Proxmox's cloud-init drive is a disk image and requires a storage with
content='images' — the same requirement as the root disk — not the
snippets storage. These are commonly different storages (e.g. 'local'
with content=snippets-only, 'local-zfs' with content=images), and real
Proxmox now creates the VM fine but fails at *start* time with "storage
'X' does not support content-type 'images'" once it tries to generate
the cloud-init ISO.
Found live: a real deployment created the VM successfully, and only
failed when the user started it manually on the Proxmox side.
nics[i]['mac'] is set via virtio=<mac>,bridge=... instead of the bare
virtio,bridge=... form, so a caller-supplied MAC actually takes effect
(needed for DHCP reservations created before the VM exists).
Real Proxmox's POST /nodes/{node}/storage/{storage}/upload only accepts
content in {iso, vztmpl, import} — content='snippets' is rejected
outright with a 400 ("does not have a value in the enumeration").
Snippets can only be written directly to the storage's filesystem path.
Found live, right after the previous multipart-upload fix: the VM
shell, disk import, and node-scoped storage selection all succeeded,
then create_vm_from_cloud_init failed with a 400 at the snippet write
step. Resolves the storage's path via the cluster storage config and
writes the file over SSH (base64-piped, to survive arbitrary YAML
content safely).
proxmoxer only builds a multipart request for io.IOBase values passed
as kwargs; a plain filename string (plus a nonexistent "data" field,
as the old code sent) goes out as an ordinary form-urlencoded POST
instead. Real Proxmox's /storage/{s}/upload endpoint expects an actual
file upload for "filename" and responds to anything else by closing
the connection with no HTTP response at all.
Found live: the VM shell, disk import, and node-scoped storage
selection all succeeded, then create_vm_from_cloud_init failed with
requests.exceptions.ConnectionError / RemoteDisconnected right at the
snippet upload step.
The cluster-wide /storage endpoint lists every storage regardless of
its "nodes" restriction, so _find_default_image_storage (and the
snippet-storage lookup) could pick a storage not actually available on
the node the VM is being created on. On a real server this stranded a
freshly-created VM shell with no disk attached: "qm importdisk" failed
with "storage 'local-lvm' is not available on node 'pve-02'" after the
VM (VMID 103) already existed. Querying /nodes/{node}/storage instead
fixes this, since Proxmox itself only lists what's available there.
Also adds get_image_storages() and an optional storage= override on
create_vm_from_cloud_init, so callers aren't stuck with auto-detection.
Proxmox's /storage API omits the "enabled" key entirely for storages that
were never explicitly toggled, rather than defaulting it to 1 — it isn't
present-and-falsy, it's just absent. Both _find_default_image_storage and
the snippet-storage discovery treated storage.get("enabled") as truthy-check,
so every storage without an explicit "enabled": 1 was silently excluded.
Confirmed live against a real Proxmox test server: local-lvm, local-zfs, and
fast-zfs all had content=images with no "enabled" key at all, causing
create_vm_from_cloud_init to always fail with "No storage with
content='images' found" despite multiple valid storages existing. All prior
tests used "enabled": 1 explicitly in their fixtures, masking the bug.
Fix: storage.get("enabled", 1) != 0 — absent or truthy means enabled, only
an explicit 0 excludes it. 4 new regression tests, 28 total pass.
Replaces the template-clone flow with: create empty VM shell, download the
cloud image on the node (cached by filename, optional checksum verification),
qm importdisk, attach as scsi0. NIC config, snippet upload, ssh keys, disk
resize, and start remain unchanged (already generic).
New helpers: _run_node_command (strict SSH exec with custom timeout and
non-zero-exit detection, unlike the best-effort _exec_ssh_command),
_download_cloud_image (idempotent download + checksum check),
_find_default_image_storage (content=images discovery, mirrors the existing
snippet-storage discovery).
24 tests pass (10 new: _run_node_command x2, _download_cloud_image x4, plus
rewrites of the 4 existing create_vm_from_cloud_init tests for the new flow).
Filters _get_node_network() to bridge/OVSBridge types only (excludes physical
NICs, bonds), plus SDN vnets from _get_sdn_vnets(). vlan_aware: Linux bridge
reflects its bridge_vlan_aware config flag; OVS bridge always true; SDN vnet
always false (VLAN already fixed by the vnet's zone/tag).
4 new tests: bridge/vnet filtering, Linux bridge vlan_aware flag, OVS bridge
always vlan_aware, SDN vnet never vlan_aware. All 15 tests in the file pass.
Test cases: single/dual NIC with VLAN tags or trunk config, per-NIC DHCP control,
disk resize parameter, snippet storage validation, IP wait timeout, destroy paths.
All TDD cases green.
