Nabble - PostgreSQL - admin

Re: Database performance problems

2009-12-23T09:31:56Z

On Wed, 2009-12-23 at 14:45 +0000, Renato Oliveira wrote:

> Dear all,
>
>
>
> I have been using Postgres but I am not an expert on the subject.
>
>
>
> I would be very grateful if someone could point to the right direction
> with regards this problem.
>
>
>
> We have a web application written in JAVA, using jaboss as a servlet.
>
> There are 100 connections open to the database server at any given
> time, apparently using hibernate.
>
>
>
> We are using Postgres 8.2.4 compiled
>
> The database size is 155GB
>
> We have 8GB of RAM
>
> 300GB Hard disk Raid1
>
> Everything is within a single Volume
>
>
>
> Today the load average was as high as 15

Linux?
Are you using elevator=deadline?
How many spindles, what type of raid?
Are you doing proper maintenance?

Joshua D. Drake

--
PostgreSQL.org Major Contributor
Command Prompt, Inc: http://www.commandprompt.com/ - 503.667.4564
Consulting, Training, Support, Custom Development, Engineering
Respect is earned, not gained through arbitrary and repetitive use or Mr. or Sir.

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Database performance problems

2009-12-23T06:45:58Z

Dear all,
 
I have been using Postgres but I am not an expert on the
subject.
 
I would be very grateful if someone could point to the right
direction with regards this problem.
 
We have a web application written in JAVA, using jaboss as a
servlet.
There are 100 connections open to the database server at any
given time, apparently using hibernate.
 
We are using Postgres 8.2.4 compiled
The database size is 155GB
We have 8GB of RAM
300GB Hard disk Raid1 
Everything is within a single Volume
 
Today the load average was as high as 15
 
Top result:
 
Tasks: 169 total,  30 running, 139
sleeping,   0 stopped,   0 zombie
Cpu(s): 69.6% us,  5.9% sy,  0.0% ni,  0.0%
id, 24.3% wa,  0.2% hi,  0.0% si
Mem:   8251404k total,  8228428k used,   
22976k free,    16296k buffers
Swap:  1052248k total,    47176k
used,  1005072k free,  6664308k cached
 
I am quite sure:
1 – we do not have enough memory
2 – our disk and RAID array setup is not good enough
3 – Postgres is not setup correctly and uses an older
version
4 – It is using Slony and replication which does not
work.
 
I would like if anyone could help with the following:
1 – Ways of proving my thoughts
2 – Gather data evidence to prove
 
I don’t have slow query check enabled, will do after
Christmas, but how can I use it to show things are not good?
 
If anyone could help me it would be very much appreciated.
 
If you need more info or details, I am very happy to give
more.
 
Thank you very much
 
Best regards
 
Renato Oliveira
 
 
Renato Oliveira

e-mail: renato.oliveira@...
 
Tel: +44 (0)1763 260811
Fax: +44 (0)1763 262410
www.grant.co.uk
 
Grant Instruments (Cambridge) Ltd 
 
Company registered in England, registration number 658133
 
Registered office address:
29 Station Road, 
Shepreth, 
CAMBS SG8 6GB 
UK
 
 
 
 
 
P Please consider the environment before printing this email
CONFIDENTIALITY: The information in this e-mail and any attachments is confidential. It is intended only for the named recipients(s). If you are not the named recipient please notify the sender immediately and do not disclose the contents to another person or take copies. 
 
VIRUSES: The contents of this e-mail or attachment(s) may contain viruses which could damage your own computer system. Whilst Grant Instruments (Cambridge) Ltd has taken every reasonable precaution to minimise this risk, we cannot accept liability for any damage which you sustain as a result of software viruses. You should therefore carry out your own virus checks before opening the attachment(s).
 
OpenXML: For information about the OpenXML file format in use within Grant Instruments please visit our website

Re: starting postgres with several Data directories

2009-12-21T23:03:58Z

Thank you, Scott & Scott :)

I appreciate your help.

--
Julius Tuskenis
Programavimo skyriaus vadovas
UAB nSoft
mob. +37068233050

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: starting postgres with several Data directories

2009-12-21T13:46:38Z

On Mon, Dec 21, 2009 at 4:35 PM, Scott Marlowe <scott.marlowe@...> wrote:

On Mon, Dec 21, 2009 at 2:26 PM, Scott Mead
<scott.lists@...> wrote:
>
> On Mon, Dec 21, 2009 at 4:17 PM, Scott Marlowe <scott.marlowe@...>
> wrote:
>>

Yeah, I create a different "admin" account for each one, that has the
PGPORT and stuff already set, and setup pg_hba.conf to reject the
admin accounts from other instances so I don't do anything too stupid.

The other useful trick is to change set your psql prompt (via .psqlrc) so that it shows which port the db you're logged into is listening on:

\set PROMPT1 '%/:%>=%# '

This will give you a prompt of:

postgres:5432=#

Take a look at http://www.postgresql.org/docs/8.3/static/app-psql.html for more prompt options.

--Scott

Re: starting postgres with several Data directories

2009-12-21T13:35:15Z

On Mon, Dec 21, 2009 at 2:26 PM, Scott Mead
<scott.lists@...> wrote:

>
> On Mon, Dec 21, 2009 at 4:17 PM, Scott Marlowe <scott.marlowe@...>
> wrote:
>>
>>
>> Yes, it's fine. Make a different data dir, point the postgresql.conf
>> gile to a different tcp port and off ya go. I do it all the time to
>> test slony setup at work.
>
> Agreed, I've actually used it for a couple of different reasons, including
> production in a couple of different architectures, no worries about it,
> other than making sure you have enough memory on the box to support them
> all, along with the remembering which instance you're logged into at a given
> time :)

Yeah, I create a different "admin" account for each one, that has the
PGPORT and stuff already set, and setup pg_hba.conf to reject the
admin accounts from other instances so I don't do anything too stupid.

