Hmm ... a consistent system. If you are solving an equation Ax = b, and we're staying in one space (what does that mean?) then we want a square matrix A. In this case, we're talking about 8 rows and 8 columns. To solve the equation Ax = b, we use (at least I use) an augmented matrix with 8 rows but 9 columns. The b (target vector) will be the 9th column. Then we row reduce to RREF. Only in the RREF (row-reduced echelon form, I assume you know what that means?) will you see those leading 1's. If the rank of A is 5, there will be 5 leading 1's, in the RREF that is. When you have the RREF, how do you know whether the system is consistent or not?
If b = the 0 vector, then yes, the system is consistent, but the only solution is the trivial one. If the rank is 8 (in this case, or n in the general case) that will be the only solution. This means the columns of A are ... you should know what. If the rank of A is less than 8 (in this case, or less than n in the general case) then the columns of A are not "you know what" and there will be how many solutions?
I hope this is clear. You seem to be using slightly different nomenclature than I'm used to.