Skip to content

Fleet inventory — October 22, 2026

Purpose: list every machine in the lab, with what runs on it and how loaded it is. Status: dated, read-only snapshot, replaced by the October 27 inventory. It replaced the October 21 inventory as the current host list. Disks are maintained in the disk map.

Collection: about 04:40 UTC, from the Kubernetes API (kubectl get nodes, kubectl top nodes, running pods), Longhorn objects and Omni MachineStatus. Queries ran separately, so this is not an atomic snapshot. CPU and memory use are point-in-time kubectl top values. Host CPU temperatures, Proxmox host RAM and NAS pools were not re-collected; the Proxmox hosts did not change.

What changed since October 21

  • HP Elite Mini 600 G9 (.22) is bare-metal Talos as hp-elite-gen13 (#3087). Proxmox, its worker VM, the Proxmox Datacenter Manager LXC, the proxmox-hp-elite provider and the hp-elite-worker machine class are gone. The node keeps the class label hp-elite-worker.
  • One 16 GB DIMM moved from the 600 G9 to the Dell Pro Micro. The Dell now has 2× 16 GiB; the 600 G9 has 1× 16 GiB.
  • New Longhorn disks: the 600 G9's Intel 660p (hp-elite-nvme), the Dell's Apple SATA SSD (dell-ultra-ssd, #3089) and the hp-sff spare 1 TB PNY (hp-sff-ssd2, #3090). Every disk except the system disks and the Wi-Fi shed node now takes replicas.
  • The USB Coral is retired. The Zigbee stick is in hp-elite-gen12, where Home Assistant runs. Frigate moves to the Dell NPU (#3088) and is Pending until then.
  • Six volumes had their only copy on the 660p. container-registry/registry was restored from kopiur. Four backup-exempt caches were recreated empty. nginx-example/storage (demo, no Argo CD owner) is still faulted. Tracking: #3071.

Physical machines

Machine Address CPU RAM Platform Runs
Threadripper (X399) .14 Ryzen TR 2950X · 32 threads 125.7 GiB Proxmox VE 9 GPU worker VM (30 vCPU, 100 GiB, 2× RTX 3090)
HP ProDesk 600 G4 SFF .21 i5-8500 · 6 threads 62.6 GiB Proxmox VE 9 control plane VM (4 vCPU, 12 GiB) + worker VM (6 vCPU, 40 GiB)
HP ProDesk 600 G4 DM (shed) .20 i5-8500T · 6 threads 31.1 GiB Proxmox VE 9 worker VM (4 vCPU, 12 GiB), USB radios, Wi-Fi bridge
Dell Pro Micro QCM1250 .181 Core Ultra 5 245T · 14 cores, iGPU + NPU 32 GiB (2× 16 GB) Talos v1.14.4 bare metal worker mini-pc-dell-ultra-worker
HP Elite Mini 800 G9 .82 i5-12500T · 12 threads 16 GiB (2× 8 GB) Talos v1.14.4 bare metal worker hp-elite-gen12, Zigbee stick
HP Elite Mini 600 G9 .88 (DHCP; was .22) i5-13500T · 20 threads 16 GiB (1× 16 GB) Talos v1.14.4 bare metal worker hp-elite-gen13
HPE ProLiant DL360 Gen9 (TrueNAS) .133 Xeon E5-2680 v4 · 28 threads 377.6 GiB TrueNAS 28.0 NFS/SMB storage, kopiur repository
Raspberry Pi 5 .15 Cortex-A76 · 4 threads 7.7 GiB Ubuntu 26.04 Omni v1.12.4, 3 Proxmox providers, booter v0.4.0 (PXE), Technitium DNS, ProxCenter

Kubernetes nodes

Cluster talos-prod-cluster-v2: 7 nodes, all Ready, Talos v1.14.4, Kubernetes v1.37.1.

