How to track Memory Leaks in 3.6.8?

I've been noticing for several updates now that I seem to be getting major memory leaks. Firefox starts off around 1-200,000 K memory usage in task manager, but within a few hours it will be past 800,000 K. Now, the things I've seen so far talk about disabling plugins and all to find the problem, but this doesn't happen immediately, so disabling things one by one isn't very feasible. How can i track the memory leak to a specific plugin, or else make it leak much, much faster so I can do the one by one thing?
== This happened ==
Every time Firefox opened
== A few updates ago

Well, It just has taken a turn for the worse... In the attached screenshot you can see firefox took 1.123.112kb out of my memory at which point the GUI became unresponsive. This has actually happened before, but not as soon...
I had 18 tabs open (but at the moment I have all of them opened again after a restart of firefox and it takes 143.988 kb at this very moment, which still is a lot actually). My computer has 4gb RAM-memory of which 46% is in use at the moment according to windows task manager.
It would be great if the whole tabstructure and maybe the garbage collector for the TraceMonkey javascript engine was reviewed... This might point out where the memory leaks are located.

Similar Messages

How to track memory leak

using an n85, i consistantly have problems when i'm in my call log. i get "log:memory full. close some applications and try again." even though i rebooted and when i check under open apps (by holding down app key), no program is open. i also have 69 megs free in phone memory (per file manager). is there a program that lists what processes are taking up memory (kinda of like MS's windows task manager or linux's TOP command)?

i did both hard resets-3key+ on button, and also 7380 method, which both work (it asks for country, time, etc), then when i restore, i get same problem. seeing how i can use other programs-web browser, gps programs, etc without any 'memory full' problem, i think it's a problem w/ the memory allocated just to the log program-and i'm running in that limit. also, i use my phone for voip calls, and the phone numbers are longer-in the 'sip:[email protected]' format, and i'm not sure if it has something to do with that. will see if the developer forum has additional info. thanks for your help!

Eclipse TPTP to track memory leak?

Hello,
I am writing a rather large program and I am collecting a lot of data. My problem is that I'm running out of heap space. I need to track large amounts of weather data over the course of the year, but I don't think that is the problem. I think I have a memory leak somewhere and I can't find it. I am working in Eclipse and I've heard a bit about the TPTP tools. I was just wondering if anyone could direct me to a tutorial for using these tools or at least give me a general set of instructions? I have everything installed but can't find a good tutorial anywhere.

JProfiler is one such tool.
However, you might try increasing the amonut of memory the heap in java is allowed to use before looking into those tools.
Perhaps your application just uses a lot of memory rather than a memory leak. Search google for 'java xmx' on how to increase heap size.

How to determine memory leaks?

I tried in XCODE, the RUN/ Start with Performance TOol / and tried out the various options. I was running my app and looking to see if it would report increasing memory use but it seemed to be looking at my total system (i was running under the simulator). In general what is the recommended procedure for determining memory leaks, which tool to use, and what tracing can i use?
How does one look at the retain count of an object? are there system routines that have knonw leaks?

You took the right path. Once instruments comes up select the Leaks tool. Turn off automatic leak detection. In your app, start off at some known state, do something, and come back to the known state and check for leaks. For instance start off in a view, do something that brings up another view then come back to the original view and check for leaks. Leaks will show you if you leaked. Since you took a very deterministic path then checked it should be straight forward to go to the code and find / fix the leaks. Leaks shows you where the code where the leak was generated.

How to detect Memory Leaking Problem in Java

I have a Java multithreading program & I suspect that there is a memory leaking issue in the program. Can someone tell me how can I know how to detect where is the memory leakage in the program?
I have download a few tools like hat, jprobe, optimizeit but not sure how to use it. I am using jdk1.3.0 in HP-UX
Thks

Not so true, you could have a memory leak if you accidental hold on to a resource, or maybe you have threads running that you have no reference to, and those threads hold some resources.
Optimizeit is pretty easy to use. You start you application with Optimize it, so Optimizeit is you JVM so to speak. In optimizeit you can get a list of active objects. If this list grows and grows, you probably have a problem. Let it run for a while, and see which objects are the most present, dubble-click it, and it lets you step down into you methods. This way you can maybe find you leak.
Good luck :)

How to detect memory leak ????

