I forgot to mention that the bug can be reproduced in a hot-standby
setup with the attached isolation spec. Note that full_page_writes must
be turned off (otherwise, the updates use full-page writes and then the
bogus code is not run). Once the spec is executed, in the replica run
SET enable_seqscan TO off;
SELECT * FROM f WHERE a=1;
and you get no rows back.
--
Álvaro Herrera http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
setup
{
DROP TABLE IF EXISTS f;
CREATE TABLE f (a int primary key, b int);
INSERT INTO f values (1, 42);
}
session "s1"
step "begin" { BEGIN; }
step "update" { UPDATE f SET b = b + 1; }
step "commit" { COMMIT; }
session "s2"
step "lock" { SELECT * FROM f FOR KEY SHARE; }
permutation "begin" "update" "update" "lock" "commit"
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
