From: | Alvaro Herrera <alvherre(at)dcc(dot)uchile(dot)cl> |
---|---|
To: | Greg Sabino Mullane <greg(at)turnstep(dot)com> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Weird prepared stmt behavior |
Date: | 2004-05-04 02:02:50 |
Message-ID: | 20040504020250.GA602@dcc.uchile.cl |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, May 04, 2004 at 01:22:53AM -0000, Greg Sabino Mullane wrote:
> What about rolling prepares back if they are in a transaction, though?
> They still have the ability to affect a transaction, despite being
> partially outside of it:
> [example ripped]
IMHO this is an oversight, not a design feature. That's why I pointed
it out and planned to correct it.
I think with the Prepare message we should _not_ abort the transaction
if it fails -- if it's outside transaction control, it shouldn't affect
the current transaction (So we would open a subtransaction, process the
message, and rollback if it fails.)
--
Alvaro Herrera (<alvherre[a]dcc.uchile.cl>)
"La fuerza no está en los medios físicos
sino que reside en una voluntad indomable" (Gandhi)
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2004-05-04 04:03:16 | Re: Weird prepared stmt behavior |
Previous Message | Greg Sabino Mullane | 2004-05-04 01:22:53 | Re: Weird prepared stmt behavior |