Hello,
I try to create an unique index for a (time)period, and my goal is to
prevent two overlapping periods in a row.
For this I created a type with following command:
CREATE TYPE period AS
("first" timestamp with time zone,
"next" timestamp with time zone);
To use the btree index I add
On Wed, 2009-08-19 at 08:58 -0400, Arturo Pérez wrote:
> 19-Aug 02:24 bacula-dir JobId 1951: Fatal error: sql_create.c:789
> Fill Path table Query failed: INSERT INTO Path (Path) SELECT a.Path
> FROM (SELECT DISTINCT Path FROM batch) AS a WHERE NOT EXISTS (SELECT
> Path FROM Path WHERE Path
On 08/19/2009 11:41 PM, Randal L. Schwartz wrote:
>> "Clemens" == Clemens Schwaighofer
>> writes:
>
> Clemens> Just in my opinion, this regex is completely too large. For basic
> Clemens> validating something like:
> Clemens>
> ^[A-Za-z0-9!#$%&'*+-\/=?^_`{|}~][A-Za-z0-9!#$%&'*+-\/=?^_`{
> "Clemens" == Clemens Schwaighofer
> writes:
Clemens> I am not going to defend any regex here, but in my opinion it helps on
Clemens> what I want to see in email addresses.
Clemens> Yes it fails on mobile, but I have not yet seen one.
And that's the problem. You get near-sighted if yo
On Wed, 19 Aug 2009, Tom Lane wrote:
BTW, we were doing full-speed tick data collection and real-time trading
analysis in the nineties, on machines that my current cell phone would
leave in the dust. The market data volume has grown a lot since then
of course, but the price of hardware has fall
On Thu, 20 Aug 2009, Sanjay Arora wrote:
What exactly the difference in layman sys admin terms between our
everyday postgres and this time series data.
There is a good introduction to where regular databases fail to solve
time-series data problems at http://cs.nyu.edu/shasha/papers/jagtalk.ht
Peter Hunsberger wrote:
On Wed, Aug 19, 2009 at 2:03 AM, Stephen Cook wrote:
Let's say I have a function that needs to collect some data from various
tables and process and sort them to be returned to the user.
In general, would it be better to create a temporary table in that function,
do the
On Wed, 19 Aug 2009, Adrian Klaver wrote:
In the its a small world category I just ran across a streaming database
called Telegraph(http://telegraph.cs.berkeley.edu/telegraphcq/v2.1/)
yesterday. It is based on Postgres 7.3.2, so approach with caution.
TelegraphCQ was the research project that
Adrian Klaver escribió:
> In the its a small world category I just ran across a streaming database
> called Telegraph(http://telegraph.cs.berkeley.edu/telegraphcq/v2.1/)
> yesterday. It is based on Postgres 7.3.2, so approach with caution.
TelegraphCQ became Truviso.
I clearly remember somebod
On Thu, Aug 20, 2009 at 4:23 AM, Tom Lane wrote:
> It does sound like you are doing pretty much exactly what we were
> doing. One thing to think about is that the real-time case is actually
> much slower and easier to deal with than back-testing. When you are
> back-testing, you'd like to test a
Sanjay Arora writes:
> Do you mean something like storing one month worth tick data in a blob
> type field and giving the contents to the CEP engine for further
> building of required data streams?
Either that or put the tick data in an external file and store that
file's pathname in the database
Thanks Tom
For this reply and many earlier ones that have guided me around postgres ;-)
On Thu, Aug 20, 2009 at 3:34 AM, Tom Lane wrote:
> Sanjay Arora writes:
> It might be worth managing your
> tick data using DBMS entries that represent, say, monthly tick data
> files. Activities like remo
On Wed, Aug 19, 2009 at 11:57:45AM -0600, Scott Marlowe wrote:
> On Wed, Aug 19, 2009 at 8:41 AM, Randal L.
> Schwartz wrote:
> >> "Clemens" == Clemens Schwaighofer
> >> writes:
> >
> > Clemens> Just in my opinion, this regex is completely too large. For basic
> > Clemens> validating some
Sanjay Arora writes:
> I want to store stock market price data in postgreSQL and some of my
> associates are saying this is not workable, not only in postgreSQL but
> in any database. Arguments given are listed below. Kindly comment and
> if possible suggest solutions and pointers to further readi
On Thu, Aug 20, 2009 at 3:11 AM, Adrian Klaver wrote:
>
>
> In the its a small world category I just ran across a streaming database
> called Telegraph(http://telegraph.cs.berkeley.edu/telegraphcq/v2.1/)
> yesterday. It is based on Postgres 7.3.2, so approach with caution.
>
Thanks Adrian,
Very
On Thu, Aug 13, 2009 at 3:00 AM, Greg Smith wrote:
>
>> buffers_backend = 740
>
> This number represents the behavior the background writer is trying to
> prevent--backends having to clean their own buffers up.
