Real-time monitoring dashboard for Proxmox VE, PBS, Docker, Kubernetes, TrueNAS and vSphere. Self-hosted, with alerts for failed backups, full disks, unhealthy containers and offline nodes.
Go aialertsdashboarddockerdocker-monitoringgohomelabinfrastructure-monitoringkubernetesmonitoringmspproxmoxproxmox-backup-serverproxmox-mail-gatewayproxmox-veself-hostedtruenastypescriptvspherewebhooks
Complete the retained stalled-dashboard help through proxy guidance
pulse-triage[bot]e000ffdb85b8
diff --git a/docs/REVERSE_PROXY.md b/docs/REVERSE_PROXY.mdindex 533d9dd580..6f01de8255 100644--- a/docs/REVERSE_PROXY.md+++ b/docs/REVERSE_PROXY.md@@ -156,7 +156,7 @@ HTTP/error status, not cookies, proxy secrets or a full network export.### Other Issues-- **"Connection lost" (Workloads) or "Disconnected", "Reconnecting...", "Sync reconnecting" or "Backend healthy" in the header status badge**: the live data stream is not connected, which can happen when the WebSocket upgrade fails. Check the proxy's `Upgrade` and `Connection` headers.+- **"Connection lost" (Workloads), "Disconnected", "Reconnecting...", "Sync reconnecting", "Backend healthy" or a stuck "Connecting to Pulse" screen:** separate the WebSocket upgrade from state delivery. If an already-observed upgrade failed through this proxy, check its `Upgrade` and `Connection` forwarding. An HTTP 101 or welcome message without subsequent state is a different failure; changing those headers is not an established remedy. Use the [stalled-dashboard guidance](TROUBLESHOOTING.md#health-is-green-but-the-dashboard-is-stalled), even if `/api/health` returns 200. Keep authentication, TLS and origin checks enabled; do not expose the backend to test a bypass.- **502 Bad Gateway**: Check Pulse's actual listener and the proxy's upstream address, port and network reachability; this does not by itself prove Pulse is stopped. Container `localhost` refers to that container, not another service.- **CORS Errors**: Do not add CORS headers in the proxy; Pulse handles them. Set **Settings → System → Network → Allowed Origins** or use `ALLOWED_ORIGINS` if needed.- **OIDC redirects fail**: Ensure `X-Forwarded-Proto` is set (see above).diff --git a/frontend-modern/public/docs/REVERSE_PROXY.md b/frontend-modern/public/docs/REVERSE_PROXY.mdindex 533d9dd580..6f01de8255 100644--- a/frontend-modern/public/docs/REVERSE_PROXY.md+++ b/frontend-modern/public/docs/REVERSE_PROXY.md@@ -156,7 +156,7 @@ HTTP/error status, not cookies, proxy secrets or a full network export.### Other Issues-- **"Connection lost" (Workloads) or "Disconnected", "Reconnecting...", "Sync reconnecting" or "Backend healthy" in the header status badge**: the live data stream is not connected, which can happen when the WebSocket upgrade fails. Check the proxy's `Upgrade` and `Connection` headers.+- **"Connection lost" (Workloads), "Disconnected", "Reconnecting...", "Sync reconnecting", "Backend healthy" or a stuck "Connecting to Pulse" screen:** separate the WebSocket upgrade from state delivery. If an already-observed upgrade failed through this proxy, check its `Upgrade` and `Connection` forwarding. An HTTP 101 or welcome message without subsequent state is a different failure; changing those headers is not an established remedy. Use the [stalled-dashboard guidance](TROUBLESHOOTING.md#health-is-green-but-the-dashboard-is-stalled), even if `/api/health` returns 200. Keep authentication, TLS and origin checks enabled; do not expose the backend to test a bypass.- **502 Bad Gateway**: Check Pulse's actual listener and the proxy's upstream address, port and network reachability; this does not by itself prove Pulse is stopped. Container `localhost` refers to that container, not another service.- **CORS Errors**: Do not add CORS headers in the proxy; Pulse handles them. Set **Settings → System → Network → Allowed Origins** or use `ALLOWED_ORIGINS` if needed.- **OIDC redirects fail**: Ensure `X-Forwarded-Proto` is set (see above).diff --git a/frontend-modern/src/features/docs/__tests__/dashboardReadinessGuide.test.ts b/frontend-modern/src/features/docs/__tests__/dashboardReadinessGuide.test.tsindex bd9d79914c..6f48ddf5e9 100644--- a/frontend-modern/src/features/docs/__tests__/dashboardReadinessGuide.test.ts+++ b/frontend-modern/src/features/docs/__tests__/dashboardReadinessGuide.test.ts@@ -71,6 +71,17 @@ describe('dashboard readiness guidance', () => {expect(rendered.querySelector('a[href="#connection-refused"]')).not.toBeNull();});+ it('keeps failed proxy upgrades separate from accepted transport without state', () => {+ const proxy = section('REVERSE_PROXY', 'Other Issues');+ expect(proxy.querySelector(`a[href="/docs/TROUBLESHOOTING#${anchor}"]`)).not.toBeNull();+ expect(text(proxy)).toMatch(/already-observed upgrade failed.*Upgrade.*Connection/);+ expect(text(proxy)).toMatch(/HTTP 101 or welcome message without subsequent state is a different failure/);+ expect(text(proxy)).toMatch(/changing those headers is not an established remedy/);+ expect(text(proxy)).toMatch(/Keep authentication, TLS and origin checks enabled/);+ expect(text(proxy)).toMatch(/do not expose the backend to test a bypass/);+ expect(read('frontend-modern/public/docs/REVERSE_PROXY.md')).toBe(guide('REVERSE_PROXY'));+ });+it('uses existing evidence without extra probes, reset advice or credential-bearing exports', () => {const rendered = stalled();const copy = text(rendered);