Osm on docker : what services are included

openrouteservice
1

Hi,

I set up an osm instance running in docker following the instructions.
The only thing I had to modify was to download and merge Germany, Austria and Italy osm.pbf and merge it and set the java memory options to 10G’s of ram on the server. So the docs are pretty good :+1:

I’m wondering which endpoints are now effectively available on this docker instance.
I read somewhere that Pelias for autocomplete must be installed separately?
Should vroom for TSM optimization also be installed aside?

If yes: are there further docs for installing those instances?

is a ryzen 5 with 64gb of ram enough for such a configuration?

Thanks!

2

Hi @Harald_Kofler

glad that you got it up and running :slight_smile:

When you install openrouteservice, you get access to the directions, isochrones and matrix endpoints. Fro the geocoder and optimization, you are correct that these are seperate things to install. As they are not things that we develop though, your best bet is to look at the documentation from the respective owners of those services. For the geocoder it is Pelias, and for optimization it is Vroom. Both of them provide some documentation on how to get up and running with Docker.

I can’t really say for sure about what resources you need for those though, but for Pelias it really depends on how much data you want to include in the geocoder.

3

Thank you @adam for your answer,

does openrouteservice host it’s own instance of pelias and vroom, or do they proxy to dedicated “external” environments?
I ramped up a Pelias environment, and now I’m working on getting the same behaviour as the autocomplete endpoint of ORS. Vromm should be easier as I’ve seen so far.

4

No worries

we host our own versions of Pelias and Vroom (the vroom instance is slightly customised due to our server setup).

5

Are your config available?
At the end I’m pretty satisfied with the complete osm stack- I just need to set up a “copy” of it to be able to issue more requests.

After setting up the two projects as dockerized systems I encountered the following issues:

  • osm provides a central front end for the API
  • ATM the search and autocomplete endpoints do not include the admin lookup data of wof (see example below), even after including the corresponding wof data in the pelias.conf.

As OSM already perfectly does this, I’m wondering if I could benefit of this configuration?

{
   "bbox" : [
      7.425038,
      50.105733,
      14.9852,
      53.535225
   ],
   "features" : [
      {
         "geometry" : {
            "coordinates" : [
               13.411382,
               52.568381
            ],
            "type" : "Point"
         },
         "properties" : {
            "accuracy" : "point",
            "addendum" : {
               "osm" : {
                  "wikidata" : "Q19963781",
                  "wikipedia" : "de:Postamt Pankow I"
               }
            },
            "gid" : "openstreetmap:address:way/141859661",
            "housenumber" : "12",
            "id" : "way/141859661",
            "label" : "12 Berliner Straße",
            "layer" : "address",
            "name" : "12 Berliner Straße",
            "postalcode" : "13187",
            "source" : "openstreetmap",
            "source_id" : "way/141859661",
            "street" : "Berliner Straße"
         },
         "type" : "Feature"
      },
      {
         "geometry" : {
            "coordinates" : [
               11.666597,
               52.128319
            ],
            "type" : "Point"
         },
         "properties" : {
            "accuracy" : "point",
            "addendum" : {
               "osm" : {
                  "operator" : "MVGM GmbH",
                  "phone" : "+49 391 593450",
                  "website" : "https://www.mvgm.de/de/getec-arena/",
                  "wheelchair" : "yes",
                  "wikidata" : "Q885030",
                  "wikipedia" : "de:GETEC Arena"
               }
            },
            "gid" : "openstreetmap:address:way/28654355",
            "housenumber" : "32",
            "id" : "way/28654355",
            "label" : "32 Berliner Chaussee",
            "layer" : "address",
            "name" : "32 Berliner Chaussee",
            "postalcode" : "39114",
            "source" : "openstreetmap",
            "source_id" : "way/28654355",
            "street" : "Berliner Chaussee"
         },
         "type" : "Feature"
      },
      {
         "geometry" : {
            "coordinates" : [
               14.983911,
               51.150795
            ],
            "type" : "Point"
         },
         "properties" : {
            "accuracy" : "point",
            "addendum" : {
               "osm" : {
                  "wikipedia" : "de:Straßburg-Passage"
               }
            },
            "gid" : "openstreetmap:address:way/158448128",
            "housenumber" : "8",
            "id" : "way/158448128",
            "label" : "8 Berliner Straße",
            "layer" : "address",
            "name" : "8 Berliner Straße",
            "postalcode" : "02826",
            "source" : "openstreetmap",
            "source_id" : "way/158448128",
            "street" : "Berliner Straße"
         },
         "type" : "Feature"
      },
      {
         "geometry" : {
            "coordinates" : [
               7.425038,
               51.347699
            ],
            "type" : "Point"
         },
         "properties" : {
            "accuracy" : "point",
            "addendum" : {
               "osm" : {
                  "wheelchair" : "no"
               }
            },
            "gid" : "openstreetmap:address:way/42874256",
            "housenumber" : "127",
            "id" : "way/42874256",
            "label" : "127 Berliner Straße",
            "layer" : "address",
            "name" : "127 Berliner Straße",
            "postalcode" : "58135",
            "source" : "openstreetmap",
            "source_id" : "way/42874256",
            "street" : "Berliner Straße"
         },
         "type" : "Feature"
      },
      {
         "geometry" : {
            "coordinates" : [
               8.763118,
               50.105733
            ],
            "type" : "Point"
         },
         "properties" : {
            "accuracy" : "point",
            "addendum" : {
               "osm" : {
                  "wikidata" : "Q882847",
                  "wikipedia" : "de:City Tower Offenbach"
               }
            },
            "gid" : "openstreetmap:address:way/41726564",
            "housenumber" : "76",
            "id" : "way/41726564",
            "label" : "76 Berliner Straße",
            "layer" : "address",
            "name" : "76 Berliner Straße",
            "postalcode" : "63065",
            "source" : "openstreetmap",
            "source_id" : "way/41726564",
            "street" : "Berliner Straße"
         },
         "type" : "Feature"
      },
      {
         "geometry" : {
            "coordinates" : [
               8.583972,
               53.535225
            ],
            "type" : "Point"
         },
         "properties" : {
            "accuracy" : "point",
            "addendum" : {
               "osm" : {
                  "website" : "https://www.denkmalpflege.bremen.de/sixcms/detail.php?template=20_denkmal_wrapper_d&obj=00001633",
                  "wikidata" : "Q46088448",
                  "wikipedia" : "de:Wohnhaus Berliner Platz 11"
               }
            },
            "gid" : "openstreetmap:address:way/184110023",
            "housenumber" : "11",
            "id" : "way/184110023",
            "label" : "11 Berliner Platz",
            "layer" : "address",
            "name" : "11 Berliner Platz",
            "postalcode" : "27570",
            "source" : "openstreetmap",
            "source_id" : "way/184110023",
            "street" : "Berliner Platz"
         },
         "type" : "Feature"
      },
      {
         "geometry" : {
            "coordinates" : [
               13.378952,
               52.514084
            ],
            "type" : "Point"
         },
         "properties" : {
            "accuracy" : "point",
            "addendum" : {
               "osm" : {
                  "opening_hours" : "24/7",
                  "wheelchair" : "limited",
                  "wikidata" : "Q160700",
                  "wikipedia" : "de:Denkmal für die ermordeten Juden Europas"
               }
            },
            "gid" : "openstreetmap:address:way/172234733",
            "housenumber" : "1",
            "id" : "way/172234733",
            "label" : "1 Cora-Berliner-Straße",
            "layer" : "address",
            "name" : "1 Cora-Berliner-Straße",
            "postalcode" : "10117",
            "source" : "openstreetmap",
            "source_id" : "way/172234733",
            "street" : "Cora-Berliner-Straße"
         },
         "type" : "Feature"
      },
      {
         "geometry" : {
            "coordinates" : [
               8.893427,
               53.081784
            ],
            "type" : "Point"
         },
         "properties" : {
            "accuracy" : "point",
            "addendum" : {
               "osm" : {
                  "website" : "https://www.denkmalpflege.bremen.de/sixcms/detail.php?template=20_denkmal_wrapper_d&obj=00000123",
                  "wheelchair" : "limited",
                  "wikidata" : "Q300972",
                  "wikipedia" : "de:Aalto-Hochhaus"
               }
            },
            "gid" : "openstreetmap:address:way/162130229",
            "housenumber" : "9",
            "id" : "way/162130229",
            "label" : "9 Berliner Freiheit",
            "layer" : "address",
            "name" : "9 Berliner Freiheit",
            "source" : "openstreetmap",
            "source_id" : "way/162130229",
            "street" : "Berliner Freiheit"
         },
         "type" : "Feature"
      },
      {
         "geometry" : {
            "coordinates" : [
               14.9852,
               51.151626
            ],
            "type" : "Point"
         },
         "properties" : {
            "accuracy" : "point",
            "addendum" : {
               "osm" : {
                  "brand" : "Nordsee",
                  "opening_hours" : "Mo-Fr 09:30-19:00; Sa 09:30-15:00",
                  "phone" : "+49 3581 649425",
                  "wheelchair" : "yes"
               }
            },
            "gid" : "openstreetmap:address:way/158386092",
            "housenumber" : "1",
            "id" : "way/158386092",
            "label" : "1 Berliner Straße",
            "layer" : "address",
            "name" : "1 Berliner Straße",
            "postalcode" : "02826",
            "source" : "openstreetmap",
            "source_id" : "way/158386092",
            "street" : "Berliner Straße"
         },
         "type" : "Feature"
      },
      {
         "geometry" : {
            "coordinates" : [
               14.984118,
               51.150917
            ],
            "type" : "Point"
         },
         "properties" : {
            "accuracy" : "point",
            "addendum" : {
               "osm" : {
                  "wikipedia" : "de:Straßburg-Passage"
               }
            },
            "gid" : "openstreetmap:address:way/158386095",
            "housenumber" : "7",
            "id" : "way/158386095",
            "label" : "7 Berliner Straße",
            "layer" : "address",
            "name" : "7 Berliner Straße",
            "postalcode" : "02826",
            "source" : "openstreetmap",
            "source_id" : "way/158386095",
            "street" : "Berliner Straße"
         },
         "type" : "Feature"
      }
   ],
   "geocoding" : {
      "attribution" : "http://localhost:4000/attribution",
      "engine" : {
         "author" : "Mapzen",
         "name" : "Pelias",
         "version" : "1.0"
      },
      "query" : {
         "lang" : {
            "defaulted" : true,
            "iso6391" : "en",
            "iso6393" : "eng",
            "name" : "English",
            "via" : "default"
         },
         "layers" : [
            "address"
         ],
         "parsed_text" : {
            "locality" : "Berlin",
            "subject" : "Berlin"
         },
         "parser" : "pelias",
         "private" : false,
         "querySize" : 20,
         "size" : 10,
         "text" : "Berlin"
      },
      "timestamp" : 1617098748134,
      "version" : "0.2"
   },
   "type" : "FeatureCollection"
}
6

We can’t give out the configs as they contain internal IP addresses etc. In general though, for pelias we use the standard approach that they provide on GitHub (similar for Vroom) which is at docker/projects/planet at master · pelias/docker · GitHub.

I also assume that when you say “OSM” you mean “ORS”?

7

Sure, I understand that.
Well, then I’ll try to fix my installation until it works :wink:

btw. in the release notes is mentioned that the proxy for geocoding has been removed somewhere in 2019. Indeed adding the geocoding section into the app.config seems not work. Are you using a different approach for your installation, or is there a a possibility to re-enable this proxying?

And yes, by “OSM” I mean “ORS”…

Thanks

8

I don’t believe it would work with the old proxy as that used a Nominatim (I believe) database whereas the Pelias one has a very specific structure and it’s own API. We use Tyk as an “proxying” system which takes the API requests and then forwards them across to the corresponding server. I think you could do something similar with either a basic Apache proxy, or something like API umbrella which is an open source alternative to Tyk.

9

OK, thanks for your feedback.
Pelias seems now to work, as a next step I’ll try to put an nginx in front to do this proxying/routing