|From:||Michael Paquier <michael(at)paquier(dot)xyz>|
|To:||Simon Riggs <simon(at)2ndquadrant(dot)com>|
|Cc:||Alvaro Herrera <alvherre(at)2ndquadrant(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>|
|Subject:||Re: pgsql: Avoid duplicate XIDs at recovery when building initial snapshot|
|Views:||Raw Message | Whole Thread | Download mbox | Resend email|
On Sat, Dec 01, 2018 at 10:51:10AM +0000, Simon Riggs wrote:
> On Fri, 30 Nov 2018 at 23:08, Michael Paquier <michael(at)paquier(dot)xyz> wrote:
>> It is not complicated to
>> handle duplicates for xacts and subxacts however holding ProcArrayLock
>> for a longer time stresses me as it is already a bottleneck.
> I hadn't realised this patch holds ProcArrayLock while removing duplicates.
> Now I know I vote against applying this patch unless someone can show that
> the performance effects of doing so are negligable, which I doubt.
Me too after more thoughts on that. Please note that I have marked the
patch as returned with feedback for now.
|Next Message||Tom Lane||2018-12-03 16:40:55||pgsql: Refactor documentation about privileges to centralize the info.|
|Previous Message||Michael Paquier||2018-12-03 05:27:40||pgsql: Add some missing schema qualifications|
|Next Message||rajan||2018-12-03 07:00:19||Re: vacuum and autovacuum - is it good to configure the threshold at TABLE LEVEL?|
|Previous Message||Tatsuo Ishii||2018-12-03 06:39:25||Re: error message when subscription target is a partitioned table|