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: starting postgres with several Data directories

2009-12-21T13:26:31Z

On Mon, Dec 21, 2009 at 4:17 PM, Scott Marlowe <scott.marlowe@...> wrote:

Yes, it's fine. Make a different data dir, point the postgresql.conf
gile to a different tcp port and off ya go. I do it all the time to
test slony setup at work.

Agreed, I've actually used it for a couple of different reasons, including production in a couple of different architectures, no worries about it, other than making sure you have enough memory on the box to support them all, along with the remembering which instance you're logged into at a given time :)

--Scott

Re: starting postgres with several Data directories

2009-12-21T13:17:43Z

On Mon, Dec 21, 2009 at 1:46 PM, Julius T <julius@...> wrote:

> Hello, pgsql guru
>
> I have windows version of postgres 8.4 for experiments.
> At first - what I have done so far:
> 1. Copied content of Data directory to Data2
> 2. Changed the port to 5433 in Data2\postgresql.conf
> 3. Started postgres from commant prompt like (F:/Program
> Files/PostgreSQL/8.4/bin/pg_ctl.exe start -N "postgresql-8.42" -D
> "F:/Program Files/PostgreSQL/8.4/data2" -w)
> 4. Started service of postgres (witch uses Data directory).
> 5. Test the connection with PgAdmin. Connection OK. Inserting into table of
> one server instance doesn't effect second one. Everything seems to work OK.
>
> I wanted to ask if it is OK to start several instances of postgres pointing
> to different data directories? I guess thats what the data directories are
> after all, but because I failed to find some assurance while google'ing I'm
> asking the question. It would be a crying shame to ruin production DB
> because of experimental stuff. :)

Yes, it's fine. Make a different data dir, point the postgresql.conf
gile to a different tcp port and off ya go. I do it all the time to
test slony setup at work.

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

starting postgres with several Data directories

2009-12-21T12:46:19Z

Hello, pgsql guru

I have windows version of postgres 8.4 for experiments.
At first - what I have done so far:
1. Copied content of Data directory to Data2
2. Changed the port to 5433 in Data2\postgresql.conf
3. Started postgres from commant prompt like (F:/Program
Files/PostgreSQL/8.4/bin/pg_ctl.exe start -N "postgresql-8.42" -D
"F:/Program Files/PostgreSQL/8.4/data2" -w)
4. Started service of postgres (witch uses Data directory).
5. Test the connection with PgAdmin. Connection OK. Inserting into table of
one server instance doesn't effect second one. Everything seems to work OK.

I wanted to ask if it is OK to start several instances of postgres pointing
to different data directories? I guess thats what the data directories are
after all, but because I failed to find some assurance while google'ing I'm
asking the question. It would be a crying shame to ruin production DB
because of experimental stuff. :)

Thank you in advance.

Julius Tuskenis

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: VACUUM FULL memory requirements

2009-12-21T09:21:13Z

I've been doing more testing on several different machines, but still haven't found a solution to my problem where VACUUM FULL is running out of memory. Besides the original case on RHEL4, I've been able to reproduce it on both Windows and OSX, with 3GB and 5GB RAM, respectively. Interestingly, it always fails on a request size of 330, even across differently-sized databases running on machines with different amounts of memory.

As far as that new tool you're describing, for my purposes it doesn't make much difference whether it's a stand-alone application or a command, though a command is a little more convenient. The ability to free space concurrent with other database operations would be nice, but I feel like the introduction of serialization failures would negate that advantage, since I at least would end up running it non-concurrently to avoid dealing with them.

I actually find the current design of VACUUM FULL reasonable enough, if I could figure out what's causing this memory issue and prevent it from happening.

David

Re: Backslashes in bytea values

2009-12-18T13:35:15Z

Now that I think of it, pg_escape_bytea operates on a handle to the
current database connection, and I believe it queries the value of
standard_conforming_strings so that it can determine how the data should
be escaped. I think I verified this by testing some time ago, although
I hadn't thought of it until just now, and I would need to test again to
make sure.

Sorry I didn't think this through more carefully before I wrote.

Thanks

Lewis

Kevin Grittner wrote:

> Lewis Kapell <lkapell@...> wrote:
>
>> I forgot to mention that all of these insertions are being done
>> via PHP scripts, and we are using the pg_escape_bytea function
>> (part of the PHP Postgres extension) to escape the data being
>> inserted into bytea columns. Rather important item, that.
>>
>> Does this address all the concerns which you raised?
>
> I'm not familiar with that extension or its functions, but it sounds
> like you're probably in better shape than I thought when I read your
> first email. :-)
>
> Perhaps someone who knows that extension can comment on the issues
> around E'' literals, the escape_string_warning setting and the
> standard_conforming_strings setting.
>
> -Kevin

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: Backslashes in bytea values

2009-12-18T12:43:44Z

Lewis Kapell <lkapell@...> wrote:

> I forgot to mention that all of these insertions are being done
> via PHP scripts, and we are using the pg_escape_bytea function
> (part of the PHP Postgres extension) to escape the data being
> inserted into bytea columns. Rather important item, that.
>
> Does this address all the concerns which you raised?

I'm not familiar with that extension or its functions, but it sounds
like you're probably in better shape than I thought when I read your
first email. :-)

Perhaps someone who knows that extension can comment on the issues
around E'' literals, the escape_string_warning setting and the
standard_conforming_strings setting.

-Kevin

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: Backslashes in bytea values

2009-12-18T12:33:45Z

Kevin,

Thanks for your reply.

I forgot to mention that all of these insertions are being done via PHP
scripts, and we are using the pg_escape_bytea function (part of the PHP
Postgres extension) to escape the data being inserted into bytea
columns. Rather important item, that.

Does this address all the concerns which you raised?

