Streaming tasks do not exit properly with Parsl and store_return_result=False
#132
Comments
WardLT
added a commit
to globus-labs/mof-generation-at-scale
that referenced
this issue
Mar 14, 2024
WardLT
added a commit
to globus-labs/mof-generation-at-scale
that referenced
this issue
Mar 14, 2024
* Update to streaming-ready version of Colmena * Switch generate function to be generator * Update parallel app to use streaming * Unrelated, but important: as_complete for LAMMPS Ensure we get results, even if a task fails * Change LAMMPS executable to "lmp" That's what it is called on conda-forge's package now * Add redis to the build enviornment, run script * Use the Anaconda version of redis The conda-forge one fails for reasons beyond my understanding. * Bug fix: Use the wrapped function in task server * Use the github branch of Colmena for now * Store return value, if though it is ``None`` See exalearn/colmena#132 * Reduce the node count used for text Also add echos so we know where the execution is
WardLT
changed the title
store_return_result=Falsestore_return_result=False
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Not sure why, but there seems to be an interaction between these two components.
The text was updated successfully, but these errors were encountered: