From oracle-l-bounce@freelists.org Mon Jun 14 10:29:00 2004 Return-Path: Received: from air189.startdedicated.com (root@localhost) by orafaq.com (8.11.6/8.11.6) with ESMTP id i5EFSTK02943 for ; Mon, 14 Jun 2004 10:28:39 -0500 X-ClientAddr: 206.53.239.180 Received: from turing.freelists.org (freelists-180.iquest.net [206.53.239.180]) by air189.startdedicated.com (8.11.6/8.11.6) with ESMTP id i5EFSC602820 for ; Mon, 14 Jun 2004 10:28:22 -0500 Received: from localhost (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 4CA9072D085; Mon, 14 Jun 2004 10:12:12 -0500 (EST) Received: from turing.freelists.org ([127.0.0.1]) by localhost (turing [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 06830-20; Mon, 14 Jun 2004 10:12:11 -0500 (EST) Received: from turing (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 7D54C72D0AB; Mon, 14 Jun 2004 10:11:33 -0500 (EST) Received: with ECARTIS (v1.0.0; list oracle-l); Mon, 14 Jun 2004 10:10:05 -0500 (EST) X-Original-To: oracle-l@freelists.org Delivered-To: oracle-l@freelists.org Received: from localhost (localhost [127.0.0.1]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id 2343272CD6A; Mon, 14 Jun 2004 10:10:04 -0500 (EST) Received: from turing.freelists.org ([127.0.0.1]) by localhost (turing [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 05981-46; Mon, 14 Jun 2004 10:10:03 -0500 (EST) Received: from ross.alise.lv (ross.alise.lv [194.19.227.242]) by turing.freelists.org (Avenir Technologies Mail Multiplex) with ESMTP id D28A872D040; Mon, 14 Jun 2004 10:09:44 -0500 (EST) In-Reply-To: <0186754BC82DD511B5C600B0D0AAC4D607B007C9@EXCHMN3> To: oracle-l@freelists.org Cc: "'oracle-l@freelists.org'" , oracle-l-bounce@freelists.org Subject: RE: OC4J Process Architecture MIME-Version: 1.0 X-Mailer: Lotus Notes Release 6.5.1 January 21, 2004 Message-ID: From: J.Velikanovs@alise.lv Date: Mon, 14 Jun 2004 18:22:53 +0300 X-MIMETrack: Serialize by Router on ROSS/IT ALISE/LV(Release 5.0.11 |July 24, 2002) at 2004.06.14 18:22:55, Serialize complete at 2004.06.14 18:22:55 Content-Type: multipart/alternative; boundary="=_alternative 005521BAC2256EB3_=" X-Virus-Scanned: by amavisd-new at freelists.org X-archive-position: 2677 X-ecartis-version: Ecartis v1.0.0 Sender: oracle-l-bounce@freelists.org Errors-To: oracle-l-bounce@freelists.org X-original-sender: J.Velikanovs@alise.lv Precedence: normal Reply-To: oracle-l@freelists.org X-list: oracle-l X-Virus-Scanned: by amavisd-new at freelists.org --=_alternative 005521BAC2256EB3_= Content-Type: text/plain; charset="windows-1257" Content-Transfer-Encoding: quoted-printable DENNIS, Thank you for good words. I=92ll take them in mind. >> execute your Java programs on a test system >> then you need to use more in-depth methods, rather than persisting in=20 simplistic solutions >> Usually their approach is to tell the DBA to "fix that" We have here, just finished J2EE application in production. ~20-40 modules. Application written by 60-100 developers during 1 year. There are ~100 concurrent users works with different applications modules=20 during working day. Quite difficult to emulate situations we have in productions (performance, = deadlock etc.) in test environment using one user connection ;). We are=20 using Rational Robot for stress testing. But there still situations in=20 productions that we are not able to reproduce in test environment. >> then you need to use more in-depth methods, rather than persisting in=20 simplistic solutions I wouldn=92t say we are fear from depth methods. I am ready to. But it seam= s=20 the problem solver need to have dip knowledge about application code for=20 successfully fighting with problems. DBA nowadays have many=20 responsibilities and looks like dipper methods will take more time from=20 daily duties. >> Usually their approach is to tell the DBA to "fix that" It look like you confirm my words.=20 DBA need to work with OC4J as with BLACK BOX.=20 =93Developers telling to fix=94 and DBA fixing=20 if =93if the problems don't to go away =93=20 then call Developers for next loop and wait for next solution, without=20 knowledge that happening inside of OC4J. Jurijs 9268222 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D http://otn.oracle.com/ocm/jvelikanovs.html DENNIS WILLIAMS Sent by: oracle-l-bounce@freelists.org 14.06.2004 17:30 Please respond to oracle-l =20 To: "'oracle-l@freelists.org'" cc:=20 Subject: RE: OC4J Process Architecture Jurijs You are correct that it would be great if someone documented=20 everything you could ever want to know about OC4J. Since you don't seem to have that, there are a couple of other approaches you can take. Here are some ideas=20 for you: - You can execute your Java programs on a test system without other=20 users and get some good traces to use with Cary's methods. If that doesn't=20 expose the problems sufficiently, then you at least have gained some experience with the types of traces you will see so that if you decide you must turn=20 on tracing on the production system you won't be totally bewildered. - While it is nice to have the "heavy artillery" for solving problems, bear in mind that most problems are of a simpler nature and can be=20 resolved without the more in-depth methods. Just keep in mind that if the problems don't seem to go away when you apply simple solutions, then you need to=20 use more in-depth methods, rather than persisting in simplistic solutions. - In my experience, few Java programmers are concerned with database issues. Usually their approach is to tell the DBA to "fix that". I think=20 the DBA needs to learn something about Java and how it tends to interact with the database. Particularly if you use an Oracle product like OC4J, this=20 will be viewed as a DBA issue. =20 The key point is that in the end the performance problems must be=20 understood so the organization can take steps to correct them. How to find the=20 problems may require some study and ingenuity on your part. Don't give up! Dennis Williams DBA Lifetouch, Inc. dwilliams@lifetouch.com=20 -----Original Message----- From: oracle-l-bounce@freelists.org=20 [mailto:oracle-l-bounce@freelists.org]On Behalf Of J.Velikanovs@alise.lv Sent: Monday, June 14, 2004 9:04 AM To: Leslie Tierstein Cc: dbvision@optusnet.com.au; oracle-l@freelists.org Subject: RE: OC4J Process Architecture Tank you for responce,=20 Unfortunately I already have book you mention (9i JDeveloper Handbook) ;( = There a little bit more information, but not so match, as I would like to ;(.=20 As I wrote, I will be indefinitely happy if it will be the information=20 about OC4J like about RDBMS.=20 Cary Millsap and others Oracle experts have written excellent books about Oracle wait interface and how to face performance issues. Unfortunately as far as I know, there no possibility to use R-Method or other response-time based performance-tuning method without written own code for debugging purposes.=20 I am just DBA and you know, I have no big expense with Java programming.=20 At the moment it looks like the good OC4J administrator will be Java top developer who perfectly know Java world. ;(=20 By my expense Looks like traditional DBA is unpowerful in many OC4J performance tuning situations.=20 Jurijs 9268222 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D http://otn.oracle.com/ocm/jvelikanovs.html "Leslie Tierstein" =20 14.06.2004 16:14=20 =20 To: =20 cc: , =20 Subject: RE: OC4J Process Architecture The tinyurl included points to the 9i JDeveloper Handbook, which is for=20 the last version of the product. The JDeveloper 10g Handbook is also listed on Amazon, but isn't published yet. The publication date is early this fall, say, September. However, the 10g book doesn't contain much information about the internals of OC4J. (I'm one of the technical editors.) But the authors, I know, have encountered similar problems. You might try contacting Paul Dorsey=20 directly (see his company's web site at http://www.dulcian.com) with specific questions. He is usually more than willing to offer advice. There is another book on Amazon about JDeveloper: Oracle JDeveloper 10G: Empowering J2EE Development One of the co-authors is Roel Stallman, who is the major Oracle evangelist (that may be his official job title) for JDeveloper. This book may contain more on the internals and architecture, although I haven't looked at it. Leslie Leslie Tierstein Senior Consultant Vision Chain, Inc. The first software to power the demand data network phone: 202-261-3549 -----Original Message----- From: oracle-l-bounce@freelists.org [mailto:oracle-l-bounce@freelists.org]On Behalf Of Nuno Souto Sent: Monday, June 14, 2004 6:58 AM To: oracle-l@freelists.org Subject: Re: OC4J Process Architecture ------- Original Message -------- Subject: Re: OC4J Process Architecture J.Velikanovs@alise.lv apparently said,on my timestamp of 14/06/2004 3:05=20 AM: > > Hope this information will help you. Yes it does. Thanks a lot for posting it, there is very little information about this library. There is a book out from Oracle Press, I think, that deals with JDeveloper in detail. Maybe you'd find some additional info there? http://tinyurl.com/234c6 will take you to it. HTH. -- Cheers Nuno Souto in sunny Sydney, Australia dbvision@optusnet.com.au ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@freelists.org put 'unsubscribe' in the subject line. -- Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html ----------------------------------------------------------------- --=_alternative 005521BAC2256EB3_= Content-Type: text/html; charset="windows-1257" Content-Transfer-Encoding: quoted-printable
DENNIS,

Thank you for good words.
I=92ll take them in mind.
>> execute your Java programs on a test system
>> then you need to use more i= n-depth methods, rather than persisting in simplistic solutions
>> Usually their approach is to tell the DBA to "fix that"

We have here, just finished J2EE app= lication in production.
~20-40 modules. Application written by 60-100 developers during 1 year.
There are ~100 concurrent users works with different applications modules during working day.
Quite difficult to emulate situations we have in productions (performance, deadlock etc.) in test environment using one user connection ;). We are using Rational Robot for stress testin= g. But there still situations in productions that we are not able to reproduce in test environment.

>> then you need to use more i= n-depth methods, rather than persisting in simplistic solutions
I wouldn=92t say we are fear from de= pth methods. I am ready to. But it seams the problem solver need to have dip knowledge about application code for successfully fighting with problems. DBA nowadays have many responsibilities and looks like dipper methods will take more time from daily duties.

>> Usually their approach is to tell the DBA to "fix that"
It look like you confirm my words. <= /font>
DBA need to work with OC4J as with B= LACK BOX.
=93Developers telling to fix=94 and = DBA fixing
if =93if the problems don't to go aw= ay =93
then call Developers for next loop a= nd wait for next solution, without knowledge that happening inside of OC4J.


Jurijs
9268222
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
http://otn.oracle.com/ocm/jvelikanovs.html




DENNIS WILLIAMS <DWILLIAMS@LIF= ETOUCH.COM>
Sent by: oracle-l-bounce@freelists.o= rg

14.06.2004 17:30
Please respond to oracle-l

       
        To:        "'oracle-l@freelists.org'" <oracle-l@freelists.org>
        cc:        
        Subject:        RE: OC4J Process Architecture



Jurijs
    You are correct that it would be great if someone documented everything
you could ever want to know about OC4J. Since you don't seem to have that,<= br> there are a couple of other approaches you can take. Here are some ideas for
you:
  - You can execute your Java programs on a test system without other users
and get some good traces to use with Cary's methods. If that doesn't expose=
the problems sufficiently, then you at least have gained some experience
with the types of traces you will see so that if you decide you must turn on
tracing on the production system you won't be totally bewildered.
  - While it is nice to have the "heavy artillery" for solving problems,
bear in mind that most problems are of a simpler nature and can be resolved=
without the more in-depth methods. Just keep in mind that if the problems don't seem to go away when you apply simple solutions, then you need to use
more in-depth methods, rather than persisting in simplistic solutions.
  - In my experience, few Java programmers are concerned with databas= e
issues. Usually their approach is to tell the DBA to "fix that". I think the
DBA needs to learn something about Java and how it tends to interact with the database. Particularly if you use an Oracle product like OC4J, this will
be viewed as a DBA issue.

The key point is that in the end the performance problems must be understoo= d
so the organization can take steps to correct them. How to find the problem= s
may require some study and ingenuity on your part. Don't give up!



Dennis Williams
DBA
Lifetouch, Inc.
dwilliams@lifetouch.com

-----Original Message-----
From: oracle-l-bounce@freelists.org [mailto:oracle-l-bounce@freelists.org]O= n
Behalf Of J.Velikanovs@alise.lv
Sent: Monday, June 14, 2004 9:04 AM
To: Leslie Tierstein
Cc: dbvision@optusnet.com.au; oracle-l@freelists.org
Subject: RE: OC4J Process Architecture



Tank you for responce,

Unfortunately I already have book you mention  (9i JDeveloper Handbook) ;(
There a little bit more information, but not so match, as I would like to
;(.
As I wrote, I will be indefinitely happy if it will be the information about
OC4J like about RDBMS.
Cary Millsap and others Oracle experts have written excellent books about Oracle wait interface and how to face performance issues. Unfortunately as
far as I know, there no possibility to use R-Method or other response-time<= br> based performance-tuning method without written own code for debugging
purposes.
I am just DBA and you know, I have no big expense with Java programming.
At the moment it looks like the good OC4J administrator will be Java top
developer who perfectly know Java world. ;(

By my expense Looks like traditional DBA is unpowerful in many OC4J
performance tuning situations.

Jurijs
9268222
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
http://otn.oracle.com/ocm/jvelikanovs.html





                "Leslie Tierstein" <Leslie.Tierstein@visionchain.com>


14.06.2004 16:14

       
       To:        <oracle-l@fre= elists.org>
       cc:        <dbvision@opt= usnet.com.au>, <J.Velikanovs@alise.lv>
       Subject:        RE: OC4J Process Architecture



The tinyurl included points to the 9i JDeveloper Handbook, which is for the
last version of the product. The JDeveloper 10g Handbook is also listed on
Amazon, but isn't published yet. The publication date is early this fall, say, September.

However, the 10g book doesn't contain much information about the internals<= br> of OC4J. (I'm one of the technical editors.) But the authors, I know, have<= br> encountered similar problems. You might try contacting Paul Dorsey directly=
(see his company's web site at http://www.dulcian.com) with specific
questions. He is usually more than willing to offer advice.

There is another book on Amazon about JDeveloper:
Oracle JDeveloper 10G: Empowering J2EE Development

One of the co-authors is Roel Stallman, who is the major Oracle evangelist<= br> (that may be his official job title) for JDeveloper. This book may contain<= br> more on the internals and architecture, although I haven't looked at it.

Leslie
Leslie Tierstein
Senior Consultant
Vision Chain, Inc.
The first software to power the demand data network
phone: 202-261-3549




-----Original Message-----
From: oracle-l-bounce@freelists.org
[mailto:oracle-l-bounce@freelists.org]On Behalf Of Nuno Souto
Sent: Monday, June 14, 2004 6:58 AM
To: oracle-l@freelists.org
Subject: Re: OC4J Process Architecture


------- Original Message --------
Subject: Re: OC4J Process Architecture
J.Velikanovs@alise.lv apparently said,on my timestamp of 14/06/2004 3:05 AM:

>
> Hope this information will help you.

Yes it does.  Thanks a lot for posting it, there is
very little information about this library.

There is a book out from Oracle Press, I think, that
deals with JDeveloper in detail.  Maybe you'd find some
additional info there?
http://tinyurl.com/234c6
will take you to it.
HTH.


--
Cheers
Nuno Souto
in sunny Sydney, Australia
dbvision@optusnet.com.au




----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to:  oracle-l-request@freelists.org
put 'unsubscribe' in the subject line.
--
Archives are at http://www.freelists.org/archives/oracle-l/
FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------

--=_alternative 005521BAC2256EB3_=-- ---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@freelists.org put 'unsubscribe' in the subject line. -- Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------