Too many atom sorting bins

Hello,
Due to the large domain, you will need to use larger bins.
You can do this by increasing 0.002 to a larger number (don't know the exact number you need).
Or you can set the bin size directly by using the neigh_modify command.

Possible reason and insight of neighbor command

1. As correctly pointed out by @mschramm, that the simulation box is too large for your case
2. since we need to understand the meaning of cutoff distance first, which is as follows

Cutoff distance = force cutoff + skin
Force cutoff >> it is the sum of the two particle radii
skin cutoff >> enlarges the cell size for neighbour list construction. As a result, the particles may migrate somewhat but the neighbour list remains correct.
(a) If the skin factor is zero, we must recreate the neighbour list in each timestep, which is computationally expensive.
(b) If the skin factor has a high value, the neighbour list is saved, but it must check many probable contact partners for each time step, which is also expensive

Hence possible ways to handle the error related to skin command is following
1. understand the dynamics of particles in the simulation
2. Reduce the time step or
3. choose the skin factor of the order of the particle radius (advisable)

Thank you and happy learning !!