[postgis-devel] Fwd: Build failed in Hudson: PostGIS-trunk-regression #1772

[Paul Ramsey]

In the current trunk I'm seeing multiple regression failures (which
I'm "pleased" to say all pre-date my commit that fired off this
mess...)

Why is the builbot returning no regression failures, except this funny
"ERROR" test that isn't even part of the standard regression? Here's
what's failing for me:

tickets.sql:

@@ -137,9 +137,9 @@
 #1320.geog.2|MULTIPOLYGON|4326
 #1320.geom.2|MULTIPOLYGON|4326
 ERROR:  Geometry type (Polygon) does not match column type (MultiPolygon)
-ERROR:  Geometry type (Polygon) does not match column type (MultiPolygon)
+ERROR:  Geometry SRID (2139062143) does not match column SRID (4326)
 #1320.geog.3|MULTIPOLYGON|4326
 #1320.geom.3|MULTIPOLYGON|4326
 </#1320>
-#1344|\001\002\000\000\000\001\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000
+#1344|\x01020000000100000000000000000000000000000000000000
 #1385|

And all the loader tests up to but not including TSTPolygon. The error
file doesn't show any error (perhaps it's been overwritten by
something without errors?)

SET
SET
BEGIN
psql:/var/folders/vQ/vQpcmrOZEXuINbyg+7Jj4k+++TI/-Tmp-/loader.out:5:
NOTICE:  CREATE TABLE will create implicit sequence
"loadedshp_gid_seq" for serial column "loadedshp.gid"
CREATE TABLE
psql:/var/folders/vQ/vQpcmrOZEXuINbyg+7Jj4k+++TI/-Tmp-/loader.out:6:
NOTICE:  ALTER TABLE / ADD PRIMARY KEY will create implicit index
"loadedshp_pkey" for table "loadedshp"
ALTER TABLE
INSERT 0 1
COMMIT
/var/folders/vQ/vQpcmrOZEXuINbyg+7Jj4k+++TI/-Tmp-//loader.err (END)

Fun, much cleaning to do.

P.





On Wed, Dec 21, 2011 at 3:49 PM, Chris Hodgson <chodgson at refractions.net> wrote:
> Hmm, actually, looking at the build artifacts, it seems like the tickets
> test is failing:
>
> http://office.refractions.net:1500/job/PostGIS-trunk-regression/1772/artifact/results_9.0.1_3.2.2_4.7.0/regress-TRUNK_8497_9.0.1_3.2.2_4.7.0.log
>
> Here is the output from the tickets diff:
>
> --- tickets_expected    2011-12-20 10:51:54.000000000 -0800
> +++
> /var/home/buildbot/opt/hudson/jobs/PostGIS-trunk-regression/workspace/results_9.0.1_3.2.2_4.7.0/test_62_out
> 2011-12-21 13:59:24.000000000 -0800
> @@ -141,5 +141,5 @@
>  #1320.geog.3|MULTIPOLYGON|4326
>  #1320.geom.3|MULTIPOLYGON|4326
> </#1320>
> -#1344|\001\002\000\000\000\001\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000
> +#1344|\x01020000000100000000000000000000000000000000000000
>  #1385|
>
> Note this is trunk postgis on postgres 9.0.1 with geos 3.2.2 and proj 4.7
>
> Chris
>
>
> On 11-12-21 02:37 PM, Sandro Santilli wrote:
>>
>> On Wed, Dec 21, 2011 at 02:20:13PM -0800, Chris Hodgson wrote:
>>>
>>> Anyone know what the file "test_62_out" is? Something is outputting
>>> it from the regression tests and then the hudson buildbot script
>>> greps all the output files for "error" and finds it in the file,
>>> thus declaring the build failed, for every build in a little while
>>> now (perhaps the last week?)
>>
>> Expecting an ERROR is not a failure (if the ERROR is obtained).
>> The only real failure is when "make check" returns non-zero.
>>
>> I guess we could give better names to log files, anyway.
>>
>> --strk;
>>
>>   ,------o-.
>>   |   __/  |    Thank you for PostGIS-2.0 Topology !
>>   |  / 2.0 |    http://www.pledgebank.com/postgistopology
>>   `-o------'
>>
>> _______________________________________________
>> postgis-devel mailing list
>> postgis-devel at postgis.refractions.net
>> http://postgis.refractions.net/mailman/listinfo/postgis-devel
>
>
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-devel