection
identifies VLANs already present on the target and offers per-VLAN
Skip/Overwrite with the port grid disabling itself when Skip is active.
Step 3 shows the full plan before a single apply call executes all
operations over one device connection and reports per-VLAN results.
Also fix CPU load display on the Devices page for hardware switches that
report 0% CPU via ASIC offloading — the previous filter (v > 0) silently
dropped all valid zero readings; now filters on field presence instead.
- DevicesPage: add statusFilter + typeFilter to filtered useMemo deps (was a real bug — filter chips had no effect until unrelated re-render)
- DevicesPage: refactor cpuHistories useMemo to pure Object.fromEntries form
- DashboardPage: suppress react-hooks/purity for intentional Date.now() chart domain
- Sidebar: move GroupHeader from inline arrow fn to module-scope named function (fixes react-hooks/static-components)
- axios: bump ^1.6.5 → ^1.16.0 to clear high-severity CVEs (prototype pollution, CRLF injection, NO_PROXY bypass)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
New Syslog page under Network Services exposes RouterOS logging actions
and routing rules with full CRUD. Logging actions define destinations
(remote syslog server, memory, disk, echo); logging rules define which
topics route to which action.
Single Device and All Devices modes are functionally identical — All
Devices applies changes across every online device in parallel and
shows per-entry coverage badges indicating which devices have each
action or rule. Toggle enable/disable works in both modes.
Remote actions display address:port with facility/severity in a single
Destination column; memory/disk/echo rows show no irrelevant remote
fields. Syslog KPI card added to the Network Services overview.
Add a right-justified refresh interval dropdown above the client table
with options for Not Updating, 30 s, 1 min, and 3 min. The chosen
interval controls both React Query polling and WebSocket-triggered
invalidations — selecting Not Updating fully freezes the list until
page reload. Preference persists via localStorage.
RouterOS vs RouterBOOT distinction:
- Rename "Firmware Updates" → "RouterOS Updates" everywhere (Config tab,
dashboard banner, devices list tooltip) to accurately reflect what is
being updated
- Add RouterBOOT bootloader upgrade support as a distinct operation:
new DeviceCollector methods checkRouterboardUpgrade() and
installRouterboardUpgrade() using /system/routerboard/print and
/system/routerboard/upgrade
- New API endpoints: POST /api/devices/:id/check-routerboard and
POST /api/devices/:id/install-routerboard
- Daily poller now also checks RouterBOOT upgrade status while already
connected during firmware check cycle, with first-discovery alerts
Config tab update panel:
- Combine RouterOS and RouterBOOT into a single "Software & Firmware
Updates" card with two subsections separated by a divider
- Pre-populate both sections from stored DB state on load — no longer
requires a manual "Check" click to see a known pending update
- Enforce update order: when both are pending, show a blue info banner
explaining RouterOS must be installed first, and disable the
RouterBOOT "Upgrade & Reboot" button until the RouterOS update clears
- Fix GET /api/devices/:id to include firmware_update_available,
latest_ros_version, routerboard_upgrade_available, upgrade_firmware_version
(these were missing from the detail endpoint)
Dashboard & devices list:
- Combined amber banner now shows two distinct rows: RouterOS (N) with
device links and RouterBOOT (N) with device links — one dismiss for both
- Devices list: amber ArrowUpCircle = RouterOS update, blue Cpu icon =
RouterBOOT upgrade, each with distinct tooltips
- Fix GET /api/devices (list) to include routerboard_upgrade_available and
upgrade_firmware_version (these were missing, causing RouterBOOT
notifications to never appear on the dashboard)
DB migration:
- Add routerboard_upgrade_available BOOLEAN DEFAULT FALSE
- Add upgrade_firmware_version VARCHAR(20)
Discovery system fixes:
- Fix duplicate entries: replace single-key (mac || ip || identity) dedup
with three cross-referenced indexes so a neighbor seen via LLDP by one
device and via MNDP by another always merges into one row
- Fix offline devices persisting: handleDeviceFailure now deletes
topology_links for the offline device (same pattern as clients)
- Add periodic stale-link sweep every 15 min that removes topology_links
rows older than 20 min as a safety net for devices that go offline
without cleanly triggering handleDeviceFailure
* fix(security): default-secret banners, login rate limiting, real-IP passthrough
- app.set('trust proxy', 1) so req.ip resolves to real client IP behind nginx
- loginRateLimit() middleware on POST /api/auth/login; limits read from
app_settings (login_rate_limit_max / login_rate_limit_window_sec) with
60-second cache; falls back to 10/60s when Redis/DB unavailable
- GET /api/auth/security-status returns warnings for default JWT_SECRET,
ENCRYPTION_KEY, and admin/admin credential — non-breaking for dev
- SecurityBanners component in AppLayout: yellow banner for default secrets,
red banner + change-password link for default admin password
- Login rate limit settings exposed in Settings > General tab
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(security): rate-limit /security-status, suppress CodeQL false positive on /login
- Add allMethods option to rateLimitRedis so it can apply to GET routes
- Apply rateLimitRedis (10 req/60s per user) to GET /auth/security-status
to prevent bcrypt-based CPU exhaustion by authenticated users
- Add lgtm suppression comment on POST /login — CodeQL does not recognize
our custom loginRateLimit() middleware; rate limiting is in place
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Denis Evers <git@evers.sh>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Documents CORS_ORIGIN env var, credential preset operator access control,
and server-side bulk device add job in README. Bumps version across all
four locations.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* devices: hide discovered neighbors that are already managed
Discovered Devices showed every unresolved MikroTik neighbor, even when
the neighbor was obviously the same device a managed node is reporting
through a different link / MAC / alias. Since we already know every
managed device interface MAC, primary IP, and system identity, we can
flag those rows.
- /api/devices/discovered now annotates each row with duplicate_of_device_id
and duplicate_of_device_name when the neighbor matches a managed device
by interface MAC, primary IP, or system identity.
- Add a Hide duplicates of managed devices toggle (default on) above the
table, with a count of how many matches were found. When enabled, the
list hides matched rows; when disabled, matches stay visible but are
dimmed and tagged Already managed so the origin is clear.
- Suppress the Add to Manager action for rows that are already a
managed device, since adding them would create a second record for the
same hardware.
Co-authored-by: Denis <denis-ev@users.noreply.github.com>
* devices: safer duplicate handling, try-all safeguards, and timezone/firmware fixes
Co-authored-by: Denis <denis-ev@users.noreply.github.com>
* Non-blocking follow-ups: parsePort util, tests, bulk-add queue, preset rate limit
Co-authored-by: Denis <denis-ev@users.noreply.github.com>
* fix(security): tighten CORS, Redis rate limits, and preset access controls
Require explicit CORS origins in production, align Socket.IO CORS with HTTP,
default Docker Compose to local HTTPS origins, and remove the in-memory
mutation limiter in favor of Redis. Also add operator-use flags for credential
presets in the UI and harden bulk-add job access and client error messages.
Co-authored-by: Denis <denis-ev@users.noreply.github.com>
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
Squash-merge of PR #28 from @denis-ev. All three review blockers addressed: encryption key management documented, string truncation replaced with skip+warn validation, and undefined return on duplicate merge fixed with proper 500. Local test stack verified against cloned prod DB — migrations, new features, and backend all clean.
2GT-RichandClaude Sonnet 4.6
