This is upsetting... It would be great if we could return a failure to the start run request, though because those are handled in different places maybe it is asking too much. Right now we get some HDF5 error, perhaps we could at least handle that and fail with a psyllid::error?
This is upsetting... It would be great if we could return a failure to the start run request, though because those are handled in different places maybe it is asking too much. Right now we get some HDF5 error, perhaps we could at least handle that and fail with a psyllid::error?