Once the R process is restarted, the system should generated the model output for missing Taxon that does not have latest output.
Steps to reproduce:
There are two models originally, generic and model for Loxodonta
Both models already produced species model output
Then I removed the Loxodonta statistical model and its model outputs will also be removed
Once the R processs is up, system does not trigger generation for Loxodonta
Current workaround
Once the R process is up, I created the species model output for Loxodonta from Generic Model. Then, I trigger the generation manually from below admin action.
We could also improve the generation of model output from a taxon by:
Checking whether the taxon has population data. This is to avoid the system triggers a request to R process when there is no population data.
When the most recent ouput record is an error, the system should not create a new output record. In this case, there should only be two output models: 1 with is_latest=True and Done status and 1 with is_latest=False and Error status.
When the most recent output record is an error, we should not update it with is_latest=True. The reason is that the system can still display the old output
System should notify the statistician when there is an error from the R code.
Once the R process is restarted, the system should generated the model output for missing Taxon that does not have latest output.
Steps to reproduce:
Current workaround Once the R process is up, I created the species model output for Loxodonta from Generic Model. Then, I trigger the generation manually from below admin action.