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

[Wrapper-user] Failed to connect to the Wrapper error, service will not start

[Buchanan, S. G. <STE...@sa...>]

Wrapper service will not start, getting the following error.  I have the
same service running on several dell boxes but having problem with on one of
the dell boxes.  

 

STATUS | wrapper  | 2006/02/13 12:40:08 | Launching a JVM...

DEBUG  | wrapper  | 2006/02/13 12:40:08 | command:
"..\j2re1.4.2_03\bin\java" -Xss10M -Xms10m -Xmx256m
-Djava.library.path="../lib" -classpath
"../lib/wrapper.jar;../SitOhiServer.jar;../"
-Dwrapper.key="gKvLC5ktpo1QecnA" -Dwrapper.port=32000 -Dwrapper.debug="TRUE"
-Dwrapper.service="TRUE" -Dwrapper.cpu.timeout="10" -Dwrapper.jvmid=5
org.tanukisoftware.wrapper.WrapperSimpleApp sitohi.server.SitOhiServer

DEBUG  | wrapper  | 2006/02/13 12:40:08 | Java Virtual Machine started
(PID=2460)

INFO   | jvm 5    | 2006/02/13 12:40:08 | Wrapper Manager: JVM #5

INFO   | jvm 5    | 2006/02/13 12:40:08 | Wrapper Manager: Registering
shutdown hook

INFO   | jvm 5    | 2006/02/13 12:40:08 | Wrapper Manager: Using wrapper

INFO   | jvm 5    | 2006/02/13 12:40:08 | Calling native initialization
method.

INFO   | jvm 5    | 2006/02/13 12:40:08 | Initializing WrapperManager native
library.

INFO   | jvm 5    | 2006/02/13 12:40:08 | Java Executable:
C:\SitOhiServer\j2re1.4.2_03\bin\java.exe

INFO   | jvm 5    | 2006/02/13 12:40:08 | Java Version   : 1.4.2_03-b02 Java
HotSpot(TM) Client VM

INFO   | jvm 5    | 2006/02/13 12:40:08 | Java VM Vendor : Sun Microsystems
Inc.

INFO   | jvm 5    | 2006/02/13 12:40:08 | 

INFO   | jvm 5    | 2006/02/13 12:40:08 | Wrapper (Version 3.0.5)

INFO   | jvm 5    | 2006/02/13 12:40:08 | 

INFO   | jvm 5    | 2006/02/13 12:40:08 | Open socket to wrapper...

INFO   | jvm 5    | 2006/02/13 12:40:08 | Failed to connect to the Wrapper.

INFO   | jvm 5    | 2006/02/13 12:40:08 | java.net.ConnectException:
Connection refused: connect

INFO   | jvm 5    | 2006/02/13 12:40:08 | Exiting JVM...

INFO   | jvm 5    | 2006/02/13 12:40:08 | Send a packet STOP : 1

INFO   | jvm 5    | 2006/02/13 12:40:09 | Send a packet STOPPED : 1

ERROR  | wrapper  | 2006/02/13 12:40:09 | JVM exited while loading the
application.

DEBUG  | wrapper  | 2006/02/13 12:40:09 | JVM was only running for 1 seconds
leading to a failed restart count of 5.

FATAL  | wrapper  | 2006/02/13 12:40:09 | There were 5 failed launches in a
row, each lasting less than 300 seconds.  Giving up.

FATAL  | wrapper  | 2006/02/13 12:40:09 |   There may be a configuration
problem: please check the logs.

STATUS | wrapper  | 2006/02/13 12:40:09 | <-- Wrapper Stopped

 

 

Re: [Wrapper-user] Linux daemon question

[Anat H. <an...@en...>]

Hi Leif,

Thanks for your response. However, it doesn't seem to be working for me :'(

Leif Mortenson wrote:
> Anat,
>    Looking at your log file, this is the cause of the application 
> being shut down:
> INFO   | jvm 1    | 2006/02/07 16:42:51 | Processing control 
> event(WRAPPER_CTRL_SHUTDOWN_EVENT)
>
>    There is a bug in 3.1.2 where the JVM is not correctly ignoring 
> TERM signals
> even if the IGNORE_SIGNALS flag is set in the sh script.  This has 
> been fixed
> for the unreleased 3.2.0.  Could you please download and try out the 
> latest
> snapshot build and verify that this fixes the problem for you?
>
>    
> http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-linux-x86-32-3.2.0-d.tar.gz 
>
>
>    You will want to use the new sh script and set the IGNORE_SIGNALS
> variable to true in the script.
I've tried the new files, with the property set to true, but it still 
shutting down when I logout. Attached are 4 logs - for a console run, 
and service run which I stopped using <app> stop - those have normal 
behavior. The other 2 logs are of service runs during which I logout - 
one running remote through VNC, and the other is run locally.
>
>    Also, from your logs, it looks like there is a GUI being 
> displayed?   That may
> why your JVM is getting the TERM signals in the first place.
Could you explain about that?
> In general, daemon
> processes don't have GUIs.  You should add a call to
> WrapperManager.isLaunchedAsService()  This method will return true if the
> Wrapper is running as a daemon process.  Only show your GUI if it is 
> false.
> If you don't want to add a dependency to the wrapper API, look for the
> "wrapper.service" system property.
I'm not actually showing the GUI, it's hidden when in service mode - 
it's just a temporary (messy) solution before I remove the GUI. Is there 
a problem with that?


Thanks,
Anat
>
> Cheers,
> Leif
>
> Anat Halpern wrote:
>> Hi Leif,
>>
>> Attached is the log file. I hope it'll help.
>>
>> Thanks,
>> Anat
>>
>>   
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log 
> files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
> +++++++++++++++++++++++++++++++++++++++++++
> This Mail Was Scanned By Mail-seCure System
> at the Tel-Aviv University CC.

Re: [Wrapper-user] Wrapper pre release 3.2.0-d

[Michael 'b. S. <msc...@gi...>]

Good morning.

Going to test these babies right away.
How long do you plan, if no major bugs appear, until the release?

I know this is an awkward question but I definetly _need_ 3.2.0 :]
Therese something I coded against a 3.2.0-b and I guess my boss starts bitching 
about using an "unstable/unreleased" version. Which he is sort of right with.

Greetings,
 Michael

On Sat, 11 Feb 2006 - 11:13am, Leif Mortenson wrote:

>Sigh. From the logs, I noticed that I screwed up these links...
>Here are the corrected ones.
>
>Leif Mortenson wrote:
>> Hi all,
>> I have been trying to get things tied up for the 3.2.0 release and have made
>> the latest pre-release builds available for testing. THESE VERSIONS WILL NOT
>> BE SUPPORTED AND SHOULD NOT BE USED IN RELEASED SOFTWARE.
>> If you could give them a try, I would appreciate hearing about any problems
>> before the actual release. Normally I don't do this, but it has been a
>> long time
>> since the last official release.
>>
>> The following are the list of platforms that I am able to build personally.
>>   
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-linux-ppc-64-3.2.0-d.tar.gz
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-linux-x86-32-3.2.0-d.tar.gz
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-linux-x86-64-3.2.0-d.tar.gz
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-macosx-ppc-32-3.2.0-d.tar.gz
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-solaris-sparc-32-3.2.0-d.tar.gz
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-solaris-x86-32-3.2.0-d.tar.gz
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-windows-x86-32-3.2.0-d.zip
>
>> Additional make files for this release exist for the following versions but
>> have not been tested for this release:
>> aix-ppc-32
>> freebsd-x86-32
>> hpux-parisc-32
>> hpux-parisc-64
>> irix-mips-32
>> osf1-alpha-32
>> solaris-sparc-64
>> windows-x86-64 (This make file needs work)
>>
>> In most cases with this release, you can simply download the prerelease
>> distribution and build a release for any of the above platforms by running
>> either "./build32.sh release" or "./build64.sh release" as appropriate.
>> This will make use of a prebuilt jar and
>>   
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper_prerelease_3.2.0-d.tar.gz
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper_prerelease_3.2.0-d.zip
>> If you need to build the java source then you should do so using the
>> following
>> source distributions by running "./build32.sh pre-release" followed
>> by "./build32.sh release".
>>   
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper_3.2.0-d_src_with_doc_src.tar.gz
>http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper_3.2.0-d_src_with_doc_src.zip
>> Cheers,
>> Leif
>>   
>
>
>
>-------------------------------------------------------
>This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
>for problems?  Stop!  Download the new AJAX search engine that makes
>searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
>http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
>_______________________________________________
>Wrapper-user mailing list
>Wra...@li...
>https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>


-- 
  I love deadlines, especially the sound they make as they go whooshing by.

Re: [Wrapper-user] Wrapper pre release 3.2.0-d

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

Sigh. From the logs, I noticed that I screwed up these links...
Here are the corrected ones.

Leif Mortenson wrote:
> Hi all,
> I have been trying to get things tied up for the 3.2.0 release and have made
> the latest pre-release builds available for testing. THESE VERSIONS WILL NOT
> BE SUPPORTED AND SHOULD NOT BE USED IN RELEASED SOFTWARE.
> If you could give them a try, I would appreciate hearing about any problems
> before the actual release. Normally I don't do this, but it has been a
> long time
> since the last official release.
>
> The following are the list of platforms that I am able to build personally.
>   
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-linux-ppc-64-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-linux-x86-32-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-linux-x86-64-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-macosx-ppc-32-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-solaris-sparc-32-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-solaris-x86-32-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper-windows-x86-32-3.2.0-d.zip

> Additional make files for this release exist for the following versions but
> have not been tested for this release:
> aix-ppc-32
> freebsd-x86-32
> hpux-parisc-32
> hpux-parisc-64
> irix-mips-32
> osf1-alpha-32
> solaris-sparc-64
> windows-x86-64 (This make file needs work)
>
> In most cases with this release, you can simply download the prerelease
> distribution and build a release for any of the above platforms by running
> either "./build32.sh release" or "./build64.sh release" as appropriate.
> This will make use of a prebuilt jar and
>   
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper_prerelease_3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper_prerelease_3.2.0-d.zip
> If you need to build the java source then you should do so using the
> following
> source distributions by running "./build32.sh pre-release" followed
> by "./build32.sh release".
>   
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper_3.2.0-d_src_with_doc_src.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.2.0-d/wrapper_3.2.0-d_src_with_doc_src.zip
> Cheers,
> Leif
>   

Re: [Wrapper-user] BackingStoreException when RUN_AS_USER on OS X

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

Sean,
> I've run into an odd situation and I was wondering if anybody had any 
> suggestions?
>
> I'm using Java Service Wrapper 3.1.2 on Mac OS X 10.4.4 (not OS X 
> Server) running Java 1.4.2_09.
>
> I've got an app wrapped using WrapperSimpleApp as per the Integration 
> Methods instructions  Since I'm on a Mac the script and wrapper file 
> live in /Library/StartupItems/<appname>. With a bit of 
> wrapper.working.dir trickery I've got wrapper running my jar which 
> lives in /Library/MediaSpan/<appname>, which has uid:guid of 
> appserver:admin, as does my jar and its lib folder.
>
> My app subclasses java.rmi.server.UnicastRemoteObject to make it 
> configurable via RMI.  It stores its configuration via 
> java.util.prefs.Preferences, using the node returned by 
> Preferences.userNodeForPackage(<myclass>.class).  No call is ever made 
> to Preferences.systemNodeForPackage().
>
> When I *do not* RUN_AS_USER, the WrapperSimpleApp wrapper starts my 
> app, and it runs just fine.  I can connect via RMI and set my app's 
> configuration, which writes to the Preferences user node via the 
> putXXX and flush methods in Preferences.  The Preferences calls all 
> work fine.
Who are you when you are running in this case?  root?
> The trouble comes when I try to RUN_AS_USER=appserver.   When I make 
> that one change everything works like before *except* the 
> Preferences.flush call, which throws a BackingStoreException:
>
> java.util.prefs.BackingStoreException: Synchronization failed for node 
> <myclass's user node>
>
> I'm confused as to why this is happening for the RUN_AS_USER case, but 
> not when wrapper runs my app as root... has anybody else had to deal 
> with this?
>
> I searched the java-dev mailing list at Apple but the only mention of 
> BackingStoreException puts the blame on making flush calls on a system 
> pref node returned by Preferences. systemNodeForPackage.
Where are the preferences being stored?  My guess is that it is 
attempting to write them to a
location which the appserver user does not have write access.

The RUN_AS_USER setting in the script is quite simple.  It just uses su 
to launch the
wrapper as the specified user.  If that user does not have the required 
permissions to
access all required resouces, you will run into problems.

Since you are switching back and forth, make sure the wrapper.log and 
other files are
all ok.  If they are created with root, they may not be getting output 
when run as
appserver.  The wrapper is designed to fail silently if it can't write 
to the log file.

Cheers,
Leif

Re: [Wrapper-user] RE: Win2k access denied on service start (Resolved)

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

Marc,
    Thanks for posting back with the solution.  File system access 
permissions can be a pain.
Let me know if you have any ideas on how I can help users resolve these 
problems more
quickly.  Problem is if the Wrapper can't be run, it can't be run...

Cheers,
Leif

Marc T. Calello wrote:
> Original thread can be found here:
> http://sourceforge.net/mailarchive/message.php?msg_id=9692073
>
>
> After hours of scouring this mailing list I was able to resolve my own problem
> with "0x5 Access is denied." errors when starting my Windows 2000 service.
>
> I installed this service without specifying the wrapper.ntservice.account, so
> it is running with the default "LocalSystem" account.  I was also able to start
> the service from the batch script without any problems.  However, running it
> from the command line inherited my currently logged in user's (Administrator)
> credentials. Also, several other (built-in) services were starting up without
> problems.
>
> The issue was that the directory structure that the service was referencing
> (C:\myproject\wrapper.exe) had full access permissions granted to the "Users"
> group, however since my service runs as "System", the directory structure and
> all the files will also need read/execute permissions granted to the "System"
> user account.
>
> Hope this helps,
> -Marc
>   

Re: [Wrapper-user] Cannot bring up a browser

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

Senthil,
    One other thing.  Unless you specify an account to run as, the 
Wrapper runs as a service
using the default SYSTEM user.  This means that any child processes will 
also be run as
the system user.   Back when I was playing with Excel, I was having a 
problem launching
Excel as the SYSTEM user because Excel likes to bring up a configuration 
wizard when
it is launched for the first time for a given user.
    I am wondering if your browser is crashing because it is not finding 
its expected
configurations for the SYSTEM user.
    I was able to get around this with Excel by running the Wrapper as a 
user for which I
had already initialized Excel.  The problem for you though is that when 
running as a user
other than SYSTEM, it is not possible to set the interactive flag.  In 
my case, I was using
Excel to run a VBS script that generated a CSV file and exited.  No user 
interaction
required so it all worked.

    Not sure what browser you are using, but you might want to try 
googling around
using this as a base.

    The memory properties of the wrapper simply set the -Xms and -Xmx 
parameters to
the JVM, they have no control over the child processes.

Cheers,
Leif

Senthil S wrote:
> Leif,
>
> I am sure its the browser launch process that fails, because even 
> after that the application runs smoothly.
> The exact statement I use is,
>
>    Process p = Runtime.getRuntime.exec(rundll32 url.dll
>    ,FileProtocolHandler http://localhost:1090);
>    and also I tried
>     Runtime.getRuntime.exec(cmd.exe /c start http://localhost:1090);
>
> Where Runtime is java.lang.Runtime.
> If not from wrapper.maxmemory=300MB (initially it had 256MB), how do I 
> set the memory for the browser launch process?(should I set it at all?)
> Without interactive flag set , I was not getting this error message 
> ("The application failed to initialize properly (0xc0000142)" ).
> In both cases(with/without interactive flag set) the browser wasn't 
> launched.
>
> Thanks,
> Senthil
>
> Leif Mortenson wrote:
>
>> Senthil,
>>
>>> I even tried using "wrapper.ntservice.interactive=true" but all I 
>>> get is a message box saying
>>> "The application failed to initialize properly (0xc0000142)"
>>
>> What is process that is crashing?  Is this your JVM or the browser 
>> process?
>>
>>> I even tried increasing the maxmemory property to 300MB. Still it 
>>> doesn't work.
>>
>> I doubt that the memory is the problem.  That just sets the memory 
>> used by the JVM
>> launched by the Wrapper.
>>
>> So without the interactive flag set, the browser gets launched but 
>> does nothing.  But
>> when the flag is set, something is crashing?
>>
>> What is the RunTime class?  Is that something that you created?  Have 
>> you tried
>> launching the browser using the java.lang.Runtime.exec method?  I 
>> haven't launched
>> a browser before, but have launched Excel without any problems in the 
>> past.
>>
>> Cheers,
>> Leif
>>
>>> On 2/10/06, *Leif Mortenson* <le...@ta... 
>>> <mailto:le...@ta...>> wrote:
>>>
>>>     Senthil,
>>>         Are you setting the wrapper.ntservice.interactive
>>>     property?  This is
>>>     needed or the service
>>>     and all of its child processes will not have access to the desktop.
>>>     ie.  They can't show the
>>>     user anything.
>>>
>>>     Cheers,
>>>     Leif
>>>
>>>     Senthil S wrote:
>>>     > Hi,
>>>     > My application launches a  browser as soon as it gets started.The
>>>     > following is the code I use
>>>     >
>>>     > RunTime.exec(rundll32 url.dll, FileHandlerProtocol
>>>     > http://localhost:1090);
>>>     >
>>>     > This one brings up a browser when I start the application in the
>>>     > console mode but doesn't invoke the browser when started as a
>>>     service
>>>     > nor does it complain of any errors. I have the following for the
>>>     path
>>>     > wrapper.java.path.1=%SYSTEM%\rundll32.dll
>>>     > wrapper.java.path.2=%SYSTEM%\url.dll
>>>     > wrapper.java.library.path.3=%SYSTEM%\rundll32.dll
>>>     > wrapper.java.library.path.4=%SYSTEM%\url.dll
>>>     >
>>>     > The %PATH% has C:\winnt\SYSTEM32 .
>>>     >
>>>     > What do I miss? Please reply me ASAP
>>>     >
>>>     > Waiting for your reply,
>>>     > Senthil
>>>     >
>>>     >
>>>     >
>>>     > -------------------------------------------------------
>>>     > This SF.net email is sponsored by: Splunk Inc. Do you grep
>>>     through log
>>>     > files
>>>     > for problems?  Stop!  Download the new AJAX search engine that
>>>     makes
>>>     > searching your log files as easy as surfing the  web.  DOWNLOAD
>>>     SPLUNK!
>>>     >
>>>     
>>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642 
>>>
>>>     
>>> <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642> 
>>>
>>>     > _______________________________________________
>>>     > Wrapper-user mailing list
>>>     > Wra...@li...
>>>     <mailto:Wra...@li...>
>>>     > https://lists.sourceforge.net/lists/listinfo/wrapper-user
>>>     >
>>>
>>>
>>>
>>>     -------------------------------------------------------
>>>     This SF.net email is sponsored by: Splunk Inc. Do you grep through
>>>     log files
>>>     for problems?  Stop!  Download the new AJAX search engine that 
>>> makes
>>>     searching your log files as easy as surfing the  web.  DOWNLOAD
>>>     SPLUNK!
>>>     
>>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642 
>>>
>>>     
>>> <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642> 
>>>
>>>     _______________________________________________
>>>     Wrapper-user mailing list
>>>     Wra...@li...
>>>     <mailto:Wra...@li...>
>>>     https://lists.sourceforge.net/lists/listinfo/wrapper-user
>>>
>>>
>>
>>
>>
>> -------------------------------------------------------
>> This SF.net email is sponsored by: Splunk Inc. Do you grep through 
>> log files
>> for problems?  Stop!  Download the new AJAX search engine that makes
>> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
>> _______________________________________________
>> Wrapper-user mailing list
>> Wra...@li...
>> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log 
> files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>