Thanks
Lewis

Kevin Grittner wrote:

> Lewis Kapell <lkapell@...> wrote:
>
>> I have updated much of our code to use the E'' syntax in order to
>> suppress these messages.
>
> If you do that, you'd better be doing something to make sure you
> have cured the problem the messages warned of; otherwise you're
> going to have corrupted documents. Are you using prepared
> statements and setting the values through those (recommended) or
> escaping the backslashes, etc. yourself?
>
>> I am trying to figure out what the repercussions would be if I
>> were to turn on the standard_conforming_strings parameter.
>
> With that on, absolutely all characters are taken as part of the
> literal except for apostrophes. That makes it a bit easier to do
> the escaping, and could result in a slight reduction in statement
> size.
>
>> None of our code actually uses a backslash to escape a double-
>> quote, but I don't know how this would affect the bytea values
>> which I discussed above. How are backslashes affected if they are
>> actually part of the raw data?
>
> If you're just throwing them inside a E'xxx' literal that you build
> yourself, it sounds like you've currently got a problem. If you are
> using parameters on prepared statements, it shouldn't matter much.
>
> -Kevin

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: Backslashes in bytea values

2009-12-18T12:23:45Z

Lewis Kapell <lkapell@...> wrote:

> My company is using Postgres 8.3.8. We have some tables with
> bytea columns that store file data such as the contents of PDF or
> RTF files.

Yeah, we do a lot of that, too.

> The escape_string_warning parameter is on, and inserting this kind
> of data causes the backslash warning to be generated, since the
> backslash character appears in the raw data of these types of
> files.

We turn that off.

> I have updated much of our code to use the E'' syntax in order to
> suppress these messages.

If you do that, you'd better be doing something to make sure you
have cured the problem the messages warned of; otherwise you're
going to have corrupted documents. Are you using prepared
statements and setting the values through those (recommended) or
escaping the backslashes, etc. yourself?

> I am trying to figure out what the repercussions would be if I
> were to turn on the standard_conforming_strings parameter.

With that on, absolutely all characters are taken as part of the
literal except for apostrophes. That makes it a bit easier to do
the escaping, and could result in a slight reduction in statement
size.

> None of our code actually uses a backslash to escape a double-
> quote, but I don't know how this would affect the bytea values
> which I discussed above. How are backslashes affected if they are
> actually part of the raw data?

If you're just throwing them inside a E'xxx' literal that you build
yourself, it sounds like you've currently got a problem. If you are
using parameters on prepared statements, it shouldn't matter much.

-Kevin

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: Foreign keys with values of zero rather than null

2009-12-18T11:46:01Z

"Campbell, Lance" <lance@...> wrote:
> I map data from rows in tables to objects in Java. When I map
> integers in the database I map them to integer primitives in Java.
> The challenge I am running into is what do I do when a table has a
> integer foreign key that is null. Ideally when a table in
> PostgreSql has a foreign key that does not exist I would like the
> value to be zero rather than null. This works really well for me
> because I NEVER use a key with the value of 0.

I would definitely use NULL in the database to indicate that a value
is unknown or not applicable. Really. Always.

How you map that to your Java objects is another issue -- if you
used an Integer object, you would have a natural mapping of NULL in
the database to a Java null and back again. That's what I normally
do. In my view, you need a really good reason to map to the
primitive, especially since Java has added the automatic boxing and
unboxing features.

If you feel compelled to micro-optimize to primitives, you need to
have some way to map the NULLs to your magic number and back again.
The former is pretty easy if you use zero as your magic number,
since getInt returns zero for a NULL, leaving it to you to check
wasNull to determine whether it's really a NULL or a zero, that
direction is easy.

http://java.sun.com/javase/6/docs/api/java/sql/ResultSet.html#getInt(int)

http://java.sun.com/javase/6/docs/api/java/sql/ResultSet.html#wasNull()

Going the other direction, you need some way to convert a zero in
that column to a NULL in the database. Exactly how you do that
depends on how you are doing your mapping -- unless you want to try
to use PostgreSQL rules to rewrite your statements or some such.

-Kevin

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Backslashes in bytea values

2009-12-18T11:33:03Z

My company is using Postgres 8.3.8. We have some tables with bytea
columns that store file data such as the contents of PDF or RTF files.
The escape_string_warning parameter is on, and inserting this kind of
data causes the backslash warning to be generated, since the backslash
character appears in the raw data of these types of files. I have
updated much of our code to use the E'' syntax in order to suppress
these messages.

I am trying to figure out what the repercussions would be if I were to
turn on the standard_conforming_strings parameter. None of our code
actually uses a backslash to escape a double-quote, but I don't know how
this would affect the bytea values which I discussed above. How are
backslashes affected if they are actually part of the raw data?

--

Thank you,

Lewis Kapell
Computer Operations
Seton Home Study School

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Foreign keys with values of zero rather than null

2009-12-18T11:07:55Z

I map data from rows in tables to objects in Java. When I map integers in the database I map them to integer primitives in Java. The challenge I am running into is what do I do when a table has a integer foreign key that is null. Ideally when a table in PostgreSql has a foreign key that does not exist I would like the value to be zero rather than null. This works really well for me because I NEVER use a key with the value of 0.

Is there some way for me to tell PostgreSql that a zero is the same as null when using foreign keys that map to integers? That way I can still use cascade delete. The only way for me to get around this issue is to remove the foreign key constraint for cascade delete.

Any help would be appreciated.

Thanks,

Lance Campbell

Software Architect/DBA/Project Manager

Web Services at Public Affairs

217-333-0382

Fwd: Problems to connect to a postgreSQL 8.4 database on windows 2003 server

2009-12-18T05:37:29Z

Hi Shakeel,

At last I got that. It was a silly mistake when setting up the ip addresses. I forgot to put the Subnet Mask. Now it's working fine.

Cheers,

Francesc

Per a qui navega sense rumb cap vent és favorable.

Sèneca

---Procureu enviar els correus massius amb les direccions en còpia carbó oculta (cco) per no col·laborar amb qui les col·lecciona per enviar SPAM---

Re: Problems to connect to a postgreSQL 8.4 database on windows 2003 server

2009-12-18T04:32:33Z

please share your hba.conf file

On Fri, Dec 18, 2009 at 3:51 PM, Francesc <francesc.b.ricart@...> wrote:

Hi,

I've a postgreSQL database in a windows 2003 server. My problem is I can't connect from other computers of the net. I've tried to configure hba.conf and postgresql.conf but I wasn't successful. Is there anything else I should do? Could anyone help me telling me how to do that?

Many thanks in advance.

Francesc

Per a qui navega sense rumb cap vent és favorable.

Sèneca

---Procureu enviar els correus massius amb les direccions en còpia carbó oculta (cco) per no col·laborar amb qui les col·lecciona per enviar SPAM---

--
Best Regards
Shakeel Abbas

Problems to connect to a postgreSQL 8.4 database on windows 2003 server

2009-12-18T02:51:58Z

Hi,

I've a postgreSQL database in a windows 2003 server. My problem is I can't connect from other computers of the net. I've tried to configure hba.conf and postgresql.conf but I wasn't successful. Is there anything else I should do? Could anyone help me telling me how to do that?

Many thanks in advance.

Francesc

Per a qui navega sense rumb cap vent és favorable.

Sèneca

---Procureu enviar els correus massius amb les direccions en còpia carbó oculta (cco) per no col·laborar amb qui les col·lecciona per enviar SPAM---

Re: Inheritance in Postgresql ?

2009-12-16T20:31:33Z

Thanks Richard Broersma for the presentation. Its really nice and explains advantages and disadvantages in a nice way.

On Wed, Dec 16, 2009 at 8:13 PM, Richard Broersma <richard.broersma@...> wrote:

On Tue, Dec 15, 2009 at 11:33 PM, ABBAS SHAKEEL

<shakeel.abbas.qau@...> wrote:

> I feel relatively clear with this concept and i stopped my mind as well from
> comparing inheritance concept with JAVA by saying that inheritance concept
> in humans, in JAVA and in Postgres is different in many aspects.
> I would like to share an example as well which describe the advantage of
> inheritance in Postgres in detail . Example can be found here.

I developed a presentation the discusses hierarchical data modeling.
One section discusses PostgreSQL's inheritance.

http://wiki.postgresql.org/images/9/91/Pguswest2008hnd.pdf

--
Regards,
Richard Broersma Jr.

Visit the Los Angeles PostgreSQL Users Group (LAPUG)
http://pugs.postgresql.org/lapug

--
Best Regards
Shakeel Abbas

Re: identify referential integrity triggers unassociated with foreign key constraints

2009-12-16T10:45:42Z

Bill MacArthur <webmaster@...> writes:
> We started using Postgres back around 7.1. We've done dump and reloads to get us up to where we are now at 8.1.

> Over time it would seem that the guts of foreign key constraints have
> changed, for the better. Soon we want to move to the latest version,
> but in the process, I would like to update all the old style "foreign
> keys" to the newest. By old style, I mean we have RI triggers on
> various tables like "RI_ConstraintTrigger_3495148". In pgadmin, these
> do not show up as triggers, neither do they show up as foreign keys as
> some of the system table relationships that exist now for identifying
> these sort of things did not exist when these "foreign keys" were
> first setup. (These constraints still work, you just can't tell they
> are they using pgadmin) Using \d tablename shows them in a list of
> triggers but not as Foreign-key constraints.

These should get updated to a "real" FK constraint automatically when
you load into 8.3 or 8.4. See commits here:
http://archives.postgresql.org/pgsql-committers/2007-11/msg00036.php
http://archives.postgresql.org/pgsql-committers/2007-11/msg00042.php
http://archives.postgresql.org/pgsql-committers/2007-11/msg00069.php
and concurrent discussions (probably on -hackers)

> Occasionally I have taken to dropping some of these triggers and
> mucking with the system tables like pg_trigger and so forth, but those
> experiences are horrid. Things have never been hosed, but I know there
> has to be a better way.

... or at least, I expect that that upgrade logic will work with
unmodified old-style FK trigger definitions. If you did something other
than drop all the old triggers and create a regular FK constraint,
I can't promise much.

regards, tom lane

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

identify referential integrity triggers unassociated with foreign key constraints

2009-12-16T09:45:39Z

Hello,

We started using Postgres back around 7.1. We've done dump and reloads to get us up to where we are now at 8.1.

Over time it would seem that the guts of foreign key constraints have changed, for the better. Soon we want to move to the latest version, but in the process, I would like to update all the old style "foreign keys" to the newest. By old style, I mean we have RI triggers on various tables like "RI_ConstraintTrigger_3495148". In pgadmin, these do not show up as triggers, neither do they show up as foreign keys as some of the system table relationships that exist now for identifying these sort of things did not exist when these "foreign keys" were first setup. (These constraints still work, you just can't tell they are they using pgadmin) Using \d tablename shows them in a list of triggers but not as Foreign-key constraints.

Occasionally I have taken to dropping some of these triggers and mucking with the system tables like pg_trigger and so forth, but those experiences are horrid. Things have never been hosed, but I know there has to be a better way.

What I would like to do is to identify these almost invisible "foreign key" triggers so that they can be dropped sanely and then create new, named constraints can be identified using ordinary tools. Is this possible?

Many thanks,
Bill MacArthur

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: Inheritance in Postgresql ?

2009-12-16T07:13:49Z

On Tue, Dec 15, 2009 at 11:33 PM, ABBAS SHAKEEL
<shakeel.abbas.qau@...> wrote:

> I feel relatively clear with this concept and i stopped my mind as well from
> comparing inheritance concept with JAVA by saying that inheritance concept
> in humans, in JAVA and in Postgres is different in many aspects.
> I would like to share an example as well which describe the advantage of
> inheritance in Postgres in detail . Example can be found here.

