[OSM-talk] Bandwidth limit/IP blocking - Error 303 on the OSM API?

[Tom Hughes]

On 01/06/11 21:11, Peter Mooney wrote:

> wanted to download an addition 2,000 ways (their full history)
> from the same machine today
> - example "wget http://www.openstreetmap.org/api/0.6/way/2048/history -O
> ./world-xml/history-2048.xml"
>
> I keep getting an ERROR 303 .. The error message doesn't make sense to me.

Actually you are getting "403 Forbidden", at least for some of your 
requests. Many of them will timeout.

> Have I found myself on a "banned" list? It would not be my intention to
> hog resources on anyone else - but I didn't think the requests were
> excessive. It took a couple of days to download the history files - i
> put an hour pause into the bash script after every 500 files downloaded.

Yes, you got blocked on 16th May by the looks of it. I don't think I did 
it so it was probably one of my colleagues.

Scraping bulk history data on the sort of scale you are talking about 
certainly is viewed as excessive by us - the API is designed to support 
editors not bulk exporting of data.

> Suggestions, work-arounds?

Use the full history dump, as somebody has already suggested.

Tom

-- 
Tom Hughes (tom at compton.nu)
http://compton.nu/