Håvard Berland
Håvard Berland
Tried poly-case with: * LSF * poly_eval.py modified to segfault with some probability * MAX_SUBMIT set to 2 * Komodo 2024.06.07 (ert 10.1) Ran ensemble_experiment and watched carefully the "Running...
The code hangs here: https://github.com/equinor/ert/blob/759f431ec6827a149663ece6691672c8b5122057/src/ert/simulator/simulation_context.py#L249
#7336 will avoid this bug for simulation_context, but is tripped by the test `test_max_runtime_while_killing()`. It is possible that 7336 is the correct solution, but that code handling the other test...
Can't reproduce either. Closing.
Might want to evaluate whether to use `watchfiles` instead of `watchdog`. https://pypi.org/project/watchfiles/
This function is already called in an async context, so the "mitigation" is only about yielding a little more often.
This is solved in the new Scheduler. The "Waiting" state is the implicit state in `job.py` before it is submitted, and if the time in waiting is nonzero, it is...
Dataclasses do look nicer!
The LSF stdout might be sufficient though, but must be fixed in #7695. Examine if there are differences when OOM strikes f.ex.
As for filename, we already have `.LSF-out` for stdout, and we might get `.LSF-err` for stderr (that is a potential issue to write). To be in line with that system,...