Skip to content

Conversation

jan-janssen
Copy link
Member

@jan-janssen jan-janssen commented Aug 1, 2024

Currently this fails with the following error message:

83.06 [Errno 28] No space left on device: '/opt/conda/lib/libomptarget.rtl.x86_64.so.18.1' -> '/opt/conda/lib/libomptarget.rtl.x86_64.so.18.1.c~'

@jan-janssen jan-janssen marked this pull request as draft August 1, 2024 07:46
@jan-janssen jan-janssen marked this pull request as ready for review August 1, 2024 11:53
@jan-janssen jan-janssen marked this pull request as draft August 1, 2024 15:57
@jan-janssen
Copy link
Member Author

As long as we restrict mlip=2.0 we cannot upgrade to a more recent LAMMPS version.

#8 28.91 The following packages are incompatible
#8 28.91 ├─ lammps 2024.06.27 *openmpi* is installable with the potential options
#8 28.91 │  ├─ lammps 2024.06.27 would require
#8 28.91 │  │  └─ python >=3.10,<3.11.0a0 , which can be installed;
#8 28.91 │  ├─ lammps 2024.06.27 would require
#8 28.91 │  │  └─ mlip * mpi_openmpi_*, which can be installed;
#8 28.91 │  ├─ lammps 2024.06.27 would require
#8 28.91 │  │  └─ python >=3.12,<3.13.0a0 , which can be installed;
#8 28.91 │  ├─ lammps 2024.06.27 would require
#8 28.91 │  │  └─ python >=3.8,<3.9.0a0 , which can be installed;
#8 28.91 │  ├─ lammps 2024.06.27 would require
#8 28.91 │  │  └─ python >=3.9,<3.10.0a0 , which can be installed;
#8 28.91 │  └─ lammps 2024.06.27 would require
#8 28.91 │     └─ __cuda, which is missing on the system;
#8 28.91 └─ mlip 2.0**  is not installable because it conflicts with any installable versions

@raynol-dsouza and @pmrv If I remember correctly you two were testing mlip 2.0 and mlip 3.0 did you come to a conclusion? Can we update the cluster to mlip 3.0 ?

@jan-janssen jan-janssen removed the request for review from niklassiemer August 1, 2024 18:25
@raynol-dsouza
Copy link

@jan-janssen MLIP 3.0 in itself had no issues, but there was (and still is) an issue with the LAMMPS-MLIP-3 interface.

@pmrv commented on the issue which was already raised, but it hasn't been addressed yet.

@pmrv
Copy link
Contributor

pmrv commented Sep 11, 2025

Upstream apparently fixed this, so updating should be fine now.

@jan-janssen jan-janssen marked this pull request as ready for review September 12, 2025 15:21
@jan-janssen
Copy link
Member Author

@pmrv I updated the LAMMPS package and the mlip package. Can you test if the stress issue is resolved?

@jan-janssen
Copy link
Member Author

We again run into the numba incompatibility error, as far as I remember pyxtal numba which requires an older numpy version compared to the latest LAMMPS release.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants