The cause of the weird issue with talknat & dissicnat probably found

[treerink]

Since a long time often the errors with talknat & dissicnat in the cmorisation log files were there, for instance in the test-all tests (and projects which took derived requests). See this #542 post. I was not able to detect the cause quickly and the real effect was not to bad, so never came to full debugging.

However, probably the cause is simply that this two Pisces variables miss in the long list of Pisces variables which only should be used for an ECE-CC configuration which is organised in the preferences admin in the prefs.py. Initial troubles with this two variables made that they were added in a later stage and then it was probably overlooked to add them to this list. The effect is that when running genecec this two pisces variables are added to the request also for non ECE-CC configurations (which should not be the case), causing the error at cmorisation because the non ECE-CC configuration does not deliver the NEMO pisces output for them.

I expect changing ouptut-control-files for non ECE-CC parts, which will be an improvement because more consistant, and causing unnecessary errors.

[treerink]

Probably solved with c07cf34. Cmorisation with the test-all case should confirm, but the changes in the output-control-files are as expected.