Overlapping oxygens in crystal structure — what's going on?

This is a screenshot of a portion of a metal organic framework (MOF). These are two oxygens of two NMP solvent molecules. What I am wondering is why they are so close and overlapping like this. I checked the CIF file, and occupancy for all atoms in this structure is 1, so it doesn't seem to be the case that there is only one NMP molecule in two equally likely positions.

Could someone explain this? Thanks

It's a bit hard to answer without having coordinates file (*.res or *.xyz), but the simplest answer that I would expect is that these oxygen atoms belong to two different molecules that are disordered over the same position (i.e. they have only partial occupation on the site that in sum is 1). It's very common for f.e. thf coordinated to alkali-metals.

If you have shelx input for the structure look for the line that defines one of oxygens and see if 6th parameter is 11.0

O1    5   -0.180348    0.245831    0.298215    11.00000    0.02208    0.02526 =
^^


This means fixed (10) occupancy of 1.

If you have cif-file look for the atom loop

loop_
_atom_site_label
_atom_site_type_symbol
_atom_site_fract_x
_atom_site_fract_y
_atom_site_fract_z
_atom_site_U_iso_or_equiv
_atom_site_occupancy
_atom_site_symmetry_multiplicity
_atom_site_calc_flag
_atom_site_refinement_flags
_atom_site_disorder_assembly
_atom_site_disorder_group
O1 O -0.18035(6) 0.24583(6) 0.29822(5) 0.02794(8) Uani 1 1 d . . .


8th parameter is occupancy (_atom_site_occupancy).

These both examples were for not disordered atoms. If they are disordered then it looks like this:

shelx-ins/res

O1    4    0.297097   -0.198418   -0.064164    10.50000    0.34896    0.34935 =
^ occup. 1/2


or

FVAR       0.22777   0.51340   0.51266
...
O1    4    0.551971    0.536866    0.295322    21.00000    0.03100    0.03031 =
^ occup. bound to 2nd free variable
here 51%


cif-file:

O1_20 O 0.6673(9) 0.3140(7) 0.1724(3) 0.0434(14) Uani 0.513(3) 1 d PDU C 1
^