Horton

Re: [HACKERS] VACUUM and ANALYZE disagreeing on what reltuples means

Tomas Vondra <tomas.vondra@2ndquadrant.com>

View thread View on postgr.es
From: Tomas Vondra <tomas.vondra@2ndquadrant.com>
To: David Steele <david@pgmasters.net>, Tom Lane <tgl@sss.pgh.pa.us>
Cc: Stephen Frost <sfrost@snowman.net>, Michael Paquier <michael.paquier@gmail.com>, Haribabu Kommi <kommi.haribabu@gmail.com>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Date: 2018-03-05T15:22:55Z
Lists: pgsql-hackers 

On 03/05/2018 04:12 PM, David Steele wrote:
> Hi Tomas,
> 
> On 1/8/18 3:28 PM, Tomas Vondra wrote:
>>
>>
>> On 01/08/2018 08:39 PM, Tom Lane wrote:
>>> Tomas Vondra <tomas.vondra@2ndquadrant.com> writes:
>>>> As I already mentioned, Tom's updated patch is better than what I
>>>> posted initially, and I agree with his approach to the remaining
>>>> issues he pointed out. But I somehow assumed that he's already
>>>> looking into that. Tom, do you plan to look into this patch soon,
>>>> or should I?
>>>
>>> No, I thought you were going to run with those ideas. I have a lot
>>> of other stuff on my plate ...
>>>
>>
>> OK, will do.
> 
> What the status of this patch?  It's been waiting on author since last
> November, though I can see there was some confusion over who was working
> on it.
> 
> Given that it's a bug fix it would be good to see a patch for this CF,
> or very soon after.
> 

Yeah, it's the next thing on my TODO.


regards

-- 
Tomas Vondra                  http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Commits

  1. Sync up our various ways of estimating pg_class.reltuples.

  2. Fix tuple counting in SP-GiST index build.

  3. Fix errors in contrib/bloom index build.

  4. When updating reltuples after ANALYZE, just extrapolate from our sample.

  5. Avoid holding AutovacuumScheduleLock while rechecking table statistics.