You should use the Amazon EMR console or CLI to manage tags on Amazon EC$2$ instances that are part of a cluster instead of the Amazon EC$2$ console or CLI.

Why is this recommended?

Because changes that you make in Amazon EC$2$ do not synchronize back to the Amazon EMR tagging system

Because changes that you make in Amazon EC$2$ only synchronize back to the Amazon EMR tagging system hourly

Because changes that you make in Amazon EC$2$ only update after a system reboot

Because changes that you make in Amazon EC$2$ only synchronize back to the Amazon EMR tagging system after a cluster reboot