Issues converging (and probably setting up) CDFT calculation

[Joanna Z]

Dear all, 

    I've been trying to run a CDFT calculation of a two-molecule system: Cu-based molecule and a Ni-based one. In my case the overall charge is zero, and I would like the Ni-based molecule to be of triplet multiplicity (so the other one is in doublet state). I have been trying to impose that constraint via CDFT in a couple different ways:
    1) by setting number of alpha and beta electrons on the Ni molecule (to 59 and 57);
    2) by setting megnetization on the Ni molecule equal to 2;
    3) by setting the number of alpha electrons and magnetization (59 and 2, correspondingly);
But in each case though the CDFT calculation starts, it has trouble converging (what's more, in cases 1 and 3 the deviation from the target in the last step was larger than in the initial one). The jobs end due to SCF not converging. 
I am attaching input file to the first type of job, output files, and the xyz file, as well as the first and last frames of the CDFT data file (it is too large to post here). 

I am attaching input file to the first type of job, output files, and the xyz file, as well as the first and last frames of the CDFT data file (it is too large to post here). 

I will be very grateful for any pointers as to how to set it up properly and what to pay attention to while running CDFT calculations.