Node Host CPU (alloc) Memory (alloc) CPU used Memory used Memory requested Pods Accelerators
talos-prod-cluster-v2-control-planes-7mxjxv .21 VM 3.95 11.0 GiB 0.72 6.7 GiB 4.2 GiB 20 —
talos-prod-cluster-v2-gpu-workers-jxfkdr .14 VM 29.95 97.6 GiB 5.77 40.5 GiB 58.3 GiB (60%) 75 nvidia.com/gpu=2
talos-prod-cluster-v2-hp-sff-workers-h76ncc .21 VM 5.95 38.6 GiB 1.75 16.5 GiB 19.0 GiB (49%) 52 gpu.intel.com/i915=4
talos-prod-cluster-v2-hp-micro-workers-pstbbd .20 VM 3.95 11.1 GiB 0.17 3.5 GiB 2.5 GiB 20 —
mini-pc-dell-ultra-worker .181 bare metal 13.25 29.2 GiB 0.87 10.2 GiB 13.5 GiB (46%) 69 gpu.intel.com/i915=4, NPU extension (no device plugin yet)
hp-elite-gen12 .82 bare metal 11.25 13.6 GiB 1.05 5.7 GiB 6.9 GiB (51%) 40 gpu.intel.com/i915=4
hp-elite-gen13 .88 bare metal 19.25 13.2 GiB 0.78 2.3 GiB 5.7 GiB (43%) 24 gpu.intel.com/i915=4
Total 87.55 214.3 GiB 11.1 85.4 GiB 110.1 GiB 300

"Memory requested" counts running pods only.

Storage

Longhorn: 66 volumes; 59 attached and healthy, 6 detached (normal), 1 faulted (nginx-example/storage). All 78 PVCs are Bound.

Node Longhorn disk Scheduling Size Free Scheduled (requested)
GPU worker gpu-bulk on 799.6 GiB 753.2 GiB 528 GiB
GPU worker ssd-flash (PLP, clone-ok) on 439.8 GiB 291.8 GiB 482 GiB (over-provisioned)
hp-sff worker hp-sff-ssd on 689.6 GiB 607.2 GiB 376 GiB
hp-sff worker hp-sff-ssd2 (new) on 919.5 GiB 898.6 GiB 61 GiB
hp-elite-gen13 hp-elite-nvme (Intel 660p, worn) on 476.7 GiB 467.4 GiB 0
mini-pc-dell-ultra-worker dell-ultra-ssd (new) on 233.6 GiB 226.0 GiB 20 GiB
hp-micro worker hp-micro-ssd off (Wi-Fi node) 849.5 GiB 833.2 GiB 0
hp-elite-gen12 — compute-only — — —

Each node also has a talos-ephemeral disk with scheduling off. Over-provisioning is 200%, so ssd-flash booking more than its size is allowed; its large users are pinned by the flash disk selector. New replicas and rebuilds land on the new disks.

GitOps health

98 Argo CD applications: 93 Synced/Healthy. Three are Progressing while pods restart after the volume recovery (database-redis-instance, my-apps-paperless-ngx, and my-apps-frigate, which waits on #3088). my-apps-posthog is OutOfSync (Healthy), and root reports Degraded from its children.

Findings

  1. Monitoring still scrapes the old .22 Proxmox host (physical-node exporters in monitoring/metrics/prometheus-stack/values.yaml, its network policy, and host-monitoring/inventory.yaml). Those targets are down now. Remove them or point them at the bare-metal nodes.
  2. hp-elite-gen13 and hp-elite-gen12 have 13.2–13.6 GiB allocatable. A second DIMM doubles either one. The Dell has 29.2 GiB after the DIMM move.
  3. Control plane is still one VM. The 1 → 3 move with the minis is the next planned change (#3071 step 4).
  4. The Intel 660p is past its rated writes. It is the first disk to replace; see the disk map.

Limits

  • Proxmox host RAM, CPU temperature and uptime were not re-collected; they did not change since the October 21 inventory.
  • Bare-metal RAM comes from Omni MachineStatus memory modules.
  • Disk serials stay out of this page. See the disk map for drive models and roles.