01 · Fleet SLOs
Every VM on its own uptime SLO.
SLO summary tiles over per-VM uptime lights: green good, red stopped, worst-VM SLO surfaced first.
Home
Proxmox Monitoring
Every guest in every PVE cluster, at fidelity. No per-node tax.
OpenTelemetry-native metrics for your Proxmox VE estate: per-guest uptime SLOs across qemu and LXC, PSI and memory depth kept at full cardinality, and rollups by PVE node, cluster, and storage pool. Scrape the Proxmox VE Exporter straight into your Collector — no proprietary agent on every node.
Eight views · one Proxmox estate
A view for every angle of the cluster.
Click any card to enlarge · esc to close
02 · Uptime trend
Availability over time, VM by VM.
Running VM count as a chart, fleet availability %, per-VM current uptime in days, and a qemu-vs-lxc guest breakdown.
03 · Node distribution
See where every VM lives.
Stacked bar of VMs per Proxmox node over time, with per-VM CPU utilisation below to spot the noisy neighbour.
04 · CPU pressure
PSI stalls, not just utilisation.
Pressure-stall CPU alongside classic utilisation and memory-used bytes. See contention before it becomes an outage.
05 · Memory depth
Top-N memory, swap, balloon, PSI.
Every memory dimension a qemu or lxc guest exposes, kept at full fidelity across the whole fleet, with no cardinality caps.
06 · Disk I/O
Bytes read, bytes written, GiB allocated.
Live disk throughput per VM plus stacked disk allocation across the fleet: capacity and contention on one screen.
07 · Network throughput
Top-N net in and net out.
Every VM's ingress and egress ranked and time-sliced. The storage-pool section starts right below.
08 · Storage pools + nodes
Every pool, every node, used and free.
Per-pool used % and free GiB tiles for every storage pool on every node, with cluster CPU and memory rolled up underneath.
Point your Collector at the Proxmox VE Exporter. Cardinal does the rest.
Scrape the Proxmox VE Exporter and your PVE node exporters into an OpenTelemetry Collector that writes to your own S3, GCS, or Azure Blob bucket — no vendor storage in the path. Per-guest SLOs, qemu- and LXC-level depth, and cluster rollups are on the moment the first object lands.