[mapserver-users] Protocol version related problems

[Rahkonen Jukka]

Hi,
 
I would suggest to read a bit about WMS 1.1.1 vs. 1.3.0 and axis order of some projections mandated by the standard, and also about WFS 1.0.0 vs. 1.1.0.  
 
-Jukka Rahkonen-

________________________________

Lähettäjä: mapserver-users-bounces at lists.osgeo.org [mailto:mapserver-users-bounces at lists.osgeo.org] Puolesta SWEYKAM at telefonica.net
Lähetetty: 25. tammikuuta 2011 21:06
Vastaanottaja: mapserver-users at lists.osgeo.org
Aihe: [mapserver-users] Protocol version related problems


I would highly appreciate some advice on the following problem:

WMS layers only display correctly when using HTTP GET requests version 1.1.1 while  1.3.0 causes a distortion of the shape and a spatial displacement.

WFS layers only display correctly when using HTTP POST requests version 1.0.0 while with 1.1.0 don't display (at least not within the view extent).

HTTP GET requests version 1.1.0 on the same WFS layer cause -no shape distortion- but an inversion and 90ºrotation of the shape plus a spatial displacement.

I have also observed that with the WFS protocol version 1.1.0 the 'lowerCorner' and 'upperCorner' coordinates are switched around: lat-long instead of lon-lat.

Looks to me as if MapServer messes up the map/layer extents, which are properly defined at MAP, LAYER and Metdadata level.

Both MAP and LAYER are in geographic coordinates.

When I request a projected WFS layer (POST + any protocol version) the server throws a 500 error.

I'm logging what MapServer does (DEBUG level 3 with PROJ_DEBUG and CPL_DEBUG on) but can't see anything suspicious.

As for the server error on POST+reprojected layer, the server logs a 'premature end of script header' in the error.log and 'HTTP/1.1" 500 538 "-" "Python-urllib/2.6" in the access.log

According to my hosting service they are using Debian Lenny, and have compiled and installed the Debian packages for MapServer 5.6.5, GDAL 1.5.4 and PROJ 4.7.0.

All services work fine on my local MS4W installation (vers. 5.2.1)

Does that make any sense?????
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/mapserver-users/attachments/20110125/6aa444ce/attachment.html