ANOVA and Unbalanced Data

Suppose I have a continuous response variable. I wish to compare the response variable between three groups while controlling for another covariate let's call it "size" which includes two levels: large and small. In this way I have a two-way ANOVA.

Y= Group_i + Size_j + error

I cannot remember how to address the issue of unbalanced data. Let's suppose Group A only consists of the large size. How do I solve this issue? Or is it an issue at all?