Closed scotthavens closed 4 years ago
It appears to be a few different possible issues. First, it can get hung up in the queue with the thermal_cloud
variable for the gridded inputs. That was an easy fix by removing the variables from the queue (not really needed at the moment).
The more complex one is in the tests. The queue doesn't appear to be properly cleaned up as the queue items like flatwind
show up in the gridded queue. Running the gridded tests by themselves don't really have a problem and will run fine. But run the test_configurations
then the gridded tests, it will have those flatwind
values in the queue.
For example, python -m unittest -v
will not work but each test can be ran individually and will pass. This is leading me to believe that there is some cleanup that isn't happening in the tests somewhere.
Closing this for now as the tests in both SMRF and AWSM are passing. Reopen if that changes after more testing.
The threading function of SMRF never worked quite right with gridded data (old issue #3). While it was fixed, with the addition of HRRR inputs, it stopped working as there are no tests to ensure that it works.