siasiesiv diagnostic: failed to merge into a single cube
I have an error message when using Earthdiagnostic on HR experiment (cf for instance this log file: /esarchive/scratch/Earth/acarreri/earthdiags_suite/suite_home/a3qa_diagnostics/19870101/Member_0/Chunks_1/run.1
):
^[[31m[ERROR] Siasiesiv Startdate: 19870101 Member: 0 Chunk: 1 Basins: Barkara,Labrador_Sea,Labrador_Sea_Nudging,Labrador_Sea_conv,Labrador_Sea_conv2,Northern_Hemisphere Omit volume: False: failed to merge into a single cube.
cube.cell_methods differ
File "/opt/conda/lib/python3.7/site-packages/earthdiagnostics/work_manager.py", line 313, in _run_job
job.compute()
File "/opt/conda/lib/python3.7/site-packages/earthdiagnostics/ocean/siasiesiv.py", line 218, in compute
self.save()
File "/opt/conda/lib/python3.7/site-packages/earthdiagnostics/ocean/siasiesiv.py", line 260, in save
self.generated[var].set_local_file(temp, diagnostic=self)
File "/opt/conda/lib/python3.7/site-packages/earthdiagnostics/datafile.py", line 318, in set_local_file
self.prepare_to_upload(rename_var)
File "/opt/conda/lib/python3.7/site-packages/earthdiagnostics/datafile.py", line 276, in prepare_to_upload
self._prepare_region()
File "/opt/conda/lib/python3.7/site-packages/earthdiagnostics/datafile.py", line 478, in _prepare_region
final_cube = cube_list.merge_cube()
File "/opt/conda/lib/python3.7/site-packages/iris/cube.py", line 396, in merge_cube
proto_cube.register(cube, error_on_mismatch=True)
File "/opt/conda/lib/python3.7/site-packages/iris/_merge.py", line 1275, in register
match = cube_signature.match(other, error_on_mismatch)
File "/opt/conda/lib/python3.7/site-packages/iris/_merge.py", line 411, in match
raise iris.exceptions.MergeError(msgs)
^[[0m^[[39m
^[[31m[ERROR] Total wasted time: 0:00:06.702166^[[0m^[[39m
It could come from the fact that because on this experiment, the EARTHDIAGNOSTICS
online tool was activated and a file already existed before running this diagnostic with earthdiagnotics suite offline. (even if I think that for other variables, it was also the case, with existing files, but I got an error only for this variable).
Another thing is that the final files seem correct, with the new region asked included. So at the end, the diagnostic seem to run correctly, despite this error message.
It seems the same kind of error that I got with the "moc" variable (cf this issue), even if the error message is not the same.
Can I trust the output results despite the error message?
Could you in that case remove this kind of "bug"?
Many thanks in advance.
Aude