Re: [PATCH] Exponential backoff for auth_delay

Hi,

On Mon, Mar 04, 2024 at 03:50:07PM -0500, Robert Haas wrote:
> I agree that two GUCs here seems to be one more than necessary, but I
> wonder whether we couldn't just say 0 means no exponential backoff and
> any other value is the maximum time.

Alright, I have changed it so that auth_delay.milliseconds and
auth_delay.max_milliseconds are the only GUCs, their default being 0. If
the latter is 0, the former's value is always taken. If the latter is
non-zero and larger than the former, exponential backoff is applied with
the latter's value as maximum delay.

If the latter is smaller than the former then auth_delay just sets the
delay to the latter, I don't think this is problem or confusing, or
should this be considered a misconfiguration?

> The idea that 0 means unlimited doesn't seem useful in practice.

Yeah, that was more how it was coded than a real policy decision, so
let's do away with it.

V5 attached.

Michael