[Geoserver-users] Bug with EPSG:41001 in Geoserver 1.4.0-RC2?

GeoServer GeoServer User
1

Hello,

i updated from the latest stable Geoserver to 1.4.0-RC2. I let Geoserver
feed my Google Maps. For this i use the script "wms-gs-1_0_1.js".

When the script uses EPSG:41001, i get an empty image, when it uses
EPSG:4326, it looks fine.

See http://freesync.ath.cx/demo/index.php , Layer Demo, when loaded
Zoomlevel is 4 and you'll see only a blue background, nothing else. Zoom 1
step up to Zoomlevel 5 and you get the map.

This works:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:4326
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=33.75,31.95216223802497,45,40.97989806962013
&WIDTH=256
&HEIGHT=256
&reaspect=false

And this not:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:41001
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=-4992581.516097316,0.052157810355956576,-2496290.758048658,2496290.8123096717
&WIDTH=256
&HEIGHT=256
&reaspect=false

I'm pretty shure this has worked for 1.3. Is this possible a bug? Or can
someone confirm this should work?

Regards,

Frank
--
View this message in context: http://www.nabble.com/Bug-with-EPSG%3A41001-in-Geoserver-1.4.0-RC2--tf2558721.html#a7130516
Sent from the GeoServer - User mailing list archive at Nabble.com.

2

I've just verified (for the lack of responses so far), that this IS a bug in
1.4.0RC2.

I've installed 1.3.4a and changed nothing else (same sourcecodes,
data-directory and db-table) and my map shows up pretty nice.

For testing / seeing the difference between 1.3.4.a and 1.4.0 RC2, i'm
willing to let both versions running here and provide links to them, if i
just know if it is possible to run 2 different Geoserver on one machine and
who/where i have to change the port of one from both Geoservers.

Note: Since friday, 2006-11-3 runs the older version on the below link,
where you will not see anything other than expected.

If some of the developers need further information, drop me a note.

Regards, Frank

dfd wrote:

Hello,

i updated from the latest stable Geoserver to 1.4.0-RC2. I let Geoserver
feed my Google Maps. For this i use the script "wms-gs-1_0_1.js".

When the script uses EPSG:41001, i get an empty image, when it uses
EPSG:4326, it looks fine.

See http://freesync.ath.cx/demo/index.php , Layer Demo, when loaded
Zoomlevel is 4 and you'll see only a blue background, nothing else. Zoom 1
step up to Zoomlevel 5 and you get the map.

This works:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:4326
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=33.75,31.95216223802497,45,40.97989806962013
&WIDTH=256
&HEIGHT=256
&reaspect=false

And this not:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:41001
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=-4992581.516097316,0.052157810355956576,-2496290.758048658,2496290.8123096717
&WIDTH=256
&HEIGHT=256
&reaspect=false

I'm pretty shure this has worked for 1.3. Is this possible a bug? Or can
someone confirm this should work?

Regards,

Frank

--
View this message in context: http://www.nabble.com/Bug-with-EPSG%3A41001-in-Geoserver-1.4.0-RC2--tf2558721.html#a7153917
Sent from the GeoServer - User mailing list archive at Nabble.com.

3

Thanks for the info frank.
We will be taking a look into this today and will keep you posted on our progress.

Brent Owens
(The Open Planning Project)

dfd wrote:

I've just verified (for the lack of responses so far), that this IS a bug in
1.4.0RC2.

I've installed 1.3.4a and changed nothing else (same sourcecodes,
data-directory and db-table) and my map shows up pretty nice.

For testing / seeing the difference between 1.3.4.a and 1.4.0 RC2, i'm
willing to let both versions running here and provide links to them, if i
just know if it is possible to run 2 different Geoserver on one machine and
who/where i have to change the port of one from both Geoservers.

Note: Since friday, 2006-11-3 runs the older version on the below link,
where you will not see anything other than expected.

If some of the developers need further information, drop me a note.

Regards, Frank

dfd wrote:
  

Hello,

i updated from the latest stable Geoserver to 1.4.0-RC2. I let Geoserver
feed my Google Maps. For this i use the script "wms-gs-1_0_1.js".