>
so what we want on busy systems is buffers_backend to be (at least)
equal or (better)
On Thu, Aug 20, 2009 at 2:45 AM, Merlin Moncure wrote:
>
> I think your first step is to stand back, think about the problem as a
> whole and see if you can't define your requirements better:
>
> *) how much data comes in each day? how much rolls out?
Need to get some data...will have this tomorr
- "Greg Stark" wrote:
> On Wed, Aug 19, 2009 at 9:22 PM, Sanjay
> Arora wrote:
> > - This is Time Series Data (I don't know what that is except that
> it
> > relates to data marked/related to time) and not suited to a RDBMS.
> > - You need it in Esper (a CEP engine used by Marketcetera, an o
On Wed, Aug 19, 2009 at 4:22 PM, Sanjay Arora wrote:
> Hello All
>
> I want to store stock market price data in postgreSQL and some of my
> associates are saying this is not workable, not only in postgreSQL but
> in any database. Arguments given are listed below. Kindly comment and
> if possible su
On Wed, Aug 19, 2009 at 9:22 PM, Sanjay Arora wrote:
> - This is Time Series Data (I don't know what that is except that it
> relates to data marked/related to time) and not suited to a RDBMS.
> - You need it in Esper (a CEP engine used by Marketcetera, an open
> source trading platform) which need
Hello All
I want to store stock market price data in postgreSQL and some of my
associates are saying this is not workable, not only in postgreSQL but
in any database. Arguments given are listed below. Kindly comment and
if possible suggest solutions and pointers to further reading.
- This is Time
Greg Stark wrote:
> On Wed, Aug 19, 2009 at 8:24 PM, Alvaro
> Herrera wrote:
> >> [1] It doesn't correctly convert °C to °F or vv, that was one of the
> >> first things I tried.
> >
> > Seems it's easy to misuse it. You need tempF(x) and tempC notation for
> > converting absolute temperature diffe
On Wed, Aug 19, 2009 at 8:24 PM, Alvaro
Herrera wrote:
>> [1] It doesn't correctly convert °C to °F or vv, that was one of the
>> first things I tried.
>
> Seems it's easy to misuse it. You need tempF(x) and tempC notation for
> converting absolute temperature differences:
>
> You have: tempF(212)
Alban Hertroys wrote:
> >There's also a Debian package which comes with a text format
> >units database:
> >
> > http://packages.debian.org/source/sid/units
> >
> >The original source for that:
> >
> >This package was put together by me, James Troup ,
> >from the GNU sources, which I obtained fro
Glen Jarvis writes:
> I found this thread online because I have the same problem.
> =# drop role ;
> ERROR: role "" cannot be dropped because some objects depend on it
> DETAIL: owner of type pg_toast.pg_toast_51797
> 1 objects in database
I think this is a known issue:
http://archives.postgr
Mark Styles-2 wrote:
>
> On Thu, Jan 29, 2009 at 02:11:37PM -0500, Tom Lane wrote:
>> Mark Styles writes:
>> > Thanks, I managed to clear out the offending dependencies. relowner was
>> > actually set correctly, but the pg_shdepend records were wrong.
>>
>> Hmm ... what actually was in the pg_
On 19 Aug 2009, at 19:20, Karsten Hilbert wrote:
Alban,
I think having an installable schema for units of measure with
definitions and functions would be a great addition to PostgreSQL.
Karsten,
Thanks for the praise and the links.
I for one know we would use it in GNUmed (wiki.gnumed.de).
"Andrus Moor" writes:
> I noticed that backups created by pg_dump are not usable if backup file size
> is greater than 2 GB.
> Backups are create in 8.1
This is a known bug that was fixed more than two years ago ... but not
in 8.1, which was already considered unsupportable on Windows at the
ti
On Wed, Aug 19, 2009 at 8:41 AM, Randal L.
Schwartz wrote:
>> "Clemens" == Clemens Schwaighofer
>> writes:
>
> Clemens> Just in my opinion, this regex is completely too large. For basic
> Clemens> validating something like:
> Clemens>
> ^[A-Za-z0-9!#$%&'*+-\/=?^_`{|}~][A-Za-z0-9!#$%&'*+-
-BEGIN PGP SIGNED MESSAGE-
Hash: RIPEMD160
> Major Replication Requirements (from the Central Office side):
>
> 1. Apply updates on certain tables like products, customers,
> users and set
> How about disabling cycling and doing your own monitoring?
> It seems a bit fiddly because sequence values are created "outside"
> of any transaction and hence if you write something into a logging
> table you're going to loose it if the transaction rolls back.
