På mandag 20. april 2015 kl. 20:27:39, skrev Jim Nasby mailto:jim.na...@bluetreble.com>>: [snip]
ISTM what would be better is allowing people to define new LO tables, so
we're not stuck trying to cram all LOs into a single table.
As for returning free space, that's a bit of a challenge period,
On 4/17/15 4:29 PM, Andreas Joseph Krogh wrote:
På fredag 17. april 2015 kl. 21:11:05, skrev Jim Nasby
mailto:jim.na...@bluetreble.com>>:
On 4/15/15 9:22 AM, Andreas Joseph Krogh wrote:
> På onsdag 15. april 2015 kl. 16:05:22, skrev Adam Hooper
> mailto:a...@adamhooper.com>>:
På fredag 17. april 2015 kl. 21:11:05, skrev Jim Nasby mailto:jim.na...@bluetreble.com>>: On 4/15/15 9:22 AM, Andreas Joseph Krogh
wrote:
> På onsdag 15. april 2015 kl. 16:05:22, skrev Adam Hooper
> mailto:a...@adamhooper.com>>:
>
> On Wed, Apr 15, 2015 at 9:57 AM, Andreas Joseph Krogh
>
On 4/15/15 9:22 AM, Andreas Joseph Krogh wrote:
På onsdag 15. april 2015 kl. 16:05:22, skrev Adam Hooper
mailto:a...@adamhooper.com>>:
On Wed, Apr 15, 2015 at 9:57 AM, Andreas Joseph Krogh
wrote:
>
> På onsdag 15. april 2015 kl. 15:50:36, skrev Adam Hooper
:
>
>
På onsdag 15. april 2015 kl. 16:05:22, skrev Adam Hooper mailto:a...@adamhooper.com>>: On Wed, Apr 15, 2015 at 9:57 AM, Andreas Joseph
Krogh
wrote:
>
> På onsdag 15. april 2015 kl. 15:50:36, skrev Adam Hooper
:
>
> On Wed, Apr 15, 2015 at 4:49 AM, Andreas Joseph Krogh
> wrote:
> >
> > I
On Wed, Apr 15, 2015 at 9:57 AM, Andreas Joseph Krogh
wrote:
>
> På onsdag 15. april 2015 kl. 15:50:36, skrev Adam Hooper
> :
>
> On Wed, Apr 15, 2015 at 4:49 AM, Andreas Joseph Krogh
> wrote:
> >
> > In other words: Does vacuumlo cause diskspace used by pg_largeobject to be
> > freed to the OS
På onsdag 15. april 2015 kl. 15:50:36, skrev Adam Hooper mailto:a...@adamhooper.com>>: On Wed, Apr 15, 2015 at 4:49 AM, Andreas Joseph
Krogh
wrote:
>
>
> In other words: Does vacuumlo cause diskspace used by pg_largeobject to be
freed to the OS (after eventually vacuumed by autovacuum)?
No
On Wed, Apr 15, 2015 at 4:49 AM, Andreas Joseph Krogh
wrote:
>
>
> In other words: Does vacuumlo cause diskspace used by pg_largeobject to be
> freed to the OS (after eventually vacuumed by autovacuum)?
No.
But that shouldn't matter in your scenario: if you create more large
objects than you de
På onsdag 15. april 2015 kl. 04:43:47, skrev Venkata Balaji N mailto:nag1...@gmail.com>>: I'm routinely vacuumlo'ing to reap orphan OIDs.
Is it necessary to manually vacuum pg_largobject or is it handled by autovacuum?
It is handled by autovacuum. What we do is, we schedule a manual VACUUM
> I'm routinely vacuumlo'ing to reap orphan OIDs. Is it necessary to
> manually vacuum pg_largobject or is it handled by autovacuum?
>
It is handled by autovacuum. What we do is, we schedule a manual VACUUM
ANALYZE nightly job on bigger tables to avoid burden on the autovacuum
during the business
Hi all I'm routinely vacuumlo'ing to reap orphan OIDs. Is it necessary to
manually vacuum pg_largobject or is it handled by autovacuum? In a system
where large objects are constantly added (and *some* rarely deleted, so it
grows every day), would I gain space (freed to the OS) by VACUUM FULL
11 matches
Mail list logo
