Note the number of exclusive locks in my first message, it is equal to the number of threads (20)
. Also, the ALTER was not running then, apologies if that was not clear.
On Wednesday, January 29, 2014, Thomas Kellerer <
spam_eater@gmx.net> wrote:
Neil Harkins wrote on 29.01.2014 23:37:
I totally understand DDL taking exclusive locks, the problem here seems to be that the *SELECTs*
are taking out exclusive locks, locking out the ALTER, which feels like a bug.
The SELECT is not holding an exclusive lock, it's holing a *shared* lock, but the ALTER is _requesting_ an exclusive lock and that can only be granted until all shared (or otherwise incompatible) locks are released.
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general