On Mon, Aug 03, 2020 at 09:46:23AM -0400, Robert Haas wrote: > On Mon, Aug 3, 2020 at 2:30 AM Noah Misch <n...@leadboat.com> wrote: > > Between (b)(2)(X) and (b)(3)(X), what are folks' preferences? Does anyone > > strongly favor some other option (including the option of changing nothing) > > over both of those two? > > I don't think we have any options here that are secure but do not > break backward compatibility.
I agree, but compatibility breaks vary in pain caused. I want to offer a simple exit to a backward-compatible configuration, and I want a $NEW_DEFAULT pleasing enough that a decent fraction of deployments keep $NEW_DEFAULT (forgo the exit). The move to default standard_conforming_strings=on is an example to follow (editing postgresql.conf was the simple exit). > I don't know how to choose between (1), (2), and (3). One way is to envision deployments you know and think about a couple of questions in the context of those deployments. If $EACH_OPTION happened, would this deployment keep $NEW_DEFAULT, override $NEW_DEFAULT to some other secure configuration, or exit to $v13_DEFAULT? Where the answer is "exit", would those deployments rate the exit recipe easy, medium, or difficult?