When the script uses EPSG:41001, i get an empty image, when it uses
EPSG:4326, it looks fine.

See http://freesync.ath.cx/demo/index.php , Layer Demo, when loaded
Zoomlevel is 4 and you'll see only a blue background, nothing else. Zoom 1
step up to Zoomlevel 5 and you get the map.

This works:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:4326
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=33.75,31.95216223802497,45,40.97989806962013
&WIDTH=256
&HEIGHT=256
&reaspect=false

And this not:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:41001
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=-4992581.516097316,0.052157810355956576,-2496290.758048658,2496290.8123096717
&WIDTH=256
&HEIGHT=256
&reaspect=false

I'm pretty shure this has worked for 1.3. Is this possible a bug? Or can
someone confirm this should work?

Regards,

Frank

4

Hi Frank,
Can you see if any errors are generated in the logs when you make a request with the 41001 projection?
Also, the two requests that you supplied, are they supposed to be in the same area? Or over different locations?

thanks,

Brent Owens
(The Open Planning Project)

dfd wrote:

I've just verified (for the lack of responses so far), that this IS a bug in
1.4.0RC2.

I've installed 1.3.4a and changed nothing else (same sourcecodes,
data-directory and db-table) and my map shows up pretty nice.

For testing / seeing the difference between 1.3.4.a and 1.4.0 RC2, i'm
willing to let both versions running here and provide links to them, if i
just know if it is possible to run 2 different Geoserver on one machine and
who/where i have to change the port of one from both Geoservers.

Note: Since friday, 2006-11-3 runs the older version on the below link,
where you will not see anything other than expected.

If some of the developers need further information, drop me a note.

Regards, Frank

dfd wrote:
  

Hello,

i updated from the latest stable Geoserver to 1.4.0-RC2. I let Geoserver
feed my Google Maps. For this i use the script "wms-gs-1_0_1.js".

When the script uses EPSG:41001, i get an empty image, when it uses
EPSG:4326, it looks fine.

See http://freesync.ath.cx/demo/index.php , Layer Demo, when loaded
Zoomlevel is 4 and you'll see only a blue background, nothing else. Zoom 1
step up to Zoomlevel 5 and you get the map.

This works:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:4326
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=33.75,31.95216223802497,45,40.97989806962013
&WIDTH=256
&HEIGHT=256
&reaspect=false

And this not:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:41001
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=-4992581.516097316,0.052157810355956576,-2496290.758048658,2496290.8123096717
&WIDTH=256
&HEIGHT=256
&reaspect=false

I'm pretty shure this has worked for 1.3. Is this possible a bug? Or can
someone confirm this should work?

Regards,

Frank

5

Hi Brent,

no, there are no errors in the logs (as far as i understand the logs, second
last in a line is always 200).

And for the requests, these are different zoomlevels of the same area, not
moving the map.

I let the Google-API write out all requests for debugging, from these i've
taken one. To show the whole map, there are a few more requests with other
bounding boxes. But all requests shows the same: nothing, even in zoomlevel
1 which should show the world. In 1.3.4 this is fine.

Let me know if i could something do to help you narrow down the bug.

Regards, Frank

Brent Owens wrote:

Hi Frank,
Can you see if any errors are generated in the logs when you make a
request with the 41001 projection?
Also, the two requests that you supplied, are they supposed to be in the
same area? Or over different locations?

thanks,

Brent Owens
(The Open Planning Project)

dfd wrote:

I've just verified (for the lack of responses so far), that this IS a bug
in
1.4.0RC2.

I've installed 1.3.4a and changed nothing else (same sourcecodes,
data-directory and db-table) and my map shows up pretty nice.

For testing / seeing the difference between 1.3.4.a and 1.4.0 RC2, i'm
willing to let both versions running here and provide links to them, if i
just know if it is possible to run 2 different Geoserver on one machine
and
who/where i have to change the port of one from both Geoservers.

Note: Since friday, 2006-11-3 runs the older version on the below link,
where you will not see anything other than expected.

If some of the developers need further information, drop me a note.

