On Mon, Nov 9, 2009 at 7:38 AM, Uwe Schroeder <uwe@oss4u.com> wrote:
> What I noticed is when I look at pg_locks, pretty much all of the processes being idle in transaction have an
exclusivelock of locktype "virtualidx".
It's "virtualxid" as in "virtual transaction id" and hopefully more
than pretty much all have a lock of this type -- *all* transactions
start with a lock on their own transaction id and hold it until they
finish. That's how other transactions wait for a transaction to
finish, by attempting to get a lock on the transaction id of the
transaction they're waiting on.
--
greg