Opened 13 years ago
Last modified 13 years ago
#3977 assigned defect
Can't WFS getFeature, nqeury, WMS Getfeatureinfo w/MapServer 6.0.1 and ArcSDE
Reported by: | myopic73 | Owned by: | sdlime |
---|---|---|---|
Priority: | high | Milestone: | 6.0.1 release |
Component: | MapServer C Library | Version: | 6.0 |
Severity: | normal | Keywords: | getfeatureinfo arcsde |
Cc: | sholl, jmckenna |
Description
Steve Lime asked me to add the ticket, Subject: Re: RE: Can't WFS getFeature w/MapServer 6.0.1
Yes, it did. The problem was that the GetFeatureInfo code in 6 does two layer open/close processes. Or, at least it tries to. It fails on the second one because the code tries to reset the connection which has already been cleared before it does the layer open. Now that it doesn't fail with the reset, the second layer open succeeds and the query is processed.
Interestingly, the problem itself has been around for a while (see http://osgeo-org.1803224.n2.nabble.com/error-while-closing- sde-layer-td5464449.html in re: MapServer 5.6.3). It's only become apparent in these OGC queries in MapServer 6 because the code for the queries have changed so much and it's peppered with these connection resets before doing any layer open. They weren't doing that before. And, apparently, most of that code was tested with Shapefiles, so the failures in the SDE connections wouldn't have shown up. Having said that, I've also found a similar problem that exhibits itself in certain circumstances with the Shapefiles. The MapServer developers are aware of that one too.
in mapsde.c's msSDELayerFreeItemInfo to handle the SDE problems.
see thread for possible resolution.. http://osgeo-org.1803224.n2.nabble.com/Can-t-WFS-getFeature-w-MapServer-6-0-1-td6600286.html
Change History (10)
comment:1 by , 13 years ago
comment:2 by , 13 years ago
Status: | new → assigned |
---|
comment:3 by , 13 years ago
Cc: | added |
---|
comment:4 by , 13 years ago
comment:5 by , 13 years ago
Steve,
I have just used the proposed patch to build a new MapServer 6.0.1 executable. I am able to successfully perform a WFS getFeature request against data in SDE when the request has a spatial filter defined AND a maxFeatures parameter limitation...which was what I couldn't do when this whole thing started.
In re: "can you refresh my memory on the similar shapefile issue?" It's the same problem. I can't make a successful WFS getFeature request that has a spatial filter defined AND a maxFeatures parameter limitation when I'm using a Shapefile either. The request will correctly respond with a maxFeatures parameter OR the spatial filter defined, but not both.
Cheers, jtm
comment:6 by , 13 years ago
I know the issue (check a new thread started on mapserver-dev) with regards to maxFeatures and hope to patch 6.0 and trunk very soon with a fix. I'll post the ticket for these changes here ASAP...
Steve
follow-up: 9 comment:8 by , 13 years ago
Cc: | added |
---|
Stephan: I have hit this truncated getcapabilities response again (on Windows). Should we open a new ticket for this? (we'll probably have to create a small test package so devs can replicate)
comment:9 by , 13 years ago
Jeff,
Replying to jmckenna:
Stephan: I have hit this truncated getcapabilities response again (on Windows). Should we open a new ticket for this? (we'll probably have to create a small test package so devs can replicate)
Yes, seems reasonable. I cannot create a Testopackage since my SDE-datasets are not available. If this applies also on not-SDE-features, I am happy to create a small shapefile-driven testcase. But I am sure it needs SDE to reproduce it.
Are you willing to create a new ticket though?
Best
Stephan
comment:10 by , 13 years ago
Stephan: yesterday I hit the exact same issue, using PostGIS, and I filed a new ticket (ticket:4039). My problem is that this issue happens on a client's machine, and I cannot duplicate locally with PostGIS.
There's a problem somewhere here...but we just need to be able to create a small testcase so the devs can tackle it. :(
I've patched branch-6-0 with what I hope is the correct fix. It's basically what was suggested earlier. The msSDELayerFreeItemInfo() was doing more work than it should compared to the other drivers so I pulled the extra code too.
Changes are in r12031, anyway you can test on your end?
Steve
myopic73, can you refresh my memory on the similar shapefile issue?