Thread Spawn

          Hi,
          What is the best way to spawn a thread when the servlet context starts and terminates
          when the context stops or ends in Weblogic 6.1? At the moment, I use a startup
          on load servlet to kick off a thread to start my background process. This process
          does a JDBC SELECT, performs some calculations and finally, JDBC INSERT or UPDATE
          resutls into a few tables. The problem I'm faing now is that when the context
          is stopped, the thread still running. Pls. advise. Thank you.
          /lim/

          I have set a flag to end the thread since interrupt() is deprecated. However, destroy()
          method is not called when I unload my webapp thru the admin console. Therefore,
          the thread still running after I have unload my webapp.
          "Perianayagam.T" <[email protected]> wrote:
          >
          >Iam not getting what u mean by stopping the context...
          >But what I suggest you is in the destroy() method of the servlet (which
          >is bascially
          >the destructor of the servlet ) you can interrupt (thread.interrupt()
          > )the thread
          >which was spawned initially...
          >
          >pls let me know if my understanding was correct..
          >Ref:http://java.sun.com/j2ee/sdk_1.3/techdocs/api/javax/servlet/Servlet.html#destroy()
          >Thanks
          >Perianayagam.T
          >
          >
          >"TH Lim" <[email protected]> wrote:
          >>
          >>Hi,
          >>
          >>What is the best way to spawn a thread when the servlet context starts
          >>and terminates
          >>when the context stops or ends in Weblogic 6.1? At the moment, I use
          >>a startup
          >>on load servlet to kick off a thread to start my background process.
          >>This process
          >>does a JDBC SELECT, performs some calculations and finally, JDBC INSERT
          >>or UPDATE
          >>resutls into a few tables. The problem I'm faing now is that when the
          >>context
          >>is stopped, the thread still running. Pls. advise. Thank you.
          >>
          >>/lim/
          >

Similar Messages

Java Logger incrementing thread ids for a thread spawnned via JNI

Hi all,
I have a Java Logger object that I am passing to a JNI dll written in C++.
If I spawn a thread from the JNI dll and log from that thread, the thread id in the log file increments for each log entry.
To test this I wrote an infinite loop on the thread to make sure that the logging originated from a single thread. The thread id keeps incrementing in the log and the memory usage of the process increases too. However, the thread count in Task Manager does not increase.
I will let the process run overnight to determine if the thread id will overflow or if an out of memory exception occurs.
Has anyone else seen this behavior? Does anyone know what I might be doing wrong? I'm using jre 1.5.1.
Thanks,
William

(1)
has anybody ever tried to use a native library from
JNI, when the library (Windows DLL) is not thread safe?
Now we want many Java clients.
That would mean each client makes its calls
to the library in its own thread. Because the library
is not thread safe, this would cause problems.Right. And therefore you have to encapsulate the DLL behind a properly synchronized interface class.
Now the details of how you have to do that depends: (a) does the DLL contain state information other than TLS? (b) do you know which methods are not thread-safe?
Depending on (a), (b) two extremes are both possible:
One extreme would be to get an instance of the interface to the DLL from a factory method you'll have to write, where the factory method will block until it can give you "the DLL". Every client thread would obtain "the DLL", then use it, then release it. That would make the whole thing a "client-driven" "dedicated" server. If a client forgets to release the DLL, everybody else is going to be locked out. :-(
The other extreme would be just to mirror the DLL methods, and mark the relevant ones as synchronized. That should be doable if (a) is false, and (b) is true.
(2)
Now we discussed to load the library several times -
separately for each client (for each thread).
Is this possible at all? How can we do that?
And do you think we can solve the problem in this
way?The DLL is going to be mapped into the process address space on first usage. More Java threads just means adding more references to the same DLL instance.
That would not result in thread-safe behavior.

Servlet with Multiple Threads spawned automatically

Hi,
I have a problem involving threads in servlet as follows:
Scenario: I am working on developing a web application using Tomcat and Netbeans IDE. I have a servlet which gets values from the session object and then constructs a server-side object and invoke a method on the same. This method on the server-side object executes an INSERT query in the mysql database.
Problem Description: The INSERT query is invoked twice => the method on the server-side object is invoked twice, and 2 records with different keys are created in the database table.
Observation: When I used the debugger in the IDE, I discovered that there are 2 different threads that are getting spawned (which I did not design/code for) and these threads call the insert query on the table twice.
I am not sure why this is happening. Can someone please throw some light on this?
Thank you in advance.

