Log message spam when storage is disabled in Boundary controller

[sym-stiller]

Describe the bug
The Boundary Controller started to log a lot of messages at error level which should be logged at debug level instead.
The message that is bothering us is the following: {"id":"7qLhsQ3dbf","source":"https://hashicorp.com/boundary/boundary-controller-5f5d7b54bd-x88m2/controller","specversion":"1.0","type":"error","data":{"error":"job \"storage_plugin_storage_bucket_secret-rewrapping-job\" not registered on scheduler","error_fields":{},"id":"e_QCwetHU7Pd","version":"v0.1","op":"scheduler.(Scheduler).schedule","info":{"msg":"error starting job"}},"datacontentype":"application/cloudevents","time":"2024-11-19T14:13:50.632082703Z"}.

To Reproduce
Steps to reproduce the behavior:

1. Run a Boundary controller instance with all storage features disabled (reproducable at all log_levels)
2. Observe the log messages

Expected behavior
This message should be logged at debug level or not logged at all. We are not intending to use the storage features, so there is no need to start this job. The code should handle this use case more gracefully and not throw errors into the log.

Additional context
The behavior started after we upgraded to Boundary version 0.17.0, and we are not using any of the storage features of Boundary (and are not planning to do so).
At 3 replicas of the Boundary controller, we are looking at over 4000 of these log messages each day, which clog up our long-term storage for logs.