Lost atoms in simulation with chains

I have system with DPD atoms, and there is some chains,
when I am trying to run in parallel I got error

ERROR: Lost atoms: original 142248 current 142016 (thermo.cpp:385)

If I run it on one processor it works fine, but even if I use command
communication cutoff to increase cutoff I have this error in parallel.
Do you have any ideas how to fix it?

You need to figure out why you are losing atoms. Likely
due to bad dynamics with atoms moving more than 1 processor
away between reneighborings.


Seems my atoms does not moving a lot, this happens even when I set cutoff equal to the size of the box,
but how to trace it?
Thanks in advance!

Then how are you losing atoms? Are they crossing a fixed boundary?
If you dump them, you should see some missing.

The cutoff size is irrelevant to losing atoms.