Regards, Frank

dfd wrote:
  

Hello,

i updated from the latest stable Geoserver to 1.4.0-RC2. I let Geoserver
feed my Google Maps. For this i use the script "wms-gs-1_0_1.js".

When the script uses EPSG:41001, i get an empty image, when it uses
EPSG:4326, it looks fine.

See http://freesync.ath.cx/demo/index.php , Layer Demo, when loaded
Zoomlevel is 4 and you'll see only a blue background, nothing else. Zoom
1
step up to Zoomlevel 5 and you get the map.

This works:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:4326
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=33.75,31.95216223802497,45,40.97989806962013
&WIDTH=256
&HEIGHT=256
&reaspect=false

And this not:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:41001
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=-4992581.516097316,0.052157810355956576,-2496290.758048658,2496290.8123096717
&WIDTH=256
&HEIGHT=256
&reaspect=false

I'm pretty shure this has worked for 1.3. Is this possible a bug? Or can
someone confirm this should work?

Regards,

Frank

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job
easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users

--
View this message in context: http://www.nabble.com/Bug-with-EPSG%3A41001-in-Geoserver-1.4.0-RC2--tf2558721.html#a7164673
Sent from the GeoServer - User mailing list archive at Nabble.com.

6

This is probably due to the switch between wkt and hsql databases for
epsg, no? So I guess it means we need to update the epsg database with
the proper well known text.

dfd wrote:

I've just verified (for the lack of responses so far), that this IS a bug in
1.4.0RC2.

I've installed 1.3.4a and changed nothing else (same sourcecodes,
data-directory and db-table) and my map shows up pretty nice.

For testing / seeing the difference between 1.3.4.a and 1.4.0 RC2, i'm
willing to let both versions running here and provide links to them, if i
just know if it is possible to run 2 different Geoserver on one machine and
who/where i have to change the port of one from both Geoservers.

Note: Since friday, 2006-11-3 runs the older version on the below link,
where you will not see anything other than expected.

If some of the developers need further information, drop me a note.

Regards, Frank

dfd wrote:

Hello,

i updated from the latest stable Geoserver to 1.4.0-RC2. I let Geoserver
feed my Google Maps. For this i use the script "wms-gs-1_0_1.js".

When the script uses EPSG:41001, i get an empty image, when it uses
EPSG:4326, it looks fine.

See http://freesync.ath.cx/demo/index.php , Layer Demo, when loaded
Zoomlevel is 4 and you'll see only a blue background, nothing else. Zoom 1
step up to Zoomlevel 5 and you get the map.

This works:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:4326
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=33.75,31.95216223802497,45,40.97989806962013
&WIDTH=256
&HEIGHT=256
&reaspect=false

And this not:

http://freesync.ath.cx:8080/geoserver/wms
?REQUEST=GetMap
&SRS=EPSG:41001
&LAYERS=topp:world
&STYLES=world
&FORMAT=image/png
&BGCOLOR=0x99B3CC
&TRANSPARENT=False
&BBOX=-4992581.516097316,0.052157810355956576,-2496290.758048658,2496290.8123096717
&WIDTH=256
&HEIGHT=256
&reaspect=false

I'm pretty shure this has worked for 1.3. Is this possible a bug? Or can
someone confirm this should work?

Regards,

Frank

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

7

Chris,

i don't know what hsql is? But it has to with the epgs 41001. Using 4326 is
ok, this is btw. the srs in the definition of the feature-layer.

Frank

Chris Holmes wrote:

This is probably due to the switch between wkt and hsql databases for
epsg, no? So I guess it means we need to update the epsg database with
the proper well known text.
--
Chris Holmes
The Open Planning Project
http://topp.openplans.org

--
View this message in context: http://www.nabble.com/Bug-with-EPSG%3A41001-in-Geoserver-1.4.0-RC2--tf2558721.html#a7168698
Sent from the GeoServer - User mailing list archive at Nabble.com.

8

Chris Holmes ha scritto:

This is probably due to the switch between wkt and hsql databases for
epsg, no? So I guess it means we need to update the epsg database with
the proper well known text.