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
- Replace fixed mgmt/capture dual-NIC parameters with generic nics list
- Loop over NICs to build net0, net1, ... config strings (access VLAN or trunk)
- Remove hardcoded 'ip link set eth1 up' Cloud-Init hack (caller responsibility)
- Add disk_resize_gb parameter for post-clone disk expansion (scsi0/virtio0/ide0/sata0)
- Make DHCP configuration per-NIC with sensible defaults (primary NIC only)
- Update docstrings and logging to reflect generic NIC architecture
_get_vm_disk_and_boot() now returns a list of disk dicts (name, size_mb)
instead of a single total. Each disk entry is read from the VM config
via /qemu/{vmid}/config or /lxc/{vmid}/config. The onboot flag is also
read from the same config endpoint.
Both QEMU VMs and LXC containers are covered. The 'disks' and 'onboot'
keys are added to every entry in the vms_snapshot.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
product_version is used as model only when it contains a space, indicating
a human-readable marketing name (e.g. "ThinkCentre M910x"). Part numbers
like "J26843-409" have no space and are skipped — product_name is used
instead (e.g. "NUC6CAYH" for Intel NUC).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The combined printf approach silently produced empty values when
product_name/version contained special chars or the shell split tokens
incorrectly. Read each /sys/class/dmi/id/ file via a separate cat,
collect lines, then apply vendor-specific model name selection:
- Intel NUC: product_name='NUC6CAYH' (marketing) preferred over
product_version='J26843-409' (part number)
- Lenovo: product_name='10MYS03U00' (type code, all-caps+digits) →
prefer product_version='ThinkCentre M910x' (marketing name)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
On Lenovo (and some other vendors) product_name contains the machine-type
code (e.g. "10MYS03U00") while product_version holds the marketing name
(e.g. "ThinkCentre M910x"). Read both and prefer product_version when it
is set and different from product_name.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
vendor, model and serial_number now come from /sys/class/dmi/id/
(sys_vendor, product_name, product_serial) via SSH, reflecting the
actual physical server rather than the Proxmox software layer.
Falls back to "Proxmox Server Solutions GmbH" / status.model if SSH
or DMI files are unavailable (e.g. bare-metal without SSH creds).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Implements get_system_config() in ProxmoxSystemMixin:
- cluster_name from /cluster/status (type=cluster entry)
- cluster_nodes list of online node hostnames
- timezone from /nodes/{node}/time
- hostname, ssh_port, ssh_password_auth with safe defaults
Used by NetOrk's sync task to name the NetBox Cluster after the
actual Proxmox cluster rather than falling back to the site name.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
After renaming /etc/hostname, Proxmox boots under the new node name and
looks for VM/CT configs in /etc/pve/nodes/<new>/qemu-server/. Without
migrating the directory first, all VMs appear missing after the reboot.
Now renames /etc/pve/nodes/<old>/ to /etc/pve/nodes/<new>/ while
pve-cluster is running (pmxcfs supports live rename). Also replaces
pvecm updatecerts -f with pvenode cert create --overwrite — pvecm
updatecerts restarts pve-cluster, unmounts /etc/pve and can crash VMs.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
netOrk's generic reboot helper (_do_reboot) calls conn._send_command()
which did not exist on ProxmoxDriver — the resulting AttributeError was
silently swallowed, so reboot commands never executed on Proxmox nodes.
Delegates to the existing _exec_ssh_command so all generic SSH-based
actions (reboot, dns_port check, etc.) work correctly.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Implements set_hostname on ProxmoxSystemMixin:
1. Writes /etc/hostname (short name, base64-safe transfer)
2. Replaces old hostname in /etc/hosts via Python regex + base64
3. Updates /etc/mailname if present
4. Updates Postfix myhostname via postconf -e if installed
5. Applies hostname immediately at runtime via hostname(1)
6. Regenerates Proxmox node TLS certificate via pvecm updatecerts -f
(falls back to pvenode cert create → pveproxy restart)
Accepts bare hostname or FQDN. A reboot is required for the Proxmox
node name to update in the web UI / cluster — the driver logs this.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
get_vlans() previously parsed 'bridge vlan show' which trunks all 4094
VIDs by default on VLAN-aware bridges, producing phantom VLAN entries.
Now derives real VLAN assignments from VM/container netN bridge=,tag=
config. get_device_warnings() now reports updates_available with
severity "warning" and a full package list/title, matching the format
used when the warning is refreshed via the updates API (previously
alternated between "info" and "warning" for the same content).
get_device_warnings() now reports lldpd_not_installed (action
install_lldpd) when lldpd is absent, matching the existing
OpenWrt pattern. run_device_action("install_lldpd") installs and
enables lldpd via apt/systemd so LLDP topology links to directly
connected switches can be discovered.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Mirrors the robust fallback pattern from napalm-opnsense so the driver
works regardless of which key name the caller passes in optional_args.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
