Proxmox VE 9 operations

The problem

Virtualisation has to stay understandable even when a host fails to boot, a disk reports bus errors, storage layouts need replacing, or guests must move to another machine.

Screenshot evidence

The landscape capture is the Proxmox VE node dashboard. The seven portrait captures are the Proxmox mobile Resources list, split across the LXC and KVM inventory. A guest name in this list is deployment evidence only; it is not the guest application’s own interface.

Proxmox LXC inventory IDs 100 to 105 — **LXC inventory, part 1.** Running LXC guests 100–105 are shown in the Proxmox Resources view.

Proxmox LXC inventory IDs 106 to 111 — **LXC inventory, part 2.** The next group of containers includes active and deliberately stopped guests.

Proxmox LXC inventory IDs 112 to 118 — **LXC inventory, part 3.** This segment records remote-access, deployment and application containers.

Proxmox LXC inventory IDs 118 to 125 — **LXC inventory, part 4.** The inventory continues through cloud, game-library and network-bridge guests.

Proxmox LXC inventory IDs 125 to 500 — **LXC inventory, part 5.** The final LXC segment includes camera, VPN proxy, DNS filtering and stopped test workloads.

Proxmox KVM inventory IDs 115 to 202 — **KVM inventory, part 1.** The Proxmox VM list records Windows, Wazuh, Ubuntu Docker and other virtual machines with their current state.

Proxmox KVM inventory IDs 201 to 234 — **KVM inventory, part 2.** The second VM segment includes the active Ubuntu Docker and 3CX production guests alongside stopped lab systems.

What I did

I operate Proxmox VE 9.2.3 as the main virtualisation platform, using both KVM virtual machines and LXC containers. The screenshot snapshot includes 28 LXC containers and 10 KVM virtual machines spanning monitoring, security, remote access, collaboration, DNS, game services, camera systems and application hosting.

Upgraded the platform to the Proxmox VE 9 generation
Diagnosed UEFI boot and ACPI/ATA failure symptoms from kernel logs
Planned backup-and-restore migrations between hosts and storage layouts
Checked bridge names, static addressing, mount points and passthrough dependencies before guest moves
Used guest separation to contain security, voice, cloud, DNS, remote-access, game and camera workloads
Maintained a mixed estate including Zabbix, Open WebUI, Matrix, Actual Budget, Wallos, Guacamole, Gitea, Pangolin, Nextcloud, Agent DVR, ProtonProxy and AdGuard Home
Kept retired and test workloads stopped rather than misrepresenting them as active production services

Captured estate

LXC containers28

KVM virtual machines10

Running25

Stopped13

The full point-in-time inventory and cropped mobile screenshots are available on the infrastructure screenshots page.

Implementation notes

Representative commands from the way I diagnose and maintain this project. Public examples use placeholders instead of credentials or sensitive addresses.

Representative operational commands

# Platform and guest inventory
pveversion -v
qm list
pct list
pvesm status

# Inspect an individual workload
qm config <VMID>
pct config <CTID>

# Evidence-led host troubleshooting
journalctl -b -p warning --no-pager
smartctl -a /dev/sdX

# Back up before a migration or storage change
vzdump <VMID> --storage <backup-store> --mode snapshot --compress zstd

Result

The result is a practical virtualisation platform I can maintain, migrate and troubleshoot rather than a collection of opaque appliances.

Current release context: Proxmox publishes a 9.2-1 installer track; the live host package level is 9.2.3.