Why not use application specific multicast listener in appserver cluster

Hello Friends,
I am using oracle9iAs. we have basically two enterprise application. and often they communicate to each other, currently we have message bean sending a message with is nothing more than 2 integer values. The second application uses these two integers to update specific data cache.
First of All I am not aware or have come across any documentation on JMS clustering in Oralce9iAS. they still are very much behind Weblogic interms of feature implementation, If I wrong correct me here.
Now coming to my requirement, we are clustering application servers now and I need to be able to send messages to all
the servers in the cluster, lets say message generated by server 1 and application A to Application B on all other servers in cluster including the host.
well My dilemma is I want to use JMS but How to subscribe to different topics on cluster if topic is not only cluster able,.. if topics themselves running in clustered servers are not clustered.
so the easiest way for me to do this is to use a Multicast Listener on Application B on all the servers in the cluster and send message to it. well I know I will using socket here,... but I do also have control over transaction scope, this message is only sent upon successful completion of transaction on Application A.
now I ask my question. In this situation do things is there any risk that I am taking by using MulticastListner in my app?
your thoughts and ideas are appreciated
Thanks
Jay

Scrap JMS. Use plain java.net or JINI.
You could create a Peer-to-Peer Chain:
Server A -> Server B -> Server C -> Server A
On each Server have a server listening for messages (from the Server to the left in the chain) and a client that sends messages (to the Server to the right in the chain) and an application that links the server and client on each Server.
So,
Server A's "server" listens for messages from the "client" on Server C, and so on.
This is vary efficent. A simple send-receive chain that is clusterd within a small (or large) group of Servers. You could also try Bi-Directional Parallel Communications BDPC. That would basically have Server A send messages to both Servers - one to the left and one to the right, then at the same time, listen for messages from the Server on the left and the Server on the right. This is better suited for large-scale clusters where a Uni-directional communications model would take too long to propogate messages to every member of the group.
But, that's just me.

Similar Messages

Why not use interfaces for constants?

Hi,
I have been getting conflicting views about why not use interfaces for constants?
Can anyone shed some light on it?
Piyush

How so?
If the constants are only used in a single classhow
does it help to put them in a separate structute?
(Ignoring the a type safe enumeration of course.)Well, mainly for readability. Suppose that you have a
web application which has a servlet class
"RedirectServlet", which takes a variable to redirect
to a given JSP file. In the JSP files, you want to
create forms which take this servlet as their action.
You could of course put all these "redirect"
constants in the RedirectServlet itself, but then
it's kind of weird that you have to import a
servlet in your jsp file to include these constants
and to be able to write
<form action="servlet/RedirectServlet?key=<%=
RedirectServlet.HOMEPAGE %>> or so.
Instead, you could create an interface called
"RedirectConstants", and then your servlet
RedirectServlet could implement this interface. Then
your JSP can also access these constants via this
interface.
But then again, this is merely a matter of style, I
fully agree that there is never a real _need_ to put
constants in a separate structure if they're only used
in a single class.Are you claiming that your example above is a single class?

Why not use notifyall

now I'm reading this book "Java Threads 2nd edition" by Scott Oaks and Henry Wong. Writer provide a class called BusyFlag to simulate the mutex.
I wonder why not use notifyall() to replay notify().
public class BusyFlag {
     protected Thread busyflag = null;
     protected int busycount = 0;
     public synchronized void getBusyFlag() {
          while (tryGetBusyFlag() == false) {
               try {
                    wait();
               } catch (Exception e) {
     public synchronized boolean tryGetBusyFlag() {
          if (busyflag == null) {
               busyflag = Thread.currentThread();
               busycount = 1;
               return true;
          if (busyflag == Thread.currentThread()) {
               busycount++;
               return true;
          return false;
     public synchronized void freeBusyFlag() {
          if (getBusyFlagOwner() == Thread.currentThread()) {
               busycount--;
               if (busycount == 0) {
                    busyflag = null;
                    *notify();*
     public synchronized Thread getBusyFlagOwner() {
          return busyflag;
}

naficbm wrote:
now I'm reading this book "Java Threads 2nd edition" by Scott Oaks and Henry Wong. Writer provide a class called BusyFlag to simulate the mutex.
I wonder why not use notifyall() to replay notify().Looks like it should work with notifyall as well.

