It looks to me, to a first approximation anyway, like a triplet of triplets, such as you would expect from coupling with two different sets of two equivalent Hs. If you number the peaks 1,2,3.. from left to right, the triplets are {1,2,4}, {3,5,7} and {6,8,9} and the peaks should have relative intensities 1:2:2:1:4:1:2:2:1. Doesn't look too far off to me. Try to measure the peak splittings accurately - evidently 0.01 ppm isn't precise enough, as the peaks are all shown as 0.01 ppm apart, and it's evident to the eye that the splittings are not all the same. Why not try simulating it, if you have software (you could even try it in excel) - estimate the coupling constants, determine peak positions and intensities, clothe with a lineshape - and see if it looks like your spectrum.