[talk-ch] Overpass server delay query

marc marc marc_marc_irc at hotmail.com
Mon Feb 10 19:25:30 CET 2020


it depends on your priorities.
if the bandwidth/ram/cpu of the machine doing the operation is less
important than the speed of the update, you can download update your
file from diffs. this allows you to update every day, hour or minute
according to your choice. you will obviously have to filter the updates
to keep only what you are interested in.

the other problem: if you only target nodes, then you won't have the POI
information mapped as a polygon.
https://taginfo.openstreetmap.org/keys/drinking_water

Once the 50MB file is produced, how does a user retrieve it into the
application?

Le 10.02.20 à 18:56, European Water Project a écrit :
> Hi Marc (or anyone else),
> 
> As suggested by Michael a while back, I am scanning the globe once a
> week for water fountains (modern and historic) in a nested loop of
> bboxes. After getting all the individual scans, I append them into one
> large geojson file ..  I scan for fountains with a bbox size of 10x10.
> 
> I seem to get a faster response using overpass turbo at
> overpass-turbo.eu <http://overpass-turbo.eu> using the web interface
> than with the API. Is this normal ?  The total size of the all the
> responses combined in one file is relatively small : about 50MB (after
> appending them all in one json file). 
> 
> It takes me on average about 3 hours to get all the data I need for my
> database.  Do you have any suggestions for how I might improve the
> process?  Or maybe, what I am doing seems ok ?
> 
> Thank you for your advice,
> 
> Stuart 
> 
> 
> 
> [out:json][timeout:150];
> 
> (
>   node["amenity"="drinking_water"]({{bbox}});
>   node["drinking_water"="yes"]({{bbox}});
> );
> // print results
> out body;
>>;
> out skel qt;
> 
> 
> On Mon, 10 Feb 2020 at 14:47, marc marc <marc_marc_irc at hotmail.com
> <mailto:marc_marc_irc at hotmail.com>> wrote:
> 
>     Hello Stuart,
> 
>     Le 10.02.20 à 14:40, European Water Project a écrit :
>     > Do both server clusters serve the same data via the API and are they
>     > updated simultaneously?
> 
>     all overpass-api servers try to keep in sync from osm.org
>     <http://osm.org> data via
>     minute "diff" update. but from time to time, that fail or take a little
>     more time.
>     every group of server have often their own sysadmin, it's not a group
>     that takes care of all the overpass-api  servers.
>     modern overpass api version return the info with a line like this :
>     <meta osm_base="2020-02-10T13:42:02Z"/>
> 
>     Regards,
>     Marc
>     _______________________________________________
>     talk-ch mailing list
>     talk-ch at openstreetmap.ch <mailto:talk-ch at openstreetmap.ch>
>     http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
> 
> 
> _______________________________________________
> talk-ch mailing list
> talk-ch at openstreetmap.ch
> http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
> 



More information about the talk-ch mailing list