The sequences are expected to
On Aug 19, 2009, at 9:46 AM, Andre Lopes wrote:
What do you think aboout this? Should I mix logic in Database and
PHP or should I control the logic only in the PHP?
There are no real hard and fast rules, of course, just rules of
thumb. Here are some of mine.
First, remember that there is
"Andrus Moor" writes:
> pg_amop in mydb contains 5 rows.
> pg_amop in template1 database contains large number of rows.
> mydb does not contain user-defined operators.
> How to repair pg_amop in mydb ?
Well, you could try copying the physical file for pg_amop from template1
to mydb (and then re
Alban,
I think having an installable schema for units of measure with
definitions and functions would be a great addition to PostgreSQL.
I for one know we would use it in GNUmed (wiki.gnumed.de).
A few points:
Would these guys be of use as a source for reference data ?
http://unitsofmeasur
On Wed, Aug 19, 2009 at 05:46:17PM +0100, Andre Lopes wrote:
> What do you think aboout this? Should I mix logic in Database and PHP or
> should I control the logic only in the PHP?
As always, it depends! I tend to put things where ever it's most
convenient, however data integrity and other invar
On Wed, 19 Aug 2009, Scott Marlowe wrote:
The only reason I noticed it was that I was building a server with a
separate /data partition for the db to live in, and went to edit
/etc/init.d/postgresql and was faced with two PGDATA assignments... I
too deleted the entire if else block when faced w
Tom,
Thank you.
Alvaro was right --- you've got damage in the system catalogs, not just
their indexes. This looks like missing entries in pg_amop.
postgres -D data mydb
PostgreSQL stand-alone backend 8.1.9
backend> select * from pg_amop
1: amopclaid (typeid = 26, len = 4, typmod =
On Wed, Aug 19, 2009 at 02:11, Randal L. Schwartz wrote:
>> "Andre" == Andre Lopes writes:
>
> Andre> I'm developing a function with some checks, for example... to check if
> the
> Andre> e-mail is valid or not.
>
> How are you hoping to do this? The regex to validate an email
> address synt
Hi all,
I've tried, in various combinations, bacula (2.4, 3.0) and pgsql 8.4
and 8.3 in an upgrade attempt. Whichever combo I try I get the
following error:
19-Aug 02:24 bacula-dir JobId 1951: Fatal error: sql_create.c:789
Fill Path table Query failed: INSERT INTO Path (Path) SELECT a.Pa
Hi,
Now I'am with some doubts on how to control the logic of a web application?
So... the web application uses PHP, object oriented and PostgresSQL
Database. I have developed some procedures and functions in the database but
I don't know if it is correct to mix logic in the database and PHP.
Wha
On Wed, Aug 19, 2009 at 03:53:29PM +0200, Marc Mamin wrote:
> I'm looking for a way to track the wrap arounds of cycle sequences.
How about disabling cycling and doing your own monitoring?
It seems a bit fiddly because sequence values are created "outside" of
any transaction and hence if you writ
2009/8/19 Tom Lane :
> Scott Marlowe writes:
>> In the init script from the PGDG rpms there's this block of code:
>
>> PGDATA=/var/lib/pgsql/data
>> if [ -f "$PGDATA/PG_VERSION" ] && [ -d "$PGDATA/base/template1" ]
>> then
>> echo "Using old-style directory structure"
>> else
>> PGDA
On Wed, Aug 19, 2009 at 4:26 AM, Alan
Millington wrote:
> I am running Postgres 8.1.4 on Windows XP Professional Service Pack 3.
In addition to the very helpful advice Craig provided, you also need
to look at upgrading your pgsql install. 8.1 is no longer supported
on windows due to issues in it
On 18 Aug 2009, at 19:59, Alban Hertroys wrote:
Hello all,
Inspired by the original discussion on aggregating quantities of
different units I made a start at a unit conversion database and the
result is here: http://solfertje.student.utwente.nl/documents/units.sql
I just uploaded an updat
Bob Gobeille writes:
> Yes, I was looking for full text searching in english. Since my
> postgresql.conf contained:
> default_text_search_config = 'pg_catalog.english'
> doesn't this specify the parser, dictionary, and template to use for
> full text searching in english?
It does, but that
"Andrus Moor" writes:
> Alvaro,
>> You can get around that particular problem by reindexing the pg_authid
>> table. But my guess is that you'll find that there's corruption
>> elsewhere that's not so easily recoverable ...
> Thank you.
> reindexing system tables and whole database succeeds.
> Af
On Aug 18, 2009, at 6:21 PM, Tom Lane wrote:
What GIN indexes are good for is indexing equality queries on the
components of something the database otherwise thinks of as a single
object. For instance you can GIN-index searches for arrays containing
a particular value as a member.
