_FillValue type mismatch

[tiemvanderdeure]

The "_FillValue" field and the missingval are identical when a Raster is read, but unlike missingval it is not updated. So something like:

using Rasters, NCDatasets
filename = tempname() * ".nc"
A = Rasters.create(filename, Float32, (X(1:10),), missingval = NaN32, force = true)
write("myraster2.nc", A .+ 1.0)

Gives a ERROR: NetCDF error: NetCDF: Not a valid data type or _FillValue type mismatch (NetCDF error code: -45)

Could we just ignore it whenever raw = false?

[rafaqz]

Yeah we probably shouldn't load it into metadata