Description of problem:
Under specific circumstances when OVN loadbalancers go into ERROR create state, Kuryr will crash with the following stack trace, in a crash loop. Along the way, it will create an unlimited amount of loadbalancers in OVN, it will add a loadbalancer every time it comes back up after having died before.

The failed load balancer creation stems from a different problem for which we do not have the root cause yet and which was fixed after a database repair. 

However, due to the failed load balancer creation, the loadbalancer object comes back as null, and our code is not checking for that. Instead, it blindy assumes that loadbalancer is an object. The KuryrLoadBalancerHandlers then crash one after another, leading to a full stop of the entire Kuryr process.

The Kuryr controller is then killed by Kubernetes, restarted again, and the same dance starts again, filling up the neutron database with in ERROR loadbalancers. That exasperates any preexisting problem by adding a new issue (namely the generation of an unlimited number of invalid loadbalancers in the database).

I checked upstream, and "/usr/lib/python3.6/site-packages/kuryr_kubernetes/controller/drivers/lbaasv2.py", line 507 does not have a try/catch logic nor does it check for type safety, even in upstream master. Regardless of the initial issue, this would need to be fixed, as otherwise KuryrLoadBalancerHandler threads simply die at that part simply because a loadbalancer goes into ERROR state at a specific point during creation.

~~~
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry [-] Report handler unhealthy KuryrLoadBalancerHandler: AttributeError: 'NoneType' object has no attribute 'id'
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry Traceback (most recent call last):
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry   File "/usr/lib/python3.6/site-packages/kuryr_kubernetes/handlers/retry.py", line 81, in __call__
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry     self._handler(event, *args, **kwargs)
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry   File "/usr/lib/python3.6/site-packages/kuryr_kubernetes/handlers/k8s_base.py", line 84, in __call__
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry     self.on_present(obj)
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry   File "/usr/lib/python3.6/site-packages/kuryr_kubernetes/controller/handlers/loadbalancer.py", line 79, in on_present
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry     if self._sync_lbaas_members(loadbalancer_crd):
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry   File "/usr/lib/python3.6/site-packages/kuryr_kubernetes/controller/handlers/loadbalancer.py", line 178, in _sync_lbaas_members
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry     if self._sync_lbaas_pools(loadbalancer_crd):
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry   File "/usr/lib/python3.6/site-packages/kuryr_kubernetes/controller/handlers/loadbalancer.py", line 439, in _sync_lbaas_pools
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry     if self._sync_lbaas_listeners(loadbalancer_crd):
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry   File "/usr/lib/python3.6/site-packages/kuryr_kubernetes/controller/handlers/loadbalancer.py", line 534, in _sync_lbaas_listeners
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry     if self._sync_lbaas_loadbalancer(loadbalancer_crd):
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry   File "/usr/lib/python3.6/site-packages/kuryr_kubernetes/controller/handlers/loadbalancer.py", line 687, in _sync_lbaas_loadbalancer
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry     provider=loadbalancer_crd['spec'].get('provider'))
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry   File "/usr/lib/python3.6/site-packages/kuryr_kubernetes/controller/drivers/lbaasv2.py", line 147, in ensure_loadbalancer
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry     response = self._ensure_loadbalancer(request)
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry   File "/usr/lib/python3.6/site-packages/kuryr_kubernetes/controller/drivers/lbaasv2.py", line 704, in _ensure_loadbalancer
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry     result = self._create_loadbalancer(loadbalancer)
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry   File "/usr/lib/python3.6/site-packages/kuryr_kubernetes/controller/drivers/lbaasv2.py", line 507, in _create_loadbalancer
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry     loadbalancer['port_id'] = self._get_vip_port(loadbalancer).id
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry AttributeError: 'NoneType' object has no attribute 'id'
2021-10-14 16:38:05.094 1 ERROR kuryr_kubernetes.handlers.retry
2021-10-14 16:38:05.096 1 ERROR kuryr_kubernetes.handlers.logging [-] Failed to handle event {'type': 'MODIFIED', 'object': {'apiVersion': 'openstack.org/v1', 'kind': 'KuryrLoadBalancer', 'metadata': {'creationTimestamp': '2021-10-12T18:31:50Z', 'finalizers': ['kuryr.openstack.org/kuryrloadbalancer-finalizers'], 'generation': 5680, 'managedFields': [{'apiVersion': 'openstack.org/v1', 'fieldsType': 'FieldsV1', 'fieldsV1': {'f:metadata': {'f:finalizers': {'.': {}, 'v:"kuryr.openstack.org/kuryrloadbalancer-finalizers"': {}}}, 'f:spec': {'.': {}, 'f:endpointSlices': {}, 'f:ip': {}, 'f:ports': {}, 'f:project_id': {}, 'f:provider': {}, 'f:security_groups_ids': {}, 'f:subnet_id': {}, 'f:type': {}}, 'f:status': {}}, 'manager': 'python-requests', 'operation': 'Update', 'time': '2021-10-14T16:37:50Z'}],...
~~~

Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info: