Xref: alice comp.databases.oracle.server:60837
Path: alice!news-feed.fnsi.net!newsfeed.icl.net!ayres.ftech.net!news.ftech.net!newshub.bart.net!newsfeed.iae.nl!news2.euro.net!news3.euro.net!usenet
From: John van der Steen <steenjoh@iquip.nl>
Newsgroups: comp.databases.oracle.server
Subject: Re: Tuning in Oracle
Date: 8 Aug 1999 22:08:37 GMT
Organization: EuroNet Internet
Lines: 53
Message-ID: <37ADFF48.7488A683@iquip.nl>
References: <37AB469A.3D06901@mail.tju.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.6 [en] (WinNT; I)
X-Accept-Language: en

Sanjeev,

The high number of messages are just messages from sqlplus or another program to
say that the process needs more data or to say that all has gone right.

You should look at the more data to client value if that is quite high and you
are running SQL*NET V2 you might want to adjust your SDU size.

SQL*NET V2 has a SDU-size of 2K it can be adjusted to 4-8-16 or 32 K
In SQLNET V1 this was no issue but in V2 it is standard 2K

If you want to know how to adjust the SDU-size let me know.

Kind Regards,

John van der Steen

Sanjeev Jain wrote:

> Hi eveyone,
>
> I was trying to tune some slow running processes. I ran utlbstat and
> utlestat and am looking at wait stats. I have couple of questions here.
>
> 1) Why is there a big number of "SQL*Net messages from client" even when
> the process is running on the same server?
> 2) If stats look like
>
> Event Name                                    Count            Total
> Time    Avg Time
> ----------------------------------------------------------------------------
>
> SQL*Net message from client        2689            1727815        642.55
>
> SQL*Net message to client            2691
> 33                    0.01
>
> OR
>
> SQL*Net message from client        288868        423307            1.47
> SQL*Net message to client            288868
> 134                    0
>
> Does it mean in the first case SQL*Net is slow and I need to do
> something? Is it slow in second case also?
>
> These stats were taken on different servers and for different processes.
>
> Any advice/ explanation is welcome.
>
> Thanks.
> Sanjeev