500 Server Error on CESM_POP xarray previews

[charlesbluca]

I'm not sure what this is related to, as other zarr-based datasets seem to load fine. I'll test this on local server and see what's wrong - another reason to look into Flask testing!

[charlesbluca]

Also noticed this problem on some of the LLC4320 datasets, and noticed that Rasterio isn't installed either which means Hydrosheds datasets can't be viewed; overall, I'd like to:

• [x] Add Rasterio to the requirements.txt (not sure if any GDAL dependencies need to be looked out for)
• [x] Check out the issues with the CESM_POP and LLC4320 datasets

[rabernat]

Can you post an error stack trace?

[charlesbluca]

Looking into the request logs of the app, we find this in the 500 response:

protoPayload.line[0].logMessage: While handling this request, the process that handled this request was found to be using too much memory and was terminated. This is likely to cause a new process to be used for the next request to your application. If you see this message frequently, you may have a memory leak in your application or may be using an instance with insufficient memory. Consider setting a larger instance class in app.yaml.

For whatever reason, the app is having difficulty handling the ds.to_dask() without killing the process.