[Geoserver-users] WFS-T oracle update faild on GS 1.7.2

2StepForward · March 4, 2009, 2:25pm

Hello List,

I have took a update from GS 1.6.3 to 1.7.2 and most things are working
great.
But one thing always will fail, the wfs-t update on oracle backend.

System is: Oracle 10g, new Oracle NG plugin and GeoServer 1.7.2

So if I send a wfs-t update post I will get an error on update.

Stacktrace have wrote this error messages:

2009-03-04 14:46:54,662 ERROR [geoserver.wfs] - Transaction failed
org.geoserver.wfs.WFSTransactionException: update error
    at
org.geoserver.wfs.UpdateElementHandler.execute(UpdateElementHandler.java:239)
    at org.geoserver.wfs.Transaction.execute(Transaction.java:333)
    at org.geoserver.wfs.Transaction.transaction(Transaction.java:108)
    at
org.geoserver.wfs.DefaultWebFeatureService.transaction(DefaultWebFeatureService.java:182)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)

and so one......

Caused by: java.lang.RuntimeException: java.sql.SQLException: ORA-29877:
failed in the execution of the ODCIINDEXUPDATE routine
ORA-13365: layer SRID does not match geometry SRID
ORA-06512: at "MDSYS.SDO_INDEX_METHOD_10I", line 243

    at org.geotools.jdbc.JDBCDataStore.update(JDBCDataStore.java:1024)
    at org.geotools.jdbc.JDBCDataStore.update(JDBCDataStore.java:997)
    at
org.geotools.jdbc.JDBCUpdateFeatureWriter.write(JDBCUpdateFeatureWriter.java:101)
    ... 74 more

Request: transaction
    handle = null
    service = WFS
    version = 1.0.0
    baseUrl = http://localhost:8081/geoserver/
    providedVersion = 1.0.0
    lockId = null
    group = [wfs:update=net.opengis.wfs.impl.UpdateElementTypeImpl@anonymised.com
(filter: [ GIS_FEATURE.4 ], handle: null, inputFormat: <unset>, srsName:
null, typeName: {http://www.features-on-gis.de/wfs\}GIS_FEATURE)]
    insert =
    update = [net.opengis.wfs.impl.UpdateElementTypeImpl@anonymised.com (filter: [
GIS_FEATURE.4 ], handle: null, inputFormat: <unset>, srsName: null,
typeName: {http://www.feature-on-gis.de/wfs\}GIS_FEATURE)]
    delete =
    native =
    releaseAction = ALL

Days before I had some problems with the different kind of XML-SRS name
style between GeoServer 1.6.3 and 1.7.2 and I had to change some parse
routines of mine, but I realy don't know if it is the possible reason for
that error on update.

Has anyone a idea what could be the reason for that problem, so I will be
very pleased,
2StepForward
--
View this message in context: http://www.nabble.com/WFS-T-oracle-update-faild-on-GS-1.7.2-tp22331042p22331042.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

Kris_Geusebroek · March 4, 2009, 2:26pm

Hi,

I think you don't fill the srsName in the request. If the database
metadata table has an srs filled in fo r the table the feature is
inserted in that might be the mismatch.

Cheers Kris

-----Original Message-----
From: 2StepForward [mailto:kwegezeder@anonymised.com]
Sent: Wednesday, March 04, 2009 3:26 PM
To: geoserver-users@lists.sourceforge.net
Subject: [Geoserver-users] WFS-T oracle update faild on GS 1.7.2

Hello List,

I have took a update from GS 1.6.3 to 1.7.2 and most things are working
great.
But one thing always will fail, the wfs-t update on oracle backend.

System is: Oracle 10g, new Oracle NG plugin and GeoServer 1.7.2

So if I send a wfs-t update post I will get an error on update.

Stacktrace have wrote this error messages:

2009-03-04 14:46:54,662 ERROR [geoserver.wfs] - Transaction failed
org.geoserver.wfs.WFSTransactionException: update error
    at
org.geoserver.wfs.UpdateElementHandler.execute(UpdateElementHandler.java
:239)
    at org.geoserver.wfs.Transaction.execute(Transaction.java:333)
    at org.geoserver.wfs.Transaction.transaction(Transaction.java:108)
    at
org.geoserver.wfs.DefaultWebFeatureService.transaction(DefaultWebFeature
Service.java:182)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)

and so one......

Caused by: java.lang.RuntimeException: java.sql.SQLException: ORA-29877:
failed in the execution of the ODCIINDEXUPDATE routine
ORA-13365: layer SRID does not match geometry SRID
ORA-06512: at "MDSYS.SDO_INDEX_METHOD_10I", line 243

    at org.geotools.jdbc.JDBCDataStore.update(JDBCDataStore.java:1024)
    at org.geotools.jdbc.JDBCDataStore.update(JDBCDataStore.java:997)
    at
org.geotools.jdbc.JDBCUpdateFeatureWriter.write(JDBCUpdateFeatureWriter.
java:101)
    ... 74 more

Request: transaction
    handle = null
    service = WFS
    version = 1.0.0
    baseUrl = http://localhost:8081/geoserver/
    providedVersion = 1.0.0
    lockId = null
    group =
[wfs:update=net.opengis.wfs.impl.UpdateElementTypeImpl@anonymised.com
(filter: [ GIS_FEATURE.4 ], handle: null, inputFormat: <unset>, srsName:
null, typeName: {http://www.features-on-gis.de/wfs\}GIS_FEATURE)]
    insert =
    update = [net.opengis.wfs.impl.UpdateElementTypeImpl@anonymised.com
(filter: [
GIS_FEATURE.4 ], handle: null, inputFormat: <unset>, srsName: null,
typeName: {http://www.feature-on-gis.de/wfs\}GIS_FEATURE)]
    delete =
    native =
    releaseAction = ALL

Days before I had some problems with the different kind of XML-SRS name
style between GeoServer 1.6.3 and 1.7.2 and I had to change some parse
routines of mine, but I realy don't know if it is the possible reason
for
that error on update.

Has anyone a idea what could be the reason for that problem, so I will
be
very pleased,
2StepForward
--
View this message in context:
http://www.nabble.com/WFS-T-oracle-update-faild-on-GS-1.7.2-tp22331042p2
2331042.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

------------------------------------------------------------------------
------
Open Source Business Conference (OSBC), March 24-25, 2009, San
Francisco, CA
-OSBC tackles the biggest issue in open source: Open Sourcing the
Enterprise
-Strategies to boost innovation and cut costs with open source
participation
-Receive a $600 discount off the registration fee with the source code:
SFAD
http://p.sf.net/sfu/XcvMzF8H
_______________________________________________
Geoserver-users mailing list
Geoserver-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-users

2StepForward · March 4, 2009, 2:56pm

Hello,

thanks Kris for fast replay. I had miss to say that the post include a
srsName and the USER_SDO_GEOM_METADATA also.

Here the post and the data from the METADATA:

<wfs:Transaction xmlns:wfs="http://www.opengis.net/wfs" version="1.0.0"
service="WFS">
  <wfs:Update xmlns:GIS_FEATURE="http://www.feature-on-gis.de/wfs"
typeName="GIS_FEATURE:GIS_FEATURE">
    <wfs:Property>
      <wfs:Name>THE_GEOM</wfs:Name>
      <wfs:Value>
        <gml:Point xmlns:gml="http://www.opengis.net/gml"
srsName="EPSG:23032">
          <gml:coordinates decimal="." cs="," ts="
">501156.53192256053,5543284.591130405
          </gml:coordinates>
        </gml:Point>
      </wfs:Value>
    </wfs:Property>
    <wfs:Property>
      <wfs:Name>FNAME</wfs:Name>
      <wfs:Value>DOSOMETHING</wfs:Value>
    </wfs:Property>
    <ogc:Filter xmlns:ogc="http://www.opengis.net/ogc">
      <ogc:FeatureId fid="GIS_FEATURE.4"/>
    </ogc:Filter>
  </wfs:Update>
</wfs:Transaction>

And the Metadata:

GIS_FEATURES
THE_GEOM
SDO_DIM_ARRAY(SDO_DIM_ELEMENT('X', 259828,702, 5230501,97, ,005),
SDO_DIM_ELEMENT('Y', 920728,245, 6084502,77, ,005))
23032

This problem only will happen on the update transaction, the insert- and
delete transactions are run successful.

Cheers 2StepForward

Kris Geusebroek wrote:

Hi,

I think you don't fill the srsName in the request. If the database
metadata table has an srs filled in fo r the table the feature is
inserted in that might be the mismatch.

Cheers Kris

--
View this message in context: http://www.nabble.com/WFS-T-oracle-update-faild-on-GS-1.7.2-tp22331042p22331661.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

Andrea_Aime3 · March 4, 2009, 3:55pm

2StepForward wrote:

Hello,

thanks Kris for fast replay. I had miss to say that the post include a
srsName and the USER_SDO_GEOM_METADATA also.

Which Oracle datastore are you using? The classic or the ng one?
The classic one has been abandoned, only the ng is going to
receive fixes.
Can you try out with ng and tell us if it works? If not, please port
a fully reproducable test case, that is, a sql script to create data
and the wfs requests to be made in order to reproduce the issue.
This will allow me to use the few time I can allocate to Oracle
to work effectively towards a solution

Cheers
Andrea

2StepForward · March 4, 2009, 4:26pm

Hello Andrea,

thanks for reply. I use the new oracle ng one, but with the old, abandoned
one, it work's.
Nothing on the backend side was changed, so there couldn't be the false.

I have saw yesterday, that in the version from 1.6.x to 1.7.x the srs
definition in the WFS GetCapabilitys had changes from:

old:urn:x-ogc:def:crs:EPSG:6.11.2:23032
to ->
new: urn:x-ogc:def:crs:EPSG:23032

but I don't know it has any relation to the srsName problem between the
layer and the geometry in the oracle backend, probable not.

Momentarily I debug the whole transaction process on the OpenLayers side and
hope to find a
little bit like kind of answer for that what's going wrong.
So if I don't find anything, I will send you tomorrow a test case.

Thanks 2StepForward

Andrea Aime-4 wrote:

2StepForward wrote:

Hello,

thanks Kris for fast replay. I had miss to say that the post include a
srsName and the USER_SDO_GEOM_METADATA also.

Which Oracle datastore are you using? The classic or the ng one?
The classic one has been abandoned, only the ng is going to
receive fixes.
Can you try out with ng and tell us if it works? If not, please port
a fully reproducable test case, that is, a sql script to create data
and the wfs requests to be made in order to reproduce the issue.
This will allow me to use the few time I can allocate to Oracle
to work effectively towards a solution

Cheers
Andrea

------------------------------------------------------------------------------
Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco,
CA
-OSBC tackles the biggest issue in open source: Open Sourcing the
Enterprise
-Strategies to boost innovation and cut costs with open source
participation
-Receive a $600 discount off the registration fee with the source code:
SFAD
http://p.sf.net/sfu/XcvMzF8H
_______________________________________________
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/WFS-T-oracle-update-faild-on-GS-1.7.2-tp22331042p22333476.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

2StepForward · March 5, 2009, 12:33pm

Hi,

now I found out a part of that what's going wrong.

The WFS-T update post was:

<wfs:Transaction xmlns:wfs="http://www.opengis.net/wfs" version="1.0.0"
service="WFS">
<wfs:Update xmlns:GIS_FEATURE="http://www.feature-on-gis.de/wfs"
typeName="GIS_FEATURE:GIS_FEATURE">
<wfs:Property>
<wfs:Name>THE_GEOM</wfs:Name>
<wfs:Value>
<gml:Point xmlns:gml="http://www.opengis.net/gml" srsName="EPSG:23032">
<gml:coordinates decimal="." cs="," ts="
">501156.53192256053,5543284.591130405
</gml:coordinates>
</gml:Point>
</wfs:Value>
</wfs:Property>
<wfs:Property>
<wfs:Name>FNAME</wfs:Name>
<wfs:Value>DOSOMETHING</wfs:Value>
</wfs:Property>
<ogc:Filter xmlns:ogc="http://www.opengis.net/ogc">
<ogc:FeatureId fid="GIS_FEATURE.4"/>
</ogc:Filter>
</wfs:Update>
</wfs:Transaction>

The geometryNode has a valid SRS "EPSG:23032"

The geoserver.log writes out under geotool development logging:

2009-03-05 10:43:09,725 TRACE [org.geotools.core] - condition:
Filter.INCLUDE
2009-03-05 10:43:09,726 DEBUG [org.geotools.jdbc] - CREATE CONNECTION
2009-03-05 10:43:09,731 TRACE [org.geotools.core] - ENTRY 22
2009-03-05 10:43:09,731 DEBUG [org.geotools.jdbc] - SELECT
GID,NAME,PATH,TYPE,THE_GEOM as THE_GEOM FROM GEO.GIS_FEATURE WHERE (GID = ?)
2009-03-05 10:43:09,731 DEBUG [org.geotools.jdbc] - 1 = 4
2009-03-05 10:43:09,736 TRACE [org.geotools.core] - ENTRY 22
2009-03-05 10:43:09,736 DEBUG [org.geotools.jdbc] - SELECT
GID,NAME,PATH,TYPE,THE_GEOM as THE_GEOM FROM GEO.GIS_FEATURE WHERE (GID = ?)
2009-03-05 10:43:09,736 DEBUG [org.geotools.jdbc] - 1 = 4
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting THE_GEOM to
POINT (501116.52692256024 5543542.401130409)
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting 3 to POINT
(501116.52692256024 5543542.401130409)
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting NAME to FSNAME
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting 0 to Ereignis
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting PATH to 1
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting 1 to 1
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting TYPE to FTYPE
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting 2 to FTYPE
2009-03-05 10:43:09,737 TRACE [org.geotools.core] - ENTRY 22
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Updating features with
prepared statement: UPDATE GEO.GIS_FEATURE SET NAME = ?,PATH = ?,TYPE =
?,THE_GEOM = ? WHERE (GID = ?)
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - 1 = FSNAME
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - 2 = 1
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - 3 = FTYPE
2009-03-05 10:43:09,740 DEBUG [org.geotools.data.oracle.sdo] - Using NULL
SRID: <---- There is the problem !!!!!
2009-03-05 10:43:09,740 DEBUG [org.geotools.jdb] - Setting paramtetr 4 as
MDSYS.SDO_GEOMETRY(2001,NULL,MDSYS.SDO_POINT_TYPE(501116.52692256024,5543542.401130409,NULL),NULL,NULL)
2009-03-05 10:43:09,740 DEBUG [org.geotools.jdbc] - 4 = POINT
(501116.52692256024 5543542.401130409)
2009-03-05 10:43:09,740 DEBUG [org.geotools.jdbc] - 1 = 4
2009-03-05 10:43:09,758 ERROR [org.geoserver.wfs] - Transaction failed

I don't know why the posted SRS will fail into the
org.geotools.data.oracle.sdo. It looks like that it will be ignored or set
to NULL, maybe also "0".
Today I have saw this Bug-description in codehouse:
http://jira.codehaus.org/browse/GEOS-781. It looks like a little bit of
what's going on with my transaction, but
I don't use a epsg of "0".

Has anybody a idea why this problem appers?

Thanks, 2StepForward

2StepForward wrote:

Hello Andrea,

thanks for reply. I use the new oracle ng one, but with the old, abandoned
one, it work's.
Nothing on the backend side was changed, so there couldn't be the false.

I have saw yesterday, that in the version from 1.6.x to 1.7.x the srs
definition in the WFS GetCapabilitys had changes from:

old:urn:x-ogc:def:crs:EPSG:6.11.2:23032
to ->
new: urn:x-ogc:def:crs:EPSG:23032

but I don't know it has any relation to the srsName problem between the
layer and the geometry in the oracle backend, probable not.

