-
Bug
-
Resolution: Unresolved
-
Major
-
None
-
None
-
None
-
None
-
Quality / Stability / Reliability
-
0.42
-
False
-
-
True
-
-
None
Description of problem:
Create cluster user defined primary network, migrate VM from vCenter to OCP cluster, map the VM network to default network (cluster user defined primary network) in cudn-ns namespace, conversion pod with correct network, the plan executed successfully.
But the VMI failed to create on target cluster, the virt-launcher pod hit error as below, it tried to add openshift-ovn-kubernetes default network to VM, but that network is not exist.
ERRORED: error configuring pod [cudn-ns/virt-launcher-mtv-feature-win2019-p4m4f] networking: Multus: [cudn-ns/virt-launcher-mtv-feature-win2019-p4m4f/84cab598-8e86-42f8-b244-624c869f3da3]: error loading k8s delegates k8s args: TryLoadPodDelegates: error in loading K8s cluster default network from pod annotation: tryLoadK8sPodDefaultNetwork: failed getting the delegate: getKubernetesDelegate: cannot find a network-attachment-definition (default) in namespace (openshift-ovn-kubernetes): networkattachmentdefinition.k8s.cni.cncf.io "default" not found
Version-Release number of selected component (if applicable):
MTV 2.10.0-6 build (IIB: forklift-fbc-prod-v420:on-pr-52e94dd9cd11eaf7e225644c31cb966f2c8e4494) with CNV 4.20.0
How reproducible:
Always
Steps to Reproduce:
1) Create the namespace: cudn-ns
$ cat cudn-ns.yaml apiVersion: v1 kind: Namespace metadata: name: cudn-ns labels: k8s.ovn.org/primary-user-defined-network: "" $ oc apply -f cudn-ns.yaml namespace/cudn-ns created $ oc label namespace/cudn-ns cluster-udn=cudn-test namespace/cudn-ns labeled
2) Create the cluster user defined network: cudn-l2-net
$ cat cudn.yaml apiVersion: k8s.ovn.org/v1 kind: ClusterUserDefinedNetwork metadata: name: cudn-l2-net spec: namespaceSelector: matchLabels: cluster-udn: cudn-test network: topology: Layer2 layer2: role: Primary ipam: lifecycle: Persistent subnets: - 10.232.18.0/24 $ oc apply -f cudn.yaml clusteruserdefinednetwork.k8s.ovn.org/cudn-l2-net created $ oc get net-attach-def -A NAMESPACE NAME AGE cudn-ns cudn-l2-net 142m
3) In cudn-ns namespace, create the vCenter provider: v7-us-cudn, OCP cluster: ocpd-04-host
apiVersion: forklift.konveyor.io/v1beta1
kind: Provider
metadata:
name: ocpd-04-host
namespace: cudn-ns
spec:
secret: {}
type: openshift
4) Create a cold migration plan to migrate VM: mtv-feature-win2019 from v7-us-cudn to ocp cluster: ocpd-04-host to target project: cudn-ns, map the VM network to default network
5) Login to the conversion pod, check the pod with pod network: 10.129.3.90/23 and cudn network: 10.232.18.51/24, and the default route: 10.232.18.1 is using udn network , the pod can ping outside network
$ oc describe pod cold-win2019-dhcp-cudn-vm-2202-2gb6p Name: cold-win2019-dhcp-cudn-vm-2202-2gb6p Namespace: cudn-ns Priority: 0 Service Account: default Node: qemtvd-04-rngzt-worker-0-nw6nx/192.168.2.81 Start Time: Mon, 22 Sep 2025 20:04:05 +0800 Labels: forklift.app=virt-v2v migration=57c602d7-3144-4f5d-9740-52b216fdbb06 plan=3806bb9a-300e-43d4-acc4-a1d934baf7c3 vmID=vm-2202 Annotations: k8s.ovn.org/open-default-ports: - protocol: tcp port: 2112 - protocol: tcp port: 8080 k8s.ovn.org/pod-networks: {"cudn-ns/cudn-l2-net":{"ip_addresses":["10.232.18.51/24"],"mac_address":"0a:58:0a:e8:12:33","gateway_ips":["10.232.18.1"],"routes":[{"des... k8s.v1.cni.cncf.io/network-status: [{ "name": "ovn-kubernetes", "interface": "eth0", "ips": [ "10.129.3.90" ], "mac": "0a:58:0a:81:03:5a", "dns": {} },{ "name": "ovn-kubernetes", "interface": "ovn-udn1", "ips": [ "10.232.18.51" ], "mac": "0a:58:0a:e8:12:33", "default": true, "dns": {} }] openshift.io/scc: forklift-controller-scc seccomp.security.alpha.kubernetes.io/pod: localhost/profiles/unshare.json security.openshift.io/validated-scc-subject-type: user Status: Succeeded
6) Plan executed successfully
7) Check the VM yaml file on target cluster as below:
$ oc get vm mtv-feature-win2019 -o yaml| grep interface -A 5
interfaces:
- binding:
name: l2bridge
macAddress: 00:50:56:b4:ad:82
model: virtio
name: net-0
$ oc get vm mtv-feature-win2019 -o yaml| grep network -A 3
networks:
- name: net-0
pod: {}
8) VM failed to boot, the pod: virt-launcher-mtv-feature-win2019-p4m4f with error below, more details in attached files: virt-launcher-mtv-feature-win2019-p4m4f-describe, virt-launcher-mtv-feature-win2019-p4m4f-yaml
$ oc get pod virt-launcher-mtv-feature-win2019-p4m4f -o yaml apiVersion: v1 kind: Pod metadata: annotations: descheduler.alpha.kubernetes.io/request-evict-only: "" k8s.ovn.org/pod-networks: '{"cudn-ns/cudn-l2-net":{"ip_addresses":["10.232.18.53/24"],"mac_address":"0a:58:0a:e8:12:35","gateway_ips":["10.232.18.1"],"routes":[{"dest":"172.30.0.0/16","nextHop":"10.232.18.1"},{"dest":"100.65.0.0/16","nextHop":"10.232.18.1"}],"ip_address":"10.232.18.53/24","gateway_ip":"10.232.18.1","tunnel_id":60,"role":"primary"},"default":{"ip_addresses":["10.129.3.105/23"],"mac_address":"00:50:56:b4:ad:82","routes":[{"dest":"10.128.0.0/14","nextHop":"10.129.2.1"},{"dest":"100.64.0.0/16","nextHop":"10.129.2.1"}],"ip_address":"10.129.3.105/23","role":"infrastructure-locked"}}' k8s.ovn.org/primary-udn-ipamclaim: mtv-feature-win2019.net-0...... v1.multus-cni.io/default-network: '[{"name":"default","namespace":"openshift-ovn-kubernetes","mac":"00:50:56:b4:ad:82","ipam-claim-reference":"mtv-feature-win2019.net-0"}]' vm.kubevirt.io/os: windows $ oc describe pod virt-launcher-mtv-feature-win2019-p4m4f Name: virt-launcher-mtv-feature-win2019-p4m4f Namespace: cudn-ns Priority: 0 Service Account: default Node: qemtvd-04-rngzt-worker-0-nw6nx/192.168.2.81 Events:....... ERRORED: error configuring pod [cudn-ns/virt-launcher-mtv-feature-win2019-p4m4f] networking: Multus: [cudn-ns/virt-launcher-mtv-feature-win2019-p4m4f/84cab598-8e86-42f8-b244-624c869f3da3]: error loading k8s delegates k8s args: TryLoadPodDelegates: error in loading K8s cluster default network from pod annotation: tryLoadK8sPodDefaultNetwork: failed getting the delegate: getKubernetesDelegate: cannot find a network-attachment-definition (default) in namespace (openshift-ovn-kubernetes): networkattachmentdefinition.k8s.cni.cncf.io "default" not found
Actual results:
In step8: The VM failed to start
Expected results:
In step8: The VM start successfully
Additional info:
