Horton

Re: pg_stat_have_stats() returns true for dropped indexes (or for index creation transaction rolled back)

Bertrand Drouvot <bertranddrouvot.pg@gmail.com>

View thread View on postgr.es
From: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>
To: Andres Freund <andres@anarazel.de>
Cc: Kyotaro Horiguchi <horikyota.ntt@gmail.com>, pgsql-hackers@lists.postgresql.org
Date: 2022-09-26T13:28:16Z
Lists: pgsql-hackers 
Hi,

On 9/26/22 10:23 AM, Drouvot, Bertrand wrote:
> Hi,
> 
> On 9/23/22 10:45 PM, Andres Freund wrote:
> 
>>
>>
>>
>> Hi,
>>
>> On 2022-09-01 08:40:54 +0200, Drouvot, Bertrand wrote:
>>> Thanks for the suggestion, I'm coming up with this proposal in v4 
>>> attached:
>>
>> I pushed the bugfix / related test portion to 15, master. Thanks!
> 
> Thanks!
> 

Forgot to say that with that being fixed, I'll come back with a patch 
proposal for the tables/indexes stats split (discovered the issue fixed 
in this current thread while working on the split patch.)

Regards,

-- 
Bertrand Drouvot
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com

Commits

  1. pgstat: Fix transactional stats dropping for indexes