[Geoserver-users] RC4 / Google Earth 4.0.2413 / Postgis not happy

GeoServer GeoServer User
1

Hi,

I'm trying to get a point layer in my postgis database to be served via Geoserver to Google Earth. I haven't been able to get it to work.

When I use

format=application/vnd.google-earth.kml+XML

The server delays for a number of seconds and I finally get a truncated XML file (see attached current_kcci.kml.gz) and nothing interesting in the log files on the server.

When I use

format=application/vnd.google-earth.kmz+XML

The server immediately responds, but Google Earth appears to have duplicate points. When I click on the point, two points appear, each with a label.

I've attached my sld (point_test.sld) and the URI to my instance is:

http://mesonet.agron.iastate.edu/geoserver//wms?service=WMS&request=GetMap&format=application/vnd.google-earth.kmz+XML&width=1024&height=1024&srs=EPSG:4326&layers=topp:current_kcci&styles=poi&bbox=-180,-90,180,90

Any ideas? thanks!

daryl

point_test.sld (1.96 KB)

current_kcci.kml.gz (1.13 KB)

2

Daryl,

  When I try the KML version, I get the behavior you described..

  But when I try the KMZ version it looks fine. I'm not getting any
duplicate points.

  Still since the google earth support is one of the big "selling
points" for using Geoserver, it would be nice to know why the straight
kml response is so funky... Any ideas?

- matt

On 11/30/06, Daryl Herzmann <akrherz@anonymised.com> wrote:

Hi,

I'm trying to get a point layer in my postgis database to be served via
Geoserver to Google Earth. I haven't been able to get it to work.

When I use

format=application/vnd.google-earth.kml+XML

The server delays for a number of seconds and I finally get a truncated XML
file (see attached current_kcci.kml.gz) and nothing interesting in the log
files on the server.

When I use

format=application/vnd.google-earth.kmz+XML

The server immediately responds, but Google Earth appears to have
duplicate points. When I click on the point, two points appear, each with
a label.

I've attached my sld (point_test.sld) and the URI to my instance is:

http://mesonet.agron.iastate.edu/geoserver//wms?service=WMS&request=GetMap&format=application/vnd.google-earth.kmz+XML&width=1024&height=1024&srs=EPSG:4326&layers=topp:current_kcci&styles=poi&bbox=-180,-90,180,90

Any ideas? thanks!

daryl

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users

--
Matthew T. Perry
GIS Analyst / Software Engineer
National Center for Ecological Analysis and Synthesis (NCEAS)
work: perry@anonymised.com
web: http://www.perrygeo.net

3

Hi Daryl

This seems really odd. Could you change the logging level to 'fine', and try the request again?

What is weird is that when you specify for a kmz result, it is returning valid kml in the zipped up kmz file. But for just regular kml it isn't working. There seem to be very few features as well, so it isn't a size issue.The style also looks fine. Although looking at the request you aren't using the style you specified.
I tried to do a getCapabilities request but that is also getting truncated part way through the response. A describe feature type request is also not working. So something weird is going on here.

Can you try the same dataset but as a shapefile instead?

Brent Owens
(The Open Planning Project)

Daryl Herzmann wrote:

Hi,

I'm trying to get a point layer in my postgis database to be served via Geoserver to Google Earth. I haven't been able to get it to work.

When I use

format=application/vnd.google-earth.kml+XML

The server delays for a number of seconds and I finally get a truncated XML file (see attached current_kcci.kml.gz) and nothing interesting in the log files on the server.

When I use

format=application/vnd.google-earth.kmz+XML

The server immediately responds, but Google Earth appears to have duplicate points. When I click on the point, two points appear, each with a label.

I've attached my sld (point_test.sld) and the URI to my instance is:

http://mesonet.agron.iastate.edu/geoserver//wms?service=WMS&request=GetMap&format=application/vnd.google-earth.kmz+XML&width=1024&height=1024&srs=EPSG:4326&layers=topp:current_kcci&styles=poi&bbox=-180,-90,180,90

Any ideas? thanks!

daryl
------------------------------------------------------------------------

<?xml version="1.0" encoding="ISO-8859-1"?>

<StyledLayerDescriptor version="1.0.0"

    xsi:schemaLocation="http://www.opengis.net/sld StyledLayerDescriptor.xsd"

    xmlns="http://www.opengis.net/sld&quot;

    xmlns:ogc="http://www.opengis.net/ogc&quot;

    xmlns:xlink="http://www.w3.org/1999/xlink&quot;

    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance&quot;&gt;

    <!-- a named layer is the basic building block of an sld document -->

  <NamedLayer>

    <Name>Default Point</Name>

    <UserStyle>

        <!-- they have names, titles and abstracts -->

      <Title>A boring default style</Title>

      <Abstract>A sample style that just prints out a purple circle</Abstract>

      <!-- FeatureTypeStyles describe how to render different features -->

      <!-- a feature type for points -->

      <FeatureTypeStyle>

        <!--FeatureTypeName>Feature</FeatureTypeName-->

        <Rule>

          <Name>Rule 1</Name>

          <Title>PurpleFill</Title>

          <Abstract>A purple fill with an 11 pixel size</Abstract>

          <!-- like a linesymbolizer but with a fill too -->

          <PointSymbolizer>

            <Graphic>

              <Mark>

                <WellKnownName>circle</WellKnownName>

                <Fill>

                  <CssParameter name="fill">#9F4DC5</CssParameter>

                  <CssParameter name="fill-opacity">1.0</CssParameter>

                </Fill>

              </Mark>

              <Size>6</Size>

            </Graphic>

          </PointSymbolizer>

                    <TextSymbolizer>

            <Label>

              <ogc:PropertyName>sname</ogc:PropertyName>

            </Label>

            <Font>

              <CssParameter name="font-style">Normal</CssParameter>

              <CssParameter name="font-size">16</CssParameter>

              <CssParameter name="font-weight">bold</CssParameter>

            </Font>

            <Fill>

              <CssParameter name="fill">#FF0000</CssParameter>

            </Fill>

          </TextSymbolizer>

        </Rule>

        </FeatureTypeStyle>

    </UserStyle>

  </NamedLayer>

</StyledLayerDescriptor>

  ------------------------------------------------------------------------

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
------------------------------------------------------------------------

_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users
  

4

Could you report it as a bug in JIRA? http://jira.codehaus.org/browse/GEOS

I don't think we were aware that KML had problems...

Chris

Matthew Perry wrote:

Daryl,

  When I try the KML version, I get the behavior you described..

  But when I try the KMZ version it looks fine. I'm not getting any
duplicate points.

  Still since the google earth support is one of the big "selling
points" for using Geoserver, it would be nice to know why the straight
kml response is so funky... Any ideas?

- matt

On 11/30/06, Daryl Herzmann <akrherz@anonymised.com> wrote:

Hi,

I'm trying to get a point layer in my postgis database to be served via
Geoserver to Google Earth. I haven't been able to get it to work.

When I use

format=application/vnd.google-earth.kml+XML

The server delays for a number of seconds and I finally get a truncated XML
file (see attached current_kcci.kml.gz) and nothing interesting in the log
files on the server.

When I use

format=application/vnd.google-earth.kmz+XML

The server immediately responds, but Google Earth appears to have
duplicate points. When I click on the point, two points appear, each with
a label.

I've attached my sld (point_test.sld) and the URI to my instance is:

http://mesonet.agron.iastate.edu/geoserver//wms?service=WMS&request=GetMap&format=application/vnd.google-earth.kmz+XML&width=1024&height=1024&srs=EPSG:4326&layers=topp:current_kcci&styles=poi&bbox=-180,-90,180,90

Any ideas? thanks!

daryl

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users

--
Chris Holmes
The Open Planning Project
http://topp.openplans.org

5

Do you guys have any sample datasets that we can use to duplicate the problem?
I've tested it with all of my data and I can't replicate it yet.

Brent Owens
(The Open Planning Project)

Matthew Perry wrote:

Daryl,

  When I try the KML version, I get the behavior you described..

  But when I try the KMZ version it looks fine. I'm not getting any
duplicate points.

  Still since the google earth support is one of the big "selling
points" for using Geoserver, it would be nice to know why the straight
kml response is so funky... Any ideas?

