[postgis-devel] regress test core dumps on Solaris 11

Regina Obe lr at pcorp.us
Mon Aug 7 11:25:57 PDT 2017


Eric,

Are you running the same version of GDAL on all servers or this is just one server we are talking about that you have 9.4,9.5, and 9.6 all installed on?

Also looks like you are testing 2.3.3 and not 2.3.4dev.  Was that a typo that you meant 2.3.3 or you really were trying to install 2.3.4dev.

In 2.3.4 (not yet released), we did put in a patch for Solaris, but looking at your trace, doesn't seem to be the same issue.

https://trac.osgeo.org/postgis/ticket/3795

Hope that helps,
Regina





-----Original Message-----
From: postgis-devel [mailto:postgis-devel-bounces at lists.osgeo.org] On Behalf Of Dechaux Eric CEN (BCQ STIG)
Sent: Monday, August 07, 2017 6:13 AM
To: postgis-devel at lists.osgeo.org
Subject: [postgis-devel] regress test core dumps on Solaris 11

Hello,

I am building postgis 2.3.4 on Solaris 11.3 SPARC and x86_64. I have no issue with postgresql 9.4.12 an 9.6.3 however on 9.5.7 the postgresql server coredumps during the make check regress step :


PATH is
/appli/stig/11.3.2017-08-07/pgsql/9.5.7/bin:/appli/stig/11.3.2017-08-07/pgsql/9.5.7/bin:/appli/stig/11.3.2017-08-07/bin:/usr/gnu/bin:/usr/gcc/5/bin:/usr/bin
Checking for shp2pgsql ... found
Checking for pgsql2shp ... found
Checking for raster2pgsql ... found
TMPDIR is /tmp/pgis_reg
Creating database 'postgis_reg'
Loading PostGIS into 'postgis_reg'
Loading Raster into 'postgis_reg'
PostgreSQL 9.5.7 on sparc-sun-solaris2.11, compiled by gcc (GCC) 5.4.0, 64-bit
   Postgis 2.3.3 - r15473 - 2017-08-07 07:20:29
   scripts 2.3.3 r15473
   raster scripts 2.3.3 r15473
   GEOS: 3.6.2-CAPI-1.10.2 4d2925d6
   PROJ: Rel. 4.9.3, 15 August 2016
   GDAL: GDAL 2.2.1, released 2017/06/23

Running tests

  check_gdal .. ok
  load_outdb ... ok
  check_raster_columns .. ok
  check_raster_overviews .. ok
  rt_io .. ok
  rt_bytea .. ok
  box3d .. ok
  rt_addband .. failed (diff expected obtained: /tmp/pgis_reg/test_8_diff)
  rt_band .. failed (diff expected obtained: /tmp/pgis_reg/test_9_diff)
  rt_tile .. failed (diff expected obtained: /tmp/pgis_reg/test_10_diff)
  rt_dimensions .. failed (diff expected obtained: 
/tmp/pgis_reg/test_11_diff)
  rt_scale .. failed (diff expected obtained: /tmp/pgis_reg/test_12_diff)
  rt_pixelsize .. failed (diff expected obtained: 
/tmp/pgis_reg/test_13_diff)
  rt_upperleft .. failed (diff expected obtained: 
/tmp/pgis_reg/test_14_diff)
  rt_rotation .. failed (diff expected obtained: /tmp/pgis_reg/test_15_diff)
  rt_georeference .. failed (diff expected obtained: 
/tmp/pgis_reg/test_16_diff)
  rt_set_properties .. failed (diff expected obtained: 
/tmp/pgis_reg/test_17_diff)
  rt_isempty .. failed (diff expected obtained: /tmp/pgis_reg/test_18_diff)
  rt_hasnoband .. failed (diff expected obtained: 
/tmp/pgis_reg/test_19_diff)
  rt_metadata .. failed (diff expected obtained: /tmp/pgis_reg/test_20_diff)
  rt_rastertoworldcoord .. failed (diff expected obtained: 
/tmp/pgis_reg/test_21_diff)

Only the rt_addband step coredumps, other tests are failing because the progress instance is restarting.


The stack trace is :

$ pstack core
core 'core' of 29251: 
/appli/stig/11.3.2017-08-07/pgsql/9.5.7/bin/postgres -D /tmp/HayW8S.sh
  ffffffff7da852d4 memcpy%sun4v-hwcap4 (0, ffffffff7fff9fe0, 
ffffffff7fff9fd0, 10098cf98, 0, 10098cf94) + 6c0
  00000001001f31f0 heap_form_tuple (10098cf68, ffffffff7fff9fe0, 
ffffffff7fff9fd0, 2, 18, 10098cf68) + 140
  ffffffff72959a24 RASTER_dumpAsPolygons (10087f258, ffffffff7fff9751, 
0, 10, 0, 10) + 85c
  000000010036abc0 ExecMakeFunctionResult (10087f1e8, 10087f088, 
100901d38, 100901e50, 1, 10087f258) + 1f4
  0000000100367bf8 ExecEvalFieldSelect (10087f198, 10087f088, 100901d38, 
100901e50, 0, 2) + 1c
  000000010036d61c ExecProject (100901e00, ffffffff7fffa46c, 10087f088, 
100901d20, 100901d38, 0) + 14c
  000000010036dad8 ExecScan (1008801e0, 100384214, 10038420c, 0, 
10087ef78, 10087f088) + 1f8
  0000000100365aac ExecProcNode (10087ef78, 2, 10087eae8, 100901c40, 
100860710, 0) + 214
  000000010036228c standard_ExecutorRun (10087ea58, 1, 0, 100400, 0, 
10087ee68) + a4
  0000000100489f74 PortalRunSelect (100860d78, 1, 0, 1008efdb8, 
10087ea58, 100860d78) + 1d0
  000000010048b5a8 PortalRun (100860d78, 7fffffffffffffff, 1, 1007b3ad8, 
1007b39c8, 0) + 34c
  0000000100487c88 PostgresMain (1007763e8, 1008efdb8, 2, 1007b7170, 
100860d78, 1007b7060) + 1aac
  00000001005de0b8 ServerLoop (100758460, 10070e000, 100000, 100186ed8, 
100000, 1007783c0) + 13b8
  0000000100421648 PostmasterMain (3, 100755fb0, 100000, 100000, 
10073f000, 100400) + 1504
  00000001005df584 main (3, ffffffff7ffff758, 64, 100755fb0, 100400, 
100750570) + 7f8
  00000001001ebe74 _start (0, 0, 0, 0, 0, 0) + 7c


Any idea why this test would crash only with this postgresql version ?


Thanks
-- 
Eric DECHAUX
BCQ / STIG
01 58 66 57 57




More information about the postgis-devel mailing list