Reconciling DatabaseSecretEngineConfig logs in twice instead of once.

[ameyp]

If logging in to the database fails, the operator sends a login to vault twice for each DatabaseSecretEngineConfig resource:

2023-01-10T15:56:30.084Z	INFO	Logging in for vault object database/postgresql/config/my-db	{"controller": "databasesecretengineconfig", "controllerGroup": "redhatcop.redhat.io", "controllerKind": "DatabaseSecretEngineConfig", "DatabaseSecretEngineConfig": {"name":"my-db","namespace":"vault"}, "namespace": "vault", "name": "my-db", "reconcileID": "2dff9e6f-dc86-4745-afac-32d13948e1de"}
2023-01-10T15:56:30.039Z	INFO	Logging in for vault object database/postgresql/config/my-db	{"controller": "databasesecretengineconfig", "controllerGroup": "redhatcop.redhat.io", "controllerKind": "DatabaseSecretEngineConfig", "DatabaseSecretEngineConfig": {"name":"my-db","namespace":"vault"}, "namespace": "vault", "name": "my-db", "reconcileID": "a582e812-04ae-466e-8006-a16cec81035b"}

There's of course an accompanying stacktrace, which I've not included above but can if it'll help. This is happening at an increasing interval, so I'm guessing there's some sort of backoff on retries, which is good. However, it does look like the operator tries to reconcile each resource twice in each retry? There are two different reconcileID values in each "cycle".

[raffaelespazzoli]

I saw the same, it seems un-armful, but we should still fix it. This resource is not different than the others, are all the resources doing it?

[ameyp]

Not sure, I only noticed it for this resource

[ameyp]

I can try adding logs for other resources to check

[raffaelespazzoli]

It might be because we update the resource with the finalizer. but that is a new change, 3 weeks ago it wasn't there.... so I don't know.