Tags give the ability to mark specific points in history as being important
-
v2.5.0
041d8cf7 · ·Probe v2.5.0 - Intel GPU Support MINOR release: New feature added **New Features:** - Intel GPU monitoring via sysfs - Support for 60+ Intel GPU models (Haswell to Raptor Lake + Arc) - Clock frequency monitoring - Power management (RC6) monitoring **Supported Intel GPUs:** - Integrated: HD Graphics 4600, UHD Graphics 770, Iris Xe, etc. - Discrete: Arc A310, A380, A580, A750, A770 **Available Metrics:** - Current/Min/Max/Boost clock frequencies - RP0/RP1/RPn frequency levels - RC6 power state residency **Limitations:** - No GPU utilization (not exposed by most Intel GPUs) - No temperature on older generations - No memory usage (integrated GPUs share system RAM) **Files Changed:** - internal/gpu/intel.go: New Intel GPU implementation - internal/gpu/gpu.go: Multi-vendor integration - probe.go: Version bump to 2.5.0
-
v2.4.1
510ce459 · ·Probe v2.4.1 - GPU filter bug fix Bug Fix: - Fixed GPU metrics not being included in gRPC responses - The filterMetricsByType function was missing the gpu case - GPU data now properly flows from probe to collector Files: - internal/grpc/metrics_server.go: Added gpu filter case This is a PATCH release (bug fix only)
-
v2.3.0
a29b7aed · ·Release v2.3.0: TOML config migration + debug HTTP API BREAKING CHANGE: Configuration format migrated from YAML to TOML Features: - TOML configuration format (probe.yaml → probe.toml) - Debug HTTP API for troubleshooting (/debug/metrics endpoint) - Enhanced bootstrap wizard with debug API configuration - GPU debug logging for troubleshooting GPU metrics Migration required for existing users - see CHANGELOG.md for guide.
-
v2.2.0
Release: V2.2.0 - Stable443f0890 · ·Release v2.2.0: AMD GPU Monitoring Complete AMD GPU monitoring implementation: - GPU utilization, memory, temperature, power, clocks, fan - Multi-GPU support with automatic device enumeration - Pure Go implementation (no CGO) - Supported: RX 7900, 6900, 6800, 6700, 6600, Vega, RX 580/570 - Configuration: metrics.enabled.gpu See CHANGELOG.md for full release notes.
-
v2.1.1
Release: V2.1.1 - Stable13dc4ac4 · ·v2.1.1 - Temperature Collection Fix Patch release fixing critical bug where temperature metrics returned empty. Fixed: - Temperature collection now works correctly with symlinks in /sys/class/hwmon/ - Removed incorrect IsDir() check that was filtering out all hwmon sensors - Temperature tables in collector now populate correctly Verified: - Collecting CPU temps (k10temp: Tctl, Tccd1, Tccd2) - Collecting system temps (NVMe, memory, WiFi, GPU) - All 35+ gRPC tests passing See CHANGELOG.md for full details.
-
v2.1.0
a2f9ede9 · ·Release v2.1.0 - Temperature Monitoring & Bootstrap Wizard Complete temperature monitoring implementation with CPU, system sensors, and disk SMART data, plus interactive setup wizard. Features: - Temperature monitoring (CPU per-core, system sensors, disk SMART) - Interactive bootstrap wizard with secure token generation - Module path updated to v2 for Go semver compliance - Comprehensive documentation updates Temperature Monitoring: - CPU temperature (per-core and package) from /sys/class/hwmon/ - System sensors (chipset, VRM, ambient) with thresholds - Disk temperature via smartctl (SATA/SCSI and NVMe) - Thermal zone monitoring via /sys/class/thermal/ - Max and critical temperature thresholds Bootstrap Wizard: - Interactive configuration setup - Cryptographically secure token generation (crypto/rand) - Server, discovery, metadata, metrics, logging configuration - Configuration summary and confirmation - Automatic backup of existing configs Breaking Changes: - Module path: gitlab.com/The-Krew/sysmon/sysmon-probe/v2 - Protocol v2.1.0 adds TemperatureMetrics field (backwards compatible) Total Metrics: 111+ fields across 6 categories