Three UX bugs in the icon picker:
1. Skeleton-forever on edit screens. The input was pre-filled with the
full CDN URL, the search query fired with that URL, and the DB only
stores bare filenames, so the response was always empty -> 15
skeletons, forever. Decouple the input display from the query:
initialize search to the committed value and only fire the query
when the user has actually typed something different.
2. Loading / error / no-results all looked identical (15 skeletons).
Split the render into four explicit branches: isError (with retry),
isLoading (skeletons, only on cold first load), results, no-results.
3. No relevance ranking and no SVG-first ordering. Add orderBy in the
icons router: relevance tier (exact > prefix > substring) -> SVG
first -> alphabetical. Patterns are bound as JS strings so the SQL
works across sqlite / postgresql / mysql without concat operator
differences.
Also add placeholderData: keepPreviousData so refetching during typing
doesn't flash the dropdown to skeletons, fix the upload pre-fill that
was leaking a local media URL into the search, drop the no-op
updateSelectedOptionIndex, and mark the file as a client component so
Next.js can render it from server-component routes.
Add common.iconPicker headerLoading / noResults / error keys to
en.json (other locales fall back to en until the next sync).
No ternaries in the changed files. Intended simplifications are marked
with ponytail: comments.
* feat(analytics): migrate from Umami to PostHog server-side analytics
Replace @umami/node with posthog-node for all server-side telemetry.
Add singleton PostHog client, feature tracking POC, DAU heartbeat,
and tRPC analytics router. Cron job is now locked (no manual trigger,
no custom interval). Update consent copy to explain what we collect
and why.
* style: auto-fix lint and format
* refactor(analytics): simplify + add client-side posthog-js init
- Inline constants.ts and track.ts into client.ts (single source)
- Remove dead exports (shutdownPostHogAsync, getPostHogClient)
- Flatten all DB queries into one Promise.all (no waterfall)
- Derive integration/widget totals from group-by (drop redundant counts)
- Add posthog-js with dynamic import, gated on enableGeneral
- Move Analytics component to body, pass consent from server settings
* style: auto-fix lint and format
* feat(cron): add hidden flag to cron jobs, hide analytics from task panel
The analytics job runs silently in the background — users don't need
to see or interact with it. Hidden jobs are filtered out of getAllAsync
so they never appear in the management UI.
* refactor(analytics): simplify to single toggle, always collect all data
Remove granular enableWidgetData/enableIntegrationData/enableUserData
toggles. Analytics is now a single on/off switch. When enabled, all
anonymous usage data is collected in one flat Promise.all call.
Side effect: existing users with enableGeneral=true who had granular
flags off will now send all data. This is by design — the granular
controls added complexity without meaningful privacy benefit since
the data is already anonymous counts.
* style: auto-fix lint and format
- Throw TRPCError instead of returning null when no integration
- Sort beszelRouter import alphabetically in mcp.ts
- Replace ternary with lookup table for background color
- Mention integration_all in MCP server instructions
- Add .default("1h") to timePeriod so MCP clients can omit it
- Include "1m" in the description's valid timePeriod values
- Clarify that getSystemStats uses only the first integrationId
- Use MantineSize type instead of plain string for itemRadius
Register beszelRouter in the MCP server with .meta({ mcp: enabled })
on getSystems, getAlerts, and getSystemStats procedures. AI assistants
can now query Beszel-monitored systems, view alerts, and fetch
historical metrics through the MCP protocol.
Ref #5968
- Restore timePeriod union cast to fix typecheck (TS2769)
- Clear live data on subscription error so error UI always shows
- Emit error when no integration is configured for live subscription
- Add error state to system select input when query fails
- Surface subscription errors in useLiveStats via onError + error state
- Show error UI in live mode when SSE connection fails
- Abort SSE reader on subscribeTopic failure instead of silent log
- Guard mock realtime against pre-aborted signal to prevent interval leak
- Remove dead includeDocker from subscription input schema
- Add subscribeSystemStats tRPC subscription bridging PocketBase SSE
- Add subscribeRealtimeMetrics to BeszelIntegration with SSE parsing
- Add error handling: try/catch in subscription, response check in topic POST
- Add cacheDurationForInput to request handler for per-period cache TTL
- Add mock subscribeRealtimeMetrics with 1s interval for dev testing
* fix(boards): add permissions to board summary schema to fix manage page crash
The getAllBoards query fetches userPermissions and groupPermissions
from the DB, but boardSummarySchema (the output validator) did not
include them. tRPC stripped the fields, causing a TypeError when
getBoardPermissionsAsync called .some() on undefined.
* fix(backup): cast Buffer to Uint8Array for NextResponse body
Node 24 narrowed the BodyInit type, making Buffer incompatible.
Wrapping in Uint8Array satisfies the type constraint.
* fix(lint): remove forbidden non-null assertions across backup and board code
Replace non-null assertions with safe guards, optional chaining, and
for-of iteration so oxlint passes in CI. Also remove unused Collapse import.
ajnart
Thomas
Louis Bailleau
GlareandManuel
ajnart
Afonso Jorge Ramos
Ashton BattyandMeier Lukas
homarr-renovate[bot]andMeier Lukas
KyleandMeier Lukas