PostgreSQL server terminated by signal 11

Hi,

My PostgreSQL server running on a Linux machine is terminated by signal
11 whenever I try to create some indexes on a table, which contains
quite a lot of data. However I succeeded in creating some other indexes
without having the PostgreSQL server terminated:

agora=> CREATE INDEX IDX_GSLOG_EVENTTIME

agora-> ON GSLOG_EVENT (EVENT_DATE_CREATED);

CREATE INDEX

Time: 152908.797 ms

agora=> explain analyze select max(event_date_created) from gslog_event;

QUERY PLAN

------------------------------------------------------------------------
------------------------------------------------------------------------
----------------------

Result (cost=3.80..3.81 rows=1 width=0) (actual time=0.218..0.221
rows=1 loops=1)

InitPlan

-> Limit (cost=0.00..3.80 rows=1 width=8) (actual
time=0.197..0.200 rows=1 loops=1)

-> Index Scan Backward using idx_gslog_eventtime on
gslog_event (cost=0.00..39338251.59 rows=10348246 width=8) (actual
time=0.188..0.188 rows=1 loops=1)

Filter: (event_date_created IS NOT NULL)

Total runtime: 0.324 ms

(6 rows)

Time: 41.085 ms

agora=> CREATE INDEX IDX_GSLOG_EVENT_SPREAD_PROTOCOL_NAME

agora-> ON GSLOG_EVENT (EVENT_DATE_CREATED)

agora-> WHERE EVENT_NAME::text <> 'player-login'::text

agora-> AND PLAYER_USERNAME IS NOT NULL

agora-> AND GAME_CLIENT_VERSION IS NULL;

server closed the connection unexpectedly

This probably means the server terminated abnormally

before or while processing the request.

The connection to the server was lost. Attempting reset: Failed.

The PostgreSQL log file doesn't give more information about what went
wrong, except that the server process has been terminated:

LOG: server process (PID 22270) was terminated by signal 11

LOG: terminating any other active server processes

LOG: all server processes terminated; reinitializing

FATAL: the database system is starting up

LOG: database system was interrupted at 2006-07-27 15:29:27 GMT

LOG: checkpoint record is at 249/179D44A8

LOG: redo record is at 249/179D44A8; undo record is at 0/0; shutdown
FALSE

LOG: next transaction ID: 543712876; next OID: 344858

LOG: next MultiXactId: 2; next MultiXactOffset: 3

LOG: database system was not properly shut down; automatic recovery in
progress

LOG: redo starts at 249/179D44EC

LOG: record with zero length at 249/179E4888

LOG: redo done at 249/179E2DFC

LOG: database system is ready

LOG: transaction ID wrap limit is 2147484146, limited by database
"postgres"

I checked the memory installed on the machine, running memtest86 during
more than one day; no error found. I checked bad blocks on every hard
drive installed in this machine, using e2fsck -c /dev/hdxx; no bad block
found.

I've already dropped the table, inserted data, and tried to create all
the indexes. The server systematically crashed when creating some
specific indexes. The only idea I have for the moment would be to setup
another machine with the same database environment. Other idea(s)?

Thanks

--

Daniel CAUNE

Ubisoft Online Technology

(514) 490 2040 ext. 3613