nsm-operator pod restart caused by segmentation violation

[szvincze]

Describe the bug Sometimes nsm-operator pod restarts when deploying the config/samples/nsm_v1alpha1_nsm-registry-k8s.yaml example. The cause of the issue is a panic because of a segmentation violation (please see the attached log for more details):

panic:` runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x18 pc=0x12c6dae]

To Reproduce I observed the fault during development of an integration test case, but it was quite rare. However it can be reproduced by deploying and deleting the above mentioned example by the following commands several times in a row:

kubectl apply -f config/samples/nsm_v1alpha1_nsm-registry-k8s.yaml
kubectl delete -n nsm nsm nsm-sample-registry-k8s

Expected behavior No pod restart.

Screenshots, CLI output and logs nsm-operator-pod-restart.log

Desktop (please complete the following information):

• Platform: Kubernetes in Docker (kind)
• Platform version: v1.23.4
• Operator version: latest

Additional context

[szvincze]

Here is the proposed fix for the issue.

I managed to reproduce the situation by the method mentioned in the above comment and now the log contains the proper info and there is no pod restart:

INFO    controller.nsm  Failed to update status {"reconciler group": "nsm.networkservicemesh.io", "reconciler kind": "NSM", "name": "nsm-sample-registry-k8s", "namespace": "nsm", "nsm": "nsm/nsm-sample-registry-k8s", "Error": "Operation cannot be fulfilled on nsms.nsm.networkservicemesh.io \"nsm-sample-registry-k8s\": the object has been modified; please apply your changes to the latest version and try again"}

[szvincze]

Fix is merged via PR#53.