Dear Zhengping,
(I am splitting my answer to your message in two messages -- I suggest
that you use a message, with a proper Subject, for each topic.)
> Another problem is from OrderN method. If I set On.ChemicalPotential,
> ON.ChemicalPotentialOrder,
> ON.ChemicalPotentialUse and ON.ChemicalPotentialTemperature in the input
> file, serial-siesta
> works well. For para-siesta, I get the following error:
>
> ordern: enum = 1984.0000
> Fatal error in MPI_Allreduce: Message truncated, error stack:
...
As the manual says, the ON.ChemicalPotential option does not work in
parallel. You could use the value you get from the serial run as input
for the parallel calculation.
Regards,
Alberto