Now type te
On Tue, Aug 18, 2009 at 08:21:49PM -0400, Tom Lane wrote:
> Sam Mason writes:
> > On Tue, Aug 18, 2009 at 03:50:47PM -0600, Bob Gobeille wrote:
> >>> CREATE INDEX "ufile_name_search" ON "public"."uploadtree" USING GIN
> >>> ("ufile_name");
> >>> ERROR: data type text has no default operator class
> "Clemens" == Clemens Schwaighofer
> writes:
Clemens> Just in my opinion, this regex is completely too large. For basic
Clemens> validating something like:
Clemens>
^[A-Za-z0-9!#$%&'*+-\/=?^_`{|}~][A-Za-z0-9!#$%&'*+-\/=?^_`{|}~\.]{0,6...@[a-za-z0-9-]+(\.[a-zA-Z0-9-]{1,})*\.([a-zA-Z]{2,
On Tuesday 18 August 2009 8:08:39 pm Craig Ringer wrote:
> On 18/08/2009 9:26 PM, Adrian Klaver wrote:
> > On Monday 17 August 2009 8:50:09 pm Andrew Bartley wrote:
> >> So the information i have thus far is that, I am not easily able to log
> >> the statements from a function.
> >>
> >> Does anyon
On Aug 18, 2009, at 16:45 , Scott Marlowe wrote:
On Tue, Aug 18, 2009 at 10:57 AM, Andre
Lopes wrote:
I need to know if it is possible to show this RAISE NOTICE when I
run this
function from PHP.
http://www.php.net/manual/en/function.pg-last-notice.php
Thanks, Scott.
Is there an equiva
Hello,
I'm looking for a way to track the wrap arounds of cycle sequences.
I also couldn't find the view/catalog where sequence definitions are
kept;
this could be a good place to store at least the timestamp of the last
wrap around.
I'm looking for a way to log all wrap arounds, thought
I gue
Alvaro,
You can get around that particular problem by reindexing the pg_authid
table. But my guess is that you'll find that there's corruption
elsewhere that's not so easily recoverable ...
Thank you.
reindexing system tables and whole database succeeds.
After that I can connect to database c
On 19/08/2009 6:26 PM, Alan Millington wrote:
2009-08-19 03:06:45 ERROR: could not read block 0 of relation
1663/52752/52896: No such file or directory
Clearly something is amiss, but I don't know what. I should be grateful
for any suggestions as to what I should check.
Got a virus scanner
On 16 Aug 2009, at 17:38, Madison Kelly wrote:
Besides that, you don't need the SELECT statement or the RECORD-
type variable as the data you need is already in the NEW and OLD
records. But, you only have an OLD record when your trigger fired
from an UPDATE, so you need to check whether your
I am running Postgres 8.1.4 on Windows XP Professional Service Pack 3.
Yesterday when attempting to select from one particular database I repeatedly
encountered errors such as the following:
2009-08-18 10:49:54 ERROR: could not open relation 1663/51796/1247: No such
file or directory
2009-08
Hello,
Somebody has solved the problem that was not present in 8.3.x ?
Thankyou
Omar
I noticed that backups created by pg_dump are not usable if backup file size
is greater than 2 GB.
Backups are create in 8.1 I tried to restore them in 8.4. Backup contains
many tables. There is a large table, attachme containing bytea field. This
table grows rapidly. If .backup file size is b
Michael Clark wrote:
> That is what Pierre pointed out, and you are both right. I
> am using the text mode.
>
> But it seems pretty crazy that a 140meg bit of data goes to
> 1.3 gigs. Does that seem a bit excessive?
>
> I avoided the binary mode because that seemed to be rather
> confusing w
On 19/08/2009 1:34 PM, Brendan Hill wrote:
Hi Craig, thanks for the analysis. If I attach a debugger on the runaway
child process, will this halt execution for all the other child processes
(ie. freeze the server)? And, can I attach Visual Studio C++ 2008, or is
there a recommended debugger for W
On Wed, 2009-08-19 at 00:01 -0600, Scott Marlowe wrote:
> In the init script from the PGDG rpms there's this block of code:
>
> PGDATA=/var/lib/pgsql/data
>
> if [ -f "$PGDATA/PG_VERSION" ] && [ -d "$PGDATA/base/template1" ]
> then
> echo "Using old-style directory structure"
> else
>
2009/8/19 Dilyan Berkovski :
> Hi All,
>
> I have a nasty table with many repeating columns of the kind
> port_ts_{i}_, where {i} is from 0 to 31, and could be
> 3 different words.
> I have made a pl/pgsql function that checks those columns from
> port_ts_1_status to port_ts_31_status and count
Let's say I have a function that needs to collect some data from various
tables and process and sort them to be returned to the user.
In general, would it be better to create a temporary table in that
function, do the work and sorting there, and return it... or keep a
permanent table for prett
63 matches
Mail list logo
