From: | Stas Kelvich <s(dot)kelvich(at)postgrespro(dot)ru> |
---|---|
To: | Stas Kelvich <s(dot)kelvich(at)postgrespro(dot)ru> |
Cc: | PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Read-only access to temp tables for 2PC transactions |
Date: | 2019-05-22 15:41:39 |
Message-ID: | 9DCFB313-8DAD-41E6-944D-8F701A11AA5B@postgrespro.ru |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
> On 14 May 2019, at 12:53, Stas Kelvich <s(dot)kelvich(at)postgrespro(dot)ru> wrote:
>
> Hi,
>
> That is an attempt number N+1 to relax checks for a temporary table access
> in a transaction that is going to be prepared.
>
Konstantin Knizhnik made off-list review of this patch and spotted
few problems.
* Incorrect reasoning that ON COMMIT DELETE truncate mechanism
should be changed in order to allow preparing transactions with
read-only access to temp relations. It actually can be be leaved
as is. Things done in previous patch for ON COMMIT DELETE may be
a performance win, but not directly related to this topic so I've
deleted that part.
* Copy-paste error with check conditions in
relation_open/relation_try_open.
Fixed version attached.
--
Stas Kelvich
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
Attachment | Content-Type | Size |
---|---|---|
2PC-ro-temprels-v2.patch | application/octet-stream | 10.6 KB |
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2019-05-22 15:52:46 | Re: "long" type is not appropriate for counting tuples |
Previous Message | Oleksii Kliukin | 2019-05-22 15:27:15 | upgrades in row-level locks can deadlock |