The problem is that you will not be able to run a monolith without a microservice. In fact, you'd be able to run an entire project without a lot of trouble.
This is so much easier to debug and debug. Even with simple tools, it's almost impossible to debug code without refactoring a few dozen layers and building up another microservice.
The question is what does the microservice plan on doing? When it needs to scale, it's an issue of how much time is spent on the development of the project. If the micro service plan is a waste of time, it will be a problem.
This is especially true if you only have 2 servers running the microservice, and you can use docker containers to build and maintain your own servers which support everything (apache, docker); for your projects, it probably takes less than 5 minutes to figure it out, and when you start getting serious about running them with one or more different pieces of infrastructure you don't think twice about spending that much time in production.
I'm not a huge fan of microservices: I think they will be hard to scale even before they become more of a problem.