Created attachment 1760891 [details]
NodeNetworkConfigurationPolicy to apply

Created attachment 1760891 [details]
NodeNetworkConfigurationPolicy to apply

Description of problem:
When doing some network configuration [1] and you have some devices marked as unmanaged at NetworkManager nmstate fails on those devices even if they are not part of the devices being configured.

Version-Release number of selected component (if applicable):
nmstate 1.0.2
NetworkManager 1.30

How reproducible: Always


Steps to Reproduce:
1.git clone https://github.com/qinqon/kubernetes-nmstate
2.cd kubernetes-nmstate
3.git checkout pin-network-manager-at-node
4.make NMSTATE_PIN=future cluster-up cluster-sync
5.cat nncp.yaml | ./cluster/kubectl.sh exec -it -n nmstate $(./cluster/kubectl.sh get pod -n nmstate -l component=kubernetes-nmstate-handler |grep nmstate |awk '{print $1}' |head -n 1) -- nmstatectl set

Actual results:
libnmstate.error.NmstateLibnmError: Activate profile uuid:a55c98f8-fadc-46cb-b79a-5864ce2bf4d5 iface:cali1a2cfcdd5b6 type: veth failed: error=nm-manager-error-quark: Connection 'cali1a2cfcdd5b6' is not available on device cali1a2cfcdd5b6 because device is strictly unmanaged (2)
command terminated with exit code 1



Expected results:
To succeed

Additional info:

Calico device are marked at unmanaged at NetworkManager configuration
[vagrant@node01 ~]$ cat /etc/NetworkManager/conf.d/001-calico.conf 
[keyfile]
unmanaged-devices=interface-name:cali*;interface-name:tunl*


To access ssh into the node you can do ./cluster/ssh.sh node01