@wangchao There are a number of problems with your input.
- for ReaxFF you need to use a shorter timestep than the default, specifically when your system contains hydrogen atoms and when you are running at a very high temperature
- please also check your log file or screen output. There should be a warning about the neighbor list settings being changed due to your settings for fix reaxff/species. averaging over 100 steps is not going to work for anything with a large-ish timestep that is not a solid at the selected temperature. So I would first test without fix reaxff/species, use a shorter timestep and tweak timestep and neighbor list settings so you can get stable dynamics that can conserve energy
- the hbondchk warning has multiple causes, one is that you run in parallel with MPI but your distribution of atoms changes a lot over the course of the simulation. The default implementation of ReaxFF assumes that the system is already somewhat equilibrated and thus makes some guess for how many hydrogen bonds might need to be stored during the entire run. The error can happen when this limit is crossed. There are several ways to address this: a) switch to the KOKKOS version of ReaxFF, which is more robust in those cases, b) run without or with fewer MPI processes and use some OPENMP parallelization (can also be applied to KOKKOS, btw), c) break down the run into multiple small chunks where you store intermediate results in data files and the use “clear” and “read_data” to start over with adjusted heuristics for the amount of hydrogen bond storage required.
From remote it is impossible state which of these points impact your simulation or whether all of them in combination are the cause of the error message.
Tracking this down in stages and carefully is essentially your job. As mentioned before, this error has been reported multiple times and more details advice has been given (beyond the one listed by @simongravelle).
Please also let me add that using fix temp/berendsen is a particularly bad choice to heat up a system. Its algorithm is meant for bulk liquids at ambient temperature, but for high temperatures, it will increase the difference between hot and cold atoms, which is the opposite of what you usually want to achieve in these scenarios. I would thus switch to a dissipative thermostat, like one of the many langevin style thermostats supported by LAMMPS. Those will try to homogenize the temperature distribution instead.