I have an app that runs monitors system.
I tested it on the Win 2000 Pro, no prob it takes 8MB top
Once I moved it to Win 2000 Server the ram usage grows constantly, it is over 30 mb now
WHY?
Are there any issues with 2000 server?
HOw to debug it?
I try JProbe , ran app for 2 hours with Profiler and it does not go over 8 MB heap???
Thank you.

JVMs are the same.
How do you limit the heap???
JProbe did not fina anything but after a few hours I
got out of vertual memory error. Java took 300MB I
asumed it was because of JProbe.
Thnks Yarik.To limit the Java heap, you specify these paramaters
java -Xmx8m YourClassNameHere
However, given that you ran out of virtual memory, it sounds to me that something else is going on outside the Java heap. The heap limit, by default, is only 64MB, so that can't be it.
What version of Java are you using (use "java -version" to find out for sure).
Chuck

How to minimize memory leaks when auto-refreshing datasets

I have a page with two Spry datasets, each of which has a
loadInterval of 30 seconds to 1 minute. Looking at the Windows Task
Manager, I see that Firefox 3 consumes 500,000kb of Working Set
memory after about 2.5 hours of no activity (other than leaving the
page open). When first loaded, Firefox was using 130,000kb, so
that's an increase of 370,000kb. Similarly, IE 7 increases from
73,000 to 130,000 in the same time period.
Is there anything that I can do programmatically to minimize
or eliminate this problem?

Thanks for the response. I need to refresh frequently because
the application is a kind of "dashboard" to show the status of
certain activities; when one of the activities occurs, the mySQL
database reflects the change, and I want the Spry dataset to show
the change to the user in more or less "real time."
I am a little unclear about your last two suggestions. When
you say, "you should generate XML files first then use it as
dataset instead directly your server side as xml on dataset url" do
you mean that I should NOT create the xml for the dataset
dynamically using a php script? (I am creating it dynamically
because that's the only way I know to get "fresh" up-to-the-moment
data. Am I missing something? And why does it matter HOW the xml is
created?
Also, can you explain how you're suggesting that I use the
javascript timeout with Spry SetURL(). I like the idea of trying to
empty the dataset, but don't understand how to implement it
(concenptually).
Thanks!

Memory leak in JCO when calling an ABAP-function that returns larg tables

Hello everybody,
I think discovered a memory leak in JCO when the calling functionions that have exporting tables with large datasets. For example the ABAP-function RFC_READ_TABLE, which in this example I use to retrieve data from a table called "RSZELTTXT", which contains ~ 120000 datasets. RFC_READ_TABLE exports the data as table "DATA".
Here a simple JUnit test:
http://pastebin.ca/1420451
When running it with Sun Java 1.6 with standard heap size of 64mb I get a heapsize OutOfMemory error:
http://pastebin.ca/1420472
Looking at the heap dump (which I unfortunately cannot post here, because of it' size), I can see that I've 65000 char[512] array objects in my heap, which don't get cleaned up. I think, each char[512] array stands for one dataset in the exporting table "DATA", since the table contains 120000 datasets, the heap is full after the first 65000 datasets are parsed. Apparently, JCO tries to read all datasets in memory instead of justing reading the dataset to which the pointer (JCoTable.setRow(i)) currently points to and releasing it from memory after the pointer moves forward ...
Did anybody else experience this?
Is SAP going to remove to issue in upcoming versions of JCO?
regards Samir

Hi,
Check Below links
1) How To Analyze Performance Problems JCO
https://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/3fbea790-0201-0010-6481-8370ebc3c17d
2) How to Avoid Memory Leaks
https://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/c3e598fe-0601-0010-e990-b8622840c8c2
Salil...
Edited by: salil chavan on Jun 2, 2009 5:21 AM

Find memory leak

Hi All ,
In our SAP application server (Solaris 10 X86 running on hardware HP ProLiant BL685c G5) physical memory available is day to day decreasing drastically. We suspect, there is a memory leak.
Could you please let us know how to find memory leak in Solaris.
Regards,
Rajesh.

Hi Rajesh
Memory leaks have a large number of possible causes, including:
•Unsupported hardware or software
•Custom Scripting
•Misconfiguration
•Third party hardware and software issues
The prstat command can be used to gather basic performance data to help identify if there are any processes consuming a large amount of memory. For example, the data below is sorted by the SIZE column. SIZE is the total virtual memory size of the process:
prstat -s size
If you are experiencing some of the symptoms described above or if you suspect a memory leak, it is particularly important that performance data is captured so the memory leak can be confirmed. It is important to gather at least the following pieces of information:
•Process ID (PID)
•Size of the process
•The executing command
•The timestamp information showing exactly when the data was captured.
while true
do
for pid in `cat pids`
do var=`date|cut -d' ' -f4`
echo "$var :\c" >> ps.log.$pid
ps -eo pid,vsz,rss,pcpu,args | grep $pid |grep -v grep >> ps.log.$pid
done
sleep 30
done
You could check the memory utilisation details through
#mdb -k
::memstat
kmastatRegards
Sadiq

Memory Leak in Swing.

Hi!
I have Real time application(JDK1.3). After each 30 sec it brings data from backend &puts into JTable.
But it works only for 6 to 8 hours. After that it gives error
in JVM.dll.
I'm using 3 threads. Main code is something like this.
public class Main extends Thread{
public void run(){
while(true)
try{
Testtable.Update();
Thread.sleep(30000);
}catch(InterruptedException e)
{ System.out.println("error occured");}
TestTable is a public class which extends AbstractTableModel. This is a inner class of Main class.
I have checked memory usage of java.exe. It increases
usage after sometime.
How to detect memory leak? Do have to do changes in my code? "vector" usage is more in this program.
Thanks,
Sujata.

Hi,
For cellrendering I have following code.
public class MultiLineHeaderRenderer extends JList implements TableCellRenderer{
public MultiLineHeaderRenderer() {
try{
setOpaque(true);
setForeground(UIManager.getColor("TableHeader.foreground"));
setBackground(UIManager.getColor("TableHeader.background"));
setBorder(UIManager.getBorder("TableHeader.cellBorder"));
ListCellRenderer renderer = getCellRenderer();
((JLabel)renderer).setHorizontalAlignment(JLabel.CENTER);
setCellRenderer(renderer);
}catch ( Exception err){
TeCILog.logTrace( this, "Failed to construct MultiLineHeaderRenderer." );
TeCILog.logError( this, err );
public Component getTableCellRendererComponent(JTable table, Object value,boolean isSelected, boolean hasFocus, int row, int column)
setFont(table.getFont());
String str = (value == null) ? "" : value.toString();
BufferedReader br = new BufferedReader(new StringReader(str));
String line;
Vector v = new Vector();
try {
while ((line = br.readLine()) != null)
v.addElement(line);
br.close();
}catch (IOException ex) {
ex.printStackTrace();
} catch(Exception err){
System.out.println("error" + err );
setListData(v);
return this;
private ListCellRenderer renderer = null;
Does this is causing leakage?
Sujata

Memory Leak when just launched and Idle.. fixes when being used ?? [HELP]

So I'm in the debugging and testing phase of my app and using this tool for tracking memory leaks ( https://github.com/mrdoob/Hi-ReS-Stats )
When I launch my app my numbers are
FPS: 61/60
MS: 17
MEM: 3.157
MAX: 3.157
Now immediately my memory starts increasing    from 3.157, 3.167, 3.177, 3.187, 3.197 and so on.
Now if I make any nav selection in my app
MEM changes back down to about 3.215
but then it starts its count again   3.215, 3.225, 3.235, 3.445, 3.455
I don't have any loops happening.
Has anyone run in to this ?
I'm almost tempted to force garbage collection every 60 seconds that the app is idle or something. Not the best way to handle this ..   I just dont know where the leak is happening.
Any support is appreciated!
Cheers!

