Possible segfault when reading summary data

[oyvindeide]

Got a report of a user reading summary data, which most likely segfaulted, and took down the model, only trace I currently have is a the final log statement:

fmu.ensemble.ensemble - ThreadPoolExecutor-9_0 - INFO - Loading smry from realization 1

[eivindjahren]

Could be related to equinor/resdata#945

[oyvindeide]

Could be, but was ran with 2020.2, so then the same problem has to be present in that version as well.

[oyvindeide]

Turns out it ran out of memory, and did not exit gracefully.