I developed a presentation the discusses hierarchical data modeling.
One section discusses PostgreSQL's inheritance.

http://wiki.postgresql.org/images/9/91/Pguswest2008hnd.pdf

--
Regards,
Richard Broersma Jr.

Visit the Los Angeles PostgreSQL Users Group (LAPUG)
http://pugs.postgresql.org/lapug

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: Looking for Good and Easy-to-use Reporting Tool

2009-12-16T06:10:40Z

2009/12/16 Su, Alan <Alan.Su@...>:
> GUI front end is nice to have but not required. We. have a system that collects network speed information in mill seconds from different hubs and this infor is stored on various Postgre tables. Throughout the day or at end of the day, I need to run a process/job that will summarize these speed data from the tables, and when the number exceeds certain threshold, say 50 ms, I want to generate an alert email to support staff for investigation.
>
> I may expand the requirements further down the road, but main processing logic should stay similar to what I described above. So what's the good tool to use for this background job?
>

Alan, as Kevin Grittner suggested, maybe you could just write a script
using bash. Otherwise, if you need more than that, and don't want to
do the coding yourself, do you have a budget?

>
> Alan. (sent from my BlackBerry Wireless Handheld)
>
> ----- Original Message -----
> From: Brian Modra <epailty@...>
> To: Su, Alan
> Cc: pgsql-admin@... <pgsql-admin@...>
> Sent: Wed Dec 16 02:43:46 2009
> Subject: Re: [ADMIN] Looking for Good and Easy-to-use Reporting Tool
>
> 2009/12/15 Su, Alan <Alan.Su@...>:
>> Brian,
>>
>> What you mentioned below is more sophisticated than what I am capable of doing now. Could you or anyone on this email recommend other tool that meets the following requirements:
>>
>> 1. Require less programming effort. I am trying to stay away from having to
>> write sophisticate programming language, such as Java or .Net
>> 2. Allow to bundle SQL commands in store procedure like style.
>> 3. Allow to evaluate conditions of SQL result set and trigger alert email based
>> on a preset threshold.
>
> Please describe the application in more detail. Do you need any sort
> of GUI front end to the application?
> Or is it just reporting via email?
>
>>
>>
>> -----Original Message-----
>> From: Brian Modra [mailto:epailty@...]
>> Sent: Friday, December 11, 2009 3:09 PM
>> To: Su, Alan
>> Cc: pgsql-admin@...
>> Subject: Re: [ADMIN] Looking for Good and Easy-to-use Reporting Tool
>>
>> 2009/12/11 Su, Alan <Alan.Su@...>:
>>> Can anyone recommend a good, easy to ease and inexpensive reporting tool for
>>> accessing and manipulating Postgre database/table? The key features I am
>>> looking for with this tool are as follows:
>>>
>>>
>>>
>>> Require less programming effort. I am trying to stay away from having to
>>> write sophisticate programming language, such as Java or .Net
>>> Allow to bundle SQL commands in store procedure like style.
>>> Allow to evaluate conditions of SQL result set and trigger alert email based
>>> on preset threshold.
>>
>>
>> Have you seen this:
>> http://www.zwartberg.com/surf.html
>>
>> You can write SQLs or calls to PLPGSQL (stored procedure) directly
>> into the XML configuration file, then generate a web front end. No
>> Java or C++ required, though you can make complex features in C++. But
>> for more complex GUI logic, you'll need Javascript. There is no
>> escaping writing programs to describe logic, but the straight-forward
>> front-end is all auto-generated from your XML configuration and HTML.
>>
>> To do email from triggers will require extra coding, but talk to me about it...
>>
>> The Karoo Project is new, and the documentation is a bit lacking, but
>> you can email me your questions "Brian Modra" <brian@...>
>>
>>>
>>>
>>> Thanks.
>>>
>>> Alan
>>>
>>> This email and all files and information transmitted with it are
>>> confidential and intended solely for the use of the individual or entity to
>>> whom the email is addressed. If you are not the intended recipient or the
>>> person responsible for delivering the email to the intended recipient(s),
>>> you have received this email in error, and use of it is strictly prohibited.
>>> Opinions expressed herein are those of the individual sender and not of the
>>> company unless specifically so indicated. If you have received this
>>> communication in error, please immediately notify the sender, delete this
>>> message, and do not print, copy, or disclose the content of this message.
>>> Inadvertent disclosure of this message does not waive any privilege or claim
>>> of confidentiality.
>>>
>>> For important conflict disclosures, please visit
>>> http://www.daiwausa.com/report_disclosure.html
>>>
>>> If your settings do not permit you to access this link directly, please use
>>> your internet browser to visit this website address.
>>>
>>
>>
>>
>> --
>> Brian Modra Land line: +27 23 5411 462
>> Mobile: +27 79 69 77 082
>> 5 Jan Louw Str, Prince Albert, 6930
>> Postal: P.O. Box 2, Prince Albert 6930
>> South Africa
>> http://www.zwartberg.com/
>>
>> This email and all files and information transmitted with it are confidential and intended solely for the use of the individual or entity to whom the email is addressed. If you are not the intended recipient or the person responsible for delivering the email to the intended recipient(s), you have received this email in error, and use of it is strictly prohibited. Opinions expressed herein are those of the individual sender and not of the company unless specifically so indicated. If you have received this communication in error, please immediately notify the sender, delete this message, and do not print, copy, or disclose the content of this message. Inadvertent disclosure of this message does not waive any privilege or claim of confidentiality.
>>
>> For important conflict disclosures, please visit http://www.daiwausa.com/report_disclosure.html
>>
>> If your settings do not permit you to access this link directly, please use your internet browser to visit this website address.
>>
>
>
>
> --
> Brian Modra Land line: +27 23 5411 462
> Mobile: +27 79 69 77 082
> 5 Jan Louw Str, Prince Albert, 6930
> Postal: P.O. Box 2, Prince Albert 6930
> South Africa
> http://www.zwartberg.com/
>
> This email and all files and information transmitted with it are confidential and intended solely for the use of the individual or entity to whom the email is addressed. If you are not the intended recipient or the person responsible for delivering the email to the intended recipient(s), you have received this email in error, and use of it is strictly prohibited. Opinions expressed herein are those of the individual sender and not of the company unless specifically so indicated. If you have received this communication in error, please immediately notify the sender, delete this message, and do not print, copy, or disclose the content of this message. Inadvertent disclosure of this message does not waive any privilege or claim of confidentiality.
>
> For important conflict disclosures, please visit http://www.daiwausa.com/report_disclosure.html
>
> If your settings do not permit you to access this link directly, please use your internet browser to visit this website address.
>

--
Brian Modra Land line: +27 23 5411 462
Mobile: +27 79 69 77 082
5 Jan Louw Str, Prince Albert, 6930
Postal: P.O. Box 2, Prince Albert 6930
South Africa
http://www.zwartberg.com/

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

password authentication failed

2009-12-16T05:19:56Z

I am newbee for Bucardo.

Whenever i am trying to add database or try to look into show all command. It shows me following error.

Connecting to database 'bucardo' as user 'bucardo'
DBI connect('dbname=bucardo;port=5433','bucardo',...) failed: FATAL: password authentication failed for user "bucardo" at ./bucardo_ctl line 191

/bucardo_ctl show all

./bucardo_ctl add db test

I have created bucardo user with password goat & it is a superuser.

Kindly suggest me to resolve this issue.

I am following below link.
http://bucardo.org/wiki/Bucardo/Installation

Thanks
Sam Jas

The INTERNET now has a personality. YOURS! See your Yahoo! Homepage.

Re: Looking for Good and Easy-to-use Reporting Tool

2009-12-16T05:19:38Z

GUI front end is nice to have but not required. We. have a system that collects network speed information in mill seconds from different hubs and this infor is stored on various Postgre tables. Throughout the day or at end of the day, I need to run a process/job that will summarize these speed data from the tables, and when the number exceeds certain threshold, say 50 ms, I want to generate an alert email to support staff for investigation.

I may expand the requirements further down the road, but main processing logic should stay similar to what I described above. So what's the good tool to use for this background job?

Alan. (sent from my BlackBerry Wireless Handheld)

----- Original Message -----
From: Brian Modra <epailty@...>
To: Su, Alan
Cc: pgsql-admin@... <pgsql-admin@...>
Sent: Wed Dec 16 02:43:46 2009
Subject: Re: [ADMIN] Looking for Good and Easy-to-use Reporting Tool

2009/12/15 Su, Alan <Alan.Su@...>:
> Brian,
>
> What you mentioned below is more sophisticated than what I am capable of doing now. Could you or anyone on this email recommend other tool that meets the following requirements:
>
> 1. Require less programming effort. I am trying to stay away from having to
> write sophisticate programming language, such as Java or .Net
> 2. Allow to bundle SQL commands in store procedure like style.
> 3. Allow to evaluate conditions of SQL result set and trigger alert email based
> on a preset threshold.

Please describe the application in more detail. Do you need any sort
of GUI front end to the application?
Or is it just reporting via email?

>
>
> -----Original Message-----
> From: Brian Modra [mailto:epailty@...]
> Sent: Friday, December 11, 2009 3:09 PM
> To: Su, Alan
> Cc: pgsql-admin@...
> Subject: Re: [ADMIN] Looking for Good and Easy-to-use Reporting Tool
>
> 2009/12/11 Su, Alan <Alan.Su@...>:
>> Can anyone recommend a good, easy to ease and inexpensive reporting tool for
>> accessing and manipulating Postgre database/table? The key features I am
>> looking for with this tool are as follows:
>>
>>
>>
>> Require less programming effort. I am trying to stay away from having to
>> write sophisticate programming language, such as Java or .Net
>> Allow to bundle SQL commands in store procedure like style.
>> Allow to evaluate conditions of SQL result set and trigger alert email based
>> on preset threshold.
>
>
> Have you seen this:
> http://www.zwartberg.com/surf.html
>
> You can write SQLs or calls to PLPGSQL (stored procedure) directly
> into the XML configuration file, then generate a web front end. No
> Java or C++ required, though you can make complex features in C++. But
> for more complex GUI logic, you'll need Javascript. There is no
> escaping writing programs to describe logic, but the straight-forward
> front-end is all auto-generated from your XML configuration and HTML.
>
> To do email from triggers will require extra coding, but talk to me about it...
>
> The Karoo Project is new, and the documentation is a bit lacking, but
> you can email me your questions "Brian Modra" <brian@...>
>
>>
>>
>> Thanks.
>>
>> Alan
>>
>> This email and all files and information transmitted with it are
>> confidential and intended solely for the use of the individual or entity to
>> whom the email is addressed. If you are not the intended recipient or the
>> person responsible for delivering the email to the intended recipient(s),
>> you have received this email in error, and use of it is strictly prohibited.
>> Opinions expressed herein are those of the individual sender and not of the
>> company unless specifically so indicated. If you have received this
>> communication in error, please immediately notify the sender, delete this
>> message, and do not print, copy, or disclose the content of this message.
>> Inadvertent disclosure of this message does not waive any privilege or claim
>> of confidentiality.
>>
>> For important conflict disclosures, please visit
>> http://www.daiwausa.com/report_disclosure.html
>>
>> If your settings do not permit you to access this link directly, please use
>> your internet browser to visit this website address.
>>
>
>
>
> --
> Brian Modra Land line: +27 23 5411 462
> Mobile: +27 79 69 77 082
> 5 Jan Louw Str, Prince Albert, 6930
> Postal: P.O. Box 2, Prince Albert 6930
> South Africa
> http://www.zwartberg.com/
>
> This email and all files and information transmitted with it are confidential and intended solely for the use of the individual or entity to whom the email is addressed. If you are not the intended recipient or the person responsible for delivering the email to the intended recipient(s), you have received this email in error, and use of it is strictly prohibited. Opinions expressed herein are those of the individual sender and not of the company unless specifically so indicated. If you have received this communication in error, please immediately notify the sender, delete this message, and do not print, copy, or disclose the content of this message. Inadvertent disclosure of this message does not waive any privilege or claim of confidentiality.
>
> For important conflict disclosures, please visit http://www.daiwausa.com/report_disclosure.html
>
> If your settings do not permit you to access this link directly, please use your internet browser to visit this website address.
>

--
Brian Modra Land line: +27 23 5411 462
Mobile: +27 79 69 77 082
5 Jan Louw Str, Prince Albert, 6930
Postal: P.O. Box 2, Prince Albert 6930
South Africa
http://www.zwartberg.com/

This email and all files and information transmitted with it are confidential and intended solely for the use of the individual or entity to whom the email is addressed. If you are not the intended recipient or the person responsible for delivering the email to the intended recipient(s), you have received this email in error, and use of it is strictly prohibited. Opinions expressed herein are those of the individual sender and not of the company unless specifically so indicated. If you have received this communication in error, please immediately notify the sender, delete this message, and do not print, copy, or disclose the content of this message. Inadvertent disclosure of this message does not waive any privilege or claim of confidentiality.

For important conflict disclosures, please visit http://www.daiwausa.com/report_disclosure.html

If your settings do not permit you to access this link directly, please use your internet browser to visit this website address.

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: Internal fragmentations statistics Was: VACUUM FULL memory requirements

2009-12-15T23:51:44Z

Le mercredi 16 décembre 2009 à 06:18:45, Gurjeet Singh a écrit :

> 2009/12/15 Guillaume Lelarge <guillaume@...>
>
> > Le mardi 15 décembre 2009 à 00:04:47, Evan Rempel a écrit :
> > > Is there a command/tool that will report on how FULL a table is
> > > getting? If there is, how intrusive is it? How computationally heavy is
> > > it?
> > >
> > > We have a database that is approx 100 million rows with
> > > approx 2 million insert/updates per day. Each day old data
> > > is purged from the database. The end result is a mostly static
> > > footprint with regards to disk space used, but I would like to
> > > know how much room is usable inside the tables as well as the
> > > OS file system (that part is easy).
> >
> > pgstattuple contrib module give such an information. But it requires an
> > exclusive lock on the table it's looking at, so it's quite intrusive. For
> > more
> > details, the 8.4 documentation is interesting:
> > http://www.postgresql.org/docs/8.4/interactive/pgstattuple.html
>
> That doc specifically says that it takes only a read lock.
>

Ouch, I always thought it took an exclusive lock. Sorry.

--
Guillaume.
http://www.postgresqlfr.org
http://dalibo.com

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: Looking for Good and Easy-to-use Reporting Tool

2009-12-15T23:43:46Z

2009/12/15 Su, Alan <Alan.Su@...>:
> Brian,
>
> What you mentioned below is more sophisticated than what I am capable of doing now. Could you or anyone on this email recommend other tool that meets the following requirements:
>
> 1. Require less programming effort. I am trying to stay away from having to
> write sophisticate programming language, such as Java or .Net
> 2. Allow to bundle SQL commands in store procedure like style.
> 3. Allow to evaluate conditions of SQL result set and trigger alert email based
> on a preset threshold.

Please describe the application in more detail. Do you need any sort
of GUI front end to the application?
Or is it just reporting via email?

Re: Inheritance in Postgresql ?

2009-12-15T23:33:56Z

Thanks Scott,

Your comments are of worth.

I feel relatively clear with this concept and i stopped my mind as well from comparing inheritance concept with JAVA by saying that inheritance concept in humans, in JAVA and in Postgres is different in many aspects.

I would like to share an example as well which describe the advantage of inheritance in Postgres in detail . Example can be found here.

On Wed, Dec 16, 2009 at 12:15 PM, Scott Marlowe <scott.marlowe@...> wrote:

On Wed, Dec 16, 2009 at 12:01 AM, ABBAS SHAKEEL

<shakeel.abbas.qau@...> wrote:

> Thanks Scott
> you are right . It inherits structure only and i don't know whether to call
> to it an advantage or disadvantage that it query the children table as well
> while select. But in case of insert and update this concepts takes a new
> turn.
> How inheritance effect data?
> Can you please briefly compare concept of inheritance in Postgres and JAVA,
> as my mind is trying to compare this with already stored concept in mind
> about inheritance ;)

I'm no Java specialist, so can't directly compare there. In the case
of inserts those go to the parent or, if you insert directly to them
the child tables. Updates operate on the appropriate child tables.

By defining triggers or rules, you can make inserts to the parent
table go to appropriate child tables, which is often done for
partitioning data, which is in turn the most common use for table
inheritance.

You can also select data from JUST the parent with the only keyword,
as in "select * from only parent_table".

Note that pgsql supports multiple inheritance, so that you can have a
table inherit structure from two or more parent tables as well.

Note that as of 8.4, according to the docs:
"All check constraints and not-null constraints on a parent table are
automatically inherited by its children. Other types of constraints
(unique, primary key, and foreign key constraints) are not inherited."

So for foreign keys and such, inheritance is much less useful than
regular tables.

--
Best Regards
Shakeel Abbas

Re: Inheritance in Postgresql ?

2009-12-15T23:15:38Z

On Wed, Dec 16, 2009 at 12:01 AM, ABBAS SHAKEEL
<shakeel.abbas.qau@...> wrote:
> Thanks Scott
> you are right . It inherits structure only and i don't know whether to call
> to it an advantage or disadvantage that it query the children table as well
> while select. But in case of insert and update this concepts takes a new
> turn.
> How inheritance effect data?
> Can you please briefly compare concept of inheritance in Postgres and JAVA,
> as my mind is trying to compare this with already stored concept in mind
> about inheritance ;)

I'm no Java specialist, so can't directly compare there. In the case
of inserts those go to the parent or, if you insert directly to them
the child tables. Updates operate on the appropriate child tables.

