Re: reloption to prevent VACUUM from truncating empty pages at the end of relation

On April 17, 2018 6:00:59 PM PDT, Michael Paquier <michael(at)paquier(dot)xyz> wrote:
>On Tue, Apr 17, 2018 at 12:12:26PM -0700, Andres Freund wrote:
>> On 2018-04-17 15:09:18 -0400, Tom Lane wrote:
>>> Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> writes:
>>>> Andres was working on a radix tree structure to fix this problem,
>but
>>>> that seems to be abandoned now, and it seems a major undertaking.
>>
>> I hope to re-ignite work on that later in the v12 cycle. But
>> realistically that means it's not going to be mergable for v12.
>
>Need a push of man-hours for that?

Not sure what you mean?

>>>> While I agree that the proposed solution is a wart, it seems much
>>>> better than no solution at all. Can we consider Fujii's proposal
>as
>>>> a temporary measure until we fix shared buffers? I'm +1 on it
>>>> myself.
>>>
>>> Once we've introduced a user-visible reloption it's going to be
>>> practically impossible to get rid of it, so I'm -1.
>>
>> It's not much work to maintain though? And even the brief AEL lock
>can
>> cause troubles, leaving the scan aside. So I'm like +0.1 or such.
>
>I would say that if the radix tree patch can make it for the first
>commit fest and has reviews, then there would be likely no need for
>this
>reloption.

There's no way it can.

Andres
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.