mpirun has exited due to process rank 0 with PID 31678 on
node node3 exiting improperly. There are two reasons this could occur:
1. this process did not call "init" before exiting, but others in
the job did. This can cause a job to hang indefinitely while it waits
for all processes to call "init". By rule, if one process calls "init",
then ALL processes must call "init" prior to termination.
2. this process called "init", but exited without calling "finalize".
By rule, all processes that call "init" MUST call "finalize" prior to
exiting or it will be considered an "abnormal termination"
This may have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).
Despite being really specific, this is really vague. This can help people on your HPC support to debug what happened, but for anyone else, this is just an error, given without any background info. I have plenty of these errors from different origins. Each HPC will have different output parameters, as well different compilations (as OpenMPI vs IntelMPI).
Good luck debugging
Thanks Gustavo Baldissera for telling me this. But, how to solve this problem in PC is this is my PC problem? Can you guide me how to solve such kind of debugging problem? Thanks
Is this error reproducible? Are you running locally or on a cluster? Did you build VASP from source yourself?
Basically this error indicates that VASP exited rather fatal, you should find the reason for this in the VASP output. Depending on how you called VASP, this output has been redirected to a log file e.g. "mpirun vasp > log.out 2> log.errors" or when running through a batch system, you may find a logfile in the working directory. Most likely this problem is related to wrong VASP input.
Thanks Norman,
I am using Xshell to connect the cluster. My input files is ok but, I am not sure what happen to this. Can you explain how to solve this problem?
Generally there is most likely more output than the message you postet. We need to see more of the output lines.
Does the problem persist. Can't you just restart VASP?
Dear Norman thanks. Can you check this file and let me how to solve the problem? I attached the whole error file.
Obviously the required library libimf.so is not installed on the machine you run VASP. Try finding it it at the compile machine, since it is a fortran lib. The environent variable LD_LIBRARY_PATH must contain the folder containing the library file.
I use Xshell the interface between cluster and my PC. So, how I use this library on VASP to run on my PC there is just Xshell so how I can find this library ? Please guide me a little ?
You either ""run and compile"" in your PC, or in the cluster. Binary files are definitely NOT a good idea to transfer, as it will not be able to find the libraries.
You still didn't tell where you got that VASP built from. If it was already installed on the cluster you try to run, they might have an "modules" environemnt where you need to load other modules VASP depends on. Maybe the user support of that cluster can help you then,
Away from that, finding a file on a linux machine in the easiest way would be somehting like:
find / | grep libimf.so
Thanks Norman,
The VASP is already installed on cluster . I am not using Linux machine . Do you know Xshell files to use as a interface between PC and cluster so, I have no idea to find that file>
It may be useful to know what you have changed since the last successful calculation. Make step by step modifications, to understand what causes the problem, because the error message is too general.
Try running on 1 core and multiple cores. Try different NPAR, KPAR. Try removing line by line from your INCAR. Try running last job that did run correctly with current INCAR.
- Badges
- Science topic
- Similar topics
- Filing