Question about remodelling

I am new to Rosetta. And we hope to redesign a protein and let it easier to dock with another two protiens.  Hope to verify if my approach is correct here.

Step 1) I applied the Rosetta protocol - remodel and create the blueprint file by "~/rosetta/main/tools/remodel/getBluePrintFromCoords.pl"

Step 2) Edit the blueprint file and apply the commands ALLAAxc to all residues. 

Step 3) run the remodel with my pdb file and chain code input by "~/rosetta/main/source/bin/remodel.linuxgccrelease -database ~/rosetta/main/database -s <pdbfile>  -run::chain <chain>  -remodel:blueprint <blueprintfile> -overwrite"

Step 4) It generated the xxxx0001.pdf file which is said the best one and it also has .sc score file.  Moreover, other 5  pdb are created while it is said the 1.pdb is the same as the xxxx0001.pdb.

Now I have some questions :   

Q1) When I put the ALLAA in the first two residues position, I got error "File: src/core/pack/task/ResfileReader.cc:1577  On line 3, the pose does not have residue with chain=D, PDBnum=1."   The log file is attached.     However, when I left the first two residues empty and apply ALLAA to all the other residues, it will not have error and Step 4) can be achieved.  How is that reason ? 

Q2) Can I fully trust the xxxx0001.pdf file is the best score one ?  Then go to the lab for further testiing.  Should I apply docking protocol to compare it with the other 4 pdb mentioned ? 

Thanks