wrapper-user — User issues for Wrapper. (Support questions, etc.)

Re: [Wrapper-user] Using the wrapper integration method 3

[Leif M. <le...@ta...>]

Jay,
Forgot to answer:

> Sorry about the no timestamps... How do I turn those on?

In the logging configuration wrapper.logfile.format should contain the
'T' symbol.
http://wrapper.tanukisoftware.org/doc/english/prop-logfile-format.html

Cheers,
Leif

Re: [Wrapper-user] Using the wrapper integration method 3

[Leif M. <le...@ta...>]

Jay,
I am not sure how Ruby would be doing this but it would surprise beif
they were daemons by default.

Try doing a thread dump.  The JVM looks like it is shutting down
fairly quickly so try adding code like the following after you start
your Ruby thread.

try {
Thread.sleep(500); // Let Ruby start.
} catch (InterruptedException e) {
//Something
}
org.tanukisoftware.wrapper.WrapperManager.requestThreadDump();

A daemon thread will look something like this:
"Reference Handler" daemon prio=10 tid=0x0003efb0 nid=0x308 in
Object.wait() [0x02b8f000..0x02b8fd68]

Non daemon like this:
"AWT-EventQueue-0" prio=7 tid=0x02d65170 nid=0x940 in Object.wait()
[0x0436f000..0x0436fd68]

If all else fails, you may need to create a simple java thread which
starts and loops until a flag is set from the WrapperListener.stop
method or something.

Cheers,
Leif

On Fri, Jun 27, 2008 at 9:29 PM,  <Jay...@sc...> wrote:
> Sorry about the no timestamps... How do I turn those on?
>
> What am I doing...
>
> In my start method I am spinning up a Jruby VM and in that VM I am spawning
> a thread. It is my understanding that in Jruby .. Ruby threads are Java
> threads... But they may not be non daemon threads I will look into that.
>
> Here are the pertinent functions:
>
> public Integer start( String[] args)
> {
> if (_service == null)
> {
> LauncherOptions opts = new LauncherOptions();
> CmdLineParser parser = new CmdLineParser(opts);
> try{
> parser.parseArgument(args);
> } catch( CmdLineException e ) {
> System.err.println(e.getMessage());
> System.err.println(" Example: java
> ServiceLauncher"+parser.printExample(ALL));
> parser.printUsage(System.err);
> System.err.println();
> return -1;
> }
> initialize(opts.serviceClassName, opts.serviceFileName,
> opts.serviceInstanceName);
>
> }
>
> _service.start();
> return null;
> }
>
> public void initialize(String serviceClassName, String serviceFile, String
> serviceName)
> {
> try {
> String filename = serviceFile.split("\\.")[0];
> // Create runtime instance of the Ruby VM
> _rubyRuntime = Ruby.newInstance();
> _rubyRuntime.evalScriptlet(String.format("require '%s'",filename));
> Object rfj =
> _rubyRuntime.evalScriptlet(String.format("%s.new",serviceClassName,
> serviceName));
> rfj = org.jruby.javasupport.JavaEmbedUtils.rubyToJava(_rubyRuntime,
> (org.jruby.runtime.builtin.IRubyObject) rfj,
> AffinityService.class); //((AffinityService)rfj).start();
>
> _service = (AffinityService)rfj;
>
> } catch (Throwable e) {
> System.err.println("Exception creating the service: ");
> e.printStackTrace(System.err);
> }
> }
>
> Here is the Ruby code that gets executed:
>
> class RubyService < tv.seachange.affininty.util.AffinityService
> attr_reader :name
> def initialize
> #configure the service for the common stuff
> @name = "RubyService"
> end
>
> def start
> @thread = Thread.new do
> @started = true
> spin_up
> end
> end
>
> def stop
> @started = false
> spin_down
> end
> end
>
>
>
> ________________________________
>
> From: wra...@li...
> [mailto:wra...@li...] On Behalf Of "Leif
> Mortenson" <le...@ta...>
> Sent: Thursday, June 26, 2008 10:48 PM
> To: Wrapper User List
> Subject: Re: [Wrapper-user] Using the wrapper integration method 3
>
>
> Jay,
> see this part of the log:
>
>> jvm 1 | WrapperManager Debug: calling WrapperListener.start()
>> jvm 1 | WrapperManager Debug: Waiting for WrapperListener.start runner
>> thread to complete.
>> jvm 1 | WrapperManager Debug: WrapperListener.start runner thread started.
>> jvm 1 | Starting up!
>> jvm 1 | WrapperManager Debug: WrapperListener.start runner thread stopped.
>> jvm 1 | hi
>> jvm 1 | WrapperManager Debug: returned from WrapperListener.start()
>> jvm 1 | WrapperManager Debug: Send a packet STARTED :
>> wrapperp | read a packet STARTED :
>> wrapper | JVM signalled that it was started.
>> jvm 1 | WrapperManager Debug: Startup runner thread stopped.
>> jvm 1 | WrapperManager Debug: ShutdownHook started
>
> The Wrapper is calling WrapperListener.start in your application.
> That returns and the WrapperManager reports that the application has
> started. The WrapperManager starts up something called the start
> runner which runs while the application is being started to keep the
> JVM from shutting itself down due to a lack of non-daemon threads.
> The problem is that as soon as that thread completes, there are no
> longer any non-daemon threads running so the JVM is initiating a
> shutdown. This is all correct behavior for the Wrapper.
>
> What exactly are you doing inside of your WrapperListener.start
> method? Are you starting any non-daemon threads which make up the
> application that should be running?
>
> If we exchange any more log files, please include time stamps. They
> are useful in understanding the timing of when things happen.
>
> Cheers,
> Leif
>
> On Fri, Jun 27, 2008 at 7:13 AM, < Jay...@sc...> wrote:
>> Leif,
>>
>> Here is the log. ...
>>
>> It could definitely be a thread thing.
>>
>> I am spinning up a thread in Jruby. (my java code launches a ruby vm and
>> in
>> ruby I am creating a thread)
>>
>> Jay
>>
> < log snip>
>>
>>
>>
>> ====
>>
>> Could you please set the wrapper.debug=true property and then reply with
>> the resulting wrapper.log attached. I only need to see a single Wrapper
>> invocation so delete the old wrapper.log file first. The log should
>> allow me to say what your exact problem is.
>>
>> Cheers,
>> Leif
>>
>> Jay...@sc... wrote:
>>> Hi,
>>>
>>> I'm playing around with the service wrapper and trying to get an example
>>> running using integration method 3. I'm running on the Mac OS X.
>>>
>>> Everything appears to work. The service starts up when I use the start
>>> script (sh.script.ln) except that after a few seconds the infrastructure
>>> appears to call a stop on my service.
>>>
>>> Does anyone know the reasons why the stop would be called? I've
>>> returned null from my start method. I spun up my logic in a thread
>>> although I did the thread start in a wierd way. (what if the thread
>>> exited (that's not necessarily happening but I'm trying to learn)?).
>>>
>>> My service is a very simple loop where I'm just printing a message to
>>> std out every second.
>>>
>>
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>

Re: [Wrapper-user] Using the wrapper integration method 3

[<Jay...@sc...>]

Sorry about the no timestamps... How do I turn those on?

What am I doing...

In my start method I am spinning up a Jruby VM and in that VM I am spawning
a thread.  It is my understanding that in Jruby .. Ruby threads are Java
threads... But they may not be non daemon threads I will look into that.

Here are the pertinent functions:

 public Integer start( String[] args)
 {
  if (_service == null)
  {
   LauncherOptions opts = new LauncherOptions();
   CmdLineParser parser = new CmdLineParser(opts);
   try{
    parser.parseArgument(args);
   } catch( CmdLineException e ) {
    System.err.println(e.getMessage());
    System.err.println("  Example: java
ServiceLauncher"+parser.printExample(ALL));
    parser.printUsage(System.err);
    System.err.println();
    return -1;
   }
   initialize(opts.serviceClassName, opts.serviceFileName,
opts.serviceInstanceName);

  }

  _service.start();
  return null;
 }

      public void initialize(String serviceClassName, String serviceFile,
String serviceName)
      {
            try {
                  String filename = serviceFile.split("\\.")[0];
                  // Create runtime instance of the Ruby VM
                  _rubyRuntime = Ruby.newInstance();
                  _rubyRuntime.evalScriptlet(String.format("require
'%s'",filename));
                  Object rfj =
_rubyRuntime.evalScriptlet(String.format("%s.new",serviceClassName,
serviceName));
                  rfj =
org.jruby.javasupport.JavaEmbedUtils.rubyToJava(_rubyRuntime,
(org.jruby.runtime.builtin.IRubyObject) rfj,

AffinityService.class);             //((AffinityService)rfj).start();

                  _service = (AffinityService)rfj;

            } catch (Throwable e) {
                  System.err.println("Exception creating the service: ");
                  e.printStackTrace(System.err);
            }
      }

Here is the Ruby code that gets executed:

class RubyService < tv.seachange.affininty.util.AffinityService
  attr_reader :name
  def initialize
    #configure the service for the common stuff
    @name = "RubyService"
  end

  def start
    @thread = Thread.new do
      @started = true
      spin_up
    end
  end

  def stop
    @started = false
    spin_down
  end
end



________________________________

From: wra...@li...
[mailto:wra...@li...] On Behalf Of "Leif
Mortenson" <le...@ta...>
Sent: Thursday, June 26, 2008 10:48 PM
To: Wrapper User List
Subject: Re: [Wrapper-user] Using the wrapper integration method 3


Jay,
see this part of the log:

> jvm 1 | WrapperManager Debug: calling WrapperListener.start()
> jvm 1 | WrapperManager Debug: Waiting for WrapperListener.start runner
> thread to complete.
> jvm 1 | WrapperManager Debug: WrapperListener.start runner thread
started.
> jvm 1 | Starting up!
> jvm 1 | WrapperManager Debug: WrapperListener.start runner thread
stopped.
> jvm 1 | hi
> jvm 1 | WrapperManager Debug: returned from WrapperListener.start()
> jvm 1 | WrapperManager Debug: Send a packet STARTED :
> wrapperp | read a packet STARTED :
> wrapper | JVM signalled that it was started.
> jvm 1 | WrapperManager Debug: Startup runner thread stopped.
> jvm 1 | WrapperManager Debug: ShutdownHook started

The Wrapper is calling WrapperListener.start in your application.
That returns and the WrapperManager reports that the application has
started. The WrapperManager starts up something called the start
runner which runs while the application is being started to keep the
JVM from shutting itself down due to a lack of non-daemon threads.
The problem is that as soon as that thread completes, there are no
longer any non-daemon threads running so the JVM is initiating a
shutdown. This is all correct behavior for the Wrapper.

What exactly are you doing inside of your WrapperListener.start
method? Are you starting any non-daemon threads which make up the
application that should be running?

If we exchange any more log files, please include time stamps. They
are useful in understanding the timing of when things happen.

Cheers,
Leif

On Fri, Jun 27, 2008 at 7:13 AM, < Jay...@sc...> wrote:
> Leif,
>
> Here is the log. ...
>
> It could definitely be a thread thing.
>
> I am spinning up a thread in Jruby. (my java code launches a ruby vm and
in
> ruby I am creating a thread)
>
> Jay
>
< log snip>
>
>
>
> ====
>
> Could you please set the wrapper.debug=true property and then reply with
> the resulting wrapper.log attached. I only need to see a single Wrapper
> invocation so delete the old wrapper.log file first. The log should
> allow me to say what your exact problem is.
>
> Cheers,
> Leif
>
> Jay...@sc... wrote:
>> Hi,
>>
>> I'm playing around with the service wrapper and trying to get an example

>> running using integration method 3. I'm running on the Mac OS X.
>>
>> Everything appears to work. The service starts up when I use the start
>> script (sh.script.ln) except that after a few seconds the infrastructure

>> appears to call a stop on my service.
>>
>> Does anyone know the reasons why the stop would be called? I've
>> returned null from my start method. I spun up my logic in a thread
>> although I did the thread start in a wierd way. (what if the thread
>> exited (that's not necessarily happening but I'm trying to learn)?).
>>
>> My service is a very simple loop where I'm just printing a message to
>> std out every second.
>>
>

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
Wrapper-user mailing list
Wra...@li...
https://lists.sourceforge.net/lists/listinfo/wrapper-user

Re: [Wrapper-user] Using the wrapper integration method 3

[Leif M. <le...@ta...>]

Jay,
see this part of the log:

> jvm 1 | WrapperManager Debug: calling WrapperListener.start()
> jvm 1 | WrapperManager Debug: Waiting for WrapperListener.start runner
> thread to complete.
> jvm 1 | WrapperManager Debug: WrapperListener.start runner thread started.
> jvm 1 | Starting up!
> jvm 1 | WrapperManager Debug: WrapperListener.start runner thread stopped.
> jvm 1 | hi
> jvm 1 | WrapperManager Debug: returned from WrapperListener.start()
> jvm 1 | WrapperManager Debug: Send a packet STARTED :
> wrapperp | read a packet STARTED :
> wrapper | JVM signalled that it was started.
> jvm 1 | WrapperManager Debug: Startup runner thread stopped.
> jvm 1 | WrapperManager Debug: ShutdownHook started

The Wrapper is calling WrapperListener.start in your application.
That returns and the WrapperManager reports that the application has
started.  The WrapperManager starts up something called the start
runner which runs while the application is being started to keep the
JVM from shutting itself down due to a lack of non-daemon threads.
The problem is that as soon as that thread completes, there are no
longer any non-daemon threads running so the JVM is initiating a
shutdown.   This is all correct behavior for the Wrapper.

What exactly are you doing inside of your WrapperListener.start
method?   Are you starting any non-daemon threads which make up the
application that should be running?

If we exchange any more log files, please include time stamps.  They
are useful in understanding the timing of when things happen.

Cheers,
Leif

On Fri, Jun 27, 2008 at 7:13 AM,  <Jay...@sc...> wrote:
> Leif,
>
> Here is the log. ...
>
> It could definitely be a thread thing.
>
> I am spinning up a thread in Jruby. (my java code launches a ruby vm and in
> ruby I am creating a thread)
>
> Jay
>
<log snip>
>
>
>
> ====
>
> Could you please set the wrapper.debug=true property and then reply with
> the resulting wrapper.log attached. I only need to see a single Wrapper
> invocation so delete the old wrapper.log file first. The log should
> allow me to say what your exact problem is.
>
> Cheers,
> Leif
>
> Jay...@sc... wrote:
>> Hi,
>>
>> I'm playing around with the service wrapper and trying to get an example
>> running using integration method 3. I'm running on the Mac OS X.
>>
>> Everything appears to work. The service starts up when I use the start
>> script (sh.script.ln) except that after a few seconds the infrastructure
>> appears to call a stop on my service.
>>
>> Does anyone know the reasons why the stop would be called? I've
>> returned null from my start method. I spun up my logic in a thread
>> although I did the thread start in a wierd way. (what if the thread
>> exited (that's not necessarily happening but I'm trying to learn)?).
>>
>> My service is a very simple loop where I'm just printing a message to
>> std out every second.
>>
>

Re: [Wrapper-user] Defining wrapper.umask on command line bug?

[Joshua S. <js...@re...>]

Thanks Leif, I finally figured out how to use it properly :)

Josh

Leif Mortenson wrote:
> Josh,
> The wrapper.umask, wrapper.anchorfile.umask, wrapper.<etc>.umask
> properties are all
> designed to set the umask of the various files created by the wrapper itself.
>
> You can specify the default umask for the Java process with wrapper.java.umask.
> These properties are set within the Wrapper and must therefor be done before the
> JVM is launched.
>
> As I understand it, you are attempting to system properties within the JVM.  The
> Wrapper will not recognize any system properties set within the JVM.
>
> Please add wrapper.java.umask to your wrapper.conf file.
> http://wrapper.tanukisoftware.org/doc/english/prop-umask.html
>
> You mentioned that you are writing a "pure java" implementation of the Wrapper?
> You will always need the native binary and JNI library for the Wrapper
> to function
> correctly.
>
> Cheers,
> Leif
>
> On Thu, Jun 26, 2008 at 11:39 PM, Joshua Sumali <js...@re...> wrote:
>   
>> Hi,
>>
>> I'm trying to use the pure java implementation of tanukiwrapper
>> (Integration Method 3) and whenever I try to specify the umask to use,
>> tanukiwrapper seems to ignore the setting. I've tried setting the umask
>> with:
>>
>> -Dwrapper.umask=0077 on the command line, and
>> System.setProperty("wrapper.umask","0077") in the actual code, but
>> neither method seems to work, as any files created by the test
>> application I'm working with still has 664 permissions instead of the
>> expected 600. Am I doing something wrong, or is this perhaps a bug?
>>
>> Any help or comments appreciated.
>>
>> Thanks,
>> Josh
>>
>> -------------------------------------------------------------------------
>> Check out the new SourceForge.net Marketplace.
>> It's the best place to buy or sell services for
>> just about anything Open Source.
>> http://sourceforge.net/services/buy/index.php
>> _______________________________________________
>> Wrapper-user mailing list
>> Wra...@li...
>> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>>
>>     
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>   

Re: [Wrapper-user] Defining wrapper.umask on command line bug?

[Leif M. <le...@ta...>]

Josh,
The wrapper.umask, wrapper.anchorfile.umask, wrapper.<etc>.umask
properties are all
designed to set the umask of the various files created by the wrapper itself.

You can specify the default umask for the Java process with wrapper.java.umask.
These properties are set within the Wrapper and must therefor be done before the
JVM is launched.

As I understand it, you are attempting to system properties within the JVM.  The
Wrapper will not recognize any system properties set within the JVM.

Please add wrapper.java.umask to your wrapper.conf file.
http://wrapper.tanukisoftware.org/doc/english/prop-umask.html

You mentioned that you are writing a "pure java" implementation of the Wrapper?
You will always need the native binary and JNI library for the Wrapper
to function
correctly.

Cheers,
Leif

On Thu, Jun 26, 2008 at 11:39 PM, Joshua Sumali <js...@re...> wrote:
> Hi,
>
> I'm trying to use the pure java implementation of tanukiwrapper
> (Integration Method 3) and whenever I try to specify the umask to use,
> tanukiwrapper seems to ignore the setting. I've tried setting the umask
> with:
>
> -Dwrapper.umask=0077 on the command line, and
> System.setProperty("wrapper.umask","0077") in the actual code, but
> neither method seems to work, as any files created by the test
> application I'm working with still has 664 permissions instead of the
> expected 600. Am I doing something wrong, or is this perhaps a bug?
>
> Any help or comments appreciated.
>
> Thanks,
> Josh
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>

[Wrapper-user] Defining wrapper.umask on command line bug?

[Joshua S. <js...@re...>]

Hi,

I'm trying to use the pure java implementation of tanukiwrapper 
(Integration Method 3) and whenever I try to specify the umask to use, 
tanukiwrapper seems to ignore the setting. I've tried setting the umask 
with:

-Dwrapper.umask=0077 on the command line, and
System.setProperty("wrapper.umask","0077") in the actual code, but 
neither method seems to work, as any files created by the test 
application I'm working with still has 664 permissions instead of the 
expected 600. Am I doing something wrong, or is this perhaps a bug?

Any help or comments appreciated.

Thanks,
Josh

[Wrapper-user] Released MacOSX 64-bit versions of 3.3.0

[Leif M. <le...@ta...>]

Hi all,
A quick update.  We released the MacOSX 64-bit versions of the Wrapper. 
These use the same codebase as 3.3.0.

Some users have asked for these so if you are one of them, please give
them a try:
http://wrapper.tanukisoftware.org/doc/english/download.jsp

Cheers,
Leif

Re: [Wrapper-user] Critical error: wait for JVM process failed

[Vincent F. <v.f...@gm...>]

Hi,

Nothing unusual with those setting, no wide chars or spaces.

1) The locale is English

2) The path is to the wrapper is
C:\GigaSpaces\Services\cmlSpace\bin\wrapper.exe

3) The path when I ran -version was
C:\GigaSpaces\Services\version\bin\wrapper.exe

Just to be paranoid I tried getting rid of the relative paths in the
-version one (it only has relative paths for wrapper.jar, wrappertest.jar
and lib), it made no difference.

    Vin


On Thu, Jun 26, 2008 at 2:38 AM, Leif Mortenson <le...@ta...>
wrote:

> Vin,
> Given that the error message is not displaying correctly, I am wondering
> if there is a general character encoding problem.  The Wrapper is not
> yet localized and double byte safe.  This is on our todo list.
>
> 1) What is the locale of your Windows?  ie.  Japanese, English, etc.
>
> 2) What is the full path to where the wrapper.exe file is located?  You
> have several relative paths in your configuration which in itself should
> be fine.
>
> 3) When you ran your "-version" test with the testwrapper configuration
> file, what was the full path to that wrapper.exe?
>
> 4) In case either of the above two paths contain non-ascii characters,
> could you try your test in a simple directory like c:\myapp\ ?
>
> Cheers,
> Leif
>
> Vincent Finn wrote:
> > Ok,
> >
> > If there is anything I can do to help narrow it down further let me know.
> >
> >     Vin
> >
> > On Tue, Jun 24, 2008 at 6:12 PM, Leif Mortenson
> > <le...@ta... <mailto:le...@ta...>> wrote:
> >
> >     Vincent,
> >     One of the lines of output from your log file does not display the
> >     error message that should be telling me what the cause of your
> >     problems is.
> >
> >     Your log file has the line:
> >     Unable to execute Java command.
> >
> >     But when I run the released Windows 3.3.0 release it generates the
> >     following message:
> >     Unable to execute Java command.  The system cannot find the file
> >     specified. (0x2)
> >
> >     That error message is generated with the following code:
> >                log_printf(WRAPPER_SOURCE_WRAPPER, LEVEL_FATAL,
> >                    "Unable to execute Java command.  %s",
> >     getLastErrorText());
> >
> >     For some reason, your log output does not include the error
> >     message and I am
> >     trying to understand why.  I had added that error message back in
> 2004
> >     to help track
> >     down the cause of problems like this.
> >
> >     If as I assume, you are indeed using the official release then I need
> >     to figure out why
> >     the getLastErrorText() function is returning "" on your machine.
>  That
> >     line in your log
> >     file does include trailing spaces so I think this is what is
> >     happening.
> >
> >     Thanks,
> >     Leif
> >
> >
> >
> >     On Tue, Jun 24, 2008 at 10:14 PM, Vincent Finn <v.f...@gm...
> >     <mailto:v.f...@gm...>> wrote:
> >     > Hi,
> >     >
> >     > Yeah, I am just running it in from the console.
> >     >
> >     > I have attached the log file but there not a lot in it.
> >     >
> >     > I also tried the trivial sample attached, all it does is run "java
> >     > -version".
> >     > That fails with the same error.
> >     >
> >     >     Vin
> >     >
> >     > On Tue, Jun 24, 2008 at 12:47 PM, Leif Mortenson
> >     <le...@ta... <mailto:le...@ta...>>
> >     > wrote:
> >     >>
> >     >> Vincent,
> >     >> To confirm, when you run with the Wrapper, you are running in a
> >     console
> >     >> and not
> >     >> as a Service correct?
> >     >>
> >     >> Looking at your wrapper.conf, I did not see anything that stood
> >     out as
> >     >> incorrect.
> >     >>
> >     >> Could you please set the wrapper.debug=true property and then
> >     reply back
> >     >> with
> >     >> the full resulting wrapper.log file.  Delete the only file
> >     before you
> >     >> start so we only
> >     >> get a single JVM run.
> >     >>
> >     >> Thanks,
> >     >> Leif
> >     >>
> >     >> On Tue, Jun 24, 2008 at 7:08 PM, Vincent Finn
> >     <v.f...@gm... <mailto:v.f...@gm...>> wrote:
> >     >> > Hi,
> >     >> >
> >     >> > I have seen this error mentioned in the forum but it doesn't
> >     to be for
> >     >> > the
> >     >> > same problem as I am having.
> >     >> >
> >     >> > I am using JSW 3.3.0 on Windows XP (32bit).
> >     >> > I am simply trying to run the batch file in the bin folder
> >     >> > (serviceName.bat)
> >     >> > and I get the error
> >     >> > "Critical error: wait for JVM process failed"
> >     >> >
> >     >> > I have attached my config file, and a file containing the
> >     command line
> >     >> > output by the wrapper (without the -Dwrapper.key argument).
> >     >> >
> >     >> > When I run the wrapper the output is:
> >     >> >
> >     >> >   wrapper  | Launching a JVM...
> >     >> >   wrapper  | Unable to execute Java command.
> >     >> >   wrapper  |     C:\...\java ...
> >     >> >   wrapper  |  Critical error: wait for JVM process failed
> >     >> >
> >     >> >
> >     >> > If I run the command directly on the command line it works
> >     perfectly, it
> >     >> > only fails when run through the wrapper.
> >     >> >
> >     >> > Any ideas as to what might be wrong?
> >     >> >
> >     >> >     Thanks, Vin
> >     >> >
> >
>
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>

Re: [Wrapper-user] Using the wrapper integration method 3

[Leif M. <le...@ta...>]

Jay,
The Wrapper will exit if all non-daemon threads have completed.  This is 
how standalone Java works as well.

Could you please set the wrapper.debug=true property and then reply with 
the resulting wrapper.log attached.  I only need to see a single Wrapper 
invocation so delete the old wrapper.log file first.   The log should 
allow me to say what your exact problem is.

Cheers,
Leif

Jay...@sc... wrote:
> Hi,
>  
> I'm playing around with the service wrapper and trying to get an example
> running using integration method 3.  I'm running on the Mac OS X.  
>  
> Everything appears to work. The service starts up when I use the start
> script (sh.script.ln) except that after a few seconds the infrastructure
> appears to call a stop on my service.
>  
> Does anyone know the reasons why the stop would be called?  I've
> returned null from my start method.  I spun up my logic in a thread
> although I did the thread start in a wierd way.  (what if the thread
> exited (that's not necessarily happening but I'm trying to learn)?).
>  
> My service is a very simple loop where I'm just printing a message to
> std out every second. 
>  
> Thanks for any help!
>

Re: [Wrapper-user] Critical error: wait for JVM process failed

[Leif M. <le...@ta...>]

Vin,
Given that the error message is not displaying correctly, I am wondering 
if there is a general character encoding problem.  The Wrapper is not 
yet localized and double byte safe.  This is on our todo list.

1) What is the locale of your Windows?  ie.  Japanese, English, etc.

2) What is the full path to where the wrapper.exe file is located?  You 
have several relative paths in your configuration which in itself should 
be fine.

3) When you ran your "-version" test with the testwrapper configuration 
file, what was the full path to that wrapper.exe?

4) In case either of the above two paths contain non-ascii characters, 
could you try your test in a simple directory like c:\myapp\ ?

Cheers,
Leif

Vincent Finn wrote:
> Ok,
>
> If there is anything I can do to help narrow it down further let me know.
>
>     Vin
>
> On Tue, Jun 24, 2008 at 6:12 PM, Leif Mortenson 
> <le...@ta... <mailto:le...@ta...>> wrote:
>
>     Vincent,
>     One of the lines of output from your log file does not display the
>     error message that should be telling me what the cause of your
>     problems is.
>
>     Your log file has the line:
>     Unable to execute Java command.
>
>     But when I run the released Windows 3.3.0 release it generates the
>     following message:
>     Unable to execute Java command.  The system cannot find the file
>     specified. (0x2)
>
>     That error message is generated with the following code:
>                log_printf(WRAPPER_SOURCE_WRAPPER, LEVEL_FATAL,
>                    "Unable to execute Java command.  %s",
>     getLastErrorText());
>
>     For some reason, your log output does not include the error
>     message and I am
>     trying to understand why.  I had added that error message back in 2004
>     to help track
>     down the cause of problems like this.
>
>     If as I assume, you are indeed using the official release then I need
>     to figure out why
>     the getLastErrorText() function is returning "" on your machine.  That
>     line in your log
>     file does include trailing spaces so I think this is what is
>     happening.
>
>     Thanks,
>     Leif
>
>
>
>     On Tue, Jun 24, 2008 at 10:14 PM, Vincent Finn <v.f...@gm...
>     <mailto:v.f...@gm...>> wrote:
>     > Hi,
>     >
>     > Yeah, I am just running it in from the console.
>     >
>     > I have attached the log file but there not a lot in it.
>     >
>     > I also tried the trivial sample attached, all it does is run "java
>     > -version".
>     > That fails with the same error.
>     >
>     >     Vin
>     >
>     > On Tue, Jun 24, 2008 at 12:47 PM, Leif Mortenson
>     <le...@ta... <mailto:le...@ta...>>
>     > wrote:
>     >>
>     >> Vincent,
>     >> To confirm, when you run with the Wrapper, you are running in a
>     console
>     >> and not
>     >> as a Service correct?
>     >>
>     >> Looking at your wrapper.conf, I did not see anything that stood
>     out as
>     >> incorrect.
>     >>
>     >> Could you please set the wrapper.debug=true property and then
>     reply back
>     >> with
>     >> the full resulting wrapper.log file.  Delete the only file
>     before you
>     >> start so we only
>     >> get a single JVM run.
>     >>
>     >> Thanks,
>     >> Leif
>     >>
>     >> On Tue, Jun 24, 2008 at 7:08 PM, Vincent Finn
>     <v.f...@gm... <mailto:v.f...@gm...>> wrote:
>     >> > Hi,
>     >> >
>     >> > I have seen this error mentioned in the forum but it doesn't
>     to be for
>     >> > the
>     >> > same problem as I am having.
>     >> >
>     >> > I am using JSW 3.3.0 on Windows XP (32bit).
>     >> > I am simply trying to run the batch file in the bin folder
>     >> > (serviceName.bat)
>     >> > and I get the error
>     >> > "Critical error: wait for JVM process failed"
>     >> >
>     >> > I have attached my config file, and a file containing the
>     command line
>     >> > output by the wrapper (without the -Dwrapper.key argument).
>     >> >
>     >> > When I run the wrapper the output is:
>     >> >
>     >> >   wrapper  | Launching a JVM...
>     >> >   wrapper  | Unable to execute Java command.
>     >> >   wrapper  |     C:\...\java ...
>     >> >   wrapper  |  Critical error: wait for JVM process failed
>     >> >
>     >> >
>     >> > If I run the command directly on the command line it works
>     perfectly, it
>     >> > only fails when run through the wrapper.
>     >> >
>     >> > Any ideas as to what might be wrong?
>     >> >
>     >> >     Thanks, Vin
>     >> >
>

Re: [Wrapper-user] Critical error: wait for JVM process failed

[Vincent F. <v.f...@gm...>]

Ok,

If there is anything I can do to help narrow it down further let me know.

    Vin

On Tue, Jun 24, 2008 at 6:12 PM, Leif Mortenson <le...@ta...>
wrote:

> Vincent,
> One of the lines of output from your log file does not display the
> error message that should be telling me what the cause of your
> problems is.
>
> Your log file has the line:
> Unable to execute Java command.
>
> But when I run the released Windows 3.3.0 release it generates the
> following message:
> Unable to execute Java command.  The system cannot find the file
> specified. (0x2)
>
> That error message is generated with the following code:
>            log_printf(WRAPPER_SOURCE_WRAPPER, LEVEL_FATAL,
>                "Unable to execute Java command.  %s", getLastErrorText());
>
> For some reason, your log output does not include the error message and I
> am
> trying to understand why.  I had added that error message back in 2004
> to help track
> down the cause of problems like this.
>
> If as I assume, you are indeed using the official release then I need
> to figure out why
> the getLastErrorText() function is returning "" on your machine.  That
> line in your log
> file does include trailing spaces so I think this is what is happening.
>
> Thanks,
> Leif
>
>
>
> On Tue, Jun 24, 2008 at 10:14 PM, Vincent Finn <v.f...@gm...> wrote:
> > Hi,
> >
> > Yeah, I am just running it in from the console.
> >
> > I have attached the log file but there not a lot in it.
> >
> > I also tried the trivial sample attached, all it does is run "java
> > -version".
> > That fails with the same error.
> >
> >     Vin
> >
> > On Tue, Jun 24, 2008 at 12:47 PM, Leif Mortenson <
> le...@ta...>
> > wrote:
> >>
> >> Vincent,
> >> To confirm, when you run with the Wrapper, you are running in a console
> >> and not
> >> as a Service correct?
> >>
> >> Looking at your wrapper.conf, I did not see anything that stood out as
> >> incorrect.
> >>
> >> Could you please set the wrapper.debug=true property and then reply back
> >> with
> >> the full resulting wrapper.log file.  Delete the only file before you
> >> start so we only
> >> get a single JVM run.
> >>
> >> Thanks,
> >> Leif
> >>
> >> On Tue, Jun 24, 2008 at 7:08 PM, Vincent Finn <v.f...@gm...>
> wrote:
> >> > Hi,
> >> >
> >> > I have seen this error mentioned in the forum but it doesn't to be for
> >> > the
> >> > same problem as I am having.
> >> >
> >> > I am using JSW 3.3.0 on Windows XP (32bit).
> >> > I am simply trying to run the batch file in the bin folder
> >> > (serviceName.bat)
> >> > and I get the error
> >> > "Critical error: wait for JVM process failed"
> >> >
> >> > I have attached my config file, and a file containing the command line
> >> > output by the wrapper (without the -Dwrapper.key argument).
> >> >
> >> > When I run the wrapper the output is:
> >> >
> >> >   wrapper  | Launching a JVM...
> >> >   wrapper  | Unable to execute Java command.
> >> >   wrapper  |     C:\...\java ...
> >> >   wrapper  |  Critical error: wait for JVM process failed
> >> >
> >> >
> >> > If I run the command directly on the command line it works perfectly,
> it
> >> > only fails when run through the wrapper.
> >> >
> >> > Any ideas as to what might be wrong?
> >> >
> >> >     Thanks, Vin
> >> >
> >>
> >>
> -------------------------------------------------------------------------
> >> Check out the new SourceForge.net Marketplace.
> >> It's the best place to buy or sell services for
> >> just about anything Open Source.
> >> http://sourceforge.net/services/buy/index.php
> >> _______________________________________________
> >> Wrapper-user mailing list
> >> Wra...@li...
> >> https://lists.sourceforge.net/lists/listinfo/wrapper-user
> >
> >
> > -------------------------------------------------------------------------
> > Check out the new SourceForge.net Marketplace.
> > It's the best place to buy or sell services for
> > just about anything Open Source.
> > http://sourceforge.net/services/buy/index.php
> > _______________________________________________
> > Wrapper-user mailing list
> > Wra...@li...
> > https://lists.sourceforge.net/lists/listinfo/wrapper-user
> >
> >
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>

[Wrapper-user] Using the wrapper integration method 3

[<Jay...@sc...>]

Hi,
 
I'm playing around with the service wrapper and trying to get an example
running using integration method 3.  I'm running on the Mac OS X.  
 
Everything appears to work. The service starts up when I use the start
script (sh.script.ln) except that after a few seconds the infrastructure
appears to call a stop on my service.
 
Does anyone know the reasons why the stop would be called?  I've
returned null from my start method.  I spun up my logic in a thread
although I did the thread start in a wierd way.  (what if the thread
exited (that's not necessarily happening but I'm trying to learn)?).
 
My service is a very simple loop where I'm just printing a message to
std out every second. 
 
Thanks for any help!
 

Re: [Wrapper-user] Critical error: wait for JVM process failed

[Leif M. <le...@ta...>]

Vincent,
One of the lines of output from your log file does not display the
error message that should be telling me what the cause of your
problems is.

Your log file has the line:
Unable to execute Java command.

But when I run the released Windows 3.3.0 release it generates the
following message:
Unable to execute Java command.  The system cannot find the file
specified. (0x2)

That error message is generated with the following code:
            log_printf(WRAPPER_SOURCE_WRAPPER, LEVEL_FATAL,
                "Unable to execute Java command.  %s", getLastErrorText());

For some reason, your log output does not include the error message and I am
trying to understand why.  I had added that error message back in 2004
to help track
down the cause of problems like this.

If as I assume, you are indeed using the official release then I need
to figure out why
the getLastErrorText() function is returning "" on your machine.  That
line in your log
file does include trailing spaces so I think this is what is happening.

Thanks,
Leif



On Tue, Jun 24, 2008 at 10:14 PM, Vincent Finn <v.f...@gm...> wrote:
> Hi,
>
> Yeah, I am just running it in from the console.
>
> I have attached the log file but there not a lot in it.
>
> I also tried the trivial sample attached, all it does is run "java
> -version".
> That fails with the same error.
>
>     Vin
>
> On Tue, Jun 24, 2008 at 12:47 PM, Leif Mortenson <le...@ta...>
> wrote:
>>
>> Vincent,
>> To confirm, when you run with the Wrapper, you are running in a console
>> and not
>> as a Service correct?
>>
>> Looking at your wrapper.conf, I did not see anything that stood out as
>> incorrect.
>>
>> Could you please set the wrapper.debug=true property and then reply back
>> with
>> the full resulting wrapper.log file.  Delete the only file before you
>> start so we only
>> get a single JVM run.
>>
>> Thanks,
>> Leif
>>
>> On Tue, Jun 24, 2008 at 7:08 PM, Vincent Finn <v.f...@gm...> wrote:
>> > Hi,
>> >
>> > I have seen this error mentioned in the forum but it doesn't to be for
>> > the
>> > same problem as I am having.
>> >
>> > I am using JSW 3.3.0 on Windows XP (32bit).
>> > I am simply trying to run the batch file in the bin folder
>> > (serviceName.bat)
>> > and I get the error
>> > "Critical error: wait for JVM process failed"
>> >
>> > I have attached my config file, and a file containing the command line
>> > output by the wrapper (without the -Dwrapper.key argument).
>> >
>> > When I run the wrapper the output is:
>> >
>> >   wrapper  | Launching a JVM...
>> >   wrapper  | Unable to execute Java command.
>> >   wrapper  |     C:\...\java ...
>> >   wrapper  |  Critical error: wait for JVM process failed
>> >
>> >
>> > If I run the command directly on the command line it works perfectly, it
>> > only fails when run through the wrapper.
>> >
>> > Any ideas as to what might be wrong?
>> >
>> >     Thanks, Vin
>> >
>>
>> -------------------------------------------------------------------------
>> Check out the new SourceForge.net Marketplace.
>> It's the best place to buy or sell services for
>> just about anything Open Source.
>> http://sourceforge.net/services/buy/index.php
>> _______________________________________________
>> Wrapper-user mailing list
>> Wra...@li...
>> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>

Re: [Wrapper-user] Double byte support

[Leif M. <le...@ta...>]

Gary,
Unfortunately no.  The Wrapper is not yet fully support double byte characters.
We use it on Japanese systems all the time, but not in paths containing double
byte characters.

This is on our to do list, but for the time being this is not supported.

Cheers,
Leif

On Tue, Jun 24, 2008 at 10:12 PM, Gary Rosenfeld
<g_r...@ho...> wrote:
> Does Wrapper support double bytes.
> We have a japanese machine and when our app with JRE and Wrapper is
> installed under a directory name that uses english letters only, everything
> works fine.  But if the install directory is named using japanese
> characters, we are getting:
> Error occurred during initialization of VM, java.lang.UnsatisfiedLinkError:
> no zip in java.library.path. in the wrapper.log
> Of course, zip.dll is in jre/bin and everything else looks good too.
> Intererstingly, I cannot use DEPENDS.exe for dlls in this directory,
> getting "invalid file path or name", but it works fine dor dlls in other
> directories with english names.
>
> Thank you very much,
>
> Gary Rosenfeld.
>
> By the way, it is a great product, easy to use, not buggy, and has a lot
> of great features.

[Wrapper-user] Double byte support

[Gary R. <g_r...@ho...>]

Does Wrapper support double bytes.We have a japanese machine and when our app with JRE and Wrapper is installed under a directory name that uses english letters only, everything works fine.  But if the install directory is named using japanese characters, we are getting: Error occurred during initialization of VM, java.lang.UnsatisfiedLinkError: no zip in java.library.path. in the wrapper.logOf course, zip.dll is in jre/bin and everything else looks good too.  Intererstingly, I cannot use DEPENDS.exe for dlls in this directory, getting "invalid file path or name", but it works fine dor dlls in other directories with english names.  Thank you very much, Gary Rosenfeld. By the way, it is a great product, easy to use, not buggy, and has a lot of great features. 
_________________________________________________________________
Earn cashback on your purchases with Live Search - the search that pays you back!
http://search.live.com/cashback/?&pkw=form=MIJAAF/publ=HMTGL/crea=earncashback

Re: [Wrapper-user] Critical error: wait for JVM process failed

[Leif M. <le...@ta...>]

Vincent,
To confirm, when you run with the Wrapper, you are running in a console and not
as a Service correct?

Looking at your wrapper.conf, I did not see anything that stood out as
incorrect.

Could you please set the wrapper.debug=true property and then reply back with
the full resulting wrapper.log file.  Delete the only file before you
start so we only
get a single JVM run.

Thanks,
Leif

On Tue, Jun 24, 2008 at 7:08 PM, Vincent Finn <v.f...@gm...> wrote:
> Hi,
>
> I have seen this error mentioned in the forum but it doesn't to be for the
> same problem as I am having.
>
> I am using JSW 3.3.0 on Windows XP (32bit).
> I am simply trying to run the batch file in the bin folder (serviceName.bat)
> and I get the error
> "Critical error: wait for JVM process failed"
>
> I have attached my config file, and a file containing the command line
> output by the wrapper (without the -Dwrapper.key argument).
>
> When I run the wrapper the output is:
>
>   wrapper  | Launching a JVM...
>   wrapper  | Unable to execute Java command.
>   wrapper  |     C:\...\java ...
>   wrapper  |  Critical error: wait for JVM process failed
>
>
> If I run the command directly on the command line it works perfectly, it
> only fails when run through the wrapper.
>
> Any ideas as to what might be wrong?
>
>     Thanks, Vin
>

[Wrapper-user] Integration Method #3, Close Event of console not handled

[Philipp K. <phi...@gm...>]

Hello all,

i am implementing a WrapperListener with this code as in the websites example:
public void controlEvent(final int event) {
	if (WrapperManager.isControlledByNativeWrapper()) {
		// The Wrapper will take care of this event
	} else {
		// We are not being controlled by the Wrapper, so
		//  handle the event ourselves.
		if (event == WrapperManager.WRAPPER_CTRL_C_EVENT
				|| event == WrapperManager.WRAPPER_CTRL_CLOSE_EVENT
				|| event == WrapperManager.WRAPPER_CTRL_SHUTDOWN_EVENT) {
			WrapperManager.stop(0);
		}
	}
}

However, the Wrapper does not take care of the
WRAPPER_CTRL_CLOSE_EVENT when I want to close the console. Instead it
does nothing and Windows thinks the application hangs. However I can
close the app by pressing CTRL-C.

Any idea whats going on?

Thanks,
Phil
-- 
The Dude Abides!

[Wrapper-user] Critical error: wait for JVM process failed

[Vincent F. <v.f...@gm...>]

"C:\Java\jdk1.6.0_05\bin\java" -Xbootclasspath/p:.;C:\GigaSpaces\xap-6.5-rc2\bin\\..\lib\xml\serializer.jar;C:\GigaSpaces\xap-6.5-rc2\bin\\..\lib\xml\xalan.jar;C:\GigaSpaces\xap-6.5-rc2\bin\\..\lib\xml\xercesImpl.jar;C:\GigaSpaces\xap-6.5-rc2\bin\\..\lib\xml\xml-apis.jar -server -Xmx512m -Dcom.gs.jini_lus.locators= -Dcom.gs.jini_lus.groups="gigaspaces-6.5-XAP-rc2" -Dsun.rmi.dgc.client.gcInterval=36000000 -Dsun.rmi.dgc.server.gcInterval=36000000 -Djava.rmi.server.hostname=""CM-DT-VFINN"" -Djava.rmi.server.RMIClassLoaderSpi=default -Djava.rmi.server.logCalls=false -Dcom.gs.home=C:\GigaSpaces\xap-6.5-rc2\bin\\.. -Dcom.gs.start-embedded-lus=true -Dcom.gs.start-embedded-mahalo=false -Dcom.gs.logging.debug=false -Djava.util.logging.config.file="C:\GigaSpaces\xap-6.5-rc2\bin\\../config/gs_logging.properties" -Djava.library.path="../lib" -classpath "../lib/wrapper.jar;../lib/wrappertest.jar;C:\GigaSpaces\xap-6.5-rc2\bin\\..\lib\jdbc\h2.jar;C:\GigaSpaces\xap-6.5-rc2\bin\\..\lib\jdbc\hsqldb.jar;C:\GigaSpaces\xap-6.5-rc2\bin\\..;C:\GigaSpaces\xap-6.5-rc2\bin\\..\lib\JSpaces.jar" -Dwrapper.port=32000 -Dwrapper.jvm.port.min=31000 -Dwrapper.jvm.port.max=31999 -Dwrapper.pid=2160 -Dwrapper.version="3.3.0" -Dwrapper.native_library="wrapper" -Dwrapper.cpu.timeout="10" -Dwrapper.jvmid=1 org.tanukisoftware.wrapper.WrapperSimpleApp com.j_spaces.core.client.SpaceFinder "/./cmlSpace"

Re: [Wrapper-user] Service that starts automatically when an application server is restarted

[pram <pra...@ya...>]

Hi all,

I figured out the solution for this in OC4J. There was no need to use the
Wrapper.

Thanks,
Pram

sstair wrote:
> 
> This sounds like something outside of the scope of the Wrapper, and more
> like you want an JBoss JMX MBean.
> 
> 
> pram wrote:
>> 
>> Hi all,
>> 
>> I have the following requirement:
>> There is a Java application deployed in an application server(OC4J).
>> I need a mechanism by which the application should be automatically
>> started and initialized as soon as the application server is
>> started/restarted, and the application should be terminated when the
>> application server goes down.
>> 
>> My question is, can this be achieved using Java Service Wrapper? Can I
>> add some kind of application server dependency in the wrapper.config
>> file, or something similar to that?
>> 
>> Any inputs are highly appreciated.
>> 
>> Thanks,
>> pram 
>> 
> 
> 

-- 
View this message in context: http://www.nabble.com/Service-that-starts-automatically-when-an-application-server-is-restarted-tp17789574p18024044.html
Sent from the Java Service Wrapper mailing list archive at Nabble.com.

[Wrapper-user] Released Itanium versions of 3.3.0

[Leif M. <le...@ta...>]

Try this again with the correct URL:

Hi all,
A quick update.  We released Windows, Linux and HPUX versions of the
Wrapper for the Itanium platform.
These use the same codebase as 3.3.0.

Some users have asked for these so if you are one of them, please give
them a try:
http://wrapper.tanukisoftware.org/doc/english/download.jsp

Cheers,

Leif

[Wrapper-user] Released Itanium versions of 3.3.0

[Leif M. <le...@ta...>]

Hi all,
A quick update.  We released Windows, Linux and HPUX versions of the
Wrapper for the Itanium platform.
These use the same codebase as 3.3.0.

Some users have asked for these so if you are one of them, please give
them a try:
http://localhost:8280/doc/english/download.jsp

Cheers,
Leif

Re: [Wrapper-user] Service that starts automatically when an application server is restarted

[Leif M. <le...@ta...>]

Pram,
The Wrapper configuration makes it possible to register
dependent Windows services when installing a service.

The Windows ServiceManager is responsible for controlling
when the Wrapper and its dependent service are started
and stopped.

I added some more documentation to the web site which
describes all of the possible states.
http://wrapper.tanukisoftware.org/doc/english/prop-ntservice-dependency-n.html

It may be possible to add the functionality you are asking
for but it would require modifications to the Wrapper.

Please let me know if you have any questions.

Cheers,
Leif

pram wrote:
> Hi all,
>
> I have the following requirement:
> There is a Java application deployed in an application server(OC4J).
> I need a mechanism by which the application should be automatically started
> and initialized as soon as the application server is started/restarted, and
> the application should be terminated when the application server goes down.
>
> My question is, can this be achieved using Java Service Wrapper? Can I add
> some kind of application server dependency in the wrapper.config file, or
> something similar to that?
>
> Any inputs are highly appreciated.
>
> Thanks,
> pram 
>   

Re: [Wrapper-user] Windows-IA64 and tanuki wrapper relationship

[Leif M. <le...@ta...>]

Lauren,
The Itanium and AMD 64-bit implementations are quite
different and require different binaries.  Thee 32-bit
instruction set is the same on both chips so the 32-bit
x86 version will work.

Great news though, We have actually be working on the
IA64 version this week and was able to release it just
now.   We did some testing on our server but would
appreciate it if you could give it a try and let me know
how it works for you.

It can be downloaded from our download page:
http://wrapper.tanukisoftware.org/doc/english/download.jsp

Cheers,
Leif


Domingo, Lauren wrote:
> I am trying to see if there could a working configuration between my IA 64 Windows system and a version of tanuki wrapper.  Off of the tanuki website, for Windows, there is only a 32 and 64 bit for x86 but when trying to run the 64 bit 86x Windows wrapper version on my IA 64 system, I get a pop-up error message:
>
> The image file C:\Program Files\wrapper-windows-x86-64.3.3.0-st\bin\wrapper.exe is vaild, but is for a machine type other than the current current machine.
>
> I would appreciate any advice on how to run this wrapper with my IA 64 Windows system and IA 64 Java service.
>
> Thank you,
> Lauren
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>   

[Wrapper-user] Windows-IA64 and tanuki wrapper relationship

[Domingo, L. <lau...@hp...>]

I am trying to see if there could a working configuration between my IA 64 Windows system and a version of tanuki wrapper.  Off of the tanuki website, for Windows, there is only a 32 and 64 bit for x86 but when trying to run the 64 bit 86x Windows wrapper version on my IA 64 system, I get a pop-up error message:

The image file C:\Program Files\wrapper-windows-x86-64.3.3.0-st\bin\wrapper.exe is vaild, but is for a machine type other than the current current machine.

I would appreciate any advice on how to run this wrapper with my IA 64 Windows system and IA 64 Java service.

Thank you,
Lauren