rorapp/cluster-knaster
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

proxmox: Site-Datei und pve-host-Rolle an die echten Knoten (pve44–pve61) angepasst
ci / test (push) Failing after 27s
Details

Browse Source 
proxmox.yaml mit gelesenen PCI-IDs/IOMMU-Gruppen, eine GPU-VM pro Knoten
mit allen Karten, DHCP, Templates je Knoten. pve-host nur für GPU-Knoten:
vfio-pci-IDs auf dem Knoten gelesen, softdeps, sysfb aus, initramfs-
Handler, Vendor-Snippet für den Gast-Agenten, Replikation erst wenn
der Container existiert.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Robert Rapp
2026-10-12 18:58:34 +02:00
co-authored by Claude Opus 5.5
parent dfb2630517
commit 908c698589
8 changed files with 130 additions and 58 deletions
Download Patch File Download Diff File Expand all files Collapse all files
sites/knaster/proxmox.yaml
+22 -20
View File
@@ -1,31 +1,33 @@
# Proxmox-Zielbild für "knaster" (Eingabe für platform/proxmox/tofu und ansible).
# ANNAHMEN (Hosts, VM-IDs, IPs, PCI-Adressen) bis zum Inventarlauf – siehe docs/IMPLEMENTATION.md.
# Stand 12.10.2026: echte Knoten pve44–pve61. PCI-Adressen, IDs und IOMMU-Gruppen per
# `pvesh get /nodes/<knoten>/hardware/pci` gelesen. Gäste vorerst per DHCP (D5 zurückgestellt).
cluster:
endpoint: https://pve1:8006/
storage_vm: tank # ZFS-Pool auf SSD 2+3 (ADR-0002)
endpoint: https://192.168.100.210:8006/
storage_iso: local
template_vmid: 9000 # Ubuntu 24.04 Cloud-Image-Template (pro Node vorhanden)
bridge: vmbr0
vlan_svc: 20
gateway: 10.20.0.1
dns: 10.20.0.1
ssh_public_key_file: ~/.ssh/id_ed25519.pub
gateway: 192.168.100.1
dns: 192.168.100.120
ssh_public_key_file: ~/.ssh/id_ed25519.pub # Schlüssel auf dem Steuerrechner ops
lxc_template: debian-13-standard_13.6-1_amd64.tar.zst
# Ubuntu-24.04-Cloud-Image-Template je Knoten (VMIDs sind clusterweit eindeutig)
templates: {pve44: 9044, pve46: 9046, pve48: 9048, pve52: 9052}
pci_mappings: # Proxmox Resource Mappings (Datacenter → Resource Mappings → PCI)
gpu-ada-1: {node: pve-ada, path: "0000:01:00", id: "10de:27b2", subsystem_id: "", iommu_group: 0}
gpu-ada-2: {node: pve-ada, path: "0000:02:00", id: "10de:27b2", subsystem_id: "", iommu_group: 0}
gpu-p2000-1: {node: pve3, path: "0000:01:00", id: "10de:1c30", subsystem_id: "", iommu_group: 0}
gpu-p2000-2: {node: pve3, path: "0000:02:00", id: "10de:1c30", subsystem_id: "", iommu_group: 0}
gpu-p2000-3: {node: pve4, path: "0000:01:00", id: "10de:1c30", subsystem_id: "", iommu_group: 0}
gpu-m4000-pve44: {node: pve44, path: "0000:01:00", id: "10de:13f1", subsystem_id: "10de:1153", iommu_group: 1}
gpu-p2000-pve44: {node: pve44, path: "0000:02:00", id: "10de:1c30", subsystem_id: "10de:11b3", iommu_group: 8}
gpu-m4000-pve46: {node: pve46, path: "0000:01:00", id: "10de:13f1", subsystem_id: "10de:1153", iommu_group: 1}
gpu-p2000-pve46: {node: pve46, path: "0000:02:00", id: "10de:1c30", subsystem_id: "10de:11b3", iommu_group: 8}
gpu-m4000-pve48: {node: pve48, path: "0000:01:00", id: "10de:13f1", subsystem_id: "10de:1153", iommu_group: 1}
gpu-p2000-pve48: {node: pve48, path: "0000:02:00", id: "10de:1c30", subsystem_id: "10de:11b3", iommu_group: 8}
gpu-m4000-pve52: {node: pve52, path: "0000:01:00", id: "10de:13f1", subsystem_id: "10de:1153", iommu_group: 1}
vms: # GPU-Rollen als VM (ADR-0001)
gpu-main: {node: pve-ada, vmid: 201, cores: 6, memory_mb: 28672, disk_gb: 40, models_disk_gb: 100, gpu: gpu-ada-1, ip: 10.20.0.201/24, role: gpu-vllm, driver_generation: modern}
gpu-flex: {node: pve-ada, vmid: 202, cores: 6, memory_mb: 28672, disk_gb: 40, models_disk_gb: 100, gpu: gpu-ada-2, ip: 10.20.0.202/24, role: gpu-vllm, driver_generation: modern}
gpu-embed-1: {node: pve3, vmid: 221, cores: 2, memory_mb: 6144, disk_gb: 30, models_disk_gb: 20, gpu: gpu-p2000-1, ip: 10.20.0.221/24, role: gpu-llamacpp, profile: embed, driver_generation: legacy}
gpu-embed-2: {node: pve4, vmid: 222, cores: 2, memory_mb: 6144, disk_gb: 30, models_disk_gb: 20, gpu: gpu-p2000-3, ip: 10.20.0.222/24, role: gpu-llamacpp, profile: embed, driver_generation: legacy}
gpu-small: {node: pve3, vmid: 231, cores: 2, memory_mb: 8192, disk_gb: 30, models_disk_gb: 20, gpu: gpu-p2000-2, ip: 10.20.0.231/24, role: gpu-llamacpp, profile: chat, driver_generation: legacy}
vms: # eine GPU-VM pro Knoten mit allen Karten des Knotens (D2)
gpu-pve44: {node: pve44, vmid: 244, cores: 4, memory_mb: 12288, disk_gb: 40, models_disk_gb: 60, storage: local-lvm, gpus: [gpu-m4000-pve44, gpu-p2000-pve44], ip: dhcp, driver_generation: legacy}
gpu-pve46: {node: pve46, vmid: 246, cores: 4, memory_mb: 12288, disk_gb: 40, models_disk_gb: 60, storage: local-lvm, gpus: [gpu-m4000-pve46, gpu-p2000-pve46], ip: dhcp, driver_generation: legacy}
gpu-pve48: {node: pve48, vmid: 248, cores: 4, memory_mb: 12288, disk_gb: 40, models_disk_gb: 60, storage: local-lvm, gpus: [gpu-m4000-pve48, gpu-p2000-pve48], ip: dhcp, driver_generation: legacy}
gpu-pve52: {node: pve52, vmid: 252, cores: 4, memory_mb: 12288, disk_gb: 40, models_disk_gb: 60, storage: local-lvm, gpus: [gpu-m4000-pve52], ip: dhcp, driver_generation: legacy}
containers: # Infrastruktur als LXC (ADR-0001), HA + ZFS-Replikation (ADR-0002)
infra: {node: pve1, vmid: 101, cores: 4, memory_mb: 12288, disk_gb: 60, ip: 10.20.0.101/24, ha: true, replicate_to: pve2}
infra: {node: pve44, vmid: 101, cores: 4, memory_mb: 8192, disk_gb: 60, storage: tank, ip: dhcp, ha: true, replicate_to: [pve52, pve48]}
replication_schedule: "*/15"