Momentarily I debug the whole transaction process on the OpenLayers side
and hope to find a
little bit like kind of answer for that what's going wrong.
So if I don't find anything, I will send you tomorrow a test case.

Thanks 2StepForward

--
View this message in context: http://www.nabble.com/WFS-T-oracle-update-faild-on-GS-1.7.2-tp22331042p22350739.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

Andrea_Aime3 · March 5, 2009, 5:11pm

2StepForward wrote:

Hi,

now I found out a part of that what's going wrong.

The WFS-T update post was:

<wfs:Transaction xmlns:wfs="http://www.opengis.net/wfs" version="1.0.0"
service="WFS">
<wfs:Update xmlns:GIS_FEATURE="http://www.feature-on-gis.de/wfs"
typeName="GIS_FEATURE:GIS_FEATURE">
<wfs:Property>
<wfs:Name>THE_GEOM</wfs:Name>
<wfs:Value>
<gml:Point xmlns:gml="http://www.opengis.net/gml" srsName="EPSG:23032">
<gml:coordinates decimal="." cs="," ts="
">501156.53192256053,5543284.591130405
</gml:coordinates>
</gml:Point>
</wfs:Value>
</wfs:Property>
<wfs:Property>
<wfs:Name>FNAME</wfs:Name>
<wfs:Value>DOSOMETHING</wfs:Value>
</wfs:Property>
<ogc:Filter xmlns:ogc="http://www.opengis.net/ogc">
<ogc:FeatureId fid="GIS_FEATURE.4"/>
</ogc:Filter>
</wfs:Update>
</wfs:Transaction>

The geometryNode has a valid SRS "EPSG:23032"

The geoserver.log writes out under geotool development logging:

2009-03-05 10:43:09,725 TRACE [org.geotools.core] - condition:
Filter.INCLUDE
2009-03-05 10:43:09,726 DEBUG [org.geotools.jdbc] - CREATE CONNECTION
2009-03-05 10:43:09,731 TRACE [org.geotools.core] - ENTRY 22
2009-03-05 10:43:09,731 DEBUG [org.geotools.jdbc] - SELECT
GID,NAME,PATH,TYPE,THE_GEOM as THE_GEOM FROM GEO.GIS_FEATURE WHERE (GID = ?)
2009-03-05 10:43:09,731 DEBUG [org.geotools.jdbc] - 1 = 4
2009-03-05 10:43:09,736 TRACE [org.geotools.core] - ENTRY 22
2009-03-05 10:43:09,736 DEBUG [org.geotools.jdbc] - SELECT
GID,NAME,PATH,TYPE,THE_GEOM as THE_GEOM FROM GEO.GIS_FEATURE WHERE (GID = ?)
2009-03-05 10:43:09,736 DEBUG [org.geotools.jdbc] - 1 = 4
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting THE_GEOM to
POINT (501116.52692256024 5543542.401130409)
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting 3 to POINT
(501116.52692256024 5543542.401130409)
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting NAME to FSNAME
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting 0 to Ereignis
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting PATH to 1
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting 1 to 1
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting TYPE to FTYPE
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Setting 2 to FTYPE
2009-03-05 10:43:09,737 TRACE [org.geotools.core] - ENTRY 22
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - Updating features with
prepared statement: UPDATE GEO.GIS_FEATURE SET NAME = ?,PATH = ?,TYPE =
?,THE_GEOM = ? WHERE (GID = ?)
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - 1 = FSNAME
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - 2 = 1
2009-03-05 10:43:09,737 DEBUG [org.geotools.jdbc] - 3 = FTYPE
2009-03-05 10:43:09,740 DEBUG [org.geotools.data.oracle.sdo] - Using NULL
SRID: <---- There is the problem !!!!!
2009-03-05 10:43:09,740 DEBUG [org.geotools.jdb] - Setting paramtetr 4 as
MDSYS.SDO_GEOMETRY(2001,NULL,MDSYS.SDO_POINT_TYPE(501116.52692256024,5543542.401130409,NULL),NULL,NULL)
2009-03-05 10:43:09,740 DEBUG [org.geotools.jdbc] - 4 = POINT
(501116.52692256024 5543542.401130409)
2009-03-05 10:43:09,740 DEBUG [org.geotools.jdbc] - 1 = 4
2009-03-05 10:43:09,758 ERROR [org.geoserver.wfs] - Transaction failed

