Re: REINDEX CONCURRENTLY 2.0
| От | Michael Paquier |
|---|---|
| Тема | Re: REINDEX CONCURRENTLY 2.0 |
| Дата | |
| Msg-id | CAB7nPqSyZhdc-gvwm3KyueoceFJSX5ypQAx_rsJ2DWr5nHtixA@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: REINDEX CONCURRENTLY 2.0 (Michael Paquier <michael.paquier@gmail.com>) |
| Ответы |
Re: REINDEX CONCURRENTLY 2.0
Re: REINDEX CONCURRENTLY 2.0 |
| Список | pgsql-hackers |
On Wed, Aug 27, 2014 at 3:53 PM, Michael Paquier <michael.paquier@gmail.com> wrote:
--
Michael
On Wed, Aug 27, 2014 at 3:41 PM, Andres Freund <andres@2ndquadrant.com> wrote:
> Can you add it to the next CF? I'll try to look earlier, but can't
> promise anything.
>
> I very much would like this to get committed in some form or another.
Added it here to keep track of it:
https://commitfest.postgresql.org/action/patch_view?id=1563
Attached is a fairly-refreshed patch that should be used as a base for the next commit fest. The following changes should be noticed:
- Use of AccessExclusiveLock when swapping relfilenodes of an index and its concurrent entry instead of ShareUpdateExclusiveLock for safety. At the limit of my understanding, that's the consensus reached until now.
- Cleanup of many comments and refresh of the documentation that was somewhat wrongly-formulated or shaped at some places
- Addition of support for autocommit off in psql for REINDEX [ TABLE | INDEX ] CONCURRENTLY
- Some more code cleanup..
I haven't been through the tab completion support for psql but looking at tab-completion.c this seems a bit tricky with the stuff related to CREATE INDEX CONCURRENTLY already present. Nothing huge though.
Regards,
Michael
Вложения
В списке pgsql-hackers по дате отправления: