Øyvind Eide

Some indications that this is the cause: ``` Fatal Python error: Illegal instruction tests/integration_tests/scheduler/test_generic_driver.py::test_kill[LsfDriver] Thread 0x000070001668e000 (most recent call first): [gw0] node down: Not properly terminated File "/Library/Frameworks/Python.framework/Versions/3.12/lib/python3.12/threading.py", line 359...

> Can be reproduced by running `ert ensemble_experiment config.ert` on the following input > > config.ert: > > ``` > NUM_REALIZATIONS 10 > GEN_KW BPR template.txt params.txt BPR.txt > ```...

I will close this as intended behavior, but added an issue for validation: #7937

> If we bump local storage to 5 what else needs to be changed? Need to add another level to the migrations, but that should be it.

This is intended behavior, so removing the bug label, have added documentation to how one can avoid updating parameters.

Note quite sure what to do with this issue, as it is intended behavior, and the parameters are updated, but still show up as constant for the user. There is...

So after discussion the solution here is to introduce an UPDATE:TRUE/FALSE flag to individual parameters (as opposed to parameter groups which already exists). If the parameter is CONST, this flag...

Note that some of this information is also available in the update snapshot: https://github.com/equinor/ert/blob/9f7ac75b4a8a6285659bb1ba641c14a2f9b79ec7/src/ert/analysis/_es_update.py#L60-L67 which is written out after an update, so might make sense to gather this information more.

Would be great to add this in the update snapshot instead of in the log, so we get a column there with the index.

Removed the bug label as this is arguably expected behavior, as all experiments should have an ensemble. Is it possible to reproduce this though the gui for example?