SCF failed to converge in CDFT

How To
#1

Hello, I have encountered this problem in CDFT, SCF failed to converge, is there any way to help convergence?
This is my input file information, please correct.
$molecule
0 1
O 1.63832 -0.46460 -0.06923
H 0.80492 -0.93632 0.04035
H 2.16541 -0.75496 0.60000
$end

$rem
JOBTYPE FORCE
METHOD MP2
BASIS aug-cc-pVDZ
SCF_PRINT TRUE
CDFT TRUE
GUI = 2
SCF_ALGORITHM DIIS
CDFT_THRESH = 8
SYMMETRY = FALSE
BECKE_SHIFT = UNSHIFTED
XC_GRID = 000099000590
SCF_MAX_CYCLES 1000
MAX_DIIS_CYCLES=1000
$end

$cdft
2
1 1 2
-1 3 3
$end

#2

You are welcome to explore alternative convergence algorithms (SCF_ALGORITHM in $rem)

#3

As being pointed out in another related post, this constraint is just way too strong to be physical. In particular, in this monomer case it is truly equivalent to setting integer charges for atoms in this 3-atom molecule:

$cdft
1
1 1 2
-1
1 3 3
$end

From the SCF run you can see that the Lagrangian multiplier got extraordinarily large, which is a sign of the constraint being unphysical.

#4

For posterity, that previous discussion is here: