Error in mtnormalise with problems with gmfod in dwi2fod

Dear experts

I am currently preprocessing an fMRI dataset and ran into an error with mtnormalise that seems to be different to the ones here discussed.
The program fails with:

mtnormalise: [ERROR] Non-positive tissue balance factor was computed. Balance factors: -nan -nan -nan

While checking the steps before, I realized that the gmfod.mif is nearly plain white as it can be seen on the image.

Hence, vf.mif does not contain grey matter at all. This seemed quite strange to me, as wmfod.mif and csffod.mif look quite normal.
is there anything wrong in my >dwi2fod< ?

dwifslpreproc 64_dir_den.mif 64_den_preproc.mif -nocleanup -pe_dir AP -rpe_pair -se_epi b0_pair.mif -eddy_options " --slm=linear --data_is_shelled"

dwibiascorrect ants 64_den_preproc.mif 64_den_preproc_unbiased.mif -bias bias.mif -force 
dwi2mask 64_den_preproc_unbiased.mif mask.mif
dwi2response dhollander 64_den_preproc_unbiased.mif wm.txt gm.txt csf.txt -voxels voxels.mif
dwi2fod msmt_csd 64_den_preproc_unbiased.mif -mask mask.mif wm.txt wmfod.mif gm.txt gmfod.mif csf.txt csffod.mif

I am quite new to MRtrix so I don’t know if I missed something obvious here, any help and comments are very welcome!

In case this helps, here is what the gm.txt looks like

Is there maybe something wrong with the gm.txt that I do not see?

I suspect your data are single-shell? In which case, it’s not appropriate to use full-blown multi-shell multi-tissue CSD with 3 tissue types. You can run a two-tissue MSMT-CSD decomposition (as suggested e.g. in this post), or use the SS3T approach (noting that this is not part of this project, and not something that we will provide support for).

Thanks a lot, you are absolutely right about the data being single-shell.
Your suggestion to use two-tissues MSMT-CSD worked very well.