On Wed, Oct 15, 2014 at 1:06 AM, Andres Freund <andres@2ndquadrant.com> wrote: > I personally think we should just disregard the race here and add a > snapshot parameter. The race is already there and not exactly > small. Let's not kid ourselves that hiding it solves anything. I, too, favor that plan.
When the snapshot specified by user is not a valid one, here is the error returned by pg_dump: $ pg_dump --snapshot 'ppo' pg_dump: [archiver (db)] query failed: ERROR: invalid snapshot identifier: "ppo" pg_dump: [archiver (db)] query was: SET TRANSACTION SNAPSHOT 'ppo'
I thinks that's fine, and it makes the code lighter to rely on the existing error machinery.