Tom Lane wrote:
> This is presuming that we abandon the notion that system catalog
> access use SnapshotNow. Which opens the question of what they should
> use instead ... to which "transaction snapshot" isn't the answer,
> because we have to be able to do system catalog accesses before
> we've set the snapshot.
I wonder if it would be possible to use SnapshotNow before the
transaction's snapshot has been established, and the transaction's
snapshot subsequently. Although it definitely makes me nervous to use
multiple snapshots over the life of a single transaction...
In response to
pgsql-hackers by date
|Next:||From: Christopher Kings-Lynne||Date: 2005-03-23 06:13:31|
|Subject: Re: "they only drink coffee at dec"|
|Previous:||From: Neil Conway||Date: 2005-03-23 05:16:36|
|Subject: Re: locks in CREATE TRIGGER, ADD FK|