Hello,
we use postgis topology for storing and updating quite complicated
polygons representing roads.
Sometimes inserting new polygon into topology causes exceptions (whis is
not nice but acceptable, I can see there is plan to add some more
usefull message https://trac.osgeo.org/postgis/ticket/5889) but
sometimes it completely shut down server which is quite bad as there are
other live databases on server.
The error is cause by our function jt_to_topo_geom() which is just
wrapper about toTopoGeom(). The inserted MultiPolygon geometry is quite
large but postgis valid with reasonable vertices. I have suspicion there
is conflict with data already inserted into topology, but have no idea
where due to not info from log.
Postgis:
POSTGIS="3.4.2 c19ce56" [EXTENSION] PGSQL="140" GEOS="3.9.0-CAPI-1.16.2"
PROJ="7.2.1 NETWORK_ENABLED=OFF URL_ENDPOINT=https://cdn.proj.org
USER_WRITABLE_DIRECTORY=/tmp/proj DATABASE_PATH=/usr/share/proj/proj.db"
LIBXML="2.9.10" LIBJSON="0.15" LIBPROTOBUF="1.3.3" WAGYU="0.5.0
(Internal)" TOPOLOGY
The code of function:
CREATE OR REPLACE FUNCTION unis_emh.jt_to_topo_geom(
dst_schema character varying,
dst_table character varying,
dst_column character varying,
geom geometry)
RETURNS topogeometry
LANGUAGE 'plpgsql'
COST 100
VOLATILE PARALLEL UNSAFE
AS $BODY$
DECLARE
layer_ layer;
topology_ topology;
BEGIN
-- get names
SELECT * FROM FindLayer(dst_schema, dst_table, dst_column) INTO layer_;
SELECT * FROM FindTopology((layer_).topology_id) INTO topology_;
-- insert geometry, no duplicate protection
RETURN toTopoGeom(geom, (topology_).name, (layer_).layer_id);
END;
$BODY$;
Database log:
2025-06-17 17:41:46.509 CET [2414906] LOG: server process (PID 2415005)
was terminated by signal 11: Segmentation fault
2025-06-17 17:41:46.509 CEST [2414906] DETAIL: Failed process was
running: INSERT INTO "j_pmh_topo_geometrie" ("id_geometrie", "topo_geom")
SELECT "id_geometrie",
"unis_emh"."jt_to_topo_geom"(current_schema::text,
'j_pmh_topo_geometrie', 'topo_geom', "geom")
FROM "j_pmh_geometrie"
WHERE "id_geometrie"=63855
2025-06-17 17:41:46.509 CEST [2414906] LOG: terminating any other
active server processes
2025-06-17 17:41:46.513 CEST [2414906] LOG: all server processes
terminated; reinitializing
2025-06-17 17:41:46.625 CEST [2415006] LOG: database system was
interrupted; last known up at 2025-06-17 17:40:42 CEST
2025-06-17 17:41:46.896 CEST [2415006] LOG: database system was not
properly shut down; automatic recovery in progress
2025-06-17 17:41:46.917 CEST [2415006] LOG: redo starts at 14F/64BCDA20
2025-06-17 17:41:46.927 CEST [2415006] LOG: unexpected pageaddr
14F/20000000 in log segment 000000010000014F00000065, offset 0
2025-06-17 17:41:46.927 CEST [2415006] LOG: redo done at 14F/64FFFFC0
system usage: CPU: user: 0.00 s, system: 0.00 s, elapsed: 0.01 s
2025-06-17 17:41:47.021 CEST [2414906] LOG: database system is ready to
accept connections
MS.