By defining triggers or rules, you can make inserts to the parent
table go to appropriate child tables, which is often done for
partitioning data, which is in turn the most common use for table
inheritance.

You can also select data from JUST the parent with the only keyword,
as in "select * from only parent_table".

Note that pgsql supports multiple inheritance, so that you can have a
table inherit structure from two or more parent tables as well.

Note that as of 8.4, according to the docs:
"All check constraints and not-null constraints on a parent table are
automatically inherited by its children. Other types of constraints
(unique, primary key, and foreign key constraints) are not inherited."

So for foreign keys and such, inheritance is much less useful than
regular tables.

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Re: Inheritance in Postgresql ?

2009-12-15T23:01:05Z

Thanks Scott

you are right . It inherits structure only and i don't know whether to call to it an advantage or disadvantage that it query the children table as well while select. But in case of insert and update this concepts takes a new turn.

How inheritance effect data?

Can you please briefly compare concept of inheritance in Postgres and JAVA, as my mind is trying to compare this with already stored concept in mind about inheritance ;)

On Wed, Dec 16, 2009 at 11:51 AM, Scott Marlowe <scott.marlowe@...> wrote:

On Tue, Dec 15, 2009 at 11:36 PM, ABBAS SHAKEEL
<shakeel.abbas.qau@...> wrote:
> Hello
> I am not an expert for database but rarely design database for Web and IVR
> applications.
> Today I wonder if there is any practical advantage of inheritance of Tables.
> Lets take the same example that is commonly referred.
>
> CREATE TABLE cities (
> name text,
> population float,
> altitude int -- in feet
> );
>
> CREATE TABLE capitals (
> state char(2)
> ) INHERITS (cities);
>
> Capital table inherit attributes from Cities but no data. I am confused with
> concept of inheritance in postgres.
>
> In These tables what i was supposing is inheritence as that in JAVA. but its
> seems to be a different scene.

The child table inherits structure from the parent, the parent then
allows you to retrieve data from it and the children together.

--
Best Regards
Shakeel Abbas

Re: Inheritance in Postgresql ?

2009-12-15T22:51:56Z

On Tue, Dec 15, 2009 at 11:36 PM, ABBAS SHAKEEL
<shakeel.abbas.qau@...> wrote:

> Hello
> I am not an expert for database but rarely design database for Web and IVR
> applications.
> Today I wonder if there is any practical advantage of inheritance of Tables.
> Lets take the same example that is commonly referred.
>
> CREATE TABLE cities (
> name text,
> population float,
> altitude int -- in feet
> );
>
> CREATE TABLE capitals (
> state char(2)
> ) INHERITS (cities);
>
> Capital table inherit attributes from Cities but no data. I am confused with
> concept of inheritance in postgres.
>
> In These tables what i was supposing is inheritence as that in JAVA. but its
> seems to be a different scene.

The child table inherits structure from the parent, the parent then
allows you to retrieve data from it and the children together.

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin

Inheritance in Postgresql ?

2009-12-15T22:36:48Z

Hello

I am not an expert for database but rarely design database for Web and IVR applications.

Today I wonder if there is any practical advantage of inheritance of Tables.

Lets take the same example that is commonly referred.

CREATE TABLE cities (
    name            text,
    population      float,
    altitude        int     -- in feet
);

CREATE TABLE capitals (
    state           char(2)
) INHERITS (cities);

Capital table inherit attributes from Cities but no data. I am confused with concept of inheritance in postgres.

In These tables what i was supposing is inheritence as that in JAVA. but its seems to be a different scene.

Can some one Please compare inheritence in JAVA and Postgresql OR make me clear about for what inheritance is introduced in Postgres.

--
Best Regards
Shakeel Abbas

Re: Internal fragmentations statistics Was: VACUUM FULL memory requirements

2009-12-15T21:18:45Z

2009/12/15 Guillaume Lelarge <guillaume@...>

Le mardi 15 décembre 2009 à 00:04:47, Evan Rempel a écrit :

> Is there a command/tool that will report on how FULL a table is getting?
> If there is, how intrusive is it? How computationally heavy is it?
>
> We have a database that is approx 100 million rows with
> approx 2 million insert/updates per day. Each day old data
> is purged from the database. The end result is a mostly static
> footprint with regards to disk space used, but I would like to
> know how much room is usable inside the tables as well as the
> OS file system (that part is easy).
>

pgstattuple contrib module give such an information. But it requires an
exclusive lock on the table it's looking at, so it's quite intrusive. For more
details, the 8.4 documentation is interesting:
http://www.postgresql.org/docs/8.4/interactive/pgstattuple.html

That doc specifically says that it takes only a read lock.

Best regards,
--
Lets call it Postgres

EnterpriseDB http://www.enterprisedb.com

gurjeet.singh@...

singh.gurjeet@{ gmail | hotmail | yahoo }.com
Twitter: singh_gurjeet
Skype: singh_gurjeet

Mail sent from my BlackLaptop device

Re: Looking for Good and Easy-to-use Reporting Tool

2009-12-15T12:38:53Z

"Su, Alan" <Alan.Su@...> wrote:

> 1. Require less programming effort. I am trying to stay away from
> having to write sophisticate programming language, such as Java or
> .Net
> 2. Allow to bundle SQL commands in store procedure like style.
> 3. Allow to evaluate conditions of SQL result set and trigger
> alert email based on a preset threshold.

Have you considered psql and grep, with maybe a little logic in bash?

-Kevin

--
Sent via pgsql-admin mailing list (pgsql-admin@...)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin