While running mdrun_mpi, I am getting following error. I have intel compiler 2018 in the cluster. And used openmpi 4.0.1 for gromacs mpi installation.
Abort(1094543) on node 0 (rank 0 in comm 0): Fatal error in PMPI_Init: Other MPI error, error stack:
MPIR_Init_thread(639)......:
MPID_Init(860).............:
MPIDI_NM_mpi_init_hook(689): OFI addrinfo() failed (ofi_init.h:689:MPIDI_NM_mpi_init_hook:No data available)
So, can anyone tell me the reason of this error? How can I solve this?
Hi Amit! Which line have you tried to type in the terminal before you got this error message? What files have you got in your folder from which you tried to simulate (cpt, tpr, mdp)?
Hi Inna Ermilova
I have used this command, mpirun -np 26 mdrun_mpi -v -multidir nvt{0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25}. And I didn't get any output file after supplying this command.
But later, instead of using mdrun_mpi I used serial mdrun like mdrun -s nvt0.tpr -deffnm nvt0. I am attaching the input file (nvt.mdp).
Now, I am trying to run remd but not getting any output or error. So, I am unable to understand what is happening and why?
Hey Amit! From what I see in your script you are trying to run REMD simulation. It might be so that you shall reinstall the software, particularly, because you get no output there. Normally if you are in trouble you shall get at least a log-file with error. It can be so that the one who have installed the software did not link libraries properly. Or maybe everything fails at grompp-stage. Did you manage to find trp-files in your folders? I would try to install separately a serial version and try to run it.
Hi Inna Ermilova
I have also installed the serial version of gromacs4.5.5 and mdrun is working fine there (but I can't do the REMD with serial version). And, yes I am getting tpr file after grompp_mpi.
Hi Amit! Then I would advice you to check out this page and see how they pass arguments to mdrun. http://www.gromacs.org/Documentation/How-tos/REMD
Thank You Inna Ermilova , I am looking at the tutorial.
Do you know how to calculate cofigurational entropy along the trajectory? Actually, in gromacs using gmx covar and gmx anaeig command I am getting entropy only for two frames. I also tried "For loop" to go through the whole trajectory but nothing changed. Do you have any idea? I also tried -b and -e flag but irrespective of the time range result was the same.
You are welcome! Actually, if you look at the function gmx anaeig you can notice that there is a default setting 2 frames. It can be changed by specifying with the flag -nframes. I am afraid that regardless what you write there it takes that default value. Perhaps, it can be a good idea to check the number of frames during the time interval and then write them after the flag. It might be a bug in the built-in routine.
Inna Ermilova ,
Nothing changed. However, changing options in gmx covar actually changing the output in gmx anaeig.
- Badges
- Science topic