Simplifies host connection configuration by removing tailscale_ip field,
connection_type preferences, and fallback retry logic. Now uses only the
ip field or hostname as fallback. Eliminates blocking TCP connectivity
tests that interfered with heartbeat processing.
This resolves intermittent host lost/found issues by removing the
connection retry timeouts that blocked the ZMQ message processing loop.
Implement maintenance_mode_file configuration option in NotificationConfig
to allow customizable file paths for suppressing email notifications.
Updates maintenance mode check to use configured path instead of hardcoded
/tmp/cm-maintenance file.
- Add dedicated heartbeat transmission every 5 seconds independent of metric collection
- Fix host offline detection by clearing metrics for disconnected hosts
- Move exclude_email_metrics to NotificationConfig for better organization
- Add cleanup_offline_hosts method to remove stale metrics after heartbeat timeout
- Ensure offline hosts show proper status icons and visual indicators
Version 0.1.63
- Add agent_heartbeat metric to agent transmission for reliable host detection
- Update dashboard to track heartbeat timestamps per host instead of general metrics
- Add configurable heartbeat_timeout_seconds to dashboard ZMQ config (default 10s)
- Remove unused timeout_ms from agent config and revert to non-blocking command reception
- Remove unused heartbeat_interval_ms from agent configuration
- Host disconnect detection now uses dedicated heartbeat metrics for improved reliability
- Bump version to 0.1.57
- Add exclude_email_metrics field to AgentConfig for filtering email notifications
- Metrics matching excluded names skip notification processing but still appear in dashboard
- Optional field with serde(default) for backward compatibility
- Bump version to 0.1.56
- 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" }
Fix notification issues for better operational experience:
Startup Notification Suppression:
- Suppress notifications for transitions from Status::Unknown during agent/server startup
- Prevents notification spam when services transition from Unknown to Warning/Critical on restart
- Only real status changes (not initial discovery) trigger notifications
- Maintains alerting for actual service state changes after startup
Recovery Notification Refinement:
- Recovery notifications only sent when ALL services reach OK status
- Individual service recoveries suppressed if other services still have problems
- Ensures recovery notifications indicate complete system health restoration
- Prevents premature celebration when partial recoveries occur
Result: Clean startup experience without false alerts and meaningful recovery
notifications that truly indicate full system health restoration.
Bump version to v0.1.48
Correct user-stopped service behavior during startup transitions:
User-Stopped Flag Timing Fix:
- Clear user-stopped flag only when service actually becomes active, not when start command succeeds
- Remove premature flag clearing from service control handler
- Add automatic flag clearing when service status metrics show active state
- Services retain user-stopped status during activating/transitioning states
Service Transition Handling:
- User-stopped services in activating state now report Status::OK instead of Status::Pending
- Prevents host warnings during legitimate service startup transitions
- Maintains accurate status reporting throughout service lifecycle
- Failed service starts preserve user-stopped flags correctly
Journalctl Popup Fix:
- Fix terminal corruption when using J key for service logs
- Correct command quoting to prevent tmux popup interference
- Stable popup display without dashboard interface corruption
Result: Clean service startup experience with no false warnings and proper
user-stopped tracking throughout the entire service lifecycle.
Bump version to v0.1.47
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
Service stop/start operations were failing because systemctl commands
were missing the .service suffix. This caused the new user-stopped
tracking feature to mark services but not actually control them.
Changes:
- Add .service suffix to systemctl commands in service control handler
- Matches pattern used throughout systemd collector
- Fixes service start/stop functionality via dashboard
Clean up legacy documentation:
- Remove outdated TODO.md, AGENTS.md, and test files
- Update CLAUDE.md with current architecture and rules only
- Comprehensive README.md rewrite with technical documentation
- Document user-stopped service tracking feature
Bump version to v0.1.44
Add comprehensive tracking for services stopped via dashboard to prevent
false alerts when users intentionally stop services.
Features:
- User-stopped services report Status::Ok instead of Warning
- Persistent storage survives agent restarts
- Dashboard sends UserStart/UserStop commands
- Agent tracks and syncs user-stopped state globally
- Systemd collector respects user-stopped flags
Implementation:
- New service_tracker module with persistent JSON storage
- Enhanced ServiceAction enum with UserStart/UserStop variants
- Global singleton tracker accessible by collectors
- Service status logic updated to check user-stopped flag
- Dashboard version now uses CARGO_PKG_VERSION automatically
Bump version to v0.1.43
- Add get_git_commit() method to read /var/lib/cm-dashboard/git-commit
- Replace NixOS build version with actual git commit hash
- Show deployed commit hash as 'Build:' value for accurate tracking
- Enable verification of which exact commit is deployed per host
- Update version to 0.1.42
- Replace hardcoded 500ms/2000ms thresholds with configurable nginx_latency_critical_ms
- Simplify status logic to only OK or Critical (no Warning status)
- Add validation for nginx latency threshold configuration
- Re-enable nginx site collection with configurable thresholds
- Resolves issue where sites showed critical at 2000ms despite 30s timeout setting
- Bump version to v0.1.38
Enhanced service discovery to properly show status for all services:
Changes:
- Use systemctl list-unit-files for complete service discovery (finds all services)
- Use systemctl list-units --all for batch runtime status fetching
- Combine both datasets to get comprehensive service list with correct status
- Services found in unit-files but not runtime are marked as inactive (Warning status)
- Eliminates 'unknown' status issue while maintaining complete service visibility
Now inactive services show as Warning (yellow ◐) and active services show as Ok (green ●)
instead of all services showing as unknown (? icon).
Changed service discovery from 'systemctl list-units --all' to 'systemctl list-unit-files'
to ensure ALL service unit files are discovered, including services that have never been started.
Changes:
- Updated systemctl command to use list-unit-files instead of list-units --all
- Modified parsing logic to handle unit file format (2 fields vs 4 fields)
- Set placeholder values in discovery cache, actual runtime status fetched during collection
- This ensures all configured services (like inactive ARK servers) appear in dashboard
The issue was that list-units --all only shows services systemd has loaded/attempted to load,
but list-unit-files shows ALL service unit files regardless of their runtime state.
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.
- Fix /tmp usage status to use proper thresholds instead of hardcoded Ok status
- Fix wear level status to use configurable thresholds instead of hardcoded values
- Add dedicated tmp_status field to SystemWidget for proper /tmp status display
- Remove host-level hourglass icon during service operations
- Implement immediate service status updates after start/stop/restart commands
- Remove active users display and collection from NixOS section
- Fix immediate host status aggregation transmission to dashboard
- Separate dashboard updates from email notifications for immediate status aggregation
- Add metric caching to MetricCollectionManager for instant dashboard updates
- Dashboard now receives cached data every 1 second instead of waiting for collection intervals
- Fix transmission to use cached metrics rather than triggering fresh collection
- Email notifications maintain separate 60-second batching interval
- Update configurable email notification aggregation interval
- Fix email notification aggregation to send batched notifications instead of individual emails
- Fix startup data collection to properly process initial status without triggering change notifications
- Maintain event-driven transmission while preserving aggregated notification batching
- Update version from 0.1.19 to 0.1.20 across all components
Major architectural improvements:
CORE CHANGES:
- Remove notification_interval_seconds - status aggregation now immediate
- Status calculation moved to collection phase instead of transmission
- Event-driven transmission triggers immediately on status changes
- Dual transmission strategy: immediate on change + periodic backup
- Real-time notifications without batching delays
TECHNICAL IMPROVEMENTS:
- process_metric() now returns bool indicating status change
- Immediate ZMQ broadcast when status changes detected
- Status aggregation happens during metric collection, not later
- Legacy get_nixos_build_info() method removed (unused)
- All compilation warnings fixed
BEHAVIOR CHANGES:
- Critical alerts sent instantly instead of waiting for intervals
- Dashboard receives real-time status updates
- Notifications triggered immediately on status transitions
- Backup periodic transmission every 1s ensures heartbeat
This provides much more responsive monitoring with instant alerting
while maintaining the reliability of periodic transmission as backup.
- Implement per-collector interval timing respecting NixOS config
- Remove all hardcoded timeout/interval values and make configurable
- Add tmux session requirement check for TUI mode (bypassed for headless)
- Update agent to send config hash in Build field instead of nixos version
- Add nginx check interval, HTTP timeouts, and ZMQ transmission interval configs
- Update NixOS configuration with new configurable values
Breaking changes:
- Build field now shows nix store config hash (8 chars) instead of nixos version
- All intervals now follow individual collector configuration instead of global
New configuration fields:
- systemd.nginx_check_interval_seconds
- systemd.http_timeout_seconds
- systemd.http_connect_timeout_seconds
- zmq.transmission_interval_seconds
- Remove all SystemRebuild command infrastructure from agent and dashboard
- Replace with direct tmux popup execution: ssh {user}@{host} {alias}
- Add configurable SSH user and rebuild alias in dashboard config
- Eliminate agent process crashes during rebuilds
- Simplify architecture by removing ZMQ command streaming complexity
- Clean up all related dead code and fix compilation warnings
Benefits:
- Process isolation: rebuild runs independently via SSH
- Crash resilience: agent/dashboard can restart without affecting rebuilds
- Configuration flexibility: SSH user and alias configurable per deployment
- Operational simplicity: standard tmux popup interface
- Add cm-rebuild systemd service for process isolation
- Add sudo permissions for service control and journal access
- Remove verbose flag for cleaner output
- Ensures reliable rebuild operations without agent crashes
- Remove auto-close behavior from terminal popup for manual review
- Fix system panel to show correct NixOS section layout
- Add missing Active users line after Agent version
- Switch agent version from nix store hash to actual version number (v0.1.11)
- Display full version string without truncation for clear version tracking
- Replace simulated progress messages with actual stdout/stderr capture
- Stream all nixos-rebuild output line-by-line to terminal popup
- Show transparent build process including downloads, compilation, and activation
- Maintain real-time visibility into complete rebuild process
- Add terminal popup UI component with 80% screen coverage and terminal styling
- Extend ZMQ protocol with CommandOutputMessage for streaming output
- Implement real-time output streaming in agent system rebuild handler
- Add keyboard controls (ESC/Q to close, ↑↓ to scroll) for popup interaction
- Fix system panel Build display to show actual NixOS build instead of config hash
- Update service filters in README with wildcard patterns for better matching
- Add periodic progress updates during nixos-rebuild execution
- Integrate command output handling in dashboard main loop
- Agent reports version via agent_version metric using nix store hash
- Dashboard displays agent version in system widget
- Foundation for cross-host version comparison
- Both agent -V and dashboard show versions
- Support multiple SATA SSD wear attributes (SSD_Life_Left, Media_Wearout_Indicator, etc.)
- Handle manufacturer differences in wear reporting
- Proper parsing of SMART table format with VALUE column
- Covers Samsung, Intel, Crucial and other common SSD types
- NVMe Percentage Used support maintained
- Consolidate SMART thresholds into DiskConfig structure
- Remove separate SmartConfig - disk collector handles all drive data
- Update NixOS configuration to use disk.temperature_* settings
- Remove hardcoded temperature thresholds in disk collector
- Logical grouping: disk collector owns all disk/drive configuration
- Remove separate smart collector implementation
- Disk collector already handles SMART data for drives
- Eliminates duplicate smartctl calls causing performance issues
- SMART functionality remains in logical place with disk monitoring
- Fixes infinite smartctl loop issue
- Add SmartCollector import and initialization
- Enable in both normal and benchmark modes
- Fixes infinite smartctl loop issue by properly managing collector
- Smart collector now active when config.smart.enabled = true
- Update to match current Metric structure
- Use correct Status enum and collector interface
- Fix MetricValue types and constructor usage
- Builds successfully with warnings only
- Rewrite smart collector to match current architecture
- Add back to mod.rs exports
- Fixes infinite smartctl loop issue
- Uses simple health and temperature monitoring
Redirect nixos-rebuild stdout/stderr to /var/log/cm-dashboard/nixos-rebuild.log
while keeping the process detached. This allows monitoring rebuild progress
and debugging why cargo builds in /tmp aren't visible when agent runs.
