Convergence in NWChem fails

[vegas667]

Using NWChem for the DFT calculations and the command line input:

nohup python -u PyDP4.py --mm t --dft n -w nomesw -s DMSO --OutputFolder output/6638 alt final 6638_NMR > 6638-test.out

This runs for about 20 minutes and then errors out because it fails to converge. Modifying the input file to NWChem xxx.nw I can increase the number of iterations significantly and it completes.

I tried to add --UseExistingInputs thinking the completed xxx.nwo files would be read but this fails.

Then I tried both --OptCycles 500 and --OptStep 500 in separate arguments and these run the same time as above before they fail to converge again.

Can anybody please offer advise on how to get this process to completion?

Thanks.

[PeteGierth]

Not a developer but I am a user who is thinking about the software a bit right now. Can you provide the structures used? And/or a relevant nwchem input file that fails, and the output file?
Which calculation is failing? Is it the geom opt, energy, or NMR?