[postgis-tickets] [PostGIS] #5106: ST_RemEdgeNewFace causes postgre to core dump, but not ST_RemEdgeModFace

[PostGIS]

#5106: ST_RemEdgeNewFace causes postgre to core dump, but not ST_RemEdgeModFace
-----------------------+---------------------------
  Reporter:  laopsahl  |      Owner:  strk
      Type:  defect    |     Status:  new
  Priority:  medium    |  Milestone:  PostGIS 3.4.0
 Component:  topology  |    Version:  3.2.x
Resolution:            |   Keywords:
-----------------------+---------------------------
Comment (by strk):

 Stack trace obtained, from https://gitlab.com/nibioopensource/resolve-
 overlap-and-gap/-/issues/23#note_1257528871:
 {{{
 (gdb) bt
 #0  0x0000557650ce5a87 in GetMemoryChunkContext (pointer=0x0)
     at ./build/../src/include/utils/memutils.h:127
 #1  pfree (pointer=0x0) at ./build/../src/backend/utils/mmgr/mcxt.c:1033
 #2  0x00007f200cb89fb4 in _lwt_RemEdge (topo=topo at entry=0x557651b26210,
 edge_id=<optimized out>,
     edge_id at entry=97135, modFace=modFace at entry=0) at lwgeom_topo.c:4052
 #3  0x00007f200cb8e37b in lwt_RemEdgeNewFace
 (topo=topo at entry=0x557651b26210,
     edge_id=edge_id at entry=97135) at lwgeom_topo.c:4221
 #4  0x00007f200cb4a5b0 in ST_RemEdgeNewFace (fcinfo=0x557651b28880) at
 postgis_topology.c:4633
 #5  0x0000557650a40de7 in ExecInterpExpr (state=0x557651b287a8,
 econtext=0x557651b284d0,
 }}}

 Happens with no side face of an edge is found in the face table.
 In normal conditions is should NOT happen because there's a foreign key on
 the edge table referencing the face table, but as those foreign keys are
 deferred it can still happen in case of concurrent transactions
-- 
Ticket URL: <https://trac.osgeo.org/postgis/ticket/5106#comment:3>
PostGIS <http://trac.osgeo.org/postgis/>
The PostGIS Trac is used for bug, enhancement & task tracking, a user and developer wiki, and a view into the subversion code repository of PostGIS project.