The servlet has a number of multiple if-else blocks. In the last if-else block of the servlet code, I am getting some attributes from the session object, creating the server class object and setting the attributes of the object.
The 2-thread problem is occurring only when the the last block is reached. The 1st thread creates the server class and the 2nd thread executes the rest of the lines in the servlet and completes the INSERT query. This is followed by the 1st thread to insert a new record again.

Ejb lookups fail from threads spawned from servlet

Hello,
We have a servlet from which we are spawning a thread. In the thread we are
trying to do a JNDI lookup for the database resource / EJBs. It does not
work as if it is not able to find the context t do the lookup.
I thought that the threads are spawned in the same JVM and context, so why
does the lookup not work.
Please need to figure out a solution to this problem, we need to call some
business logic in the EJBS from the thread and we do not want to do RMI-IIOP
lookups becuase they could be slower.
Appreciate your help in advance.
Regards,
Sakib

Custom threading is not supported. I think a workaround was posted on
this alias (search for custom threading), where you can call the
proprietary API's of the container to recreate the context, but I'm
pretty sure that it's not a supported or recommended solution.
The better (and more scalable) solution is to avoid spawning your own
threads.
David
Sakib Mehasanewala wrote:
Hello,
We have a servlet from which we are spawning a thread. In the thread we are
trying to do a JNDI lookup for the database resource / EJBs. It does not
work as if it is not able to find the context t do the lookup.
I thought that the threads are spawned in the same JVM and context, so why
does the lookup not work.
Please need to figure out a solution to this problem, we need to call some
business logic in the EJBS from the thread and we do not want to do RMI-IIOP
lookups becuase they could be slower.
Appreciate your help in advance.
Regards,
Sakib

Threads spawning in Servlets allowed?

Hello to all servlet afficionados,
          I am following thread discussions already a while and there's always
          discussion whether to user threads or not (I am not talking about a
          servlet thread but about threads manually instantiated for other
          reasons like asynchroneous handling of tasks for loose coupling or for
          long running purposes).
          So here's my question:
          Is it definitely allowed to spawn threads from a servlet (and even put
          the reference into the application context to allow others to later on
          access that thread)?
          Does the jdk specify on this?
          Or Is it even specific for each servlet engine? What is BEA saying
          about this?
          And even further: Is this different on application servers that
          include an ejb-container? Again, does the specification say something
          about this? What does BEA say about this?
          Thanks for trying to find out the real truth :-)
          regards
          Stefan


Stefan,
          There is nothing in the Servlet spec that specifically prohibits the use
          of threads within a servlet, but most will tell you that you should avoid it
          if possible. Having said that, I've seen quite a few implimentations that
          use background threads for tasks such as log queue processing and have not
          really had any issues. To me it's a matter of degree. As long as your
          architecture makes sense and you are careful to avoid potential deadlocks,
          you should be ok. FWIW, BEA recommends not creating threads within a web
          application.
          -chris
          "Stefan Hoehn" <[email protected]> wrote in message
          news:[email protected]...
          > Hello to all servlet afficionados,
          >
          > I am following thread discussions already a while and there's always
          > discussion whether to user threads or not (I am not talking about a
          > servlet thread but about threads manually instantiated for other
          > reasons like asynchroneous handling of tasks for loose coupling or for
          > long running purposes).
          >
          > So here's my question:
          >
          > Is it definitely allowed to spawn threads from a servlet (and even put
          > the reference into the application context to allow others to later on
          > access that thread)?
          >
          > Does the jdk specify on this?
          >
          > Or Is it even specific for each servlet engine? What is BEA saying
          > about this?
          >
          > And even further: Is this different on application servers that
          > include an ejb-container? Again, does the specification say something
          > about this? What does BEA say about this?
          >
          >
          > Thanks for trying to find out the real truth :-)
          >
          > regards
          > Stefan

Extproc agtctl thread spawning

Hi all,
I have an external procedure (dll) which is called by a .NET application to undertake some functions. However, although it works the first time, creating an extproc thread on the DB server and returning the results, after the first execution it does not drop the extproc process thread and the second time i run the procedure it fails.
If i run the same procedure within an Oracle debugger (TOAD, SQL Developer etc.) at the end of the execution the process thread is removed, hence it works if you want to run it again. I am having trouble understanding how the process threads are handled, from what I have read an extproc process thread is generated on first execution for a given session and stays with until said session is ended - why does this not happen in TOAD / SQL Developer?
I have tried implementing the agtctl service to handle multi-threading for this extproc process, but my attempts don't seem to have bourne much fruit thus far, would this be a correct way to go? Any information on this would be most welcome, I have read through the official Oracle documentation but other sources of information seem a little light on the ground.
TIA
Dan

Hi all,
As a follow-up to my earlier post, I have identified that I would need to manually kill the extproc process thread at the end of processing (due to session length extproc spawns), otherwise the next user that hits the database and tries to run the process fails. However, how do i identify the Windows PID for the extproc procedure and subsequently stop it from within Oracle? I know how to do this using other languages, but does Oracle support this kind of functionality or am I going to have to write something externally to do so. The problem I have is that the extporc process runs on a seperate server to the application itself (using a DB_Link), so I cannot identify the PID directly from the application. This is because the application server is a Windows 2003 x64 server, which doesn't support extproc 32 (the shared dll is 32-bit).
If anyone can give me any pointers I would be most grateful.
Regards
Dan

Does setQueryTimeout still spawn a thread with Microsoft JDBC Driver 4.0?

Hi,
Does setQueryTimeout still spawn a thread with Microsoft JDBC Driver 4.0?
This link here indicates this was fixed awhile ago:
http://msdn.microsoft.com/en-us/sqlserver/aa937724.aspx
However, we are using Microsoft JDBC Driver 4.0 but are noticing that a setQueryTimeout thread is being spawned (of class com.microsoft.sqlserver.jdbc.TimeoutTimer) for every query we execute.
And, the thread seems to stay around even after the query has finished executing.
Can anyone confirm whether this is actually fixed?
Or, does anyone have any ideas how we can avoid having a new thread spawned for monitoring for a query timeout?
Thanks!

Hi,
Does setQueryTimeout still spawn a thread with Microsoft JDBC Driver 4.0?
This link here indicates this was fixed awhile ago:
http://msdn.microsoft.com/en-us/sqlserver/aa937724.aspx
However, we are using Microsoft JDBC Driver 4.0 but are noticing that a setQueryTimeout thread is being spawned (of class com.microsoft.sqlserver.jdbc.TimeoutTimer) for every query we execute.
And, the thread seems to stay around even after the query has finished executing.
Can anyone confirm whether this is actually fixed?
Or, does anyone have any ideas how we can avoid having a new thread spawned for monitoring for a query timeout?
Thanks!

Why can't I interrupt the main thread from a child thread with this code?

I am trying to find an elegant way for a child thread (spawned from a main thread) to stop what its doing and tell the main thread something went wrong. I thought that if I invoke mainThread.interrupt() from the child thread by giving the child thread a reference to the main thread, that would do the trick. But it doesn't work all the time. I want to know why. Here's my code below:
The main class:
* IF YOU RUN THIS OFTEN ENOUGH, YOU'LL NOTICE THE "Child Please!" MESSAGE NOT SHOW AT SOME POINT. WHY?
public class InterruptingParentFromChildThread
     public static void main( String args[] )
          Thread child = new Thread( new ChildThread( Thread.currentThread() ) );
          child.start();
          try
               child.join();
          catch( InterruptedException e )
// THE LINE BELOW DOESN'T GET PRINTED EVERY SINGLE TIME ALTHOUGH IT WORKS MOST TIMES, WHY?
               System.out.println( "Child please!" );
          System.out.println( "ALL DONE!" );
The class for the child thread:
public class ChildThread implements Runnable
     Thread mParent;
     public ChildThread( Thread inParent )
          mParent = inParent;
     public void run()
          System.out.println( "In child thread." );
          System.out.println( "Let's interrupt the parent thread now." );
          // THE COMMENTED OUT LINE BELOW, IF UNCOMMENTED, DOESN'T INVOKE InterruptedException THAT CAN BE CAUGHT IN THE MAIN CLASS' CATCH BLOCK, WHY?
          //Thread.currentThread().interrupt();
          // THIS LINE BELOW ONLY WORKS SOMETIMES, WHY?
          mParent.interrupt();
}

