Let me know which errors remain after you set up the 3D support.
We are again trying to get 3D Geometry support on the Oracle test database.
Consequently, I have re-enabled the oracle test fixture and the Oracle integration test on ares, so expect a few more build failure messages on the mailing list. FOr oracle, we appear to be getting a couple of legitimate test failures, plus a few related to 3D geometry support that should be fixed once we get this working.
[INFO] ------------------------------------------------------------------------
[INFO] Building Application Schema Integration Online Test with Oracle Database 2.9-SNAPSHOT
[INFO] ------------------------------------------------------------------------
...
Results :
Failed tests: testQueryBboxLongLat(org.geoserver.test.BBoxFilterTest): expected:<[0]> but was:<>
testQueryBboxLongLatURN(org.geoserver.test.BBoxFilterTest): expected:<[0]> but was:<>
testQueryBboxLatLongEPSGCode(org.geoserver.test.BBoxFilterTest): expected:<[0]> but was:<>
Tests in error:
testBbox1(org.geoserver.test.BBox3DTest): java.sql.SQLSyntaxErrorException: ORA-00904: "ARES"."SC4O"."ST_GEOMFROMEWKT": invalid identifier(..)
testBbox2(org.geoserver.test.BBox3DTest): java.sql.SQLSyntaxErrorException: ORA-00904: "ARES"."SC4O"."ST_GEOMFROMEWKT": invalid identifier(..)
testBboxPost(org.geoserver.test.BBox3DTest): java.sql.SQLSyntaxErrorException: ORA-00904: "ARES"."SC4O"."ST_GEOMFROMEWKT": invalid identifier(..)
testReprojection(org.geoserver.test.Reprojection3DTest): java.sql.SQLSyntaxErrorException: ORA-00904: "ARES"."SC4O"."ST_GEOMFROMEWKT": invalid identifier(..)
Tests run: 169, Failures: 3, Errors: 4, Skipped: 0
On Tue, May 12, 2015 at 8:50 AM, Torben Barsballe <tbarsballe@anonymised.com <mailto:tbarsballe@anonymised.com>> wrote:
Hi Neils
Oracle tests are going fine so far, there has been a bit of a
delay since the DBA who manages the Oracle DB we are using has
been out for the last couple weeks.
The only real question I had was about the TESTTBS tablespace -
based on the errors we are getting from the app-schema tests it
looks like this is an undocumented requirement? Was this
tablespace maybe included by default in an earlier version of oracle?
Thanks,
Torben
On Tue, May 12, 2015 at 5:15 AM, Niels Charlier <niels@anonymised.com
<mailto:niels@anonymised.com>> wrote:
Hello Torben,
Sorry for my late response. Picking up after a holiday.
The oracle online tests are not so straight-forward to set up.
Can I be of any assistance or have you sorted it out in the
meantime?
Regards
Niels
On 24-04-15 01:50, Torben Barsballe wrote:
Excellent, good to know we are making progress.
When I go through and update the documentation, I'll make
these points a bit more clear, rather than just showing up as
an example.
SC4O hasn't been added yet, so I think that is why you are
seeing that error.
Torben
On Thu, Apr 23, 2015 at 4:15 PM, Ben Caradoc-Davies
<ben@anonymised.com <mailto:ben@anonymised.com>> wrote:
Aha!
Oh, this sucks horribly, but the good news is that I had
a peek into the Jenkins pre-build script and I can
reproduce the failures locally against my own database
and I think I have fixed them for you:
(1) For both database types, both "password" and "passwd"
are required. I have added "passwd" which was missing.
Yes, different fixtures look for different keys (*thunch*
(sound of forehead hitting keyboard)), so having only one
will cause only some to fail. Similar brain damage can be
found in GeoTools. The documentation, while appearing to
contain a bizarre duplication, is in fact correct!
(2) Postgis tests appear to *require* dbtype=postgisng,
and do not work with dbtype=postgis or this key not
present, which is also curious. Probably a dodgy
assumption somewhere, or need for consistency with
mapping files which contain this alternative name
(connection pooling, perhaps?). I have added
dbtype=postgis to the Jenkins pre-build script. This
second bizarre detail in the documentation is also correct!
I have applied these two changes to Jenkins and have
kicked the build. Fingers crossed.
Kind regards,
Ben.
On 24/04/15 10:35, Torben Barsballe wrote:
I took a look the actuall test script
(Oracle_Data_ref_set.sql) extracted
from the test data jar, and it references the TESTTBS
tablespace
frequently, without ever creating it, so it seems
like an undocumented
requirement. I'll try and get that tablespace added
to the Oracle DB.
I've added -DquietTests=false to the mvn command.
Based on the test
failures it look like data is missing somewhere, but
I can't see any log
messages suggesting why this might be occuring.
- Are you able to run the tests in Eclipse
against the same database? This
might give you console output that surefire does
not capture.
No, the test databases are firewalled to be
accessible from ares only. I
have very limited access to them.
Torben
-- Ben Caradoc-Davies <ben@anonymised.com
<mailto:ben@anonymised.com>>
Director
Transient Software Limited <http://transient.nz/>
New Zealand
------------------------------------------------------------------------------
One dashboard for servers and applications across Physical-Virtual-Cloud
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
<mailto:Geoserver-devel@lists.sourceforge.net>
https://lists.sourceforge.net/lists/listinfo/geoserver-devel
