Updated the dashboard system widget to properly display MergerFS storage
pools in the exact format described in CLAUDE.md:
- Pool header showing "mergerfs (2+1):" format
- Total usage line: "├─ Total: ● 63% 2355.2GB/3686.4GB"
- Data Disks section with tree structure
- Individual drive entries: "│ ├─ ● sdb T: 24°C W: 5%"
- Parity drives section: "├─ Parity: ● sdc T: 24°C W: 5%"
- Mount point footer: "└─ Mount: /srv/media"
The dashboard now processes both data_drives and parity_drives arrays from
the agent data correctly and renders the complete MergerFS pool hierarchy
with proper status indicators, temperatures, and wear levels.
Storage display now matches the enhanced tree structure format specified
in documentation with correct Unicode tree characters and spacing.
Replace timestamp parsing with direct display of start_time from backup TOML file to ensure timestamp always appears regardless of format. Remove empty line spacing above backup section for compact layout.
Changes:
- Remove parsed timestamp fields and use raw start_time string from TOML
- Display backup time directly from TOML file without parsing
- Remove blank line above backup section for tighter layout
- Simplify BackupData structure by removing last_run and next_scheduled fields
Version bump to v0.1.150
Replace standalone backup widget with compact backup section in system widget displaying disk serial, temperature, wear level, timing, and usage information.
Changes:
- Remove standalone backup widget and integrate into system widget
- Update backup collector to read TOML format from backup script
- Add BackupDiskData structure with serial, usage, temperature, wear fields
- Implement compact backup display matching specification format
- Add time formatting utilities for backup timing display
- Update backup data extraction from TOML with disk space parsing
Version bump to v0.1.149
- Remove duplicate status string fields from ServiceData and SubServiceData
- Use only Status enum as single source of truth for service status
- Agent calculates Status enum using calculate_service_status()
- Dashboard converts Status enum to display text for UI
- Implement flexible metrics system for sub-services with label/value/unit
- Fix status icon/text mismatches (inactive services now show gray circles)
- Ensure perfect alignment between service icons and status text
Fully restored CM Dashboard as a complete monitoring system with working
status evaluation and email notifications.
COMPLETED PHASES:
✅ Phase 1: Fixed storage display issues
- Use lsblk instead of findmnt (eliminates /nix/store bind mount)
- Fixed NVMe SMART parsing (Temperature: and Percentage Used:)
- Added sudo to smartctl for permissions
- Consistent filesystem and tmpfs sorting
✅ Phase 2a: Fixed missing NixOS build information
- Added build_version field to AgentData
- NixOS collector now populates build info
- Dashboard shows actual build instead of "unknown"
✅ Phase 2b: Restored status evaluation system
- Added status fields to all structured data types
- CPU: load and temperature status evaluation
- Memory: usage status evaluation
- Storage: temperature, health, and filesystem usage status
- All collectors now use their threshold configurations
✅ Phase 3: Restored notification system
- Status change detection between collection cycles
- Email alerts on status degradation (OK→Warning/Critical)
- Detailed notification content with metric values
- Full NotificationManager integration
CORE FUNCTIONALITY RESTORED:
- Real-time monitoring with proper status evaluation
- Email notifications on threshold violations
- Correct storage display (nvme0n1 T: 28°C W: 1%)
- Complete status-aware infrastructure monitoring
- Dashboard is now a monitoring system, not just data viewer
The CM Dashboard monitoring system is fully operational.
Implements clean structured data collection eliminating all string metric
parsing bugs. Collectors now populate AgentData directly with type-safe
field access.
Key improvements:
- Mount points preserved correctly (/ and /boot instead of root/boot)
- Tmpfs discovery added to memory collector
- Temperature data flows as typed f32 fields
- Zero string parsing overhead
- Complete removal of MetricCollectionManager bridge
- Direct ZMQ transmission of structured JSON
All functionality maintained: service tracking, notifications, status
evaluation, and multi-host monitoring.
Update storage display to match CLAUDE.md specification:
- Show drive temp/wear on main line: nvme0n1 T: 25°C W: 4%
- Display individual filesystems as sub-items: /: 55% 250.5GB/456.4GB
- Remove Total usage line in favor of filesystem breakdown
Clean up code warnings:
- Remove unused heartbeat methods and fields
- Remove unused backup widget fields and methods
- Add allow attributes for legacy methods
Remove parentheses from drive temperature/wear display to match the
hierarchical format specified in documentation. Drive details now show
directly with status icons as 'nvme0n1 T: 25°C W: 4%' format.
Replace fragile string-based metrics with type-safe JSON data structures.
Agent converts all metrics to structured data, dashboard processes typed fields.
Changes:
- Add AgentData struct with CPU, memory, storage, services, backup fields
- Replace string parsing with direct field access throughout system
- Maintain UI compatibility via temporary metric bridge conversion
- Fix NVMe temperature display and eliminate string parsing bugs
- Update protocol to support structured data transmission over ZMQ
- Comprehensive metric type coverage: CPU, memory, storage, services, backup
Version bump to 0.1.131
- Display wear percentage in storage headers for single physical drives
- Remove redundant drive type indicators, show wear data instead
- Fix wear metric parsing for physical drives (underscore count issue)
- Add NVMe temperature parsing support (Temperature: format)
- Add raw metrics debugging functionality for troubleshooting
- Clean up physical drive display to remove redundant information
- Display actual drive name (e.g., nvme0n1) instead of mount point for physical drives
- Fix health status parsing for physical drives to show proper status icons
- Update pool name extraction to handle disk_{drive}_health metrics correctly
- Improve storage widget rendering for physical drive identification
- Eliminate hardcoded mappings like 'root' -> '/' and 'steampool' -> '/mnt/steampool'
- Use device names directly for physical drives
- Rely on mount_point metrics from agent for actual mount paths
- Implement zero-configuration architecture as specified in CLAUDE.md
- Remove fallback logic that could extract incorrect pool names
- Simplify pool suffix matching to use explicit arrays
- Ensure only valid metric patterns create pools
- Use actual device names (sdb, sdc) instead of data_0, parity_0
- Fix physical drive naming to show device names instead of mount points
- Update pool name extraction to handle new device-based naming
- Ensure Drive: line shows temperature and wear data for physical drives
- Add SnapRAID parity drive detection to mergerfs discovery
- Remove Pool Status health line as discussed
- Update drive display to always show wear data when available
- Include /mnt/parity drives as part of mergerfs pool structure
- Restructure storage rendering logic to prevent drive duplication
- Use specific mergerfs check instead of generic multi-drive condition
- Ensure drives only appear once under organized data/parity sections
- Update extract_pool_name to handle data_/parity_ drive metrics correctly
- Fix extract_drive_name to parse mergerfs drive roles properly
- Prevent srv_media_data from being parsed as separate pool
- Improve pool name extraction in dashboard parsing
- Use consistent mergerfs pool naming in agent
- Add mount_point metric parsing to use actual mount paths
- Fix pool consolidation to prevent duplicate entries
1. Add missing _fs_ filter to usage_percent parsing in dashboard
2. Fix agent to use calculated fs_status instead of hardcoded Status::Ok
This completes the disk collector auto-discovery by ensuring filesystem
usage percentages and status indicators display correctly.
1. Prevent filesystem _fs_ metrics from overwriting pool totals
2. Fix filesystem name extraction to properly parse boot/root names
This resolves both the pool total display (showing 0.1GB instead of 220GB)
and individual filesystem display (showing —% —GB/—GB).
Prevent string slicing panic in extract_filesystem_metric when
parsing individual filesystem metrics. This resolves the issue
where filesystem entries show —% —GB/—GB instead of actual usage.
Fixed critical bug where dashboard crashed with 'begin <= end' slice error
when parsing disk metrics with new naming format. Added bounds checking
to prevent invalid string slicing operations.
- Fixed extract_pool_name string slicing bounds check
- Removed ineffective panic handling that caused infinite loop
- Dashboard now handles new disk collector metrics correctly
Added comprehensive error handling to storage metrics parsing to prevent
dashboard crashes when encountering unexpected metric formats or parsing
errors. Dashboard now continues gracefully with empty storage display
instead of crashing, improving reliability during metric format changes.
- Wrapped storage metric parsing in panic recovery
- Added logging for metric parsing failures
- Dashboard shows empty storage on errors instead of crashing
- Ensures dashboard remains functional during agent updates
- Make filesystem display more forgiving - show partial data if available
- Will display usage% even if GB values are missing, or vice versa
- This should help identify which specific metrics aren't being populated
- Debug version to identify filesystem data population issues
- Fix extract_filesystem_metric() to handle multi-underscore metric names correctly
- Parse known metric suffixes (usage_percent, mount_point, available_gb, etc.)
- Prevent incorrect parsing like boot_mount_point -> fs_name='boot_mount', metric_type='point'
- Should now correctly show /boot and / instead of /boot/mount and /root/mount
- Allow filesystem entries to be created with any metric, not just mount_point
- Ensure filesystem children appear under physical drive pools
- Improve mount point fallback logic for better compatibility
- Fix extract_pool_name() to handle filesystem metrics (_fs_) correctly
- Prevent individual filesystem pools (nvme0n1_fs_boot, nvme0n1_fs_root) from being created
- Fix incorrect mount point names (was showing /root/mount instead of /)
- Only create filesystem entries when receiving mount_point metrics
- Add available_gb field to FileSystem struct for proper available space handling
- Ensure filesystem children show correct usage data instead of —% —GB/—GB
- Implement filesystem children display under physical drive pools
- Agent generates individual filesystem metrics for each mount point
- Dashboard parses filesystem metrics and displays as tree children
- Add filesystem usage, total, and available space metrics
- Support target format: drive info + filesystem children hierarchy
- Fix compilation warnings by properly using available_bytes calculation
- Add support for mergerfs pool grouping with data and parity disk separation
- Implement pool health monitoring (healthy/degraded/critical status)
- Create hierarchical tree view for multi-disk storage arrays
- Add automatic pool type detection and member disk association
- Maintain backward compatibility for single disk configurations
- Support future extension for RAID and ZFS pool types
- Add disk wear percentage collection from SMART data in backup script
- Add backup_disk_wear_percent metric to backup collector with thresholds
- Display wear percentage in backup widget disk section
- Fix storage section overflow handling to use consistent "X more below" logic
- Update maintenance mode to return pending status instead of unknown
- Fix get_selected_service to always return parent service names
- Prevent selection of container sub-items when managing docker services
- Ensure service commands operate on correct systemd service names
- Simplify service selection logic to only consider parent services
- Update version to 0.1.92
- Remove scroll offset fields from HostWidgets struct
- Replace scrolling with simple "X more below" indicators in all widgets
- Remove user-stopped service tracking from agent (now uses SSH control)
- Inactive services now consistently show Status::Inactive with empty circles
- Simplify widget render methods by removing scroll parameters
- Clean up unused imports and legacy scrolling infrastructure
- Fix journalctl command to use -fu for proper log following
- Add new Status::Inactive variant to enum for better service state representation
- Agent now assigns Status::Inactive instead of Status::Warning for inactive services
- Dashboard displays inactive services with empty circle (○) icon in gray color
- User-stopped services still show as Status::Ok with green filled circle
- Inactive services treated as OK for host status aggregation
- Improves visual clarity between active (●), inactive (○), and warning (◐) states
- Remove all transitional icon infrastructure (CommandType, pending transitions)
- Clean up ZMQ command system remnants after SSH migration
- Add real-time log streaming for service start operations
- Show final logs and status for service stop operations
- Fix compilation warnings by removing unused methods
- Simplify UI architecture with pure SSH-based service control
Display the connection IP address that the dashboard is configured to use
for each host below the Agent version information. Shows which network
path (local/Tailscale) is being used for connections based on host
configuration.
Features:
- Display detected IP below Agent row in system widget
- Uses existing host configuration connection logic
- Shows actual IP being used for dashboard connections
- Add Status::Offline enum variant for disconnected hosts
- All configured hosts now always visible showing offline status when disconnected
- Add WakeOnLAN support using wake-on-lan Rust crate
- Implement w key binding to wake offline hosts with MAC addresses
- Simplify configuration to single [hosts] section with MAC addresses only
- Change critical status icon from ◯ to ! for better visibility
- Add proper MAC address parsing and error handling
- Silent WakeOnLAN operation with logging for success/failure
Configuration format:
[hosts]
hostname = { mac_address = "AA:BB:CC:DD:EE:FF" }
Improve user-stopped service tracking behavior:
Service Display Fix:
- Services widget now shows actual systemctl status (active/inactive)
- Use info.status instead of hardcoded text based on widget_status
- User-stopped services correctly display 'inactive' with green OK icon
- Prevents misleading 'active' display for stopped services
User-Stopped Flag Timing Fix:
- Clear user-stopped flag AFTER successful service start, not when command sent
- Prevents warnings during service startup transition period
- Service remains Status::OK during 'activating' state for user-stopped services
- Flag only cleared when systemctl start command actually succeeds
- Failed start attempts preserve user-stopped flag
Result: Clean service state tracking with accurate display and no false alerts
during intentional user operations.
