Re: Pg_upgrade and toast tables bug discovered

Поиск
Список
Период
Сортировка
От Fabrízio de Royes Mello
Тема Re: Pg_upgrade and toast tables bug discovered
Дата
Msg-id CAFcNs+rCBwns8pcsqT2fV0zBKxKZ7mGpinHCevJk5tSnyua79A@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Pg_upgrade and toast tables bug discovered  (Robert Haas <robertmhaas@gmail.com>)
Список pgsql-hackers
<div dir="ltr"><div class="gmail_extra"><br />On Mon, Jul 14, 2014 at 12:26 PM, Robert Haas <<a
href="mailto:robertmhaas@gmail.com">robertmhaas@gmail.com</a>>wrote:<br />><br />> On Fri, Jul 11, 2014 at
9:55AM, Bruce Momjian <<a href="mailto:bruce@momjian.us">bruce@momjian.us</a>> wrote:<br /> > > On Fri, Jul
11,2014 at 09:48:06AM -0400, Bruce Momjian wrote:<br />> >> > Uh, why does this need to be in ALTER TABLE?
 Can'tthis be part of<br />> >> > table creation done by pg_dump?<br /> > >><br />> >>
Uh,I think you need to read the thread.  We have to delay the toast<br />> >> creation part so we don't use an
oidthat will later be required by<br />> >> another table from the old cluster.  This has to be done after
all<br/> > >> tables have been created.<br />> >><br />> >> We could have pg_dump spit out
thoseALTER lines at the end of the dump,<br />> >> but it seems simpler to do it in pg_upgrade.<br />>
>><br/> > >> Even if we have pg_dump create all the tables that require pre-assigned<br />> >>
TOASToids first, then the other tables that _might_ need a TOAST table,<br />> >> those later tables might
createa toast oid that matches a later<br /> > >> non-TOAST-requiring table, so I don't think that fixes the
problem.<br/>> ><br />> > What would be nice is if I could mark just the tables that will need<br />>
>toast tables created in that later phase (those tables that didn't have<br /> > > a toast table in the old
cluster,but need one in the new cluster).<br />> > However, I can't see where to store that or how to pass that
backinto<br />> > pg_upgrade.  I don't see a logical place in pg_class to put it.<br /> ><br />>
reloptions?<br/>><br /><br /></div><div class="gmail_extra">Is this another use case to "custom reloptions" idea?<br
/></div><divclass="gmail_extra"><br /></div><div class="gmail_extra">Regards,<br /></div><div class="gmail_extra"><br
/>--<br/>Fabrízio de Royes Mello<br />Consultoria/Coaching PostgreSQL<br />>> Timbira: <a
href="http://www.timbira.com.br">http://www.timbira.com.br</a><br/>>> Blog sobre TI: <a
href="http://fabriziomello.blogspot.com">http://fabriziomello.blogspot.com</a><br/> >> Perfil Linkedin: <a
href="http://br.linkedin.com/in/fabriziomello">http://br.linkedin.com/in/fabriziomello</a><br/>>> Twitter: <a
href="http://twitter.com/fabriziomello">http://twitter.com/fabriziomello</a></div></div>

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Steve Singer
Дата:
Сообщение: Re: 9.4 logical replication - walsender keepalive replies
Следующее
От: Peter Geoghegan
Дата:
Сообщение: Re: B-Tree support function number 3 (strxfrm() optimization)