I received your sql script and test case by private mail and managed to fix the issue. The patch has been committed a few minutes ago, you should find it in tomorrow's nightly build of GeoServer/Oracle plugin
at:
http://gridlock.openplans.org/geoserver/1.7.x/
(make sure whatever you download is marked March 6, or later, at the moment it seems the nightly build script is down, we're looking into
it)

Cheers
Andrea

2StepForward · March 6, 2009, 12:29pm

Hi,

realy thanks Andrea for build up the test case and fix this issue.
I will try it on monday and wish you at least a nice weekend.

Greetings 2StepForward

Andrea Aime-4 wrote:

I received your sql script and test case by private mail and managed to
fix the issue. The patch has been committed a few minutes ago, you
should find it in tomorrow's nightly build of GeoServer/Oracle plugin
at:
http://gridlock.openplans.org/geoserver/1.7.x/
(make sure whatever you download is marked March 6, or later, at the
moment it seems the nightly build script is down, we're looking into
it)

Cheers
Andrea

--
View this message in context: http://www.nabble.com/WFS-T-oracle-update-faild-on-GS-1.7.2-tp22331042p22371570.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

2StepForward · March 9, 2009, 10:52am

2StepForward wrote by private E-Mail:

Hello Andrea,

sorry, but I get the same error:

2009-03-09 11:07:36,585 DEBUG [org.geotools.jdbc] - Updating features with
prepared statement: UPDATE VGEO.GIS_OCCATIONS SET THE_GEOM = ?,NAME = ?,PATH
= ?,TYPE = ? WHERE (GID = ?)
2009-03-09 11:07:36,588 DEBUG [org.geotools.data.oracle.sdo] - Using NULL
SRID:
2009-03-09 11:07:36,588 DEBUG [org.geotools.jdb] - Setting paramtetr 1 as
MDSYS.SDO_GEOMETRY(2001,NULL,MDSYS.SDO_POINT_TYPE(500982.4000825604,5543535.770290409,NULL),NULL,NULL)
2009-03-09 11:07:36,588 DEBUG [org.geotools.jdbc] - 1 = POINT
(500982.4000825604 5543535.770290409)
2009-03-09 11:07:36,588 DEBUG [org.geotools.jdbc] - 2 = FEATURE
2009-03-09 11:07:36,588 DEBUG [org.geotools.jdbc] - 3 = 1
2009-03-09 11:07:36,588 DEBUG [org.geotools.jdbc] - 4 = FTYPE
2009-03-09 11:07:36,588 DEBUG [org.geotools.jdbc] - 1 = 4
2009-03-09 11:07:36,605 ERROR [org.geoserver.wfs] - Transaction failed

I have took the whole nightly version with the new nightly oracleng plugin,
it seems to play the same game as befor.

Here the insert output. There is no problem and the SRID could determine:
2009-03-09 11:13:49,287 DEBUG [org.geotools.jdbc] - Inserting new feature
with ps: INSERT INTO VGEO.GIS_OCCATIONS ( NAME,PATH,TYPE,THE_GEOM,GID )
VALUES ( ?,?,?,?,?)
2009-03-09 11:13:49,287 DEBUG [org.geotools.jdbc] - 1 = TEST
2009-03-09 11:13:49,287 DEBUG [org.geotools.jdbc] - 2 = 1
2009-03-09 11:13:49,288 DEBUG [org.geotools.jdbc] - 3 = asdf
2009-03-09 11:13:49,288 DEBUG [org.geotools.data.oracle.sdo] - Using layer
SRID: 23032

I don't understand this. There is also no problem if I delet some features..

