nvkelso/natural-earth-vector

Cannot download naciscdn.org data from naturalearthdata.com

Closed this issue · 21 comments

Hi all. Any help would be greatly appreciated.

When I attempt to download https://naciscdn.org/naturalearth/110m/physical/ne_110m_ocean.zip, , the request times out.

I assume you're still working on this because I can't get to https://www.naturalearthdata.com/http//www.naturalearthdata.com/download/10m/raster/OB_LR.zip
Responds with This site can’t be reached naciscdn.org took too long to respond.
Any ETA for the repoint???

Hello, are there any news on the naciscdn.org outage?

The original private server backing naciscdn.org seems to be busted.

I'm in the process of redirecting to S3. Here are some file resource patterns in the meantime:

Official S3 public data: (EVERYONE SHOULD MIGRATE HERE, ESPECIALLY IF YOU USE CI)
https://naturalearth.s3.amazonaws.com/10m_cultural/10m_cultural.zip
(note 10m_cultural where naturalearthdata.com uses 10m/cultural with / not _ so you'll need to update more than your domain)

Eventually this will get naciscdn.org pretty domain:
https://s3.us-west-2.amazonaws.com/naciscdn.org/naturalearth/10m/cultural/10m_cultural.zip
(same file structure as naturalearthdata.com)

Is there a way to download a raster file with an AWS link?

Oops, I missed the rasters the first attempt, thanks for flagging that. Adding the 10m and 50m rasters now.

@dwnadler The rasters should all be there now.

Is this still down? I currently can't access the server at naciscdn.org. Thanks :)

FWIW I downloaded data from the S3 URL using geopandas
gpd.read_file("https://naturalearth.s3.amazonaws.com/10m_cultural/ne_10m_populated_places.zip")

If you need to find one of these links and made it this far, here they are in a gist, until the site is restored. Taken from the index at https://naturalearth.s3.amazonaws.com/

https://gist.github.com/DanielJWood/b71237cc200831acf8e637c05ce2c375

Downloaded the files successfully from https://naturalearth.s3.amazonaws.com/10m_physical/ne_10m_coastline.zip - thank you!

Just wanted to say thanks for rehosting it on https://naturalearth.s3.amazonaws.com/
The same url structure worked for me to download the files I needed for cartopy.
Appreciate it!

Hey!
First of all thanks for sharing the downloads! I have all of the data saved locally now, but cartopy still wants to download the packages from naciscdn.org. How do I tell the programm to use the local files instead of trying to download?
I tried changing the config with cartopy.config['pre_existing_data_dir']=expanduser('~/cartopy-data') but it still won't work. Might be a stupid question, but help would be appreciated!

Julia

@https://github.com/julia-remke I'm not a cartopy user, but SciTools/cartopy#1833 and SciTools/cartopy#1849 suggests that a new release of cartopy (v0.20.0) should address this. If you're using that and it still doesn't work, that's probably something to raise at https://github.com/SciTools/cartopy/issues .

@julia-remke I'm in the same situation. It still seems to be pulling from the non-AWS site and I'm not sure how to change that.

@ttruksa I ended up just downloading them manually and putting them into cartopy's offline data. ...You may also need the cartopy offline data package as well. (https://github.com/willirath/cartopy_offlinedata) Not sure, as I ended up going with folium...

>>> import cartopy.config
>>> cartopy.config['data_dir']
'/home/pelson/.local/share/cartopy'

The http version of the naciscdn.org links should now work again.

Next up the https version (which is used by naturalearthdata.com download redirects).

The https versions should also now work again, meaning downloads from naturalearthdata.com should be operational again.

For posterities sake, the nameservers configs were scrambled / not auto generated correctly. Manually setting them to the same 4 values in all the places corrected the problem, and unblocked adding the SSL cert.