Hi Scot,
scott.marlowe said:
> So, to start with, try changing random page cost. you can change it for
As "unrealistic" as it should be, I need <1 before Postgres takes the
bait. Initialy 0.7, to be exact, but later It also worked at a little
higher setting of 1. I have given PG 96Mb of memory to
"muteki muteki" <[EMAIL PROTECTED]> writes:
> And you have pointed out my concern. Even though we have WAL
> enable, we have intentionally disabled both fsync and fdatasync
> inside the kernel because of other reasons.
[blinks...] Surely you're joking?
If you insist on running a configuration a
Eric Ridge wrote:
> On Feb 11, 2004, at 10:00 PM, Bruce Momjian wrote:
> > No one really has thought of that before. We could do it, though there
> > are admin reasons for restricting that ability. If we said only
> > superusers could change it, it wouldn't be very useful.
>
> That's a good poin
On Thu, 12 Feb 2004, Bas Scheffers wrote:
> Hi Scot,
>
> As "unrealistic" as it should be, I need <1 before Postgres takes the
> bait. Initialy 0.7, to be exact, but later It also worked at a little
> higher setting of 1. I have given PG 96Mb of memory to play with, so
> likely all my data will b
On Feb 12, 2004, at 11:54 AM, Bruce Momjian wrote:
Well, let's see if someone else like the feature because adding it
might
reduce its usability as a guaranteed value for admins.
Maybe making the ability to do this a configuration option (off by
default, of course) would make everyone happy?
eri
scott.marlowe said:
> Yes. drop cpu_tuple_index_cost by a factor of 100 or so
No effect.
> Also up effective_cache_size. It's measured in 8k blocks, so for a
That's better, set to 9000, which seems reasonable for my current setup,
it will start using the index when RANDOM_PAGE_COST <= 1.5.
> No
Eric Ridge wrote:
> On Feb 12, 2004, at 11:54 AM, Bruce Momjian wrote:
> > Well, let's see if someone else like the feature because adding it
> > might
> > reduce its usability as a guaranteed value for admins.
>
> Maybe making the ability to do this a configuration option (off by
> default, of
I have the below query written. I have removed a lot from the select
Statement for simplicity sake. The query takes way too long.
I am moving from an Access backend to a Postgres back with
Access in the front. The below query is taking like 14-20 seconds.
Is their a better way I can write the j
Hello:
In kylix3 I made a SQLConnection component (called 'SQLConnection1') to
connect to a PostgreSQL database. I started the PostgreSQL server so
Object inspector shows 'Connected: True' on that object.
Then I made a SQLDataSet component (called 'SQLDataset1').
On 'SQLDataset1' I set SQLConnecti
Clinging to sanity, [EMAIL PROTECTED] (John Gibson) mumbled into her beard:
> I need to upgrade my dual Xeon PostgreSQL engine.
>
> Assuming similar memory and disk sub-systems, I am considering a
> Quad Xeon system vs. a Dual Itanium for PostgreSQL. I believe that
> the PostgreSQL code is written
A long time ago, in a galaxy far, far away, [EMAIL PROTECTED] (Christoffer Gurell)
wrote:
>> I think you should use a cursor; you declare it, and then you fetch the
>> rows as you need them.
>
> thanks this works really nice.. just one more question .. how do i check the
> number of rows in a curs
Quoth [EMAIL PROTECTED] ("Mark Cubitt"):
> is there a way of finding out which table is using a certain amount of CPU
> power?
That doesn't entirely make sense.
Tables normally use up disk space.
What consumes CPU power is the computation of queries, and, for big,
expensive queries, it is pretty
Hi, all.
We are running PostgreSQL 7.2.2 (7.4.1 in a few weeks). We have a
pesky legacy application which periodically leaves open idle
connections. (I know... fix the application). In the short term, I
would like to find out if the database engine has a time-out
configuration option for i
Martha Stewart called it a Good Thing when [EMAIL PROTECTED] ("Ed L.") wrote:
> On Tuesday February 10 2004 3:48, scott.marlowe wrote:
>> On Tue, 10 Feb 2004, Ed L. wrote:
>> > Interesting. Why leave very large tables to the kernel instead
>> > of the db cache? Assuming a dedicated DB server and
The world rejoiced as [EMAIL PROTECTED] (Tom Lane) wrote:
> Martijn van Oosterhout <[EMAIL PROTECTED]> writes:
>> Currently you can create temporary tables that are deleted at the
>> end of the session. But how about temporary views? It's just a
>> table with a rule so I don't imagine it would be t
[EMAIL PROTECTED] ("scott.marlowe") wrote:
> no kidding. I have to admit that in my mad scientist moments, I have
> envisioned writing a new connect lib that's BSD/LPGL licensed. Primarily
> just because I can be a bit antisocial... :)
A group of users, distressed at license changes, proposed
John DeSoi wrote:
Scott,
On Feb 10, 2004, at 2:06 PM, scott.marlowe wrote:
Now, if you don't mind using the ODBC connector, you're scott free.
but
you WILL be bound by the GPL, and the GPL (not MySQL's interpretation,
just the GPL in general) being applied to connect libs seriously limits
your
On Tue, 10 Feb 2004 15:32:11 -0500 Chris Browne <[EMAIL PROTECTED]> wrote:
> There's certainly room for PostgreSQL to have a sub-motto something
> like:
> _PostgreSQL: Free software means no need to fear license audits._
here's something that Theo de Raadt says (in the context of OpenBSD)
that
Hi
> > I have the below query written. I have removed a lot from the
select
> > Statement for simplicity sake. The query takes way too long.
> > I am moving from an Access backend to a Postgres back with
> > Access in the front. The below query is taking like 14-20 seconds.
> > Is their a bette
On Thu, 12 Feb 2004, Jason Tesser wrote:
> I have the below query written. I have removed a lot from the select
> Statement for simplicity sake. The query takes way too long.
> I am moving from an Access backend to a Postgres back with
> Access in the front. The below query is taking like 14-20
I'm flummoxed on this one. I have a class that's building a query which
selects data from 1-n tables based on a common indexed id, io_id. These
tables may contain 1-n rows of data keyed on io_id. What I want the
query to do is return nulls for replicated columns rather than just
replicating
> You didn't answer my question about whether they are the same types.
Are
> they?
Yes but testclass is a view. But the field it is referencing is of the
same type.
---(end of broadcast)---
TIP 7: don't forget to increase your free space map set
> -Original Message-
> From: Jason Tesser [mailto:[EMAIL PROTECTED]
> Sent: Thursday, February 12, 2004 10:03 AM
> To: [EMAIL PROTECTED]
> Subject: [GENERAL] help with query speed
>
>
> I have the below query written. I have removed a lot from the select
> Statement for simplicity sake
On Thu, 12 Feb 2004, Jason Tesser wrote:
> Hi
>
> > > I have the below query written. I have removed a lot from the
> select
> > > Statement for simplicity sake. The query takes way too long.
> > > I am moving from an Access backend to a Postgres back with
> > > Access in the front. The below
On Thursday 12 February 2004 19:32, Jason Tesser wrote:
> Hi
>
> > Is there an index on:
> > testclass.classid
> I created the indexes and it speed up a little. Still a little slower
> than Access though. Which I think should not be. The testclass is a view
> not A query so I cannot make an ind
25 matches
Mail list logo