Why not use "new" operator with strings

why we not use new when declaring a String .because in java String are treated as objects. we use new operator for creating an object in java .
and same problem wiht array when we declare array as well as initialize .here we are alse not using new for array
why

Strings aren't just treated as objects, Strings are Objects.
As for why not using new for Strings, you can, if you want.:
String str = "this is a string";
and
String str = new String("this is a string");
do the same thing (nitty-gritty low level details about literals aside). Use whatever you like, but isn't it simpler not to type new String(...) since you still need to type the actual string?
As for arrays, you can use new:
int[] ints = new int[10];
or
int[] ints = { 0, 1, 2, 3, ..., 9 };
But the difference here is you are creating an empty array in the first one, the second creates and fills the array with the specified values. But which to you often depends on what you are doing.

Why not use paper label on finished DVD project?

I also use printed paper labels for my DVD covers
why not use them?
please advise?

Hi
The DVD get's unbalanced = Laser can't read track
DVD are several times more sencitive than CD (not advicable to put lables on them either)
And DVDs/CDs that are unbalanced also give more strain to the player mechanism
that eventually stops working.
Yours Bengt W

Why not use iPhone in car as active noise canceller.

The iPhone uses noise cancellation during conversations. Why not make that available in the car through the car audio to quiet the interior at all times?

I have one of those stereos bought from Ebay, no badge, no name, and with multitude of functions,and a mic attached to my visor. Once it is paired with Iphone, all my music and calls go through the car stereo.
for applications that do not direct Audio through BT such as CoPilot (GPS navigation) I connect the Iphone through headset socket and into the stereo AUX input and works great (all sounds, voice,calls) go through it. One disadvantage of this method is that I can't listen to the radio. Well you cannot have it all. Or perhaps there is a Radio application where it plays in the background while I still use my GPS application.

Why not use Jsp for control in MVC2 ?

hi
i am rajendra , i am preparing interview FAQ, so please give me my Que. Ans.
1) we are using servlet for control in MVC2 Module,then why we can not use
Jsp for control in mvc2
2) if we used jsp for control in mvc2 then what is disadvantage compare to if using servlet for control in mvc2
Please give me ans. as soon as..
Thanks.

Are you preparing a FAQ for others or are you just cramming before an upcoming interview and someone gave you a set of questions? It looks more like the latter to me.
If you haven't actually written any MVC based code, you won't be able to snow the interviewer. It's better to just relax, get a good night's sleep and impress them with your willingness to learn, your fit with the team and whatever real accomplishments you have to show.
If you really want an answer to this I suggest that you try writing an MVC2 system using JSP for the controller, which is possible, and then try doing the same in a servlet, or better yet using a framework like Spring MVC or Struts to provide the servlet for you and let you concentrate on your application.

Why not use Redo log for consistent read

Oracle 11.1.0.7:
This might be a stupid question.
As I understand if a select was issued at 7:00 AM and the data that select is going to read has changed at 7:10 AM even then Oracle will return the data that existed at 7:00 AM. And for this Oracle needs the data in Undo segments.
My question is since redo also has past and current information why can't redo logs be used to retreive that information? Why is undo required when redo already has all that information.

