GLIBC mismatch on RHEL/Rocky when loading PythonFMU-built FMU (request RHEL/Rocky CI support) #249
Replies: 2 comments
-
I assume the mismatch is due to use using a newer version of Ubuntu in the publish pipeline: https://github.com/NTNU-IHB/PythonFMU/blob/master/.github/workflows/publish.yml An older version is ideally better, and would likely fix the issue (?) but then we need to pull in docker images to build. I'm not in a position to look at this at the moment. I ought to publish version 0.7.0 at some point and should probably also then update the conda build. |
Beta Was this translation helpful? Give feedback.
-
Mac support is another thing. There are probably no obstacles for it, but the pipeline needs to be updated with mac. It steals build time, but likely not an issue for this project. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
We’re building a MOOSE → FMU interface using PythonFMU (see PR: idaholab/moose#30912). When we add
pythonfmu
to our test YAML and run the FMU, we hit a GLIBC version mismatch on RHEL/Rocky Linux.I have tried building and installing
PythonFMU
from source in Rocky Linux machine and it fixes the issue. I also noticed there is a condaPythonFMU
package but it is not updated to0.6.9
which includes two fixes I need. I can work around this by including the build and installation in MOOSE FMU package for our downstream users, but native support would largely simplify the distribution and future maintenance of MOOSE FMU interface. So we wonder if it is possible for pythonFMU to add CI coverage for RHEL/Rocky (e.g., Rocky 8/9) and if the team plans to support more environments other than Ubuntu and windows like Mac?Thanks a lot!
Beta Was this translation helpful? Give feedback.
All reactions