On 12/7/2010 2:10 PM, Kenneth Marshall wrote:
On Tue, Dec 07, 2010 at 11:56:51AM -0800, Richard Broersma wrote:
On Tue, Dec 7, 2010 at 11:43 AM, Andy Colson<andy@xxxxxxxxxxxxxxx> wrote:
In PG the first statement you fire off (like an "insert into" for example)
will start a transaction. ?If you dont commit before you disconnect that
transaction will be rolled back. ?Even worse, if your program does not
commit, but keeps the connection to the db open, the transaction will stay
open too.
Huh - is this new? I always thought that every statement was wrapped
in its own transaction unless you explicitly start your own. So you
shouldn't need to commit before closing a connection if you never
opened a transaction to begin with.
--
Regards,
Richard Broersma Jr.
The default of autocommit unless explicitly starting a transaction with
BEGIN is the normal behavior that I have seen as well.
Cheers,
Ken
Crikey! You're right. I need to be more careful with my assumptions.
I maintain that people need to be more careful with pg transactions.
I've seen several posts about "idle in transaction". But its not as bad
as I made out. My confusion comes from the library I use to hit PG,
which fires off a "begin" for me, and if I dont explicitly commit, it
gets rolled back.
sorry, it was confused between framework and PG.
-Andy
--
Sent via pgsql-performance mailing list (pgsql-performance@xxxxxxxxxxxxxx)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance
