Closed UncleSamSwiss closed 1 year ago
This could of course be a bug in the provider, but I somehow doubt it.
But, I should perhaps note that I have a service in the cluster that is of type LoadBalancer which of course created a new OpenStack load balancer.
@UncleSamSwiss This is definitely a bug of the OpenStack provider. However I never manage to get it reproduced consistently and never had an actual effect (I suspect this is when reading the state after an apply). When it happens, you can apply a second time and everything should be fine.
I had the following configuration:
Then I changed the nodes_count for the agent pool (and removed the bootstrap flag):
This lead to an inconsistent plan: