fcbis_restart_node_during_sync_with_pre_images.js fails here if the previous FCBIS test case fails to kill the backup cursor. Since we use a fire and forget command to kill the backup cursor, we don't find out if it failed for some reason. The node crashes after it exhausts all its FCBIS attempts trying to establish a backup cursor on the sync source.

In production this shouldn't be much of a problem since the backup cursor will timeout if we stop sending getMores as a keep-alive and because we don't expect FCBIS to be started so soon after a previously failed attempt (so we'd expect the backup cursor timeout to clean up the sync source).

Even if this does happen in production, the impact is low since the customer will just have to wait a little before retrying the FCBIS attempt.