EJP wrote:
I'm not convinced about that. The wording in join() suggests that, but the wording in interrupt() definitely does not.Thread.join() doesn't really provide much in the way of details, but Object.wait() does:
"throws InterruptedException - if any thread interrupted the current thread +before+ or while the current thread was waiting for a notification. The interrupted status of the current thread is cleared when this exception is thrown."
every jdk method i've used which throws InterruptedException will always throw if entered while a thread is currently interrupted. admitted, i rarely use Thread.join(), so it's possible that method could be different. however, that makes the thread interruption far less useful if it's required to hit the thread while it's already paused.
a simple test with Thread.sleep() confirms my expected behavior (sleep will throw):
Thread.currentThread().interrupt();
Thread.sleep(1000L);

How to make Graphics a separate thread.

Hi all!
I am performing dense 2D Delaunay triangulation on the Java Graphics context. I don't want to add any "sleeps" there in my code, so my app eats about 95-98% of CPU resources. I am becoming unable to normally use any components (i.e. buttons, combos etc) in my app because they are stuck and will responde after 10 secs or more. I tried to get out and did the things specified below.
Assuming I have the following code:
import javax.swing.*;
import java.awt.*;
public class Test extends JPanel {
int count = 0;
int loop = 100;
Graphics g0 = null;
Thread thr = null;
public void paint(Graphics g)
     super.paint(g);
     g0 = g;
      thr = new Thread(new Runnable() {
          public void run() {
           while (loop < 600) {
                * Really I am doing a lot of things here,
                * so there is no problem with such a blind loop causing 100%CPU usage
                * or "too much threads" errors
               g0.drawString(""+(count++), loop, 200);
               if(count>9)count=1;
                * Uncomment this after one try.
                * Your JVM will crash instantly or a little later
                * depending on the sleep value below.
               //g0.drawLine(loop-10,100, loop-10, 200);
               /*Uncomment this for test and manipulate with the sleep time in range 1..500*/
               //try {Thread.sleep(500);}catch (InterruptedException ex) {}
               loop+=10;
               repaint();
           loop = 100;
      thr.setPriority(Thread.MAX_PRIORITY);
      thr.start();
public static void main(String[] args) {
     JFrame fr = new JFrame();
     Test test1 = new Test();
     test1.setDoubleBuffered(true);
     fr.setSize(700,500);
     fr.getContentPane().add(test1);
     fr.setLocation(70,70);
     fr.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
     fr.show();
}After the crash my JVM(1.4.2_01-b06 on WinXP) says me "good bye" with this message or a similar long one with the whole dll list:
Another exception has been detected while we were handling last error.
Dumping information about last error:
ERROR REPORT FILE = (N/A)
PC = 0x02eb7516
SIGNAL = -1073741819
FUNCTION NAME = Java_sun_print_Win32PrintJob_endPrintRawData
OFFSET = 0x13E2
LIBRARY NAME = C:\JBuilderX\jdk1.4\jre\bin\awt.dll
Please check ERROR REPORT FILE for further information, if there is any.
Good bye.
An unexpected exception has been detected in native code outside the VM.
Unexpected Signal : EXCEPTION_ACCESS_VIOLATION (0xc0000005) occurred at PC=0x2EB7516
Function=Java_sun_print_Win32PrintJob_endPrintRawData+0x13E2
Library=C:\JBuilderX\jdk1.4\jre\bin\awt.dll
Current Java thread:
at sun.awt.windows.Win32DDRenderer.doDrawLineDD(Native Method)
at sun.awt.windows.Win32DDRenderer.clipAndDrawLine(Win32DDRenderer.java:76)
at sun.awt.windows.Win32DDRenderer.drawLine(Win32DDRenderer.java:126)
As I described, the crash happens only when I use drawLine()/draw()/drawRect() or another standard method of Graphics/Graphics2D context except drawString() method.
Did you try to solve a problem like mine is? Can you provide a RIGHT(I know that mine is simple, wrong and dummy ) alternative for separation of Graphics thread from the main one?
Thank you guys for at least reading my post
Waiting for a shameful reply to this message!
Goodbye!

Don't draw the graphics in the AWT event queue thread.
Draw them in the main thread and repaint the
component that displays them as they become
available. The basics of a component like that are
shown in message 15 this thread:
http://forum.java.sun.com/thread.jsp?forum=31&thread=55
795&start=15&range=15&hilite=false&q=
In the same thread under the reply #19 u wrote:
The problem with drawing in the paintComponent() method is that it executes in the GUI thread and this >slows down the rest of the GUI. Many of my GUI apps involve mathematical transformations to images >which take too long to run inside paintComponent().I have the same problem though I am overriding the paint() method, i.e. my GUI thread is being hung even I use the paint() method. Forget about the code above, so there is no thread spawning, no MAX_PRIORITY and no paint()/repaint() recursivity. Did you fixed that problem by such a double buffering like action you specified in the reply #15?
Thx!

Can a thread created in an applet still run after the browser is closed?

Dear all,
I was able to make an applet's thread keep running even the web page changed to other page. Is it possible to make a thread spawned by an applet keep running even if the browser is close? I am using signed applet and JRE1.3.1_02. Thank for answer.
[email protected]
Arthur Pan

No. A thread is part of a process and if the browser closes that process will (generally) be destroyed.

In this case, can I modify swing GUI out of swing thread?

I know the Swing single thread rule and design (Swing is not thread safe).
For time-consuming task, we are using other thread to do the task and
we use SwingUtilities.invokeAndWait() or SwingUtilities.invokeLater (or SwingWorker) to update Swing GUI.
My problem is, my time-consuming task is related to Swing GUI stuff
(like set expanded state of a huge tree, walk through entire tree... etc), not the classic DB task.
So my time-consuming Swing task must executed on Swing thread, but it will block the GUI responsivity.
I solve this problem by show up a modal waiting dialog to ask user to wait and
also allow user to cancel the task.
Then I create an other thread (no event-dispatch thread) to do my Swing time-consuming tasks.
Since the modal dialog (do nothing just allow user to cancel) is a thread spawn by
Swing event-dispatch thread and block the Swing dispatch-thread until dialog close.
So my thread can modify Swing GUI stuff safely since there are no any concurrent access problem.
In this case, I broke the Swing's suggestion, modify Swing stuff out of Swing event-dispatch thread.
But as I said, there are no concurrent access, so I am safe.
Am I right? Are you agree? Do you have other better idea?
Thanks for your help.

If you drag your modal dialog around in front of the background UI, then there is concurrent access: paint requests in your main window as the foreground window moves around.

Can multiple threads use same transaction concurrently?

          Is it possible that same transaction is being used by multiple threads concurrently
          in WLS? If each thread is doing suspend and resume how does it work? Does the
          transaction implementation support it? Is there a way to do it?


Why you don't tell us some more about your application?
          I'm assuming this is a relatively long running transaction if run
          serially. One common solution is to break this type of workflow into
          several separate transactions with queues between them.
          -- Rob
          Karambir Singh wrote:
          > Is there any workaround for this? I mean to something like explicitly associating
          > the txn with user created threads.
          >
          > "krishna" <[email protected]> wrote:
          >
          >>Transaction context cannot be propagated to user created Threads.
          >>-Krishna
          >>"Karambir Singh" <[email protected]> wrote in message
          >>news:[email protected]...
          >>
          >>>I'm starting a transaction in main thread and this thread spawns three
          >>
          >>threads.
          >>
          >>>I want updates done in main transaction and updates done in three child
          >>
          >>threads
          >>
          >>>to be part of the same transaction. The transaction is finally commited
          >>
          >>by
          >>main
          >>
          >>>thread.
          >>>
          >>>
          >>>
          >>>
          >>>
          >>>"Dimitri I. Rakitine" <[email protected]> wrote:
          >>>
          >>>>What are you trying to do ?
          >>>>
          >>>>Karambir Singh <[email protected]> wrote:
          >>>>
          >>>>
          >>>>>Is there any workaround to do this?
          >>>>
          >>>>>"Dimitri I. Rakitine" <[email protected]> wrote:
          >>>>>
          >>>>>>No, it is associated with a thread which started it.
          >>>>>>
          >>>>>>Karambir Singh <[email protected]> wrote:
          >>>>>>
          >>>>>>
          >>>>>>>Is it possible that same transaction is being used by multiple
          >>
          >>threads
          >>
          >>>>>>concurrently
          >>>>>>
          >>>>>>>in WLS? If each thread is doing suspend and resume how does it
          >>
          >>work?
          >>
          >>>>>>Does the
          >>>>>>
          >>>>>>>transaction implementation support it? Is there a way to do it?
          >>>>>>
          >>>>>>--
          >>>>>>Dimitri
          >>>>>>
          >>>>
          >>>>--
          >>>>Dimitri
          >>>>
          >>>
          >>
          >

