Dear Vasp developers,
I am facing a problem with some of my calculations. Some of them are showing the following err. message:
srun: error: hyperion-116: tasks 48-95: Killed
s for job step to finish.
14T23:36:14 ***
o object at address 0x1c)
total write size = 800, bytes this sub-write = 800, bytes actually written = 18446744073709551615, offset = 0
Surprisingly, one of them showed a bug report in the err.* file asking me to submit a bug report. Here it is the zip file containing the required input and output files of the bugged one. I hope you can bring some light on this, about what is happening here.
Thank you very much in advance for your kind reply.
Ramon
Bug report in Pt cluster relaxation
Moderators: Global Moderator, Moderator
-
ramon_bergua
- Newbie
- Posts: 5
- Joined: Thu Jan 12, 2023 6:02 pm
Bug report in Pt cluster relaxation
You do not have the required permissions to view the files attached to this post.
-
alexey.tal
- Global Moderator
- Posts: 314
- Joined: Mon Sep 13, 2021 12:45 pm
Re: Bug report in Pt cluster relaxation
Dear Ramon,
It seems that the issue has to do with the HDF5 library as it crushed when VASP was trying to write something into the vaspout.h5 file. I was not able to reproduce this error on our machine.
What version of HDF5 do you use? Could you provide the list of libraries and compilers VASP was built with?
I would also suggest that you try to compile the code without HDF5 support and run this calculation again.
It seems that the issue has to do with the HDF5 library as it crushed when VASP was trying to write something into the vaspout.h5 file. I was not able to reproduce this error on our machine.
What version of HDF5 do you use? Could you provide the list of libraries and compilers VASP was built with?
I would also suggest that you try to compile the code without HDF5 support and run this calculation again.
-
ramon_bergua
- Newbie
- Posts: 5
- Joined: Thu Jan 12, 2023 6:02 pm
Re: Bug report in Pt cluster relaxation
Dear Alexey,
I am using VASP/6.4.1 in a High Performance Computing Center, so I do not know the version of HDF5 they have installed, I will contact them and after that I will give you an answer.
Thank you very much for your reply.
By the way, What about this other error what appears in several crashed calculations? This not send BUG report message but are the most common crash errors in the calculations. Here I attach the zipfile
Ramon
I am using VASP/6.4.1 in a High Performance Computing Center, so I do not know the version of HDF5 they have installed, I will contact them and after that I will give you an answer.
Thank you very much for your reply.
By the way, What about this other error what appears in several crashed calculations? This not send BUG report message but are the most common crash errors in the calculations. Here I attach the zipfile
Ramon
You do not have the required permissions to view the files attached to this post.
-
ramon_bergua
- Newbie
- Posts: 5
- Joined: Thu Jan 12, 2023 6:02 pm
Re: Bug report in Pt cluster relaxation
With respect to the Hfd5 version installed in the HPC center,
HDF5/1.12.2-iimpi-2022a
This includes:
intel-compilers/2022.1.0
impi/2021.6.0-intel-compilers-2022.1.0
imkl-FFTW/2022.1.0-iimpi-2022a
imkl/2022.1.0
They are already compiling without HFd5 to try to reproduce the error.
HDF5/1.12.2-iimpi-2022a
This includes:
intel-compilers/2022.1.0
impi/2021.6.0-intel-compilers-2022.1.0
imkl-FFTW/2022.1.0-iimpi-2022a
imkl/2022.1.0
They are already compiling without HFd5 to try to reproduce the error.
-
alexey.tal
- Global Moderator
- Posts: 314
- Joined: Mon Sep 13, 2021 12:45 pm
Re: Bug report in Pt cluster relaxation
This error message comes from the HDF5 library. The exact cause of this error I can't tell as so far I was not able to reproduce it.By the way, What about this other error what appears in several crashed calculations? This not send BUG report message but are the most common crash errors in the calculations. Here I attach the zipfile