[Wrapper-user] BackingStoreException when RUN_AS_USER on OS X

[Sean M. <sea...@ba...>]

I've run into an odd situation and I was wondering if anybody had any  
suggestions?

I'm using Java Service Wrapper 3.1.2 on Mac OS X 10.4.4 (not OS X  
Server) running Java 1.4.2_09.

I've got an app wrapped using WrapperSimpleApp as per the Integration  
Methods instructions  Since I'm on a Mac the script and wrapper file  
live in /Library/StartupItems/<appname>. With a bit of  
wrapper.working.dir trickery I've got wrapper running my jar which  
lives in /Library/MediaSpan/<appname>, which has uid:guid of  
appserver:admin, as does my jar and its lib folder.

My app subclasses java.rmi.server.UnicastRemoteObject to make it  
configurable via RMI.  It stores its configuration via  
java.util.prefs.Preferences, using the node returned by  
Preferences.userNodeForPackage(<myclass>.class).  No call is ever  
made to Preferences.systemNodeForPackage().

When I *do not* RUN_AS_USER, the WrapperSimpleApp wrapper starts my  
app, and it runs just fine.  I can connect via RMI and set my app's  
configuration, which writes to the Preferences user node via the  
putXXX and flush methods in Preferences.  The Preferences calls all  
work fine.

The trouble comes when I try to RUN_AS_USER=appserver.   When I make  
that one change everything works like before *except* the  
Preferences.flush call, which throws a BackingStoreException:

java.util.prefs.BackingStoreException: Synchronization failed for  
node <myclass's user node>

I'm confused as to why this is happening for the RUN_AS_USER case,  
but not when wrapper runs my app as root... has anybody else had to  
deal with this?

I searched the java-dev mailing list at Apple but the only mention of  
BackingStoreException puts the blame on making flush calls on a  
system pref node returned by Preferences. systemNodeForPackage.

Sean

[Wrapper-user] RE: Win2k access denied on service start (Resolved)

[Marc T. C. <mar...@in...>]

Original thread can be found here:
http://sourceforge.net/mailarchive/message.php?msg_id=9692073


After hours of scouring this mailing list I was able to resolve my own problem
with "0x5 Access is denied." errors when starting my Windows 2000 service.

I installed this service without specifying the wrapper.ntservice.account, so
it is running with the default "LocalSystem" account.  I was also able to start
the service from the batch script without any problems.  However, running it
from the command line inherited my currently logged in user's (Administrator)
credentials. Also, several other (built-in) services were starting up without
problems.

The issue was that the directory structure that the service was referencing
(C:\myproject\wrapper.exe) had full access permissions granted to the "Users"
group, however since my service runs as "System", the directory structure and
all the files will also need read/execute permissions granted to the "System"
user account.

Hope this helps,
-Marc

Re: [Wrapper-user] Cannot bring up a browser

[Senthil S <sen...@gm...>]

Leif,

I am sure its the browser launch process that fails, because even after 
that the application runs smoothly.
The exact statement I use is,

    Process p = Runtime.getRuntime.exec(rundll32 url.dll
    ,FileProtocolHandler http://localhost:1090);
    and also I tried
     Runtime.getRuntime.exec(cmd.exe /c start http://localhost:1090);

Where Runtime is java.lang.Runtime.
If not from wrapper.maxmemory=300MB (initially it had 256MB), how do I 
set the memory for the browser launch process?(should I set it at all?)
Without interactive flag set , I was not getting this error message 
("The application failed to initialize properly (0xc0000142)" ).
In both cases(with/without interactive flag set) the browser wasn't 
launched.

Thanks,
Senthil

Leif Mortenson wrote:

> Senthil,
>
>> I even tried using "wrapper.ntservice.interactive=true" but all I get 
>> is a message box saying
>> "The application failed to initialize properly (0xc0000142)"
>
> What is process that is crashing?  Is this your JVM or the browser 
> process?
>
>> I even tried increasing the maxmemory property to 300MB. Still it 
>> doesn't work.
>
> I doubt that the memory is the problem.  That just sets the memory 
> used by the JVM
> launched by the Wrapper.
>
> So without the interactive flag set, the browser gets launched but 
> does nothing.  But
> when the flag is set, something is crashing?
>
> What is the RunTime class?  Is that something that you created?  Have 
> you tried
> launching the browser using the java.lang.Runtime.exec method?  I 
> haven't launched
> a browser before, but have launched Excel without any problems in the 
> past.
>
> Cheers,
> Leif
>
>> On 2/10/06, *Leif Mortenson* <le...@ta... 
>> <mailto:le...@ta...>> wrote:
>>
>>     Senthil,
>>         Are you setting the wrapper.ntservice.interactive
>>     property?  This is
>>     needed or the service
>>     and all of its child processes will not have access to the desktop.
>>     ie.  They can't show the
>>     user anything.
>>
>>     Cheers,
>>     Leif
>>
>>     Senthil S wrote:
>>     > Hi,
>>     > My application launches a  browser as soon as it gets started.The
>>     > following is the code I use
>>     >
>>     > RunTime.exec(rundll32 url.dll, FileHandlerProtocol
>>     > http://localhost:1090);
>>     >
>>     > This one brings up a browser when I start the application in the
>>     > console mode but doesn't invoke the browser when started as a
>>     service
>>     > nor does it complain of any errors. I have the following for the
>>     path
>>     > wrapper.java.path.1=%SYSTEM%\rundll32.dll
>>     > wrapper.java.path.2=%SYSTEM%\url.dll
>>     > wrapper.java.library.path.3=%SYSTEM%\rundll32.dll
>>     > wrapper.java.library.path.4=%SYSTEM%\url.dll
>>     >
>>     > The %PATH% has C:\winnt\SYSTEM32 .
>>     >
>>     > What do I miss? Please reply me ASAP
>>     >
>>     > Waiting for your reply,
>>     > Senthil
>>     >
>>     >
>>     >
>>     > -------------------------------------------------------
>>     > This SF.net email is sponsored by: Splunk Inc. Do you grep
>>     through log
>>     > files
>>     > for problems?  Stop!  Download the new AJAX search engine that
>>     makes
>>     > searching your log files as easy as surfing the  web.  DOWNLOAD
>>     SPLUNK!
>>     >
>>     
>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
>>     
>> <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642> 
>>
>>     > _______________________________________________
>>     > Wrapper-user mailing list
>>     > Wra...@li...
>>     <mailto:Wra...@li...>
>>     > https://lists.sourceforge.net/lists/listinfo/wrapper-user
>>     >
>>
>>
>>
>>     -------------------------------------------------------
>>     This SF.net email is sponsored by: Splunk Inc. Do you grep through
>>     log files
>>     for problems?  Stop!  Download the new AJAX search engine that makes
>>     searching your log files as easy as surfing the  web.  DOWNLOAD
>>     SPLUNK!
>>     
>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
>>     
>> <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642> 
>>
>>     _______________________________________________
>>     Wrapper-user mailing list
>>     Wra...@li...
>>     <mailto:Wra...@li...>
>>     https://lists.sourceforge.net/lists/listinfo/wrapper-user
>>
>>
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log 
> files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>

Re: [Wrapper-user] Wrapper stdout redirection in realtime

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

Daniel,
    You are quite welcome.  It is already all checked in for the 3.2.0 
release.  I will leave
the default value false for now until it has been tested more and I am 
sure that it does not
have any adverse effects on performance on any platforms.
    Be sure to upgrade as soon as the release is out.

Cheers,
Leif

Daniel Mace wrote:
> Leif,
>
> IT WORKS! Awesome. :)
>
> Will this feature/workaround be going into the next release? I'll be
> using this snapshot in the meantime. 
>
> It's been a pleasure working on this with you. Many thanks. If there are
> any more specific details you need from me regarding this problem, I'd
> be glad to help.
>
> Thanks again.
>
> Daniel Mace
> Software Engineer, Payroll Integration
> benefitfocus.com
> 843-849-7476 x393
>  
>
>   
>> -----Original Message-----
>> From: wra...@li... 
>> [mailto:wra...@li...] On Behalf 
>> Of Leif Mortenson
>> Sent: Friday, February 10, 2006 9:43 AM
>> To: wra...@li...
>> Subject: Re: [Wrapper-user] Wrapper stdout redirection in realtime
>>
>> That link should have been:
>> http://wrapper.tanukisoftware.org/tmp/3.2.0-e/wrapper-windows-
>> x86-32-3.2.0-e.zip
>>
>> Leif Mortenson wrote:
>>     
>>> Daniel,
>>>    It is possible to explicitly flush stdout by calling 
>>> fflush(stdout).  I added a new wrapper.console.flush=true property, 
>>> which when set to true will do this after
>>> each line of output logged to the console.   Could you try out the 
>>> snapshot below
>>> with and without the above property and let me know if it makes any 
>>> difference.
>>> When using the wrapper in normal operation, it doesn't seem to do 
>>> anything at all.  The performance tests are unaffected one 
>>>       
>> way or the 
>>     
>>> other.  That may be because the command prompt is causing 
>>>       
>> an automatic 
>>     
>>> flush or something.
>>>
>>>
>>>       
>> http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-windows-x86-32-3
>>     
>>> .2.0-e.zip
>>>
>>>
>>> Cheers,
>>> Leif

RE: [Wrapper-user] Wrapper stdout redirection in realtime

[Daniel M. <dan...@be...>]

Leif,

IT WORKS! Awesome. :)

Will this feature/workaround be going into the next release? I'll be
using this snapshot in the meantime.=20

It's been a pleasure working on this with you. Many thanks. If there are
any more specific details you need from me regarding this problem, I'd
be glad to help.

Thanks again.

Daniel Mace
Software Engineer, Payroll Integration
benefitfocus.com
843-849-7476 x393
=20

> -----Original Message-----
> From: wra...@li...=20
> [mailto:wra...@li...] On Behalf=20
> Of Leif Mortenson
> Sent: Friday, February 10, 2006 9:43 AM
> To: wra...@li...
> Subject: Re: [Wrapper-user] Wrapper stdout redirection in realtime
>=20
> That link should have been:
> http://wrapper.tanukisoftware.org/tmp/3.2.0-e/wrapper-windows-
> x86-32-3.2.0-e.zip
>=20
> Leif Mortenson wrote:
> > Daniel,
> >    It is possible to explicitly flush stdout by calling=20
> > fflush(stdout).  I added a new wrapper.console.flush=3Dtrue =
property,=20
> > which when set to true will do this after
> > each line of output logged to the console.   Could you try out the=20
> > snapshot below
> > with and without the above property and let me know if it makes any=20
> > difference.
> > When using the wrapper in normal operation, it doesn't seem to do=20
> > anything at all.  The performance tests are unaffected one=20
> way or the=20
> > other.  That may be because the command prompt is causing=20
> an automatic=20
> > flush or something.
> >
> >=20
> http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-windows-x86-32-3
> > .2.0-e.zip
> >
> >
> > Cheers,
> > Leif
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep=20
> through log files for problems?  Stop!  Download the new AJAX=20
> search engine that makes searching your log files as easy as=20
> surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D103432&bid=3D230486&
> dat=3D121642
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>=20


*************************************************************************=
***************
BENEFITFOCUS.COM CONFIDENTIALITY NOTICE: This electronic message is =
intended only for the individual or entity to which it is addressed and =
may contain information that is confidential and protected by law. =
Unauthorized review, use, disclosure, or dissemination of this =
communication or its contents in any way is prohibited and may be =
unlawful. If you are not the intended recipient or a person responsible =
for delivering this message to an intended recipient, please notify the =
original sender immediately by e-mail or telephone, return the original =
message to the original sender or to bfp...@be..., and =
destroy all copies or derivations of the original message. Thank you.  =
(BFeComNote Rev. 08/01/2005)
*************************************************************************=
**************

Re: [Wrapper-user] Wrapper stdout redirection in realtime

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

That link should have been:
http://wrapper.tanukisoftware.org/tmp/3.2.0-e/wrapper-windows-x86-32-3.2.0-e.zip

Leif Mortenson wrote:
> Daniel,
>    It is possible to explicitly flush stdout by calling 
> fflush(stdout).  I added a new
> wrapper.console.flush=true property, which when set to true will do 
> this after
> each line of output logged to the console.   Could you try out the 
> snapshot below
> with and without the above property and let me know if it makes any 
> difference.
> When using the wrapper in normal operation, it doesn't seem to do 
> anything at
> all.  The performance tests are unaffected one way or the other.  That 
> may be
> because the command prompt is causing an automatic flush or something.
>
> http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-windows-x86-32-3.2.0-e.zip 
>
>
> Cheers,
> Leif

Re: [Wrapper-user] Wrapper stdout redirection in realtime

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

Daniel,
    It is possible to explicitly flush stdout by calling 
fflush(stdout).  I added a new
wrapper.console.flush=true property, which when set to true will do this 
after
each line of output logged to the console.   Could you try out the 
snapshot below
with and without the above property and let me know if it makes any 
difference.
When using the wrapper in normal operation, it doesn't seem to do 
anything at
all.  The performance tests are unaffected one way or the other.  That 
may be
because the command prompt is causing an automatic flush or something.

http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-windows-x86-32-3.2.0-e.zip

Cheers,
Leif

Daniel Mace wrote:
> Leif,
>
>   
>> Strange.  I assume you are using the console output of the Wrapper.   
>> That output should
>> be getting sent to stdout in pretty much real time.  It shows 
>> up in the console immediately
>> on all platforms at least.   I wonder if there is some kind 
>> of a flush 
>> that needs to be done
>> that is not needed the way the wrapper is normally used.  You 
>> might want to try poking
>> around in the logger.c file to see if you have any ideas.   
>> The Windows 
>> version makes
>> use of the WriteConsole function to write to the console if 
>> running as a service and a console is being displayed.  Other 
>> cases simply write to stdout.
>>     
>
> I will take a peek at it today to see if I notice anything obvious...
>
>   
>> I'll poke around, but other than flushing ???  Not sure what 
>> it would be.  Let me know how your program is running.  If 
>> you are running as a service, then you may need to set the 
>> wrapper.ntservice.console property to get a console to capture from.
>>     
>
> I apologize! I should have mentioned in my first message: I am running
> in console mode, NOT as a service. Hopefully this will be a clue. 
>
> Thanks for the tips. The search continues...
>
> Daniel Mace
> Software Engineer, Payroll Integration
> benefitfocus.com
> 843-849-7476 x393
>   

RE: [Wrapper-user] Wrapper stdout redirection in realtime

[Daniel M. <dan...@be...>]

Leif,=20

I found a person with a similar flushing problem on usenet when fprintf
is invoked under a JVM:
http://groups.google.com/group/comp.soft-sys.matlab/browse_thread/thread
/1c4b786fe609c2ff/d30d06cccda7e21c?lnk=3Dst&q=3Dfprintf+java+flush&rnum=3D=
3&hl
=3Den#d30d06cccda7e21c

Looks promising. Might just have to force the flush (with fflush? I
haven't programmed C in for quite some time). If I have time, I might
try and force the flush and compile it myself to test it. From my quick
look at logger.c, it seems like the fprintf call of interest is at
logger.c:729 when running in console mode. Sound reasonable? Do you know
of any drawbacks to flushing after every write?

Daniel Mace
Software Engineer, Payroll Integration
benefitfocus.com
843-849-7476 x393
=20

> -----Original Message-----
> From: wra...@li...=20
> [mailto:wra...@li...] On Behalf=20
> Of Daniel Mace
> Sent: Friday, February 10, 2006 8:50 AM
> To: wra...@li...
> Subject: RE: [Wrapper-user] Wrapper stdout redirection in realtime
>=20
> Leif,
>=20
> > Strange.  I assume you are using the console output of the=20
> Wrapper.  =20
> > That output should
> > be getting sent to stdout in pretty much real time.  It shows up in=20
> > the console immediately
> > on all platforms at least.   I wonder if there is some kind=20
> > of a flush
> > that needs to be done
> > that is not needed the way the wrapper is normally used.  You might=20
> > want to try poking
> > around in the logger.c file to see if you have any ideas.  =20
> > The Windows
> > version makes
> > use of the WriteConsole function to write to the console if=20
> running as=20
> > a service and a console is being displayed.  Other cases=20
> simply write=20
> > to stdout.
>=20
> I will take a peek at it today to see if I notice anything obvious...
>=20
> > I'll poke around, but other than flushing ???  Not sure=20
> what it would=20
> > be.  Let me know how your program is running.  If you are=20
> running as a=20
> > service, then you may need to set the wrapper.ntservice.console=20
> > property to get a console to capture from.
>=20
> I apologize! I should have mentioned in my first message: I=20
> am running in console mode, NOT as a service. Hopefully this=20
> will be a clue.=20
>=20
> Thanks for the tips. The search continues...
>=20
> Daniel Mace
> Software Engineer, Payroll Integration
> benefitfocus.com
> 843-849-7476 x393
>=20
>=20
> > -----Original Message-----
> > From: wra...@li...
> > [mailto:wra...@li...] On Behalf Of Leif=20
> > Mortenson
> > Sent: Thursday, February 09, 2006 10:33 PM
> > To: wra...@li...
> > Subject: Re: [Wrapper-user] Wrapper stdout redirection in realtime
> >=20
> > Daniel,
> >=20
> > Daniel Mace wrote:
> > > I have a C#/.NET 2.0 GUI application I use as a front end=20
> to a Java=20
> > > application I have developed. Traditionally, this Java app
> > was invoked
> > > via a DOS batch script which constructed classpaths and
> > executed the
> > > Java command directly. Within the GUI application, I was able to=20
> > > invoke the batch script and redirect stdout and stderr to
> > capture the
> > > Java app's output in realtime and display it to the user.=20
> > In order to
> > > gain more control over the indirectly-invoked Java app, I have=20
> > > encapsulated it in the Service Wrapper, to good effect (For
> > instance,
> > > I can now capture the pid of the Java/wrapper instance=20
> and kill it=20
> > > from the C# application). I realize this isn't necessarily
> > the intent
> > > of the wrapper, but launch4j can't do all the classpath
> > stuff I need,
> > > and I don't have access to an exe4j license. So bear with me...
> > >  =20
> > Fine.  viva open source. :-)
> > > My problem is this: since I modified the batch script to=20
> invoke the=20
> > > wrapper rather than the Java command itself to launch the
> > application,
> > > my C# application no longer gets realtime output from=20
> stdout/stderr=20
> > > when the wrapper is running. The C# code itself hasn't=20
> changed; in=20
> > > fact, output from the batch script itself gets redirected
> > properly and
> > > in realtime as before. It's just the wrapper output which
> > is funky: it
> > > doesn't actually get passed back to stdout until it's
> > completely run
> > > its course.
> > >  =20
> > Strange.  I assume you are using the console output of the=20
> Wrapper.  =20
> > That output should
> > be getting sent to stdout in pretty much real time.  It shows up in=20
> > the console immediately
> > on all platforms at least.   I wonder if there is some kind=20
> > of a flush
> > that needs to be done
> > that is not needed the way the wrapper is normally used.  You might=20
> > want to try poking
> > around in the logger.c file to see if you have any ideas.  =20
> > The Windows
> > version makes
> > use of the WriteConsole function to write to the console if=20
> running as=20
> > a service and a console is being displayed.  Other cases=20
> simply write=20
> > to stdout.
> > > Any ideas what could be causing this? Obviously it's redirecting=20
> > > somewhat properly, because I do eventually get the stdout=20
> stream...
> > > but not until the end of the wrapper execution. I searched and=20
> > > couldn't find any configuration properties related to
> > output redirection.
> > >  =20
> > I'll poke around, but other than flushing ???  Not sure=20
> what it would=20
> > be.  Let me know how your program is running.  If you are=20
> running as a=20
> > service, then you may need to set the wrapper.ntservice.console=20
> > property to get a console to capture from.
> >=20
> > Cheers,
> > Leif
> > > Heeeelllp! :)
> > >
> > > Daniel Mace
> > > Software Engineer, Payroll Integration benefitfocus.com
> > > 843-849-7476 x393
> > >
> > >
> > >=20
> > **************************************************************
> > **************************
> > > BENEFITFOCUS.COM CONFIDENTIALITY NOTICE: This electronic
> > message is intended only for the individual or entity to=20
> which it is=20
> > addressed and may contain information that is confidential and=20
> > protected by law. Unauthorized review, use, disclosure, or=20
> > dissemination of this communication or its contents in any way is=20
> > prohibited and may be unlawful. If you are not the intended=20
> recipient=20
> > or a person responsible for delivering this message to an intended=20
> > recipient, please notify the original sender immediately by=20
> e-mail or=20
> > telephone, return the original message to the original sender or to=20
> > bfp...@be..., and destroy all copies or=20
> derivations=20
> > of the original message. Thank you.
> > (BFeComNote Rev. 08/01/2005)
> > >=20
> > **************************************************************
> > *************************
> > >
> > >
> > > -------------------------------------------------------
> > > This SF.net email is sponsored by: Splunk Inc. Do you grep
> > through log files
> > > for problems?  Stop!  Download the new AJAX search engine=20
> that makes=20
> > > searching your log files as easy as surfing the  web.
> > DOWNLOAD SPLUNK!
> > > =
http://sel.as-us.falkag.net/sel?cmd=3Dk&kid=103432&bid#0486&dat=121642
> > > _______________________________________________
> > > Wrapper-user mailing list
> > > Wra...@li...
> > > https://lists.sourceforge.net/lists/listinfo/wrapper-user
> > >
> > >  =20
> >=20
> >=20
> >=20
> > -------------------------------------------------------
> > This SF.net email is sponsored by: Splunk Inc. Do you grep=20
> through log=20
> > files for problems?  Stop!  Download the new AJAX search=20
> engine that=20
> > makes searching your log files as easy as surfing the  web.
> > DOWNLOAD SPLUNK!
> > http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D103432&bid=3D230486&
> dat=3D121642
> > _______________________________________________
> > Wrapper-user mailing list
> > Wra...@li...
> > https://lists.sourceforge.net/lists/listinfo/wrapper-user
> >=20
>=20
>=20
> **************************************************************
> **************************
> BENEFITFOCUS.COM CONFIDENTIALITY NOTICE: This electronic=20
> message is intended only for the individual or entity to=20
> which it is addressed and may contain information that is=20
> confidential and protected by law. Unauthorized review, use,=20
> disclosure, or dissemination of this communication or its=20
> contents in any way is prohibited and may be unlawful. If you=20
> are not the intended recipient or a person responsible for=20
> delivering this message to an intended recipient, please=20
> notify the original sender immediately by e-mail or=20
> telephone, return the original message to the original sender=20
> or to bfp...@be..., and destroy all copies=20
> or derivations of the original message. Thank you. =20
> (BFeComNote Rev. 08/01/2005)
> **************************************************************
> *************************
>=20
>=20
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep=20
> through log files for problems?  Stop!  Download the new AJAX=20
> search engine that makes searching your log files as easy as=20
> surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=3Dk&kid=103432&bid#0486&dat=121642
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>=20


*************************************************************************=
***************
BENEFITFOCUS.COM CONFIDENTIALITY NOTICE: This electronic message is =
intended only for the individual or entity to which it is addressed and =
may contain information that is confidential and protected by law. =
Unauthorized review, use, disclosure, or dissemination of this =
communication or its contents in any way is prohibited and may be =
unlawful. If you are not the intended recipient or a person responsible =
for delivering this message to an intended recipient, please notify the =
original sender immediately by e-mail or telephone, return the original =
message to the original sender or to bfp...@be..., and =
destroy all copies or derivations of the original message. Thank you.  =
(BFeComNote Rev. 08/01/2005)
*************************************************************************=
**************

Re: [Wrapper-user] Cannot bring up a browser

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

Senthil,
> I even tried using "wrapper.ntservice.interactive=true" but all I get 
> is a message box saying
> "The application failed to initialize properly (0xc0000142)"
What is process that is crashing?  Is this your JVM or the browser process?
> I even tried increasing the maxmemory property to 300MB. Still it 
> doesn't work.
I doubt that the memory is the problem.  That just sets the memory used 
by the JVM
launched by the Wrapper.

So without the interactive flag set, the browser gets launched but does 
nothing.  But
when the flag is set, something is crashing?

What is the RunTime class?  Is that something that you created?  Have 
you tried
launching the browser using the java.lang.Runtime.exec method?  I 
haven't launched
a browser before, but have launched Excel without any problems in the past.

Cheers,
Leif

> On 2/10/06, *Leif Mortenson* <le...@ta... 
> <mailto:le...@ta...>> wrote:
>
>     Senthil,
>         Are you setting the wrapper.ntservice.interactive
>     property?  This is
>     needed or the service
>     and all of its child processes will not have access to the desktop.
>     ie.  They can't show the
>     user anything.
>
>     Cheers,
>     Leif
>
>     Senthil S wrote:
>     > Hi,
>     > My application launches a  browser as soon as it gets started.The
>     > following is the code I use
>     >
>     > RunTime.exec(rundll32 url.dll, FileHandlerProtocol
>     > http://localhost:1090);
>     >
>     > This one brings up a browser when I start the application in the
>     > console mode but doesn't invoke the browser when started as a
>     service
>     > nor does it complain of any errors. I have the following for the
>     path
>     > wrapper.java.path.1=%SYSTEM%\rundll32.dll
>     > wrapper.java.path.2=%SYSTEM%\url.dll
>     > wrapper.java.library.path.3=%SYSTEM%\rundll32.dll
>     > wrapper.java.library.path.4=%SYSTEM%\url.dll
>     >
>     > The %PATH% has C:\winnt\SYSTEM32 .
>     >
>     > What do I miss? Please reply me ASAP
>     >
>     > Waiting for your reply,
>     > Senthil
>     >
>     >
>     >
>     > -------------------------------------------------------
>     > This SF.net email is sponsored by: Splunk Inc. Do you grep
>     through log
>     > files
>     > for problems?  Stop!  Download the new AJAX search engine that
>     makes
>     > searching your log files as easy as surfing the  web.  DOWNLOAD
>     SPLUNK!
>     >
>     http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
>     <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642>
>     > _______________________________________________
>     > Wrapper-user mailing list
>     > Wra...@li...
>     <mailto:Wra...@li...>
>     > https://lists.sourceforge.net/lists/listinfo/wrapper-user
>     >
>
>
>
>     -------------------------------------------------------
>     This SF.net email is sponsored by: Splunk Inc. Do you grep through
>     log files
>     for problems?  Stop!  Download the new AJAX search engine that makes
>     searching your log files as easy as surfing the  web.  DOWNLOAD
>     SPLUNK!
>     http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
>     <http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642>
>     _______________________________________________
>     Wrapper-user mailing list
>     Wra...@li...
>     <mailto:Wra...@li...>
>     https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>

RE: [Wrapper-user] Wrapper stdout redirection in realtime

[Daniel M. <dan...@be...>]

Leif,

> Strange.  I assume you are using the console output of the Wrapper.  =20
> That output should
> be getting sent to stdout in pretty much real time.  It shows=20
> up in the console immediately
> on all platforms at least.   I wonder if there is some kind=20
> of a flush=20
> that needs to be done
> that is not needed the way the wrapper is normally used.  You=20
> might want to try poking
> around in the logger.c file to see if you have any ideas.  =20
> The Windows=20
> version makes
> use of the WriteConsole function to write to the console if=20
> running as a service and a console is being displayed.  Other=20
> cases simply write to stdout.

I will take a peek at it today to see if I notice anything obvious...

> I'll poke around, but other than flushing ???  Not sure what=20
> it would be.  Let me know how your program is running.  If=20
> you are running as a service, then you may need to set the=20
> wrapper.ntservice.console property to get a console to capture from.

I apologize! I should have mentioned in my first message: I am running
in console mode, NOT as a service. Hopefully this will be a clue.=20

Thanks for the tips. The search continues...

Daniel Mace
Software Engineer, Payroll Integration
benefitfocus.com
843-849-7476 x393


> -----Original Message-----
> From: wra...@li...=20
> [mailto:wra...@li...] On Behalf=20
> Of Leif Mortenson
> Sent: Thursday, February 09, 2006 10:33 PM
> To: wra...@li...
> Subject: Re: [Wrapper-user] Wrapper stdout redirection in realtime
>=20
> Daniel,
>=20
> Daniel Mace wrote:
> > I have a C#/.NET 2.0 GUI application I use as a front end to a Java=20
> > application I have developed. Traditionally, this Java app=20
> was invoked=20
> > via a DOS batch script which constructed classpaths and=20
> executed the=20
> > Java command directly. Within the GUI application, I was able to=20
> > invoke the batch script and redirect stdout and stderr to=20
> capture the=20
> > Java app's output in realtime and display it to the user.=20
> In order to=20
> > gain more control over the indirectly-invoked Java app, I have=20
> > encapsulated it in the Service Wrapper, to good effect (For=20
> instance,=20
> > I can now capture the pid of the Java/wrapper instance and kill it=20
> > from the C# application). I realize this isn't necessarily=20
> the intent=20
> > of the wrapper, but launch4j can't do all the classpath=20
> stuff I need,=20
> > and I don't have access to an exe4j license. So bear with me...
> >  =20
> Fine.  viva open source. :-)
> > My problem is this: since I modified the batch script to invoke the=20
> > wrapper rather than the Java command itself to launch the=20
> application,=20
> > my C# application no longer gets realtime output from stdout/stderr=20
> > when the wrapper is running. The C# code itself hasn't changed; in=20
> > fact, output from the batch script itself gets redirected=20
> properly and=20
> > in realtime as before. It's just the wrapper output which=20
> is funky: it=20
> > doesn't actually get passed back to stdout until it's=20
> completely run=20
> > its course.
> >  =20
> Strange.  I assume you are using the console output of the Wrapper.  =20
> That output should
> be getting sent to stdout in pretty much real time.  It shows=20
> up in the console immediately
> on all platforms at least.   I wonder if there is some kind=20
> of a flush=20
> that needs to be done
> that is not needed the way the wrapper is normally used.  You=20
> might want to try poking
> around in the logger.c file to see if you have any ideas.  =20
> The Windows=20
> version makes
> use of the WriteConsole function to write to the console if=20
> running as a service and a console is being displayed.  Other=20
> cases simply write to stdout.
> > Any ideas what could be causing this? Obviously it's redirecting=20
> > somewhat properly, because I do eventually get the stdout stream...=20
> > but not until the end of the wrapper execution. I searched and=20
> > couldn't find any configuration properties related to=20
> output redirection.
> >  =20
> I'll poke around, but other than flushing ???  Not sure what=20
> it would be.  Let me know how your program is running.  If=20
> you are running as a service, then you may need to set the=20
> wrapper.ntservice.console property to get a console to capture from.
>=20
> Cheers,
> Leif
> > Heeeelllp! :)
> >
> > Daniel Mace
> > Software Engineer, Payroll Integration benefitfocus.com
> > 843-849-7476 x393
> >
> >
> >=20
> **************************************************************
> **************************
> > BENEFITFOCUS.COM CONFIDENTIALITY NOTICE: This electronic=20
> message is intended only for the individual or entity to=20
> which it is addressed and may contain information that is=20
> confidential and protected by law. Unauthorized review, use,=20
> disclosure, or dissemination of this communication or its=20
> contents in any way is prohibited and may be unlawful. If you=20
> are not the intended recipient or a person responsible for=20
> delivering this message to an intended recipient, please=20
> notify the original sender immediately by e-mail or=20
> telephone, return the original message to the original sender=20
> or to bfp...@be..., and destroy all copies=20
> or derivations of the original message. Thank you. =20
> (BFeComNote Rev. 08/01/2005)
> >=20
> **************************************************************
> *************************
> >
> >
> > -------------------------------------------------------
> > This SF.net email is sponsored by: Splunk Inc. Do you grep=20
> through log files
> > for problems?  Stop!  Download the new AJAX search engine that makes
> > searching your log files as easy as surfing the  web. =20
> DOWNLOAD SPLUNK!
> > =
http://sel.as-us.falkag.net/sel?cmd=3Dk&kid=103432&bid#0486&dat=121642
> > _______________________________________________
> > Wrapper-user mailing list
> > Wra...@li...
> > https://lists.sourceforge.net/lists/listinfo/wrapper-user
> >
> >  =20
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep=20
> through log files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web. =20
> DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D103432&bid=3D230486&
dat=3D121642
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>=20


*************************************************************************=
***************
BENEFITFOCUS.COM CONFIDENTIALITY NOTICE: This electronic message is =
intended only for the individual or entity to which it is addressed and =
may contain information that is confidential and protected by law. =
Unauthorized review, use, disclosure, or dissemination of this =
communication or its contents in any way is prohibited and may be =
unlawful. If you are not the intended recipient or a person responsible =
for delivering this message to an intended recipient, please notify the =
original sender immediately by e-mail or telephone, return the original =
message to the original sender or to bfp...@be..., and =
destroy all copies or derivations of the original message. Thank you.  =
(BFeComNote Rev. 08/01/2005)
*************************************************************************=
**************

Re: [Wrapper-user] Cannot bring up a browser

[Senthil S <sen...@gm...>]

Hi Leif,

I even tried using "wrapper.ntservice.interactive=3Dtrue" but all I get is =
a
message box saying
"The application failed to initialize properly (0xc0000142)"
I even tried increasing the maxmemory property to 300MB. Still it doesn't
work.

Thanks for early reply,
Senthil


On 2/10/06, Leif Mortenson <le...@ta...> wrote:
>
> Senthil,
>     Are you setting the wrapper.ntservice.interactive property?  This is
> needed or the service
> and all of its child processes will not have access to the desktop.
> ie.  They can't show the
> user anything.
>
> Cheers,
> Leif
>
> Senthil S wrote:
> > Hi,
> > My application launches a  browser as soon as it gets started.The
> > following is the code I use
> >
> > RunTime.exec(rundll32 url.dll, FileHandlerProtocol
> > http://localhost:1090);
> >
> > This one brings up a browser when I start the application in the
> > console mode but doesn't invoke the browser when started as a service
> > nor does it complain of any errors. I have the following for the path
> > wrapper.java.path.1=3D%SYSTEM%\rundll32.dll
> > wrapper.java.path.2=3D%SYSTEM%\url.dll
> > wrapper.java.library.path.3=3D%SYSTEM%\rundll32.dll
> > wrapper.java.library.path.4=3D%SYSTEM%\url.dll
> >
> > The %PATH% has C:\winnt\SYSTEM32 .
> >
> > What do I miss? Please reply me ASAP
> >
> > Waiting for your reply,
> > Senthil
> >
> >
> >
> > -------------------------------------------------------
> > This SF.net email is sponsored by: Splunk Inc. Do you grep through log
> > files
> > for problems?  Stop!  Download the new AJAX search engine that makes
> > searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> > http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D103432&bid=3D230486&dat=
=3D121642
> > _______________________________________________
> > Wrapper-user mailing list
> > Wra...@li...
> > https://lists.sourceforge.net/lists/listinfo/wrapper-user
> >
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log
> files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D103432&bid=3D230486&dat=
=3D121642
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>

Re: [Wrapper-user] FreeDSD loading native library

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

Got this reply off list. For the archives:

Hi Leif

I actually figured it out! I'm running the Linux java VM in Linux
Compatibility mode, when I switched to the Linux version it executed
perfectly!

Cheers + thanks for your trouble,
Malan


Leif Mortenson wrote:
> Malan,
>    Sorry for the slow response on this.  I don't have access to a 
> freebsd box personally to
> test this out.  Could you try building from source using the 3.2.0-d 
> pre release version that
> I posted about today?   I did a search for your "ELF file OS ABI 
> invalid" error.  It may be
> a dependency conflict between your system and the system that the 
> version you were
> using was built on...  I would be interested to hear the results of 
> you building from source
> for the latest version.
>
> Cheers,
> Leif
>
> Malan Joubert wrote:
>> Hi
>>
>> I'm running Java service wrapper on FreeBSD 5.3.
>>
>> I've configured the wrapper, but it fails to load:
>>
>> In debug mode I get the following error:
>>
>> jvm 1    | Initializing...
>> jvm 1    | WrapperManager class initialized by thread: main  Using
>> classloader: sun.misc.Launcher$AppClassLoader@1813fac
>> jvm 1    | Wrapper Manager: JVM #1
>> jvm 1    | Wrapper Manager: Registering shutdown hook
>> jvm 1    | Wrapper Manager: Using wrapper
>> jvm 1    | Loading native library failed: libwrapper.so  Cause:
>> java.lang.UnsatisfiedLinkError:
>> /home/fireflight/smsgateway/lib/libwrapper.so:
>> /home/fireflight/smsgateway/lib/libwrapper.so: ELF file OS ABI invalid
>> jvm 1    |
>> jvm 1    | WARNING - Unable to load the Wrapper's native library
>> 'libwrapper.so'.
>> jvm 1    |           The file is located on the path at the following
>> location but
>> jvm 1    |           could not be loaded:
>> jvm 1    |             /home/fireflight/smsgateway/lib/libwrapper.so
>> jvm 1    |           Please verify that the file is readable by the 
>> current user
>> jvm 1    |           and that the file has not been corrupted in any 
>> way.
>> jvm 1    |           System signals will not be handled correctly.
>> jvm 1    |
>> jvm 1    | Java Version   : 1.4.2_10-b03 Java HotSpot(TM) Client VM
>> jvm 1    | Java VM Vendor : Sun Microsystems Inc.
>> jvm 1    |
>> jvm 1    | Wrapper (Version 3.1.2) http://wrapper.tanukisoftware.org
>>
>>
>> What can I do? Should I try to recompile the libwrapper.so file 
>> manually?
>>
>> Thank you,
>> Malan
>>   

Re: [Wrapper-user] Wrapper ending after running Timer (TimerTask) once as a service

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

David,
    Could you go back and give this a try with the 3.2.0-d prerelease 
version:
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-windows-x86-32-3.2.0-d.zip

    There have been a few modifications which may fix the problems you were
seeing.  The WrapperSimpleApp and WrapperStartStopApp classes were modified
so this should no longer happen, and a wrapper.thread_count_delay 
property was
added for other cases.

Cheers,
Leif

DAVID TURNER wrote:
>
> Hi Leif,
>
> The log was for a non-daemon.  The log for the daemon method also 
> produced an identical log like the previous log with the same line it 
> as the below.
>
> INFO   | jvm 1    | 2005/12/15 08:58:15 | All non-daemon threads have
> stopped.  Exiting.
>
> Here's the main method of the main class.  Should this be done 
> differently in order to avoid the wrapper from shutting down early?
>
>         public static void main(String[] args) {
>                     Timer timer = new Timer("ConnectionTestTimer", true);
>                     timer.schedule(new Task1(), 0, 1*60*1000); // 1 
> minute
>                     timer.schedule(new Task2(), 0, 1*60*1000); // 1 
> minute
>         }
>
>
> Why, if I set up the timer to be a daemon, is the wrapper shutting 
> down early?  Am I overlooking something?
>
>
>
>
>
> *Leif Mortenson <le...@ta...>*
> Sent by: wra...@li...
>
> 12/15/2005 09:27 AM
> Please respond to
> wra...@li...
>
>
> 	
> To
> 	wra...@li...
> cc
> 	
> Subject
> 	Re: [Wrapper-user] Wrapper ending after running Timer (TimerTask) 
> once as a service
>
>
>
> 	
>
>
>
>
>
> David,
>    From your log, the Wrapper is exiting because it has determined that
> all non-daemon
> threads have terminated.  You mentioned that you tried running your
> program with
> the timer thread set as a daemon and as a non-daemon.  Which was the 
> log you
> posted from?
>
>    Here is the line in the log which gives the reason why the Wrapper
> is exiting the JVM:
>
> INFO   | jvm 1    | 2005/12/15 08:58:15 | All non-daemon threads have
> stopped.  Exiting.
>
>    Have you tried adding debug output into your Timer thread to make
> sure that it is
> indeed still running?   I can't think of anything that would be
> different in this case about
> running as a service vs running in a console.
>    Is it possible that an exception is being thrown that is causing the
> timer thread to be
> terminated?   Is there any code which would be preventing any exceptions
> from being
> logged?
>    I usually do the following in my runner threads just to make sure
> that they never exit
> without me knowing about it.  (I use a logging tool however)
>
> public void run() {
>    System.out.println( "Runner started." );
>    try {
>        while( !stopping ) {
>            // Loop
>        }
>    } finally {
>        System.out.println( "Runner terminated" );
>    }
> }
>
> Cheers,
> Leif
>
> DAVID TURNER wrote:
> >
> > I've included the contents of wrapper.log after setting the
> > wrapper.debug to true.
> >
> > One thing I want to mention is that I tried to set the java.util.Timer
> > both as a daemon and a non-daemon.  The outcome is the same for both,
> > and that the nt service (wrapper) stops after one execution of the
> > TimerTask.
> >
> >
> > DEBUG  | wrapper  | 2005/12/15 08:57:21 | Service command:
> > 
> C:\MyProjects\connection-monitor\packaged\connection-monitor\bin\wrapper.exe 
>
> > -s
> > 
> C:\MyProjects\connection-monitor\packaged\connection-monitor\bin\..\conf\wrapper.conf 
>
> >
> > STATUS | wrapper  | 2005/12/15 08:57:22 | Connection Monitor installed.
> > STATUS | wrapper  | 2005/12/15 08:58:06 | --> Wrapper Started as Service
> > DEBUG  | wrapper  | 2005/12/15 08:58:06 | Using system timer.
> > DEBUG  | wrapperp | 2005/12/15 08:58:06 | server listening on port 
> 32000.
> > STATUS | wrapper  | 2005/12/15 08:58:06 | Launching a JVM...
> snip
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log 
> files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>
>
>
> ========================= NMB (USA) EMAIL NOTICE 
> ============================
> The information contained herein may be confidential and proprietary 
> and is intended only for
> the use of the addressee. Any unauthorized dissemination or disclosure 
> of this communication
> is strictly prohibited.  If you have received this communication in 
> error, please notify the sender
> immediately. The information herein is not warranted to be free of 
> virus or any other defect.
> ============================================================================== 
>

Re: [Wrapper-user] Wrapper stdout redirection in realtime

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

Daniel,

Daniel Mace wrote:
> I have a C#/.NET 2.0 GUI application I use as a front end to a Java
> application I have developed. Traditionally, this Java app was invoked
> via a DOS batch script which constructed classpaths and executed the
> Java command directly. Within the GUI application, I was able to invoke
> the batch script and redirect stdout and stderr to capture the Java
> app's output in realtime and display it to the user. In order to gain
> more control over the indirectly-invoked Java app, I have encapsulated
> it in the Service Wrapper, to good effect (For instance, I can now
> capture the pid of the Java/wrapper instance and kill it from the C#
> application). I realize this isn't necessarily the intent of the
> wrapper, but launch4j can't do all the classpath stuff I need, and I
> don't have access to an exe4j license. So bear with me...
>   
Fine.  viva open source. :-)
> My problem is this: since I modified the batch script to invoke the
> wrapper rather than the Java command itself to launch the application,
> my C# application no longer gets realtime output from stdout/stderr when
> the wrapper is running. The C# code itself hasn't changed; in fact,
> output from the batch script itself gets redirected properly and in
> realtime as before. It's just the wrapper output which is funky: it
> doesn't actually get passed back to stdout until it's completely run its
> course. 
>   
Strange.  I assume you are using the console output of the Wrapper.   
That output should
be getting sent to stdout in pretty much real time.  It shows up in the 
console immediately
on all platforms at least.   I wonder if there is some kind of a flush 
that needs to be done
that is not needed the way the wrapper is normally used.  You might want 
to try poking
around in the logger.c file to see if you have any ideas.   The Windows 
version makes
use of the WriteConsole function to write to the console if running as a 
service and a
console is being displayed.  Other cases simply write to stdout.
> Any ideas what could be causing this? Obviously it's redirecting
> somewhat properly, because I do eventually get the stdout stream... but
> not until the end of the wrapper execution. I searched and couldn't find
> any configuration properties related to output redirection.
>   
I'll poke around, but other than flushing ???  Not sure what it would 
be.  Let me know
how your program is running.  If you are running as a service, then you 
may need to
set the wrapper.ntservice.console property to get a console to capture from.

Cheers,
Leif
> Heeeelllp! :)
>
> Daniel Mace
> Software Engineer, Payroll Integration
> benefitfocus.com
> 843-849-7476 x393
>
>
> ****************************************************************************************
> BENEFITFOCUS.COM CONFIDENTIALITY NOTICE: This electronic message is intended only for the individual or entity to which it is addressed and may contain information that is confidential and protected by law. Unauthorized review, use, disclosure, or dissemination of this communication or its contents in any way is prohibited and may be unlawful. If you are not the intended recipient or a person responsible for delivering this message to an intended recipient, please notify the original sender immediately by e-mail or telephone, return the original message to the original sender or to bfp...@be..., and destroy all copies or derivations of the original message. Thank you.  (BFeComNote Rev. 08/01/2005)
> ***************************************************************************************
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=k&kid3432&bid#0486&dat1642
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>   

Re: [Wrapper-user] FreeDSD loading native library

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

Malan,
    Sorry for the slow response on this.  I don't have access to a 
freebsd box personally to
test this out.  Could you try building from source using the 3.2.0-d pre 
release version that
I posted about today?   I did a search for your "ELF file OS ABI 
invalid" error.  It may be
a dependency conflict between your system and the system that the 
version you were
using was built on...  I would be interested to hear the results of you 
building from source
for the latest version.

Cheers,
Leif

Malan Joubert wrote:
> Hi
>
> I'm running Java service wrapper on FreeBSD 5.3.
>
> I've configured the wrapper, but it fails to load:
>
> In debug mode I get the following error:
>
> jvm 1    | Initializing...
> jvm 1    | WrapperManager class initialized by thread: main  Using
> classloader: sun.misc.Launcher$AppClassLoader@1813fac
> jvm 1    | Wrapper Manager: JVM #1
> jvm 1    | Wrapper Manager: Registering shutdown hook
> jvm 1    | Wrapper Manager: Using wrapper
> jvm 1    | Loading native library failed: libwrapper.so  Cause:
> java.lang.UnsatisfiedLinkError:
> /home/fireflight/smsgateway/lib/libwrapper.so:
> /home/fireflight/smsgateway/lib/libwrapper.so: ELF file OS ABI invalid
> jvm 1    |
> jvm 1    | WARNING - Unable to load the Wrapper's native library
> 'libwrapper.so'.
> jvm 1    |           The file is located on the path at the following
> location but
> jvm 1    |           could not be loaded:
> jvm 1    |             /home/fireflight/smsgateway/lib/libwrapper.so
> jvm 1    |           Please verify that the file is readable by the current user
> jvm 1    |           and that the file has not been corrupted in any way.
> jvm 1    |           System signals will not be handled correctly.
> jvm 1    |
> jvm 1    | Java Version   : 1.4.2_10-b03 Java HotSpot(TM) Client VM
> jvm 1    | Java VM Vendor : Sun Microsystems Inc.
> jvm 1    |
> jvm 1    | Wrapper (Version 3.1.2) http://wrapper.tanukisoftware.org
>
>
> What can I do? Should I try to recompile the libwrapper.so file manually?
>
> Thank you,
> Malan
>   

Re: [Wrapper-user] Console messages

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

Anastasios Angelidis wrote:
> Thanks it's just that the 3rd party app I'm trying to wrap is very 
> verbose on the console...
>
> But it has it's own logs... I'm sure it loags it's own exceptions! 
> Unless you saying I wont see crash duimps of the wrapper...
Well hopefully the Wrapper itself won't crash.   But if it did, it 
wouldn't be able to
log itself in most cases.   I am talking about crashes of the JVM.   
Those happen at
a low level and it is not possible to log such things from within the 
JVM.   Thread
dumps will also not be logged by java based logging.  Same if you have 
any code
which uses System.out or System.err

Cheers,
Leif
>
>
> Leif Mortenson wrote:
>
>> Anastasios,
>>    Try this:
>> wrapper.logfile.loglevel=STATUS
>>
>>    You will still see the output in the console, but it will not be 
>> shown in the log file.
>> The problem with doing this is that you may lose other information 
>> that you really do
>> want to see.  For example crash dumps etc.
>>
>>    A better solution may be to do something like this:
>> wrapper.logfile.maxsize=1m
>> wrapper.logfile.maxfiles=1
>>
>>    By doing this, the wrapper.log file will always be renamed to 
>> wrapper.log.1
>> when it exceeds 1Mb in size.  The max files specifies the maximum 
>> number of
>> rolled log files that will be kept around.  By doing this, you will 
>> be able to see
>> all recent log output, but will also be able to guarantee that the 
>> wrapper's log
>> files will never use more than 2Mb of disk space.
>>
>> Cheers,
>> Leif
>>
>> Anastasios Angelidis wrote:
>>
>>> Is there a way to stop the wrapper from outputting console messages 
>>> to the log? I'm using a 3rd party app which outputs alot to the 
>>> console. So everything gets logged to the wrapper log which I dont 
>>> want it to! Thanks
>>
>>
>>
>>
>> -------------------------------------------------------
>> This SF.net email is sponsored by: Splunk Inc. Do you grep through 
>> log files
>> for problems?  Stop!  Download the new AJAX search engine that makes
>> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
>> _______________________________________________
>> Wrapper-user mailing list
>> Wra...@li...
>> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>>
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log 
> files
> for problems?  Stop!  Download the new AJAX search engine that makes
> searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>

[Wrapper-user] Wrapper pre release 3.2.0-d

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

Hi all,
I have been trying to get things tied up for the 3.2.0 release and have made
the latest pre-release builds available for testing. THESE VERSIONS WILL NOT
BE SUPPORTED AND SHOULD NOT BE USED IN RELEASED SOFTWARE.
If you could give them a try, I would appreciate hearing about any problems
before the actual release. Normally I don't do this, but it has been a
long time
since the last official release.

The following are the list of platforms that I am able to build personally.
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-linux-ppc-64-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-linux-x86-32-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-linux-x86-64-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-macosx-ppc-32-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-solaris-sparc-32-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-solaris-x86-32-3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-windows-x86-32-3.2.0-d.zip

Additional make files for this release exist for the following versions but
have not been tested for this release:
aix-ppc-32
freebsd-x86-32
hpux-parisc-32
hpux-parisc-64
irix-mips-32
osf1-alpha-32
solaris-sparc-64
windows-x86-64 (This make file needs work)

In most cases with this release, you can simply download the prerelease
distribution and build a release for any of the above platforms by running
either "./build32.sh release" or "./build64.sh release" as appropriate.
This will make use of a prebuilt jar and

http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper_prerelease_3.2.0-d.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper_prerelease_3.2.0-d.zip

If you need to build the java source then you should do so using the
following
source distributions by running "./build32.sh pre-release" followed
by "./build32.sh release".

http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper_3.2.0-d_src_with_doc_src.tar.gz
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper_3.2.0-d_src_with_doc_src.zip

Cheers,
Leif

Re: [Wrapper-user] Linux daemon question

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

Anat,
    Looking at your log file, this is the cause of the application being 
shut down:
INFO   | jvm 1    | 2006/02/07 16:42:51 | Processing control 
event(WRAPPER_CTRL_SHUTDOWN_EVENT)

    There is a bug in 3.1.2 where the JVM is not correctly ignoring TERM 
signals
even if the IGNORE_SIGNALS flag is set in the sh script.  This has been 
fixed
for the unreleased 3.2.0.  Could you please download and try out the latest
snapshot build and verify that this fixes the problem for you?

    
http://wrapper.tanukisoftware.org/tmp/3.1.2-d/wrapper-linux-x86-32-3.2.0-d.tar.gz

    You will want to use the new sh script and set the IGNORE_SIGNALS
variable to true in the script.

    Also, from your logs, it looks like there is a GUI being 
displayed?   That may
why your JVM is getting the TERM signals in the first place.  In 
general, daemon
processes don't have GUIs.  You should add a call to
WrapperManager.isLaunchedAsService()  This method will return true if the
Wrapper is running as a daemon process.  Only show your GUI if it is false.
If you don't want to add a dependency to the wrapper API, look for the
"wrapper.service" system property.

Cheers,
Leif

Anat Halpern wrote:
> Hi Leif,
>
> Attached is the log file. I hope it'll help.
>
> Thanks,
> Anat
>
>