- API tokens now work even when DISABLE_AUTH is set
- Added API token section back to Security tab in settings
- API tokens can protect API access for automation while keeping UI open
- Invalid tokens are rejected even with auth disabled
- Export/import endpoints still require valid API token
- PBS Servers table now appears right after Proxmox Nodes table
- Better logical flow: infrastructure servers → guest resources → storage
- Makes more sense to have all server-level monitoring grouped together
- Global thresholds section now starts collapsed to reduce visual clutter
- Added dedicated PBS Servers table for monitoring PBS nodes separately
- PBS servers show CPU and Memory thresholds (when metrics are available)
- PBS servers have distinctive indigo color badge for easy identification
- Only shows PBS servers that have metric data (requires Sys.Audit permission)
- PBS nodes don't belong in the 'Proxmox Nodes' table
- They're fundamentally different infrastructure (backup vs hypervisor)
- The mixed presentation was confusing and out of place
- Reverting the previous hasty implementation
- PBS nodes now appear in the Nodes table when configured with proper permissions
- Shows CPU and memory metrics for PBS nodes (disk metrics not available)
- PBS nodes are clearly labeled with 'PBS:' prefix and indigo color badge
- Only shows PBS nodes that have metric data available (requires Sys.Audit permission)
- Disk threshold column is automatically hidden for PBS nodes
- Guests table now groups resources by node with visual separators
- Each node group has a header row showing the node name
- Resources within each node group are sorted by VMID
- Maintains clean visual hierarchy across all three tables
- Created reusable ResourceTable component for rendering resource tables
- Split single large table into 3 distinct tables: Nodes, VMs & Containers, Storage
- Each table only shows relevant metrics (nodes don't show I/O, storage only shows usage)
- Improved organization and visual clarity of threshold configuration
- Reverted to original grouping where guests are under their node names
- Removed "on <node>" text for guests (redundant since they're grouped by node)
- Kept "on <node>" for storage devices (still useful there)
- Order is now: Nodes first, then node groups (delly, minipc), then Storage last
- Changed from node-based grouping to type-based grouping
- Now shows three clear sections: Nodes, Guests, Storage (in that order)
- Guests now show "on <node>" to identify which node they belong to
- Within each section, resources are sorted by node first, then by name/vmid
- Cleaner visual hierarchy that's easier to understand
This makes the table structure more logical - you see all nodes together,
all guests together (but can still see which node they're on), and all
storage together.
- Storage devices now display "on <node>" next to their name
- Makes it clear which node each storage device belongs to
- Helps users identify storage when multiple nodes have similarly named pools
- Added Storage column to threshold table for individual storage thresholds
- Storage devices now appear in their own "Storage" group
- Each storage device can have its own usage threshold override
- Storage shows "-" for CPU/Memory/Disk/IO metrics (not applicable)
- Added orange badge styling for storage device type
- Updated Override interface to support storage type and usage threshold
- Fixed colspan values for new 13-column layout
This allows users to set different thresholds for different storage pools
(e.g. tighter monitoring for system storage vs bulk storage)
- Added migration logic to set I/O thresholds to 0 when loading saved configs
- Only migrates if values match old defaults (150 MB/s disk, 200 MB/s network)
- Ensures existing users get the new "Off by default" behavior
- Prevents saved configs from overriding the new defaults
- Changed default thresholds for diskRead/diskWrite/networkIn/networkOut to 0 (Off)
- UI now shows "Off" instead of "0 MB/s" for disabled I/O metrics
- Reduces alert noise for users who don't need I/O monitoring
- Advanced users can still enable them by setting custom thresholds
- Prevents common false positives like hourly pihole database writes
This makes Pulse focus on the core metrics (CPU, memory, disk space) by default
while keeping I/O monitoring available for those who need it.
- Added 40% opacity to rows where alerts are disabled
- Dimmed text color for disabled resource names
- Provides clear visual feedback that alerts are off for that guest
- Row remains interactive so users can re-enable alerts
- Removed cog/bell-slash icons next to guest names
- Users can see alert configuration in the Alerts settings tab
- Cleaned up unused alert override loading code
- Simplified dashboard display to reduce visual clutter
- Fixed alert units display (MB/s vs %) in notifications and UI
- Added missing threshold controls for disk I/O and network metrics
- Redesigned threshold table with active alert indicators
- Added ability to disable node connectivity alerts
- Improved visual distinction between defaults and overrides
- Added keyboard shortcuts for search (type to search, ESC to clear)
- Better threshold management with automatic cleanup of empty overrides
- Enhanced UI with proper units display for all metrics
addresses #336
- redesigned global defaults section with compact table layout
- fixed bug where editing without changes created unnecessary overrides
- improved visual consistency across the alerts interface
- replaced slider-based UI with clean table layout similar to Guest URLs tab
- shows all resources in a searchable table with inline threshold editing
- added collapsible global defaults section at the top
- enable/disable alerts directly in the table for each guest
- visual indicators for custom overrides
- preserves all existing threshold configurations and overrides
- Guest URLs now stored in backend's guest_metadata.json
- Automatically included in backup/restore operations
- Survives browser data clearing
- Migrates existing localStorage URLs to backend automatically
- Uses existing GuestMetadata infrastructure that was already in place
- Backend API endpoints at /api/guests/metadata
- Matches main dashboard layout with node header rows
- Guests sorted by VMID within each node
- Removed redundant Node column since it's shown in headers
- Better visual organization and consistency
- Reduced padding from py-3 to py-1.5 for data rows
- Reduced header padding from py-3 to py-2
- Input fields now have py-0.5 instead of py-1
- More guests visible without scrolling
- Name, Type, VMID, and Node now in separate columns
- Prevents long text from making rows too wide
- Cleaner, more organized table layout
- Matches the style of the main dashboard table
- Users can now paste complete URLs directly
- Much simpler and more intuitive
- Maintains backward compatibility with old format
- Example: just paste 'https://192.168.1.100:8006'
- Remove inline URL editing from dashboard guest rows
- Create dedicated Guest URLs section in Settings with structured input
- Support protocol (HTTP/HTTPS), host, and port fields separately
- Clean up dashboard by removing unnecessary edit icons
- Guest names remain clickable when URLs are configured
- Add ability to completely disable alerts for specific guests in Custom Overrides
- Refactor override editing to use single form instead of inline editing
- Add dashboard indicators for guests with custom overrides (blue cog for custom thresholds, grey bell-slash for disabled)
- Remove complex Proxmox tag-based alert control system in favor of simpler UI controls
- Improve layout and UX for alert override management
Move service stop to happen before attempting to copy the binary file.