1.4.0 is still running against epsg-wkt, only trunk is running against epsg-hsql. So I guess that's not the problem.
Cheers
Andrea

9

dfd ha scritto:

Hi Brent,

no, there are no errors in the logs (as far as i understand the logs, second
last in a line is always 200).

Please go to the geoserver configuration, set level to fine, enable logging to file, rertart geoserver, and then look again at the log
file... Is there an exception this time?
Cheers
Andrea

10

Andrea,

no, i can't see any exception.

You can take a look at http://freesync.ath.cx/demo/geoserver_log.txt (ca. 680 k) or download as zip http://freesync.ath.cx/demo/geoserver.zip (32k)

Strange enougth, i've done a test using the shape-based us-states demo, works like a charm. So now i think this is not related to EGPS:41001.
I took the generated sql and put it in pgadmin, no errors there, got some records.
Then i've used another SLD (the default polygon) to see if it's SLD-related, but no luck, nothing to display, only the background comes back..
What's next?

Regards,

        Frank

Andrea Aime schrieb am 04.11.2006 14:52:

dfd ha scritto:

Hi Brent,

no, there are no errors in the logs (as far as i understand the logs, second
last in a line is always 200).

Please go to the geoserver configuration, set level to fine, enable logging to file, rertart geoserver, and then look again at the log
file... Is there an exception this time?
Cheers
Andrea

---
avast! Antivirus auf durstewitz.biz: Eingehende Nachricht virenfrei.
Getestet am: 04.11.2006 14:52:33
Virus-Datenbank (VPS): 0645-4, 03.11.2006

11

Did you set the srs of your data right when you loaded it up in to postgis? That could lead to results that are a bit off...

Frank Durstewitz wrote:

Andrea,

no, i can't see any exception.

You can take a look at http://freesync.ath.cx/demo/geoserver_log.txt (ca. 680 k) or download as zip http://freesync.ath.cx/demo/geoserver.zip (32k)

Strange enougth, i've done a test using the shape-based us-states demo, works like a charm. So now i think this is not related to EGPS:41001.
I took the generated sql and put it in pgadmin, no errors there, got some records.
Then i've used another SLD (the default polygon) to see if it's SLD-related, but no luck, nothing to display, only the background comes back..
What's next?

Regards,

       Frank

Andrea Aime schrieb am 04.11.2006 14:52:

dfd ha scritto:

Hi Brent,

no, there are no errors in the logs (as far as i understand the logs, second
last in a line is always 200).

Please go to the geoserver configuration, set level to fine, enable logging to file, rertart geoserver, and then look again at the log
file... Is there an exception this time?
Cheers
Andrea

---
avast! Antivirus auf durstewitz.biz: Eingehende Nachricht virenfrei.
Getestet am: 04.11.2006 14:52:33
Virus-Datenbank (VPS): 0645-4, 03.11.2006

!DSPAM:1003,454cccf8322438365517736!

---
avast! Antivirus auf durstewitz.biz: Ausgehende Nachricht virenfrei.
Getestet am: 04.11.2006 18:24:09
Virus-Datenbank (VPS): 0645-4, 03.11.2006

!DSPAM:1003,454cccf8322438365517736!

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

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

!DSPAM:1003,454cccf8322438365517736!

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

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

!DSPAM:1003,454cccf8322438365517736!

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

12

Chris,

you got it.

I've imported the shapes via shp2pgsql, which does srs -1 and in the table-definition i also defined -1 instead of 4326.
Just wondering version in 1.3.4 does work with srs -1?

Thank you, Chris.

Best regards,

    Frank

Chris Holmes schrieb am 04.11.2006 21:21:

Did you set the srs of your data right when you loaded it up in to postgis? That could lead to results that are a bit off...

13

Frank Durstewitz ha scritto:

Chris,

you got it.

I've imported the shapes via shp2pgsql, which does srs -1 and in the table-definition i also defined -1 instead of 4326.
Just wondering version in 1.3.4 does work with srs -1?

Hmm... no exact idea, since we don't have many insights about were the problem lies in the codebase, yet, I'd say geotools got pickier about srs definitions... In the past gt has been very sloppy towards CRS handling, and got "better" during time. Yet I would have expected a
stack trace or some kind, not just an empty image...

Cheers
Andrea