Dear Andrew 

Your right, the postgresql.conf and the Server OS kernel parameters are
not tuned. After tuning the replication working fine.
Thank you for your tech-help.

Regards,
G.Muniswamy

-----Original Message-----
From: Andrew Hammond [mailto:[EMAIL PROTECTED] 
Sent: Wednesday, March 07, 2007 6:20 AM
To: G Muniswamy
Cc: Andreas Kostyrka; [email protected]
Subject: Re: [Slony1-general] Slave system becomes Slow

It sounds like you haven't done much work tuning your postgresql.conf
file. If that is the case, then I suggest you read the following and
follow the steps they recommend.

http://revsys.com/writings/postgresql-performance.html
http://www.postgresql.org/docs/current/static/performance-tips.html
http://www.powerpostgresql.com/Docs

Finally, if you still have performance issues, please try the
pgsql-performance mailing list.
http://archives.postgresql.org/pgsql-performance/

Once you have the basic performance tuning ironed out, it might make
sense to come back to this list if you have slony specific performance
questions.

Andrew


On 2/27/07, G Muniswamy <[EMAIL PROTECTED]> wrote:
> Dear Andreas
>
> We have around 15-20 big table tables, each has around more than 2000
> rows in that tables.We need to set any parameter in postgresql.conf
file
> or
> Parameter needs to set for the slony.
>
>
> Best regards,
>
> g.Muniswamy
>
>
> -----Original Message-----
> From: Andreas Kostyrka [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, February 27, 2007 3:38 PM
> To: G Muniswamy
> Cc: [email protected]
> Subject: Re: [Slony1-general] Slave system becomes Slow
>
> Just out of curiosity, how many rows are touched by the updates on the
> master?
>
> One important thing is:
>
> select count(*) from tablea => 1000
> update tablea set fieldb=fieldb + 1 => 1 update statement on master,
>                                     => BUT 1000 updates on slave
>
> So it's easy to overwhelm a slony system with one statement that
> touches many rows.
>
> Andreas
>
> --
> This message has been scanned for viruses and
> dangerous content by OpenProtect(http://www.openprotect.com), and is
> believed to be clean.
>
>
> --
> This message has been scanned for viruses and
> dangerous content by OpenProtect(http://www.openprotect.com), and is
> believed to be clean.
>
> _______________________________________________
> Slony1-general mailing list
> [email protected]
> http://gborg.postgresql.org/mailman/listinfo/slony1-general
>

-- 
This message has been scanned for viruses and
dangerous content by OpenProtect(http://www.openprotect.com), and is
believed to be clean.


-- 
This message has been scanned for viruses and
dangerous content by OpenProtect(http://www.openprotect.com), and is
believed to be clean.

_______________________________________________
Slony1-general mailing list
[email protected]
http://gborg.postgresql.org/mailman/listinfo/slony1-general

Reply via email to