Hi there - I just had the same query a couple of days ago (http://forums.adobe.com/thread/977174?tstart=30).
I saw the same symptoms on my app so I built a blank app with just the profiler on stage. I've been monitoring it for a few days now and notice that memory does creep up even when the app is left idle (apart from the profiler) - but ... and this is the important bit ... it does periodically get reduced back to the starting point (when the garbage collector kicks in and memory is released).
When I was monitoring my app the time through this cycle could be well over 5 mins.
If you actually use the monitor when putting your app through it's paces you'll see memory being gobbled up more rapidly and hopefully (if you've no leaks) the garbage collection kicking in more regularly and bringing the reported usage back down.

Best Practices for EJB--memory leak avoidance

I an researching a memory leak. From what I can tell, we do not want to put any instance variables at the class level. This would incur a memory leak when the ejb containers creates the ejb.
I am wondering if there are any more 'best practices' with EJBs that prevent memory leaks. A good site would help me too
Russ

Thank for your reply.
You are right. I was referring to stateless session
beans.
What I was thinking was the following:
1. Making parameters final unless the parameter
r state is changed inside a method.Won't help with memory leaks. There are benefits for an object being immutable, but I don't think that it eliminates the possibility of the object being leaked.
2. All instance variables in stateless session beans
s should be set to null upon ejbPassivate and
ejbRemove operations. The variables should be
populated on ejbCreate and ejbActivate.Optimizing compilers don't need the hint of setting the reference to null. I don't think that helps, but I'm not 100% certain.
3. Before throwing an exception at the session bean
level, clear up all the instance variables by setting
them to null.Shouldn't scope make it clear to the GC that the objects aren't needed? If they're all primitives, how does this help memory leaks?
I am using a profiler to identify the objects still
being held onto once GC, but I was hoping others with
more experience than I would share some of their tips
on how they avoid memory leaks.I don't know where you're getting these tips, but I don't think they're helpful.
Collections and Singletons would be the places where I'd look. If you add a reference to an object that holds onto it in a collection and never lets go, the GC won't reclaim it.
Singletons aren't cleaned up, because their instance is static. Anything the Singleton refers to won't be cleaned up unless the Singleton relinquishes the reference.
Look for things like that. I think they matter more.
%

How to deal with Memory Leaks, that are caused by Binding

Hi, I recently noticed (huge?) memory leaks in my application and suspect bindings to be the cause of all the evil.
I made a little test case, which confirms my suspicion:
import javafx.application.Application;
import javafx.beans.property.SimpleStringProperty;
import javafx.beans.property.StringProperty;
import javafx.scene.Scene;
import javafx.scene.control.Button;
import javafx.scene.layout.VBox;
import javafx.stage.Stage;
public class TestAppMemoryLeak extends Application {
    public static void main(String[] args) {
        launch(args);
    @Override
    public void start(Stage stage) throws Exception {
        VBox root = new VBox();
        Button button = null;
        for (int i = 0; i < 100000; i++) {
            button = new Button();
            button.textProperty().bind(text);
            button.textProperty().unbind(); // if you don't call this, you can notice the increased memory of the java process.
        root.getChildren().add(button);
        Scene scene = new Scene(root);
        stage.setScene(scene);
        stage.show();
    private StringProperty text = new SimpleStringProperty("test");
}Now the problem is, HOW can I know, when a variable is no longer needed or overwritten by a new instance.
Just an example:
I have a ListView with a Cell Factory. In the updateItem method, I add a ContextMenu. The textProperty of each MenuItem is bound to a kind of global property, like in the example above. I have to do it in the updateItem method, since the ContextMenu differs depending on the item.
So every time the updateItem method is called a new ContextMenu is created, which binds some properties, but the old context menus remain in memory.
I guess there could be many more example.
How can I deal with it?

I've dealt with this situation and created a Jira issue for it, but I also have a work-around that is a bit unwieldy but works. I'll share it with you.
The bug that deals with this (in part atleast): http://javafx-jira.kenai.com/browse/RT-20616
The solution is to use weak invalidation listeners, however they are a bit of a pain to use as you cannot do it with something simplistic as "bindWeakly"... and you need to keep a reference around to the wrapped listener otherwise it will just get garbage collected immediately (as it is only weakly referenced). Some very odd bugs can surface if weak listeners disappear randomly because you forgot to reference them :)
Anyway, see this code below, it shows you some code that is called from a TreeCell's updateItem method (I've wrapped it in some more layers in my program, but it is essentially the same as an updateItem method):
public class EpisodeCell extends DuoLineCell implements MediaNodeCell {
private final WeakBinder binder = new WeakBinder();
@Override
public void configureCell(MediaNode mediaNode) {
    MediaItem item = mediaNode.getMediaItem();
    StringBinding episodeRange = MapBindings.selectString(mediaNode.dataMapProperty(), Episode.class, "episodeRange");
    binder.unbindAll();
    binder.bind(titleProperty(), MapBindings.selectString(mediaNode.dataMapProperty(), Media.class, "title"));
    binder.bind(ratingProperty(), MapBindings.selectDouble(mediaNode.dataMapProperty(), Media.class, "rating").divide(10));
    binder.bind(extraInfoProperty(), Bindings.when(episodeRange.isNull()).then(new SimpleStringProperty("Special")).otherwise(episodeRange));
    binder.bind(viewedProperty(), item.viewedProperty());
    subtitleProperty().set("");
}This code makes use of a class called WeakBinder -- it is a helper class that can make Weak bindings and can keep track of them. When you call unbindAll() on it, all of the bindings it created before are released immediately (although they will also disappear when the Cell itself is garbage collected, which is possible because it only makes weak references).
I've tested this extensively and it solves the problem of Cells keeping references to objects with much longer life cycles (in my case, the MediaNode passed in has a longer lifecycle than the cells and so it is important to bind weakly to it). Before this would create huge memory leaks (crashing my program within a minute if you kept refreshing the Tree)... now it survives hours atleast and the Heap usage stays in a fixed range which means it is correctly able to collect all garbage).
The code for WeakBinder is below (you can consider it public domain, so use it as you see fit, or write your own):
package hs.mediasystem.util;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import javafx.beans.InvalidationListener;
import javafx.beans.Observable;
import javafx.beans.WeakInvalidationListener;
import javafx.beans.property.Property;
import javafx.beans.value.ObservableValue;
public class WeakBinder {
private final List<Object> hardRefs = new ArrayList<>();
private final Map<ObservableValue<?>, WeakInvalidationListener> listeners = new HashMap<>();
public void unbindAll() {
    for(ObservableValue<?> observableValue : listeners.keySet()) {
      observableValue.removeListener(listeners.get(observableValue));
    hardRefs.clear();
    listeners.clear();
public <T> void bind(final Property<T> property, final ObservableValue<? extends T> dest) {
    InvalidationListener invalidationListener = new InvalidationListener() {
      @Override
      public void invalidated(Observable observable) {
        property.setValue(dest.getValue());
    WeakInvalidationListener weakInvalidationListener = new WeakInvalidationListener(invalidationListener);
    listeners.put(dest, weakInvalidationListener);
    dest.addListener(weakInvalidationListener);
    property.setValue(dest.getValue());
    hardRefs.add(dest);
    hardRefs.add(invalidationListener);
}Let me know if this solves your problem.

How to root out memory leak with Java JNI & Native BDB 11g ?

We are testing a web application using the 32-bit compiled native 11g version of BDB (with replication) under 32-bit IBM 1.5 JVM via JNI under 64-bit RedHat Linux. We are experiencing what appears to be a memory leak without a commensurate increase in Java heap size. Basically the process size continues to grow until the max 32-process size is reached (4Gb) and eventually stops running (no core). Java heap is set to 2Gb min/max. GCs are nominal, so the leak appears to be native and outside Java bytecode.
We need to determine whether there is a memory leak in BDB, or the IBM JVM or simply a mis-use of BDB in the Java code. What tools/instrumentation/db statistic should be used to help get to root cause? Do you recommend using System Tap (with some particular text command script)? What DB stats should we capture to get to the bottom of this memory leak? What troubleshooting steps can you recommend?
Thanks ahead of time.
JE.
Edited by: 787930 on Aug 12, 2010 5:42 PM

That's troublesome... DB itself doesn't have stats that track VM in any useful way. I am not familiar with SystemTap but a quick look at it seems to imply that it's better for kernel monitoring than user space. It's pretty hard to get DB to leak significant amounts of memory. The reason is that it mostly uses shared memory carved from the environment. Also if you are neglecting to close or delete some object DB generally complains about it somewhere.
I don't see how pmap would help if it's a heap leak but maybe I'm missing something.
One way to rule DB out is to replace its internal memory allocation functions with your own that are instrumented to track how much VM has been allocated (and freed). This is very easy to do using the interfaces:
db_env_set_func_malloc()
db_env_set_func_free()
These are global to your process and your functions will be used where DB would otherwise call malloc() and free(). How you get usage information out of the system is an exercise left to the reader :-) If it turns out DB is the culprit then there is more thinking to do to isolate the problem.
Other ideas that can provide information if not actual smoking guns:
-- accelerate reproduction of the problem by allocating nearly all of the VM to the JVM and the DB cache (or otherwise limit the allowable VM in your process)
-- change the VM allocated to the JVM in various ways
Regards,
George

How to define the memory leak in stability test

Hi
Usually, we run our system with 70% CPU load for 72 hours for stability test (on solaris 10), because some plugin of our system is using mtmalloc, and we use prstat to monitor the memory of each plugin. But because of the complex memory usage of our application, we don't know when the application will use the Max memory, and because of the "mtmalloc", the memory showed by "prstat" will not released but continually increased.
So fro the test point of view, there is the risk of memory leak, but actually, there may be no memory leak, so my question is how to define the memory leak in such condition.

kevin wrote:
Thank you for the input and all the info.
isn't java heap the same as memory allocated to the java process in my weblogic
starup script ?The heap is sized by the -Xmx and -Xms parameters you pass on the java
command-line. The permanent generation is separate.
>
let me also download jprobe and try to run it and see what it gives me.
I'd start by running with -verbose:gc. I'd want to know whether you're
running out of heap or permgen space.
-- Rob
kevin
Rob Woollen <[email protected]> wrote:
Unfortunately memory leaks are not fun to track down even with tools.
I'd first suggest determining whether you're running out of space in
the
permanent area (where classes are loaded), or you've exhausted the java
help space.
I'd start by adding -verbose:gc. Look at the gc messages right before
you hit the OutOfMemoryError. If there's plenty of space left, I'd
suspect you're running out of perm space. Search these newsgroups and
the web for MaxPermSize, and you should see plenty of info.
If you're running out of java heap, tools like jprobe and OptimizeIt
are
helpful. If you can tell me a little more about your application and
how you're testing it, I can offer some more tips.
-- Rob
kevin wrote:
Iam new to JAVA and weblogic. I have an application that runs out ofmemory time
and again.
please let me know how to pin point this problem and moreover, howto interpret
or understand that there is a problem. I have downloaded JPROFILE tool,but it
is very confusing to understand what is goin on in this tool.
If somebody can let me know how to interpret and understand the memoryleak, that
will be great !!!
thank you.

How to track Memory Leaks in 3.6.8?

Similar Messages

Maybe you are looking for