I don't know why but now when I created this howto recreate crash I used new cluster (on these same server) and now system crash file is not created.
So now I have only info that postgres process has Segmentation fault.
I have had latest postgis from ubuntu apt update (POSTGIS="2.5.3 r17699" [EXTENSION] PGSQL="100" GEOS="3.7.1-CAPI-1.11.1 27a5e771" PROJ="Rel. 4.9.2, 08 September 2015" GDAL="GDAL 1.11.3, released 2015/09/16" LIBXML="2.9.3" LIBJSON="0.11.99" LIBPROTOBUF="1.2.1" RASTER).
So maybe I should report it to postgis bug tracking system?
On Fri, Feb 28, 2020 at 02:23:21PM +0000, PG Bug reporting form wrote:
>The following bug has been logged on the website:
>
>Bug reference: 16283
>Logged by: Przemysław Szustak
>Email address: przemyslaw.szustak@gmail.com
>PostgreSQL version: 10.12
>Operating system: Ubuntu 16.04.6 LTS
>Description:
>
>Postgresql crash segmentation fault on create index.
>More info https://github.com/przemyslaw-szustak/postgresql_postgis_crash.
>
This seems more like a bug in postgis, considering the last frame before
the segfault looks like this:
#0 0x000055fb9015c45b in pfree ()
#1 0x00007f64ee7f2b5e in ?? () from /usr/lib/postgresql/10/lib/postgis-2.5.so
#2 0x00007f64ee7f4367 in gserialized_gist_picksplit () from /usr/lib/postgresql/10/lib/postgis-2.5.so
#3 0x000055fb9013a272 in FunctionCall2Coll ()
#4 0x000055fb8fd7f526 in gistSplitByKey ()
so it's a call from GiST, but the last two calls are somewhere in
PostGIS library. If I had to guess, I'd say this looks like double-free
or something like that.
Which PostGIS version are you using? Are you sure you have the latest
2.5.x version?
Also, maybe try to install packages with debug symbols, that should give
us a bit more context (parameters, line numbers, ...).
regards
--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
--
Z poważaniem,
Przemysław Szustak