Andrea Aime worte back:

Anyways, I don't really know. I double checked my commits, the patch
went in March 5, along with tests making sure geometries can be updated,
so the GeoServer nightly build from March 9 should contain the
fix.

Justin, is there any chance the nightly builds are not using a up to
date version of the GeoTools jars (in which the fix was made?)

Cheers
Andrea

2StepForward wrote:

Hi,

realy thanks Andrea for build up the test case and fix this issue.
I will try it on monday and wish you at least a nice weekend.

Greetings 2StepForward

Andrea Aime-4 wrote:

I received your sql script and test case by private mail and managed to
fix the issue. The patch has been committed a few minutes ago, you
should find it in tomorrow's nightly build of GeoServer/Oracle plugin
at:
http://gridlock.openplans.org/geoserver/1.7.x/
(make sure whatever you download is marked March 6, or later, at the
moment it seems the nightly build script is down, we're looking into
it)

Cheers
Andrea

2StepForward wrote by private E-Mail:2StepForward wrote by private E-Mail:
--
View this message in context: http://www.nabble.com/WFS-T-oracle-update-faild-on-GS-1.7.2-tp22331042p22410871.html
Sent from the GeoServer - User mailing list archive at Nabble.com.

Andrea_Aime3 · March 9, 2009, 7:06pm

2StepForward ha scritto:

Andrea Aime worte back:

Anyways, I don't really know. I double checked my commits, the patch
went in March 5, along with tests making sure geometries can be updated,
so the GeoServer nightly build from March 9 should contain the
fix.

Ok, managed to reproduce your issue. When I fixed the bug I checked
only by adding a new unit test to the datastore, but the unit test
did not cover the whole set of conditions GeoServer puts the datastore
into.
Today I augmented the fix so that it works in GeoServer as well,
and double checked I can actually run the WFS-T request from
GeoServer and have it work fine.

Sorry for the hassle. You'll find the fix in tomorrow's nightly.
Unfortunately 1.7.3 release packages are already built, so we've
lost this release train but you'll be able to use tomorrow's nightly,
which is really close to the released GeoServer
(only 2-3 patches went into it in the meantime, all bugfixes, no new
features).

Cheers
Andrea

--
Andrea Aime
OpenGeo - http://opengeo.org
Expert service straight from the developers.

2StepForward · March 10, 2009, 8:35am

Hi,

yes the update transaction work's great. If have took the last nightly build
from 10-Mar-2009 and also the oracle new generation plugin from same
nightly.

Realy thank to Andrea for you action to fix this bug and also to Justin who
had control the nightly build.

That's good work,

Greetings Klaus

Andrea Aime-4 wrote:

2StepForward ha scritto:

Andrea Aime worte back:

Anyways, I don't really know. I double checked my commits, the patch
went in March 5, along with tests making sure geometries can be updated,
so the GeoServer nightly build from March 9 should contain the
fix.

Ok, managed to reproduce your issue. When I fixed the bug I checked
only by adding a new unit test to the datastore, but the unit test
did not cover the whole set of conditions GeoServer puts the datastore
into.
Today I augmented the fix so that it works in GeoServer as well,
and double checked I can actually run the WFS-T request from
GeoServer and have it work fine.

Sorry for the hassle. You'll find the fix in tomorrow's nightly.
Unfortunately 1.7.3 release packages are already built, so we've
lost this release train but you'll be able to use tomorrow's nightly,
which is really close to the released GeoServer
(only 2-3 patches went into it in the meantime, all bugfixes, no new
features).

Cheers
Andrea

--
Andrea Aime
OpenGeo - http://opengeo.org
Expert service straight from the developers.

------------------------------------------------------------------------------
Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco,
CA
-OSBC tackles the biggest issue in open source: Open Sourcing the
Enterprise
-Strategies to boost innovation and cut costs with open source
participation
-Receive a $600 discount off the registration fee with the source code:
SFAD
http://p.sf.net/sfu/XcvMzF8H
_______________________________________________
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/WFS-T-oracle-update-faild-on-GS-1.7.2-tp22331042p22429906.html
Sent from the GeoServer - User mailing list archive at Nabble.com.