user628400 wrote:
Thanks. I get that piece but isn't it the same problem with UNDO? It's written as it expires and there is no guranteee until we specifically ask oracle to gurantee the UNDO retention? I guess I am trying to understand that UNDO was created for effeciency purposes so that there is less performance overhead as compared to reading and writing from redo.And this also you said,
>
If data was changed to 100 to 200 wouldn't both the values be there in redo logs. As I understand:
1. Insert row with value 100 at 7:00 AM and commit. 100 will be writen to redo log
2. update row to 200 at 8:00 AM and commit. 200 will be written to redo log
So in essence 100 and 200 both are there in the redo logs and if select was issued at 7:00 data can be read from redo log too. Please correct me if I am understanding it incorrectly.I guess you didnt understand the explaination that I did. Its not the old data that is kept. Its the changed vector of Undo that is kept which is useful to "recover" it when its gone but not useful as such for a select statement. Whereas in an Undo block, the actual value is kept. You must remember that its still a block only which can contain data just like your normal block which may contain a table like EMP. So its not 100,200 but the change vectors of these things which is useful to recover the transaction based on their SCN numbers and would be read in that order as well. And to read the data from Undo, its quite simple for oracle to do so using an Undo block as the transaction table which holds the entry for the transaction, knows where the old data is kept in the Undo Segment. You may have seen XIDSEQ, XIDUSN, XIDSLOT in the tranaction id which are nothing but the information that where the undo data is kept. And to read it, unlke redo, undo plays a good role.
About the expiry of Undo, you must know that only INACTIVE Undo extents are marked for expiry. The Active Extents which are having an ongoing tranaction records, are never marked for it. You can come back after a lifetime and if undo is there, your old data would be kept safe by oracle since its useful for the multiversioning. Undo Retention is to keep the old data after commit, something which you need not to do if you are on 11g and using Total Recall feature!
HTH
Aman....

Using Application - Specific Upgrade tools

Hi,
This is in refrence to point 3.25 of the ECC 6.0 SR1 upgrade document (Relevant text is below). Is this step really required to be performed? Can a Basis resource do it independently or will require ABAP/Functional help? If we ignore this step, how will it impact the upgraded system? If we have to run this step, at what stage we have to do it?
Any insights would be helpful.
Thanks & Regards,
Abdul
===Relevant text from Upgrd Doc===
Application-Specific Upgrade Toolbox
The application-specific upgrade toolbox (ASU toolbox) provides you with application-related upgrade guides and tools:
n Application-specific upgrade guides
These upgrade guides contain collections of known application-specific upgrade problems together with
the related SAP Notes, which saves you the time when searching for SAP Notes on SAP Service Marketplace.
n Automatic repair toolbox
This tool automatically starts application-specific repair and conversion programs.
n Additional repair toolbox
This program contains tools that assist you in adjusting customer data to the new release. Particularly
useful is the variant restore tool that can adjust the report variants to the new interfaces of the reports. In
the past, a lot of variants had to be created again, without giving you the option of displaying the old values.
For more information, see SAP Note 623723.
=== End of section ===

The ASU toolbox is a mixture of Basis and application.
The way you can go would be
- upgrade a test system
- let the modules have a look if they need parts from the toolbox
- upgrade production
- run the parts that are necessary
Since this is implementation dependent a general advice can´t be given. On our upgrade from 4.6c to 4.7 we needed only five of the given tools, your mileage may vary.
Markus

Apache VS Netscape Why Not Use Apache....

Is this correct?
          I should use Netscape as a proxy server to proxy jsp and servlet requests to
          a weblogic cluster because the libproxy.so module has special functionality
          in dealing with weblogic clusters.
          I should NOT use Apache because it does not handle the in-memory replication
          of Sessions to the weblogic cluster.
          Why shouldn't I use Apache in a clustered environment?
          Thanks
          Michael Schreiber


               Your assumption is correct. Currently there is no Apache plugin to
          handle session replication for a clustered environment in 4.51. You can
          use Apache's mod_proxy for simple proxying, but that will not handle
          failover or load balancing. An apache plugin for clustered setups will
          be available in Weblogic 5.0 (currently in public beta).
                              - Mike
          Michael Schreiber wrote:
          >
          > Is this correct?
          >
          > I should use Netscape as a proxy server to proxy jsp and servlet requests to
          > a weblogic cluster because the libproxy.so module has special functionality
          > in dealing with weblogic clusters.
          >
          > I should NOT use Apache because it does not handle the in-memory replication
          > of Sessions to the weblogic cluster.
          >
          > Why shouldn't I use Apache in a clustered environment?
          >
          > Thanks
          > Michael Schreiber

Xletview: why not use sun's javaTV source code directly ?

xletview : one emulator for viewing MHP Xlets on a PC;
I see many difference from sun's code about the part of javaTV.
Why the author Martin Sveden need to rewrite them and not use sun's javaTV source code directly ?
Thank you for your any answer.
Sorry for my horrible english.

