monitoring: auto-discover SMART devices instead of hardcoding
...
Remove smartDevices option and per-host device lists. Telegraf will
now scan all block devices automatically, so disks can be added or
removed without config changes.
Co-Authored-By: Claude Opus 4.6 <[email protected]>
Custom script reports zpool health as numeric metric (0=ONLINE,
1=DEGRADED, 2=FAULTED, etc.) via Telegraf inputs.exec, enabling
Grafana alerting on pool degradation.
Co-Authored-By: Claude Opus 4.6 <[email protected]>
monitoring: fix InfluxDB URL and add nvme-cli to Telegraf PATH
...
Use door1's LAN IP (10.0.91.30) instead of WireGuard IP (172.16.1.1)
for InfluxDB endpoint. Add nvme-cli to Telegraf's PATH for NVMe SMART
attribute collection.
Co-Authored-By: Claude Opus 4.6 <[email protected]>
skydick: add Telegraf monitoring with SMART, ZFS, and system metrics
...
Sends metrics to door1 InfluxDB (bucket: skydick) via Telegraf.
Monitors all 5 Mach2 SAS drives, NVMe P4500, and boot SSD via SMART.
InfluxDB token encrypted with agenix.
Co-Authored-By: Claude Opus 4.6 <[email protected]>
