On Thu, 2011-02-24 at 18:13 -0800, Daniel Farina wrote:
> I have also reproduced this. Notably, things seem fine as long as
> pgbench is confined to one backend, but as soon as two are used (-c 2)
> by the feature I can get segfaults.
Sorry that you all experienced this. I wasn't able to get concurrent
queue accesses even with -c 8, so I spent about half a day last week
investigating a possible spinlock locking flaw. That meant the code in
that area was untested, which is most obvious now. I guess that means I
should test on different hardware in future.
-- Simon Riggs http://www.2ndQuadrant.com/books/PostgreSQL Development, 24x7 Support, Training and Services