- matt

On 11/30/06, Daryl Herzmann <akrherz@anonymised.com> wrote:
  

Hi,

I'm trying to get a point layer in my postgis database to be served via
Geoserver to Google Earth. I haven't been able to get it to work.

When I use

format=application/vnd.google-earth.kml+XML

The server delays for a number of seconds and I finally get a truncated XML
file (see attached current_kcci.kml.gz) and nothing interesting in the log
files on the server.

When I use

format=application/vnd.google-earth.kmz+XML

The server immediately responds, but Google Earth appears to have
duplicate points. When I click on the point, two points appear, each with
a label.

I've attached my sld (point_test.sld) and the URI to my instance is:

http://mesonet.agron.iastate.edu/geoserver//wms?service=WMS&request=GetMap&format=application/vnd.google-earth.kmz+XML&width=1024&height=1024&srs=EPSG:4326&layers=topp:current_kcci&styles=poi&bbox=-180,-90,180,90

Any ideas? thanks!

daryl

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users

6

Hi Brent,

Thanks for your email.

This seems really odd. Could you change the logging level to 'fine', and try the request again?

For the kml request I get:

8399097 [FINE] org.vfny.geoserver.servlets.AbstractService - reading request: service=WMS&request=GetMap&format=application/vnd.google-earth.kml+XML&width=1024&height=1024&srs=EPSG:4326&styles=snet8&layers=topp:current_kcci&bbox=-180,-90,180,90&BBOX=-93.84408985736377,41.49898648304644,-93.66650786674322,41.61480866506157
8399097 [FINE] org.vfny.geoserver.global.Data - getting type topp:current_kcci
8399098 [FINE] org.geotools.data.postgis.PostgisDataStore - get Feature source called on current_kcci
8399098 [FINE] org.geotools.data.postgis.PostgisDataStore - returning pg feature locking
8399098 [FINE] org.geotools.styling.StyleImpl - number of fts set 1
8399098 [FINE] org.vfny.geoserver.wms.requests.GetMapKvpReader - establishing snet8 style for topp:current_kcci
8399099 [INFO] org.vfny.geoserver.servlets.AbstractService - handling request: org.vfny.geoserver.wms.requests.GetMapRequest@anonymised.com
8399099 [FINE] org.vfny.geoserver.servlets.AbstractService - strategy is: PARTIAL-BUFFER
8399100 [FINE] org.vfny.geoserver.wms.responses.GetMapResponse - setting up map
8399100 [FINE] org.geotools.data.postgis.PostgisDataStore - get Feature source called on current_kcci
8399100 [FINE] org.geotools.data.postgis.PostgisDataStore - returning pg feature locking
8399100 [FINE] org.vfny.geoserver.servlets.AbstractService - mime type is: application/vnd.google-earth.kml+xml
8399101 [FINE] org.vfny.geoserver.servlets.AbstractService - content encoding is: null
8399101 [FINE] org.geotools.data.jdbc.ConnectionPool - Getting available connection.
8399103 [FINE] org.geotools.data.jdbc.ConnectionPool$ConnectionListManager - Connection closed - adding to available connections.
8399103 [INFO] org.vfny.geoserver.wms.responses.map.kml.EncodeKML - Layer () rendered with KML vector output.
8399104 [FINE] org.geotools.styling.StyleImpl - number of fts set 1
8399104 [FINE] org.geotools.data.jdbc.JDBC1DataStore - calling sql builder with filter [ geom bbox POLYGON ((-180 -90, -180 90, 180 90, 180 -90, -180 -90)) ]
8399105 [FINE] org.geotools.data.jdbc.JDBC1DataStore - sql is SELECT "oid", "station", encode(AsBinary(force_2d("geom"), 'XDR'),'base64'), "network", "sname", "tmpf", "dwpf", "drct", "sknt" FROM "public"."current_kcci" WHERE "geom" && GeometryFromText('POLYGON ((-180 -90, -180 90, 180 90, 180 -90, -180 -90))', 4326)
8399105 [FINE] org.geotools.data.jdbc.JDBC1DataStore - About to execute query: SELECT "oid", "station", encode(AsBinary(force_2d("geom"), 'XDR'),'base64'), "network", "sname", "tmpf", "dwpf", "drct", "sknt" FROM "public"."current_kcci" WHERE "geom" && GeometryFromText('POLYGON ((-180 -90, -180 90, 180 90, 180 -90, -180 -90))', 4326)
8399105 [FINE] org.geotools.data.jdbc.ConnectionPool - Getting available connection.
8399108 [INFO] org.geotools.renderer.style.SLDStyleFactory - Bad file name in SLDStyleFactoryserif
java.io.FileNotFoundException: serif: No such file or directory
8399109 [INFO] org.geotools.renderer.style.SLDStyleFactory - null input stream

What is weird is that when you specify for a kmz result, it is returning valid kml in the zipped up kmz file. But for just regular kml it isn't working. There seem to be very few features as well, so it isn't a size issue.

Should only be 71 features.

The style also looks fine. Although looking at the request you aren't using the style you specified.

Good point. It should be 'snet8'. I think I just gave you an older URL i have been testing with.

I tried to do a getCapabilities request but that is also getting truncated part way through the response. A describe feature type request is also not working. So something weird is going on here.

Hopefully it isn't me!

Can you try the same dataset but as a shapefile instead?

Ok, I dumped the shapefile out of the database and added it, I changed the style to use the UPPERCASE attribute.

I'm getting the same errors. kml hangs, kmz has double :frowning:

http://mesonet.agron.iastate.edu/geoserver/wms?service=WMS&request=GetMap&format=application/vnd.google-earth.kml+XML&width=1024&height=1024&srs=EPSG:4326&styles=snet8_shp&layers=topp:current_kcci_shp&bbox=-180,-90,180,90&

thanks and sorry,
   daryl

--
/**
  * Daryl Herzmann, RHCE
  * Program Assistant -- Iowa Environmental Mesonet
  * http://mesonet.agron.iastate.edu
  */

7

On Thu, 30 Nov 2006, Brent Owens wrote:

Do you guys have any sample datasets that we can use to duplicate the problem? I've tested it with all of my data and I can't replicate it yet.

My shapefile can be found here:

http://mesonet.agron.iastate.edu/pickup/geoserver.zip

I've been using Google Earth on Linux this afternoon. Will test on windows this evening and see if it does the same. As I recall, this is the same behaviour I was seeing a couple months ago, but the kmz was working at the time. I was hitting other bugs at the time and didn't follow up with you guys, sorry.

daryl

--
/**
  * Daryl Herzmann, RHCE
  * Program Assistant -- Iowa Environmental Mesonet
  * http://mesonet.agron.iastate.edu
  */

8

Ok I tested it, and after fixing a couple of other bugs, I still can't replicate your problem.
Are you using jetty or tomcat, or another container?

It's very strange to have the request successfully completed, but not returned fully. So something is prematurely terminating the output and closing the stream. Especially since it is doing this to most requests like the getCapabilities.
I'm going to do a code review to see what I can find.

If you guys have any other information that might help, send it our way and we will try to fix it as soon as possible.
cheers,

Brent Owens
(The Open Planning Project)

Daryl Herzmann wrote:

On Thu, 30 Nov 2006, Brent Owens wrote:

Do you guys have any sample datasets that we can use to duplicate the problem? I've tested it with all of my data and I can't replicate it yet.

My shapefile can be found here:

http://mesonet.agron.iastate.edu/pickup/geoserver.zip

I've been using Google Earth on Linux this afternoon. Will test on windows this evening and see if it does the same. As I recall, this is the same behaviour I was seeing a couple months ago, but the kmz was working at the time. I was hitting other bugs at the time and didn't follow up with you guys, sorry.

daryl

9

On Fri, 1 Dec 2006, Brent Owens wrote:

Ok I tested it, and after fixing a couple of other bugs, I still can't replicate your problem. Are you using jetty or tomcat, or another container?

I believe Jetty from the deployment zip file. Perhaps the only difference is that my geoserver is behind a reverse proxy? I just did a test and that doesn't seem to be an issue.

If you guys have any other information that might help, send it our way and we will try to fix it as soon as possible. cheers,

Thanks. Will keep trying here too!

daryl

--
/**
  * Daryl Herzmann, RHCE
  * Program Assistant -- Iowa Environmental Mesonet
  * http://mesonet.agron.iastate.edu
  */