Increasing the number of thread results in core dump

Hi all ,
I am having application in C++ on Solaris and WIn NT OS . In Win Nt if the Number of Threads Spawned is 50 it's working fine but if we try the same it results in Core , but if we reduce the number of Threads spawned to 40 the application works just fine . The System configuration for WIn NT is 512 RAM and for Solaris is 1GB . Any pointers to what can be the possible reason and solution to the cause .
Regards
Rahul

Hi,
The basis sets the maximum number of pages that can be printed by the user in authorization object S_SPO_PAGE.
Please discuss with your basis guy.
Tyken

Start a new thread from JSP ???

suppose my JSP code is like this,
<%
Obj.heavyProcess();
%>
How can I start a new thread to do
out.println(" ");
out.flush("");
while the heavyProcess is going on? I need to do this because my ISP will cut any Idle connection after certain time and my heavyProcess takes a long time to finish.
for certain reasons, I can not do waiting page, refreshing, etc.

The same way as you would from a normal java application. Write your thread class
package com.yourpkg;
public class YourHeavyProcess implements Runnable{
   private YourObject obj;
   public YourHeavyProcess(YourObject obj){
            this.obj = obj;
   public void run(){
       obj.heavyProcess();
}and create and invoke this thread from your jsp.
<%
        YourObject obj = new YourObject();
        Thread t = new Thread(new YourHeavyProcess(obj));
        t.start();
%>
out.println()..;
out.flush()....;Note that the thread cannot write anything to the output stream as the response is over and done with. It can only perform backend tasks (like data archival for instance)
You could consider using a Timer and a TimerTask class which provides a slightly higher abstraction layer over threads . Look at [url http://forum.java.sun.com/thread.jspa?threadID=758874]this post (reply 48) for an example.
And finally you wouldnt want a large number of threads to be generated and competing for cpu time on your web server (whose primary responsibility is to receive http requests and send responses). You can use a thread pool to manage the number of threads spawned by your application. If you use java5, take a look at the [url http://java.sun.com/j2se/1.5.0/docs/api/java/util/concurrent/Executors.html]java.util.concurrent.Executor class for thread pool implementations that you can use off the shelf.
cheers,
ram.

Waiting for many threads

Hey guys,
I know how to make one thread wait for another named thread to complete, but how do u make a thread wait for many threads to complete with the added problem of not knowing the names of the threads your waiting for? Looked and been trying for ages but can't get anything to work.
Thanx
Lisa

No i saw it, pehaps i should rephrase with a question, how would you go about giving all these threads a reference? If it ain't obvious already am not great with java and if someone could tell me how to give the threads spawned references it would be great.
As ive explained the code is really long so am not going to waist peoples time by posting it. Here is basically how i am spawning the threads "willy-nilly" style.
     while ((p < searchTerms.size())&&(p < 30)){
         term = (searchTerms.elementAt(p)).toString();
         NetChecker nc = new NetChecker(term,excludedAdds,refWeb);
         searchClasses.addElement(nc);
         new Thread(nc).start();
         p++;
     } the classes all return web addresses in a vector, thats why i need all the threads to complete so i can collect all the results
Thanx
Lisa

Thread Spawn

Similar Messages

Maybe you are looking for