XleView is a fine effort, but it has its limitations. I tried using it for the GunBunny demo available in the JavaME SDK 3.0 Eary Access available in
http://java.sun.com/javame/downloads/sdk30ea.jsp
and I encountered some of thost limitations. What Sveden does it several tricky manipulations in a classloader for the Xlet. In particular, he translates bytecode as he loads the Xlet's classes, changing some class names, such as changing
java.awt.Toolkit
tto
xjava.awt.Toolkit
This works up to a point. This version of Toolkit usually just turns around calls java.awt.Toolkit's corresponding routine, but he deliberately he does not implement the ubiquitous
public Image createImage(URL url)
When I downloaded his source and tried adding ths into his code, I then encountered at runtime
Uncaught error fetching image:
java.lang.NullPointerException
at sun.awt.image.URLImageSource.getConnection(URLImageSource.java:97)
at sun.awt.image.URLImageSource.getDecoder(URLImageSource.java:107)
at sun.awt.image.InputStreamImageSource.doFetch(InputStreamImageSource.java:240)
at sun.awt.image.ImageFetcher.fetchloop(ImageFetcher.java:172)
at sun.awt.image.ImageFetcher.run(ImageFetcher.java:136)
I think that this is some optimization within J2SE where they do not wait for the image to load and do the fetching in a separate thread. I do not know if there is some way to force the Sun's JVM to disable this optimization and I have not investigated furher, but I did email Svenden. I noticed that the current version of XleTView is 0.3.6 which is dated June 2004 and the web site http://www.xletview.org/ has had not news since 2005 .
I note that Svenden provides an elegant user interface but I advise you not try to learn the API's from his classes: he makes no effort to keep his implementation separate from the API standards so, for instance, his version of HScene is
package org.havi.ui;
public class HScene extends Container
implements HComponentOrdering, ImageObserver, MenuContainer, Serializable {
whereas the real standard is
public class HScene extends Container
implements HComponentOrdering {
I hate to critize such a fine effort, but it is best to keep an imlementation separate from the public API.
I agree that having a lightweight, simple and Java feature-complete Xlet viewer running would be very helpful, especially if BD-J gains much popularity. I have found trying to work with the vendor players described at
http://wiki.java.net/bin/view/Mobileandembedded/Blu-RayDiscHelloWorld
to be frustrating. I tried all four mentioned and I have gotten none of them to work for me yet. These are huge downloads and the "Intro Version" of Arcsoft pointed to at the SDK 3.0 Early Access page
http://www.arcsoft.com/products/totalmediatheatre/
just gives a "File not found" HTML page. This forum post mentions that perhaps it was recently taken down?
http://www.arcsoft.com/forum/forum_posts.asp?TID=1084
I have already emailed Michael Downs of Arcsoft since he is mentioned here:
http://wiki.java.net/bin/view/Mobileandembedded/BDJPCPlayers
What did work for me was to take Sun's JavaTV 1.1 Reference Implementation and PowerDVD's BDJ.jar and hack/fix a few classes to get GunBunny to work within the RI's RunXlet program. This is not trivial but I will will describe:
Download the JavaME 3.0 SDK EA and then download the correct version of PowerDVD that the SDK page points to:
http://www.brothersoft.com/powerdvd-download-50794.html
Download the RI binary and source of Java TV API 1.1
http://java.sun.com/javame/technology/javatv/index.jsp
Since I am too lazy to use anything but the latest 1.6 JRE as a runtime, I have to fix
com.sun.tv.media.util.MediaThread
to comment out its stop() method. This is becase stop() was deprecated and made final in J2SE 1.5
http://java.sun.com/j2se/1.5.0/docs/api/java/lang/Thread.html#stop()
http://java.sun.com/j2se/1.5.0/docs/guide/misc/threadPrimitiveDeprecation.html
