> > edge. I'll put together a barebones .map and pass around > > that will hopefully be able to reproduce the problem.
How's this for interesting . . . The problem persists, but for the purposes of this little .tar.Z I dumped the table to a shapefile. And here is what happens: Doing a GetCapabilities on everything like normal (shapefile where it should be), http://your.box.here/cgi-bin/mapserv?map=/I/love/mapserver/m.map&request=get capabilities&service=sos returns an empty XML just like I promised. Now do one of two things: * remove ANY layer, and you should get back a nice XML doc (this is in keeping w/ what I reported earlier, i.e. this # of layers has sent mapserver over the edge). * or leave all the layers and simply remove the shapefile goodies. The getcaps works fine! Strange. I was surprised that it hit the data source at all for a getcaps. Oddly, something gets a little foobared when doing a getobs against this test dataset, and I'm not sure why. But I'm not really concerned, since it was able to return nice getobs from my real db no matter how many layers I had and whether or not the getcaps failed. Thanks for the help, guys. Charlton
sos_test.tar.Z
Description: Unix compressed data
