[Feature] Temperature Monitoring of PVE hosts

[jostrasser]

Describe the feature

Please integrate Temperature Monitoring for PVE hosts.
It would be nice if you can also save and display the data in a chart to visualize the data over time.

Use Case

Temp Monitoring and alerting

Alternatives considered

none

How important is this for you?

It´s important for a stable cluster (to avoid overheating on high-load clusters)

Checklist

• I have searched existing issues and discussions to make sure this hasn't been requested before

[mkellermann97]

Good news — per-host temperature monitoring already ships. Each node's detail view has a Sensors panel (lm-sensors) with CPU package/core temps, fan RPM and voltages, and cells get flagged when they cross the max/critical thresholds; NVMe drive temps also surface under SMART. What's genuinely missing is the two extras you ask for: (1) storing those readings for a chart over time, and (2) temperature alerting — our metrics-history collector and alert engine currently cover CPU/RAM/disk/backup-SLA, not sensors. Both are sensible additions (feed temps into metrics history + a new temperature alert metric with a crit threshold), so I've flagged them as roadmap candidates. If you already run lm-sensors on the hosts, the Sensors panel should cover the live day-to-day while the history/alert side gets built. Thanks for the clear use case.

If PegaProx helps your setup, donations keep it moving: https://opencollective.com/pegaprox 💛

— Marcus / MK