Next, I had to mae a change to Sun's JavaTV implementation to keep the GunBunny demo happy by providing a ServiceContext. The RunXlet program calls SIEmulator, but that environment does not provide a ServiceContex, so I added one back in:
com.sun.tv.receiver.SIEmulator
private static ServiceContext svcctx = null;
Add before every call to new AppSignalEvent() and add the svcctx as the second-to-last argument:
if(svcctx == null){ try { svcctx = new com.sun.tv.ServiceContextImpl(); }catch (Exception e){ e.printStackTrace(); } }
I then had to hack a few PowerDVD classes, which is evil of me. First, I had to build an empty class by the name of
sun.util.BDJPlugin
becuase some PowerDVD class extends it but it is not in its BDJ.jar and then I had to reverse-compile (evil!) and modify one of their internal classes to avoid a dependency on their native methods. What I ended up with was a hacked version of com.cl.bdj.helper.CUtil to avoid the native methods which access the registry where they query for registry settings but also provide a default value (which I return).
com.cl.bdj.helper.CUtil
private static String pGetRegistryString(long i ,String string ,String string3) {
return string3;
private static int pGetRegistryInt(long i ,String string ,int j) {
return j;
The last thing I had to do was to change com.hdccookbook.gunbunny.BaseXlet, to use the JavaTV container rather than the org.havi one:
import javax.tv.graphics.TVContainer;
import java.awt.Container;
change
// protected HScene scene;
protected Container scene;
public final void run() {
// waitForPresenting();
// scene = HSceneFactory.getInstance().getDefaultHScene();
scene = TVContainer.getRootContainer(xletContext);
In order to having working keys, I had to add the KeyListener interface to BaseXet and
run(){
addKeyListener(this);
and then add a near duplicate of the org.dvb.event.UserEvent* stuff that is already there:
public void keyPressed(KeyEvent e){
switch(e.getKeyCode()){
case KeyEvent.VK_0:
case KeyEvent.VK_1:
case KeyEvent.VK_2:
case KeyEvent.VK_3:
case KeyEvent.VK_4:
case KeyEvent.VK_5:
case KeyEvent.VK_6:
case KeyEvent.VK_7:
case KeyEvent.VK_8:
case KeyEvent.VK_9:
numberKeyPressed(e.getKeyCode() - KeyEvent.VK_0);
break;
case KeyEvent.VK_ENTER:
enterKeyPressed();
break;
case KeyEvent.VK_LEFT:
arrowLeftKeyPressed();
break;
case KeyEvent.VK_RIGHT:
arrowRightPressed();
break;
case KeyEvent.VK_UP:
arrowUpPressed();
break;
case KeyEvent.VK_DOWN:
arrowDownPressed();
break;
public void keyReleased(KeyEvent e){
// System.out.println("Released: e="+e+" code="+e.getKeyCode());
public void keyTyped(KeyEvent e) {
// System.out.println("Typed: e="+e+" code="+e.getKeyCode());
It then comes up and functions, but the rendering is not very good. Again, this is complete hack. Running the Java TV 1.1 RI in a JRE 1.6 is not supported. If you try to run any of the other examples, you will discover that the XML parsing is broken in JRE 1.6 with this message:
Parsing failed: com.sun.org.apache.xerces.internal.impl.io.MalformedByteSequence
Exception: Invalid byte 1 of 1-byte UTF-8 sequence., file: lib\JavaTVSampleFile01.xml
and you have to go and change the lines in
lib\JavaTV.properties
from
ServiceFileHandler=com.sun.tv.receiver.ReceiverFile
# ServiceFileHandler=SampleData_01
to
# ServiceFileHandler=com.sun.tv.receiver.ReceiverFile
ServiceFileHandler=SampleData_01
and build samples\db\SampleData_01.java and put it in your classpath along with all these other hacks.
Nonetheless, if you want to develop a somewhat generic Xlet that uses some BD-J features, that is how you might do it. But the moment you exercise anything else in the vendor's BDJ.jar that simply doe snot work outside of their environment or invokes one of their native methods, you are again stuck.
It would be nice if Sun would at least bother to update the JavaTV RI with the fixes I hae mentioned. Their "jmflite" implementation does not render perfecly but at least it would give the programmer something to work with without having to deal with an older JRE, etc.
Andrew
Edited by: AndrewMorrow on Oct 20, 2008 2:48 AM

