On Fri, Dec 03, 2004 at 09:24:48AM -0700, Ed L. wrote:
> Sounds like 24x7x365 operations on a single cluster is maybe too
> tall of an order under these loads. Maybe time for slony & friends
> for maintenance switchovers.
24x7x365 was never the true operation point. All systems have/need
regular
On Saturday December 4 2004 1:01, Ed L. wrote:
>
> ...I have a
> shell game using rules that goes like this for bloated table foo:
> ...
> begin
> drop view foo
> alter table slim_foo rename to foo
> commit
>
> Some basic testing for my particulars suggests this works, but anyone see
>
On Friday December 3 2004 9:09, Tom Lane wrote:
> "Ed L." <[EMAIL PROTECTED]> writes:
> > [SIDEBAR: Vacuum + fsm is not working as
> > I expected; it is clearly not reclaiming space resulting from the
> > UPDATEs. If I UPDATE 1 rows and then run vacuum, I was
> > expecting/hoping that a subseq
"Ed L." <[EMAIL PROTECTED]> writes:
> [SIDEBAR: Vacuum + fsm is not working as
> I expected; it is clearly not reclaiming space resulting from the UPDATEs.
> If I UPDATE 1 rows and then run vacuum, I was expecting/hoping that a
> subsequent UPDATE of 9000 rows would largely reuse the space
On Friday December 3 2004 1:54, Martijn van Oosterhout wrote:
>
> For this, check the FSM settings. It can only keep track of a limited
> number of pages. So if your tuples are large it may not be able to
> track it all...
>
> The FSM should be set big enough to cover all the space that might be
>
Ed L. wrote:
I need to reclaim the diskspace from a heavily updated 7.3.4 table which has
grown 95% bloated to 20gb of disk (indices and toast included), and I need
to reclaim it while the table continues to get updates, and without
interrupting/delaying access more than a few seconds. This is
I can't help you with your general enquiry, but some comments:
On Thu, Dec 02, 2004 at 11:06:08PM -0700, Ed L. wrote:
> If I UPDATE 1 rows and then run vacuum, I was expecting/hoping that a
> subsequent UPDATE of 9000 rows would largely reuse the space reclaimed from
> the 1-row UPDATE.]
I need to reclaim the diskspace from a heavily updated 7.3.4 table which has
grown 95% bloated to 20gb of disk (indices and toast included), and I need
to reclaim it while the table continues to get updates, and without
interrupting/delaying access more than a few seconds. This is an
explanat
