Re: END_OF_RECOVERY shutdowns and ResetUnloggedRelations()

Поиск
Список
Период
Сортировка
От Abhijit Menon-Sen
Тема Re: END_OF_RECOVERY shutdowns and ResetUnloggedRelations()
Дата
Msg-id 20141027103930.GA1629@toroid.org
обсуждение исходный текст
Ответ на Re: END_OF_RECOVERY shutdowns and ResetUnloggedRelations()  (Andres Freund <andres@2ndquadrant.com>)
Ответы Re: END_OF_RECOVERY shutdowns and ResetUnloggedRelations()
Список pgsql-hackers
At 2014-09-25 22:41:18 +0200, andres@2ndquadrant.com wrote:
>
> On 2014-09-24 17:06:05 +0530, Abhijit Menon-Sen wrote:
> >
> > 1. Move the call to ResetUnloggedRelations(UNLOGGED_RELATION_INIT) to
> >    earlier in StartupXLOG.
> >
> > 2. Inside that function, issue fsync()s for the main forks we create by
> >    copying the _init fork.
>
> These two imo should definitely be backpatched.

I've attached updated versions of these two patches. The only changes
are to revise the comments as you suggested.

> > 3. A small fixup to add a const to "typedef char *FileName", because the
> >    earlier patch gave me warnings about discarding const-ness. This is
> >    consistent with many other functions in fd.c that take const char *.
>
> I'm happy with consider that one for master (although I seem to recall
> having had a patch for it rejected?), but I don't think we want to do
> that in the backbranches.

(I gave up on this for now as an unrelated diversion.)

> > 4. Issue an fsync() on the data directory at startup if we need to
> > perform crash recovery.
>
> I personally think this one should be backpatched too. Does anyone
> believe differently?

I revised this patch as well, I'll rebase and send it separately along
with the initdb -S patch shortly.

-- Abhijit

Вложения

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Petr Jelinek
Дата:
Сообщение: Re: Better support of exported snapshots with pg_dump
Следующее
От: Pavel Stehule
Дата:
Сообщение: Re: Function array_agg(array)