Use: tail -f /var/log/cm-dashboard/nixos-rebuild.log to monitor progress.
Run nixos-rebuild with nohup in background to prevent the agent
from killing itself during system rebuild. The rebuild process
now runs independently, allowing the agent to return success
immediately and avoid crashes during binary updates.
This fixes the issue where agent would crash during rebuild
and restart with the old binary due to missing daemon-reload.
- Replace git commit hash with nix store hash extraction
- Read from /run/current-system symlink target
- Extract first 8 characters of nix store hash: d8ivwiar
- Shows actual deployed configuration, not just source
- Enables proper rebuild completion detection
- Accurate deployment verification
Fixes kernel namespace sandboxing issues when running as systemd service.
The --no-sandbox flag disables Nix build sandboxing which requires
kernel namespaces not available in restricted service environments.
Remove sudo -u cm wrapper that was causing git repository ownership
mismatch. Now cm-agent runs nixos-rebuild directly as root, avoiding
the ownership conflict between cm-agent (git clone) and cm user.
Updated sudo rules to allow cm-agent -> root nixos-rebuild access.
Use explicit /run/current-system/sw/bin/nixos-rebuild path instead of
'nixos-rebuild' command to match sudo rules exactly. This resolves
'command not allowed' errors when the command resolves to nix store paths.
- Add nixos_config_api_key_file option to NixOS configuration
- Support reading API token from file for private repositories
- Automatically inject token into HTTPS URLs (https://token@host/repo.git)
- Graceful fallback to original URL if key file missing/empty
- Default key file location: /var/lib/cm-dashboard/git-api-key
Usage: echo 'your-api-token' | sudo tee /var/lib/cm-dashboard/git-api-key
Replace direct directory access with git clone/pull approach:
- Add git configuration options (url, branch, working_dir) to NixOS module
- Update SystemConfig and AgentCommand to use git parameters
- Implement ensure_git_repository() method for clone/pull operations
- Agent clones nixosbox to /var/lib/cm-dashboard/nixos-config
- Maintains security while solving permission denied issues
The agent now manages its own copy of the configuration without
needing access to /home/cm directory.
- Add sudo permissions for systemctl and nixos-rebuild commands
- Use sudo in agent command execution for proper privileges
- Fix backup collector to handle missing status files gracefully
- Eliminate backup error spam when no backup system is configured
This implements the core functionality for executing remote commands through
the dashboard and providing real-time visual feedback to users.
Key Features:
- Remote service control (start/stop/restart) via existing keyboard shortcuts
- System rebuild command with maintenance mode integration
- Real-time visual feedback with service status transitions
- ZMQ command protocol extension for service and system operations
Implementation Details:
- Extended AgentCommand enum with ServiceControl and SystemRebuild variants
- Added agent-side handlers for systemctl and nixos-rebuild execution
- Implemented command status tracking system for visual feedback
- Enhanced services widget to show progress states (⏳ restarting)
- Integrated command execution with existing keyboard navigation
Keyboard Controls:
- Services Panel: Space (start/stop), R (restart)
- System Panel: R (nixos-rebuild switch)
- Backup Panel: B (trigger backup)
Technical Architecture:
- Command flow: UI → Dashboard → ZMQ → Agent → systemctl/nixos-rebuild
- Status tracking: InProgress/Success/Failed states with visual indicators
- Maintenance mode: Automatic /tmp/cm-maintenance file management
- Service feedback: Icon transitions (● → ⏳ → ● with status text)
