Hi,

I have a query that uses a PL/R function to run a statistical model on data in a postgresql table. The query runs the function 4 times, each of which generates about 2 million lines of results, generating a final table that has about 8 million lines. Each time the function is called, it takes about ~3-5 minutes to finish its work. I know that the function is running fine as it gives me progress and I have had no trouble using it on slightly smaller datasets. However, this time it does the 4 runs, but ends with psql saying the connection to the server has been lost. In the postgresql log file, I have:

2013-05-16 14:05:47 CEST LOG:  server process (PID 27488) was terminated by 
signal 9: Killed

2013-05-16 14:05:47 CEST LOG:  terminating any other active server processes

2013-05-16 14:05:47 CEST WARNING:  terminating connection because of crash of 
another server process

2013-05-16 14:05:47 CEST DETAIL:  The postmaster has commanded this server 
process to roll back the current transaction and exit, because another server 
process exited abnormally and possibly corrupted shared memory


The process ID it mentions is the postgresql query process:

postgres 27488 87.1 26.1 2272848 2101804 ?     Rs   13:44   1:01 postgres: 
dkaplan fads [local] SELECT


Does anyone know what might be causing this problem and how to fix it? Is something timing out? Or is it a memory problem? How would I go about getting more diagnostic information?

Thanks for the help,
David


--
**********************************
David M. Kaplan
Charge de Recherche 1

Institut de Recherche pour le Developpement
Centre de Recherche Halieutique Mediterraneenne et Tropicale
av. Jean Monnet
B.P. 171
34203 Sete cedex
France

Phone: +33 (0)4 99 57 32 27
Fax: +33 (0)4 99 57 32 95

http://www.umr-eme.org/team/dkaplan/
http://www.amped.ird.fr/
**********************************



--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general

Reply via email to