-
Bug
-
Resolution: Not a Bug
-
Minor
-
None
-
4.12.z
-
Quality / Stability / Reliability
-
False
-
-
3
-
Moderate
-
No
-
None
-
None
-
None
-
Metal Platform 245
-
1
-
None
-
None
-
None
-
None
-
None
-
None
-
None
Description of problem:
[BM-IPI] The effect of this bug are the same of: https://bugzilla.redhat.com/show_bug.cgi?id=2094090 In this case using a custom VLAN name rather then the "name.number" standard leave again the VM in provisiong state.
Version-Release number of selected component (if applicable):
OCP 4.12.z
How reproducible:
Install a new cluster using a VLAN on top of a bond without a 'dot' in the name.
Actual results:
Machines keeps the Provisoned State
Expected results:
Machines move on Running state
Additional info:
The workaround described on https://access.redhat.com/solutions/6964105 works also on our scenario.
Looking at: https://review.opendev.org/c/openstack/ironic-python-agent/+/848866/1/ironic_python_agent/hardware.py
The blow funcion seems only contemplate a VLAN with a ('.') in it:
======================================
def _is_vlan(self, interface_name):
# A VLAN interface does not have /device, check naming convention
# used when adding VLAN interface
interface, sep, vlan = interface_name.partition('.') <---------- THIS
return vlan.isdigit()
======================================
- account is impacted by
-
RFE-4976 Enable the usage of custom VLAN names in OCP4
-
- Backlog
-