This prevents the error when the binary is still in use by the running service.
Added 2-second delay to ensure process fully releases the file.
- ProxmoxVE community script uses pulse-backend.service
- Our install script now detects and preserves existing service name
- Removed pulse-backend.service from pre-v4 detection (it's used by v4 too)
- Install script stops the correct service during upgrades
- New installs use 'pulse', existing use whatever they have
This ensures compatibility with all deployment methods without breaking changes.
The service name pulse-backend.service is used by both v4 and pre-v4 installations,
so it's not a reliable indicator. Only check for Node.js artifacts (package.json,
node_modules, etc) which are exclusive to pre-v4 versions.
- Add dismissible X button to the Quick Tip in Alerts overview
- Persist dismissed state in localStorage
- Tip can be dismissed and won't show again on page refresh
- pulse-no-alerts: completely silent, no alerts generated at all
- pulse-monitor-only: alerts show in UI for manual checking, but no notifications sent
- updated descriptions to make the distinction clearer
- improved use case examples to show when to use each
- added prominent help box in Alerts overview showing available tags
- updated Custom Rules tab to show tags in priority order
- made feature discoverable without reading documentation
- emphasized use cases (dev VMs, TrueNAS, maintenance windows)
removed incorrect 'temporary override' framing - users want permanent control for VMs with special needs (TrueNAS, Samba, Frigate, etc). tags are now presented as direct per-VM control that can be used for both permanent configuration and temporary needs
key insight: tags let users say 'this specific VM is special' vs custom rules for patterns
- updated documentation to clearly distinguish between:
- Custom Rules: permanent alert policies configured in UI
- Tags: temporary operational overrides applied in Proxmox
- improved log messages to indicate tags are temporary overrides
- added clear use cases and best practices for each approach
- emphasized that tags complement (not replace) custom rules
changed pulse-relaxed behavior to override with fixed values (95% CPU/RAM, 98% disk) rather than adding to existing thresholds. this avoids confusing interactions with custom alert rules and provides more predictable behavior
also updated docs to clarify the priority order of tags vs custom rules
documented the new tag-based alert suppression feature that allows controlling alert behavior per VM/container without UI changes. includes pulse-no-alerts, pulse-monitor-only, and pulse-relaxed tags with examples
addresses #336 - diskRead/diskWrite/networkIn/networkOut metrics are measured in MB/s not percentages, so alert messages now correctly show "X MB/s" instead of "X%"
- pulse-no-alerts: suppress all alerts for VM/CT
- pulse-monitor-only: show alerts in UI but skip email notifications
- pulse-relaxed: use relaxed thresholds (+15% for CPU/memory, +10% for disk)
Tags are read from Proxmox VM/CT configuration and applied during alert checks.
This provides a simple way to manage alert behavior without adding UI complexity.
- Frontend checks for disabled: true in security status
- When auth is disabled, goes straight to dashboard
- Fixes issue where login screen showed even with DISABLE_AUTH=true
- Added DISABLE_AUTH env var to completely bypass authentication
- Useful for reverse proxy setups (Authentik, Authelia, etc.)
- Updated documentation in CONFIGURATION.md and REVERSE_PROXY.md
- Fixed security tests to handle auth disabled state
Frontend (TypeScript):
- Eliminated all 'any' types (7 → 0)
- Added proper types for event system with generics
- Fixed event data interfaces with specific types
- Replaced any with unknown where appropriate
Backend (Go):
- Created central types.go with 30+ typed API structures
- Eliminated all interface{} in /internal/api package (158 → 0)
- Replaced map[string]interface{} with typed structs:
- ChartResponse, VMChartData, NodeChartData, StorageChartData
- DiagnosticsInfo with NodeDetails, ClusterInfo, PBSDetails
- StorageChartsResponse with StorageMetrics
- Improved compile-time type safety for all API responses
Benefits:
- Better IDE support and autocomplete
- Compile-time error detection
- Clearer API contracts
- Improved maintainability
All tests passing, service running successfully with typed code.
- Added test-release.sh for core functionality testing
- Added test-edge-cases.sh for URL and header edge cases
- Added test-proxy-scenarios.sh for reverse proxy testing
- Added test-security.sh for security vulnerability testing
- Added test-installation-methods.sh for deployment validation
- Added test-all.sh master script to run all tests
- These tests would have caught issue #334 and prevent similar issues
Pulse Monitor