Bump version to v0.1.45
- Change system panel title from 'NixOS:' to 'NixOS hostname:'
- Make main dashboard title 'cm-dashboard' bold in top bar
- Remove unused Typography::title() function to fix warnings
- Update SystemWidget::render_with_scroll to accept hostname parameter
- Update version to 0.1.41 in all Cargo.toml files and dashboard code
- Fix disk drive name extraction for mount points with underscores (e.g., /mnt/steampool)
- Replace confusing "1" and "2" drive names with proper device names like "sda1", "sda2"
- Update title bar with blue background and dark text styling
- Right-align host list in title bar while keeping "cm-dashboard" on left
- Bump version to v0.1.35
- Fix service icons to use background color when selected for better visibility against blue selection background
- Combine start/stop service help text entries into single "s/S: Start/Stop Service"
- Change help text keys to lowercase (r: Rebuild Host, q: Quit)
- Bump version to v0.1.34
Major UI simplification and navigation improvements:
Changes:
- Removed panel selection concept entirely (no more Shift+Tab)
- Service selection always visible with blue highlighting
- Up/Down arrows now directly control service selection
- Added j/k vi-style navigation keys as alternatives to arrow keys
- Removed panel focus borders - all panels look uniform
- Service commands (s/S) work without panel focus requirements
- Updated keyboard shortcuts to reflect simplified navigation
Navigation:
- Tab: Switch hosts
- ↑↓/jk: Select service (always works)
- R: Rebuild host
- s: Start service
- S: Stop service
- q: Quit
The interface is now much simpler and more intuitive with direct service control.
Simplified keyboard controls by removing service restart functionality:
- Removed 'r' key restart functionality from Services panel
- Made 'R' key always trigger system rebuild regardless of focused panel
- Updated context shortcuts to show 'R: Rebuild Host' globally
- Removed all ServiceRestart enum variants and associated code:
- UiCommand::ServiceRestart
- CommandType::ServiceRestart
- ServiceAction::Restart
- Cleaned up pending transition logic to only handle Start/Stop commands
The 'R' key now consistently rebuilds the current host from any panel,
while 's' and 'S' continue to handle service start/stop in Services panel.
Resolved issues with transitional service icons not being properly visible:
- Removed 3-second timeout that was clearing pending transitions prematurely
- Fixed selection highlighting disappearing when transitional icons appeared
- Implemented conditional coloring for transitional icons:
- Blue when service is not selected
- Dark background color when service is selected (for visibility against blue selection)
- Transitions now persist until actual service status changes occur
Both selection highlighting and transitional icons are now visible simultaneously.
The root cause of transitional service icons not showing was that service names
were stored as raw names (e.g., "sshd") in pending_transitions but looked up
against formatted display lines (e.g., "sshd active 1M ").
Changes:
- Modified display_lines structure to include both formatted text and raw service names
- Updated rendering loop to use raw service names for pending transition lookups
- Fixed get_selected_service() method to use the new tuple structure
- Transitional icons (↑ ↓ ↻) should now appear correctly when pressing s/S/r keys
- Force sshd service to always show "↑ starting" for debugging
- Test if basic directional arrow rendering works in services widget
- Temporary debug change to isolate rendering vs logic issues
- Will help determine if problem is in pending transitions or rendering
If arrow appears: pending transitions logic issue
If no arrow: basic rendering path issue
- Store pending transitions even for redundant commands (start active service)
- Add 3-second timeout for redundant command visual feedback
- Include timestamp in pending transitions to enable timeout clearing
- Show directional arrows immediately regardless of command validation result
- Fix core issue where state validation prevented visual feedback storage
Now pressing s/S/r always shows immediate directional arrows, even for
redundant operations, providing consistent visual feedback to users.
- Prevent selection highlighting when service has pending transition
- Allow directional arrows (↑ ↓ ↻) to show through on selected services
- Fix core issue where selection styling was overwriting transitional icons
- Transitional icons now properly visible during service command execution
The selection highlighting logic now skips services with pending transitions,
ensuring that directional arrows are visible when executing service commands.
- Replace timeout-based command status with pending service transitions
- Show immediate directional arrows when pressing service commands (↑ ↓ ↻)
- Track original service status and command type for each pending operation
- Automatically clear transitional icons when real status updates arrive
- Remove unused TerminalPopup and CommandStatus infrastructure
- Simplify visual feedback system using state-based approach
Service commands now provide instant visual feedback that persists until
the actual service state changes, eliminating timing issues and improving UX.
