Re: [Proposal] Fully WAL logged CREATE DATABASE - No Checkpoints

Andrew Dunstan <andrew@dunslane.net>

From: Andrew Dunstan <andrew@dunslane.net>
To: Dilip Kumar <dilipbalaut@gmail.com>, Robert Haas <robertmhaas@gmail.com>
Cc: Heikki Linnakangas <hlinnaka@iki.fi>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-02-11T17:11:40Z
Lists: pgsql-hackers
On 2/10/22 07:32, Dilip Kumar wrote:
> On Wed, Feb 9, 2022 at 9:31 PM Robert Haas <robertmhaas@gmail.com> wrote:
>> On Wed, Feb 9, 2022 at 10:59 AM Dilip Kumar <dilipbalaut@gmail.com> wrote:
>>> On Wed, Feb 9, 2022 at 9:25 PM Andrew Dunstan <andrew@dunslane.net> wrote:
>>>> On 6/16/21 03:52, Dilip Kumar wrote:
>>>>> On Tue, Jun 15, 2021 at 7:01 PM Andrew Dunstan <andrew@dunslane.net> wrote:
>>>>>> Rather than use size, I'd be inclined to say use this if the source
>>>>>> database is marked as a template, and use the copydir approach for
>>>>>> anything that isn't.
>>>>> Yeah, that is possible, on the other thought wouldn't it be good to
>>>>> provide control to the user by providing two different commands, e.g.
>>>>> COPY DATABASE for the existing method (copydir) and CREATE DATABASE
>>>>> for the new method (fully wal logged)?
>>>> This proposal seems to have gotten lost.
>>> Yeah, I am planning to work on this part so that we can support both methods.
>> But can we pick a different syntax? In my view this should be an
>> option to CREATE DATABASE rather than a whole new command.
> Maybe we can provide something like
>
> CREATE DATABASE..WITH WAL_LOG=true/false ? OR
> CREATE DATABASE..WITH WAL_LOG_DATA_PAGE=true/false ? OR
> CREATE DATABASE..WITH CHECKPOINT=true/false ? OR
>
> And then we can explain in documentation about these options?  I think
> default should be new method?
>
>

The last one at least has the advantage that it doesn't invent yet
another keyword.

I can live with the new method being the default. I'm sure it would be
highlighted in the release notes too.


cheers


andrew


--
Andrew Dunstan
EDB: https://www.enterprisedb.com




Commits

  1. When using the WAL-logged CREATE DATABASE strategy, bulk extend.

  2. Avoid using a fake relcache entry to own an SmgrRelation.

  3. Fix data-corruption hazard in WAL-logged CREATE DATABASE.

  4. initdb: When running CREATE DATABASE, use STRATEGY = WAL_COPY.

  5. Simplify a needlessly-complicated regular expression.

  6. In 020_createdb.pl, change order of command-line arguments.

  7. Add new block-by-block strategy for CREATE DATABASE.

  8. Fix replay of create database records on standby

  9. Refactor code for reading and writing relation map files.

  10. Replace RelationOpenSmgr() with RelationGetSmgr().

  11. Refactor the fsync queue for wider use.