The optimizer recently experienced a glitch where prices weren't calculated correctly when a customer is asking an instance that cannot provide all of the resources the costumer's component requires (CPU / Memory, etc. )- that is, customer requests more CPUs than any instance can provide (for example, the component needs 100 CPUs, but the largest instance currently has only 96 CPUs). As a result, the optimizer returns the best result for the other components, but ignores the large machine - without returning any error.
The optimizer recently experienced a glitch where prices weren't calculated correctly when a customer is asking an instance that cannot provide all of the resources the costumer's component requires (CPU / Memory, etc. )- that is, customer requests more CPUs than any instance can provide (for example, the component needs 100 CPUs, but the largest instance currently has only 96 CPUs). As a result, the optimizer returns the best result for the other components, but ignores the large machine - without returning any error.