- Fix stale alerts not clearing when nodes/hosts go offline in CheckNode and HandleHostOffline
- Fix stale alerts persisting when thresholds are disabled or set to 0 in CheckGuest and CheckNode
- Fix CheckHost to properly clear disk alerts when overrides disable them
- Update audit_report.md with findings from the Alert System Reliability Audit
The reporting engine held a direct pointer to the metrics store, which
becomes invalid after a monitor reload (settings change, node config
save, etc.) closes and recreates the store. Use a dynamic getter closure
that always resolves to the current monitor's active store.
Also adds diagnostic logging when report queries return zero metrics,
and integration tests covering the full metrics-to-report pipeline
including reload scenarios.
Fixes#1186
Proxmox's FreeMem field reports free memory relative to the balloon's
guest-visible total (total_mem), not relative to MaxMem. When ballooning
is active and the VM's memory has been reduced, subtracting FreeMem from
MaxMem produces wildly inflated usage (e.g. 97% when actual usage is 20%).
Proxmox's Mem field is already calculated as (total_mem - free_mem),
giving the correct used bytes regardless of balloon state. Swap the
priority so Mem is checked before FreeMem.
Related to #1185
UI/UX Improvements for AI-skeptical users:
- Only show 'Ideas' button if AI is enabled AND configured
- Renamed 'Suggest Profile' to 'Ideas' with lightbulb icon
- Moved 'New Profile' button to primary position
- Changed AI button styling from prominent purple to subtle gray
- Updated modal title to 'Profile Ideas' with neutral language
Multi-tenant bug fix:
- ProfileSuggestionHandler now uses MultiTenantPersistence
- Properly resolves tenant-specific persistence from request context
- Fixes potential nil pointer panic in multi-tenant deployments
- Existing profiles are now correctly loaded per-tenant for AI context
Tests updated to use MultiTenantPersistence with org context injection.
"Maybe later" previously only hid the modal for the current session,
causing it to reappear on every subsequent visit. Now it snoozes the
modal for 7 days before showing again.
The previous commit added FreeBSD binaries to the Docker build and
release directory but didn't update the manifest generator or validation
script. Without this, FreeBSD binaries would be built but not tracked
or validated, and release validation would fail on the new entries.
Related to #1051
Two fixes for FreeBSD agent support:
1. The Docker image never built or included FreeBSD agent binaries, causing
404 errors when FreeBSD clients requested the download. Added FreeBSD
amd64/arm64 cross-compilation for both host-agent and unified-agent,
plus COPY statements to include them in the image. Also added bare
FreeBSD binaries to GitHub release assets for the redirect fallback.
2. pfSense does not use the standard FreeBSD rc.d boot system — scripts
in /usr/local/etc/rc.d/ must end in .sh to run at boot. The installer
now detects pfSense and creates a .sh boot wrapper alongside the
standard rc.d script. Also added -r flag to daemon for auto-restart.
Related to #1051
