I am using 7.1.2 on red hat 7.1
I compiled postgres myself
On Wed, Aug 08, 2001 at 08:16:35PM -0400, [EMAIL PROTECTED] wrote:
> man creatdb says -D supposed to be specify the
> alternative location
>
> I try (as postgres user)
>
> $ createdb -D /bla bla
>
> and it says
>
> absolute pat
If anyone cares I have figured out how to
do this. I use SPI_getbinval
and it works perfectly
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTE
Can someone explain why pg is doing
a sequential scan on table item with the following
statement
-
q=> explain select distinct h.id,i.item,i.heading,i.finish from item i ,body_index h
where h.id=i.item and (h.string='book') order by finish;
NOTICE: QUERY PLAN:
Unique (cost=6591.46..6606.5
Two estimates I undestand are quite
good.
select distinct id on body_index where string='book'
returns about 1500 rows. That matches with
the bottom line of the plan
There are 5139 rows in table item. It is
the same number of rows in the plan for
sequential scan
If I were doing a maual join
On Thu, Aug 16, 2001 at 08:10:41PM -0400, [EMAIL PROTECTED] wrote:
> Ok I set enable_hashjoin and enable_mergejoin to off
> and performance is much much worse: just over 1 second
> job becomes a minute job
>
> Perhaps I should re-check if the database
> gets bigger.
>
> Thanks a lot
>
> On Thu,
Hello
I have statements (highly simplified just to get
the point across) like
select a,b,c from a where d=2 order by e limit 10;
Now I think that because of "order by" the above query
already "knows" the result of the below query
select count(*) from a where d=2;
The point is that I want to k
On Mon, Aug 20, 2001 at 04:56:29PM -0700, Tony Reina wrote:
> Perhaps GROUP BY will get you where you want to go:
>
> select count(*), a, b, c from a where d=2 group by a, b, c order by e limit 10;
>
>
Here count(*) doesn't give total count i.e. grand total
count if there is no "limit."
Wha
On Tue, Aug 21, 2001 at 07:19:42PM -0400, Tom Lane wrote:
>
> What I suspect is that "autocommit off" causes the DBD driver to send a
> fresh BEGIN immediately after the COMMIT. You might be better off with
> "autocommit on" which I think suppresses any automatic issuance of
> BEGIN/COMMIT. The
Did you vacuum after
populating the tables?
If not you should do it
On Wed, Aug 22, 2001 at 11:08:55AM -0400, Paul C. wrote:
> Greetings,
> I am trying to test out the performance of the contrib/fulltextindex
> package and I am getting horrid performance results.
> The Setup:
> I created a
While people are discussing mysql vs pg
I wonder if anyone of the two
support raw partition. If not
is it on the todo list?
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
On Mon, Aug 27, 2001 at 09:40:13AM -0400, Alex Pilosov wrote:
> For the people who really really want PL/java, you can fake it with
> untrusted pl/perl (in 7.2) and Inline::Java.
>
Off topics --
I am very interested in this plperlu
Can plperlu be used in triggers? Any idea
how I can go about
On Mon, Aug 27, 2001 at 11:12:40AM -0400, Tom Lane wrote:
> > child process lost (is SIGCHLD ignored or trapped?)
>
> It's ignored in a backend, see src/backend/tcop/postgres.c.
>
> Current sources change the SIG_IGN setting to SIG_DFL, which may
> well solve your problem; you could try patching
Interesting. I think I might
be completely mistaken exec or eval exec in
tcl with the power of backtick operator
in perl.
Let's say my perl script is
$a=`$a`;
print $a
--
Then tcl $data variable gets not only
$a but also that error message
However I change perl to
---
If you are going to install DBD::Pg you need lib and include directories
just to install the module
On Tue, Feb 20, 2001 at 04:29:34PM +0100, Jose Manuel Lorenzo Lopez wrote:
> Hello PG's,
>
> I have a question concerning the DBI module for postgresql.
>
> I want to use the DBI interface for ac
Thank you. Look what I get..
--
$ pg_ctl start -o "-F -S 2048"
postmaster successfully started up.
$ usage: /home/newsreader/pgsql/bin/postmaster [options]
-B nbufsset number of shared buffers
-D datadir set data directory
-S s
extra -o
to make it work.
On Wed, Feb 21, 2001 at 11:43:46PM -0500, [EMAIL PROTECTED] wrote:
> Thank you. Look what I get..
> --
> $ pg_ctl start -o "-F -S 2048"
> postmaster successfully started up.
> $ usage: /home/newsreader/pgsql/bin/postmaster [options]
I tried at one point though I don't know the cygwin version.
I didn't work; said c compiler cannot produce
executeable
On Wed, Feb 28, 2001 at 10:43:54PM -0500, Jeff wrote:
> I've tried fruitlessly to install cygwin1.1.8 work with postgresql7.03
>
> Has any body out there done it?
>
> Where ca
I think I have been using -F option since
but now I'm not sure. I either start
with pg_ctl or postmaster directly. In any case
man pages suggest that if you can pass optional parameters
to postgres via -o switch from postmaster. I would
think that if I do it correctly such options will
show up
I dump database from 7.0.3 and attempting
to restore to 7.1rc4. And get the following
messages. I can reproduce them by
dropping new db and recreating new ones.
Line number varies from one restore to
the next and 'MoveOfflineLogs' message
was seen only once
Should I worry? How do I increase
W
Is it me or man vacumm is broken in 7.1 release?
Later pages seem to be corrupted. I don't
think my download is to be blamed as
md5sum checks just fine.
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
I really don't want to read man vacuum.
man vacuumdb is good.
On Sun, Apr 15, 2001 at 04:12:12PM -0400, [EMAIL PROTECTED] wrote:
> Is it me or man vacumm is broken in 7.1 release?
> Later pages seem to be corrupted. I don't
> think my download is to be blamed as
> md5sum checks just fine.
>
>
On Tue, Apr 17, 2001 at 09:23:02AM -0300, Marcelo Pereira wrote:
> Hi All,
>
>
> Now I would like to select all employees which name begins with the letter
> "P".
>
> > Select * from employee where "name-begin-with-letter-P" :-)
>
select * from employee where email ~ '^P';
or if cas
psql no longer works the systax
$ psql dbname@hostname
In my DBI scripts I had
$d=DBI->connect('dbi:Pg:dbname=dbname@hostname','user')
That must be rewritten as
$d=DBI->connect('dbi:Pg:dbname=dbname;host=hostname','user')
---(end of broadcast)-
I recently upgraded my production server
to 7.1 because people keep saying how
performance is so much better than 7.0.3
and because 8k row length limitation was removed
postgres is accessed by mod_perl processes
maintaining persistent connections. Before
I can count the number of mod_perl proce
On Wed, May 02, 2001 at 10:45:09AM -0400, Bruce Momjian wrote:
> > I wish these two tables to live on two separately.
>
> I just wrote two articles, one on performance and the other on writing
> PostgreSQL applications. You can get them at:
>
> http://candle.pha.pa.us/main/writings/pgsql/
On Fri, May 04, 2001 at 12:47:02PM -0400, Joel Burton wrote:
>
> Hmmm... this raises an interesting question.
>
> Would it be possible to hook into (via trigger or other mechanism) so that
> we could execute a function on commit? There are PG triggers to do things
> like send email, etc., which,
Thanks everyone for very quick reply.
The reason I found odd was I had
created another able with the same
field name and don't recall having
problems at the time with 7.0.3.
Or did I get the same problem
but I just forgot? I dumped
and reloaded 7.0.3 table to 7.1
without problem though.
On Thu,
27 matches
Mail list logo