feat: NetworkTargetDict.uplinks, the NICs a bridge leaves through (NetOrk/netork#1325)
driver / standard (pull_request) Successful in 18s
driver / lint (pull_request) Successful in 14s
driver / types (pull_request) Successful in 1m1s
driver / test (3.12) (pull_request) Successful in 1m3s
driver / test (3.13) (pull_request) Successful in 59s
driver / lowest (pull_request) Successful in 36s
driver / build (pull_request) Successful in 32s
driver / security (pull_request) Successful in 1m12s
CI / driver (pull_request) Successful in 2m47s
driver / publish (pull_request) Skipped

A VM on a VLAN-aware bridge without a VLAN tag only gets an address when
the switch port behind the bridge carries an untagged network. To find
that port, the caller needs the bridge's physical NICs: their LLDP
neighbours name the switch and the port.

uplinks lists them in groups, one per way out of the bridge: a bond's
slaves together, a single NIC alone. A bond is one link, so a caller can
let one slave's neighbour speak for the others when lldpd sees only one.
Internal, VLAN and virtual ports are no uplinks. A missing key means the
driver cannot tell; an empty list is a bridge without uplink.
This commit is contained in:
christianmanivong committed 2026-10-30 09:27:23 +01:00
1 parent 145c2c668b
commit a5b7b540c6
5 files changed
+47 -6

No files matched your search

+6
View File
@@ -14,6 +14,12 @@ recorded here; their history is the git log.
3.12 venv no longer has.
### Added
- `NetworkTargetDict.uplinks` (optional): the physical NICs a bridge leaves the
host through, one group per way out -- a bond's slaves together, a single NIC
alone (`[["eno1", "eno2"]]`). A missing key means unknown, an empty list a
bridge without uplink. NetOrk follows their LLDP neighbours to the switch port
and its untagged VLAN; one slave's neighbour speaks for its bond
(NetOrk/netork#1325).
- `HostShutdownMixin.shutdown_host()`, on `DeviceTypeDriver`: shut the device itself
down cleanly through its own means; declared, so `hasattr` says whether a driver
can. No forced variant (NetOrk/netork#1636).
+8
View File
@@ -503,6 +503,14 @@ class ProxmoxDriver(HypervisorDriver):
def create_vm_snapshot(self, name, snapshot, description="", include_memory=False): ...
def get_network_targets(self):
# return List[NetworkTargetDict]: bridges and vnets a new VM's NIC may
# attach to; a bridge may name its physical NICs in `uplinks`, one group
# per way out: [["eno1", "eno2"]] for a bond, [["eno1"]] for a single NIC,
# no internal or VLAN ports -- leave the key out when they cannot be
# determined
...
def get_vm_cpu_types(self):
# optional — return List[VMCpuTypeDict]: the CPU models a new VM may get
# on this node, each with its cpuinfo flags and whether the node can run
+3 -1
View File
@@ -613,7 +613,9 @@ class HypervisorDriver(PackageManagementMixin, HealthMetricsMixin, DeviceTypeDri
List[NetworkTargetDict] - each entry's ``vlan_aware`` flag tells the
caller whether a ``NICConfigDict.vlan_tag`` may additionally be set
for a NIC using that target (see ``NetworkTargetDict`` for the
per-kind rules).
per-kind rules). ``uplinks``, where the driver can tell, names
the physical NICs a bridge leaves the host through, grouped:
a bond's slaves in one group, a single NIC in its own.
"""
...
+5 -5
View File
@@ -1310,17 +1310,17 @@ class NetworkTargetDict(TypedDict):
Distinguishes real bridges (Linux or OVS) from SDN network segments (vnets),
and tells the caller whether a separate ``vlan_tag`` may be applied on top:
- Linux bridge: vlan_aware reflects the bridge's own ``bridge_vlan_aware`` flag.
- OVS bridge: always vlan_aware (OVS bridges tag per-port regardless of a
dedicated "VLAN aware" setting).
- SDN vnet: never vlan_aware — the VLAN is already fixed by the vnet's zone/tag,
so a NIC attached to it must not also carry a ``vlan_tag``. That fixed VLAN
is surfaced via ``fixed_vlan_tag`` instead, for display purposes.
- OVS bridge: always vlan_aware (OVS tags per port, with no "VLAN aware" setting).
- SDN vnet: never vlan_aware — its zone/tag fixes the VLAN (``fixed_vlan_tag``).
``uplinks``: the physical NICs it leaves the host through, one group per way out
(a bond's slaves together, a single NIC alone); no internal, VLAN or virtual ports.
"""
name: str # Bridge, vnet or port group name, usable directly as NICConfigDict.bridge
kind: str # "bridge", "vnet" or "portgroup" (VMware: VLAN fixed like a vnet's)
vlan_aware: bool # True if a NICConfigDict.vlan_tag may be set on top of this target
fixed_vlan_tag: NotRequired[int | None] # vnet only: the VLAN ID already baked into it
uplinks: NotRequired[list[list[str]]] # NIC groups it leaves through; absent = unknown
class StorageTargetDict(TypedDict):
+25
View File
@@ -0,0 +1,25 @@
"""A network target can say which physical NICs it leaves the host through.
A VM on a VLAN-aware bridge without a VLAN tag only gets an address when the
switch port behind the bridge carries an untagged network. To look that port up,
the caller needs the bridge's physical NICs: their LLDP neighbours name the
switch and the port (NetOrk/netork#1325). They come in groups, one per way out
of the bridge: a bond's slaves together, a single NIC alone. A bond has no
neighbour of its own, and its slaves are one link, so one slave's neighbour
speaks for the others when they report none.
"""
from __future__ import annotations
from typing import get_type_hints
from napalm_device_types.models import NetworkTargetDict
def test_uplinks_is_an_optional_list_of_nic_groups():
assert "uplinks" in NetworkTargetDict.__optional_keys__
assert get_type_hints(NetworkTargetDict)["uplinks"] == list[list[str]]
def test_contract_core_stays_required():
assert {"name", "kind", "vlan_aware"} <= NetworkTargetDict.__required_keys__