Why not use Fourier transforms for filtering?

Are there any resources for constructing filters using discrete Fourier transforms (DFTs)? Or is it as easy as it seems. For example, for lowpass filtering I obtain the DFT of my signal, zero out components above the selected cutoff frequency and then transform back to get my signal minus the high frequency (noise) components. It seems to behave correctly for the signal I've applied it to, but am I missing something. Upsides would appear to be (1) no time delay between the input and output and (2) the filter is sharp with no transition band. Downside is that this method is slower than using impulse filters, but are there other upsides, downsides, or considerations I'm missing?

The technique of filtering using frequency domain truncation and IFFT is used in specific areas where computation time is essential (to avoid the time domain convolution with a long impulse response), like image processing. However the technique also has some downsides you should be aware of.
The frequency domain filtering performed on a data record is totally equivalent to a circular convolution made on your time signal with the impulse response of your ideal low-pass filter (typically a sin(x)/x function). The fact what we are doing a circular convolution will result in unwanted artifacts and you'll see some "parts" of the beginning of your signal at the end of the filtered signal and vice-versa.
Try for example to create a short signal (like a sharp pul
se) and extend it with zeros. After filtering you will see a sort of "pre-shoot" of your original signal at the very end of your signal. If you then rotate your time signal you'll discover that the signal is actually continuous (in a circular way).
You can reduce the problem by, for example, adding "some" zeros at the beginning and at the end of your signal (zero-pad) before your filtering operation (before the FFT) and then remove these additional samples again from you filtered signal (after the IFFT).

Why not use DateFormat?

I've noticed a few posts recently that deal with how to calculate the differences between dates (example, example), and it seems like folks are shying away from DateFormat to parse dates for these calculations. Incidentally, the OPs are running into month offset problems as well when trying to calculate using Calendars.
DateFormat seems much more simple to me than all the other parsing rigmarole, but I wonder if I'm missing something. Is there a good reason to not use this technique in the listed examples, or can it be chalked up to unfamiliarity with the API?

In "Literate Programming," Donald Knuth wrote "We
should forget about small efficiencies, about 97% of
the time. Premature optimization is the root of all
evil." That'll garner the remaining Dukes. Excellent words to program by, even (especially) for newbies such as I.

Why jdbc api ?why not use odbc api from java application?

hi guys....am pretty confused abt these jdbc drivers...am unable to understand why we r using jdbc api when odbc api has been in market prior to jdbc.....please clarify my doubt

chari wrote:
hi guys....am pretty confused abt these jdbc drivers...am unable to understand why we r using jdbc api when odbc api has been in market prior to jdbc.....please clarify my doubtbecause jdbc is the way that java does it.
use it or don't.
C++ was in the market before java - why was java developed at all? cobol was in the market before C++ - why was C++ developed at all?
%

Why not use the MS works calendar?

I'm a little confused as to why I can sync my contacts with MS Works, but I cant sync my calendar with MS Works. I understand the iphone doesnt support this, but why? Life would have been so much simpler if I had both my contacts and calendar staring me in the face through one program. I only have outlook express on my computer, and that does not have a calendar to go along with it. If I have to purchase a program to be able to sync both contacts and calendar, which do you suggest? Is outlook my only choice? I'm a recent college grad who just entered the professional world and more then 50% of my work deals with meeting clients. The calendar feature is a biggy for me. So what do you folks suggest.
Thanks in advance,
Dave

You don't have many options.
You must get Outlook - especially if a calendar feature is a biggy for you. With Outlook, you have access to email accounts, address book contacts and calendar events with a single application and is probably what 90% of Windows users use in an enterprise/business environment.
Syncing contacts or calendars to your iPhone requires that the latest version of iTunes is installed.
*Contacts must be stored in one of the following compatible contact managers:*
Outlook Express (Windows XP only)
Windows Contacts (Windows Vista only)
Outlook 2003
Outlook 2007
Yahoo!
*Calendars must be stored in one of the following compatible calendar applications:*
Outlook 2003
Outlook 2007

Why not use application specific multicast listener in appserver cluster

Similar Messages

Maybe you are looking for