I had a look into the attachment and your input file
(water_TIP3P_gcmc_100ps_muVT_300K.in) contains a line:
# molecules to add/remove
molecule h2omol 1water_mod.data
but the files you sent in your contribution do not contain such
a file. The closest file in the attachment is "1water.data"
If I include *this* molecule file you provided, the simulation
does indeed run (see log attached, version 2018.08)
Your system (1800 TIP3P at liquid density, but with one large empty
bubble at a box vertex) is probably too dense and too large for getting
meaningful results in an acceptable time with GCMC. The huge neighbor
list (neighbor 3.0 bin) and the high Ewald accuracy won't help either.
M.
log.lammps (7.41 KB)