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

Re: Request for opinions. Re: [Wrapper-user] wrapper.java.additional.11 with # in value

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

Paul Casanova wrote:

>Hi Leif,
>
>The latter idea (##) seems the most logical since most programming
>conventions use a double whatever to escape and use the whatever!
>
>Eg, in VB an & is used to specify an accelarator key in menus buttons etc,
>so && escapes this and shows "&" in the menu button etc.
>
>The only issue might be where people have a string of #'s used to highlight
>an area of code in a conf file, eg
>########################### IMPORTANT CODE ###########################.
>
It should be ok, each double ## would be reduced to single #, then first 
single #
would be interpreted as a comment.
In addition, any line which starts with a # would still be a comment 
unless it was #include
I'll have to test it, but I think the above example should be ok.

Cheers,
Leif

Re: Request for opinions. Re: [Wrapper-user] wrapper.java.additional.11 with # in value

[Paul C. <cas...@au...>]

Hi Leif,

The latter idea (##) seems the most logical since most programming
conventions use a double whatever to escape and use the whatever!

Eg, in VB an & is used to specify an accelarator key in menus buttons etc,
so && escapes this and shows "&" in the menu button etc.

The only issue might be where people have a string of #'s used to highlight
an area of code in a conf file, eg
########################### IMPORTANT CODE ###########################.

Regards,

Paul Casanova
IBM RSSC-Ballarat
UB01
Locked Bag 1999
Ballarat
Victoria 3350

Telephone:
61-3-53273458
Fax:
61-3-53273599



                                                                                                                                                   
                      Leif Mortenson                                                                                                               
                      <le...@ta...>            To:       wra...@li...                                            
                      Sent by:                             cc:                                                                                     
                      wra...@li...        Subject:  Request for opinions. Re: [Wrapper-user] wrapper.java.additional.11 with # in 
                      ceforge.net                           value                                                                                  
                                                                                                                                                   
                                                                                                                                                   
                      25/07/2003 11:41 AM                                                                                                          
                      Please respond to                                                                                                            
                      wrapper-user                                                                                                                 
                                                                                                                                                   



I have been looking into the options for how to make it possible to
include #
characters in a property value.
The problem is that property declarations like the following do not work
because
The Wrapper clips everything from the # as a comment.
wrapper.java.additional.1=-Dtest=my#test
The Wrapper sees the above as:
wrapper.java.additional.1=-Dtest=my

1) One option is to ignore # characters inside of quotes, but this
causes problems
on *NIX platforms.  It is not possible to strip the quotes because
changing that
would cause compatibility problems with users who are already using quotes
and expecting them to remain unmodified by the Wrapper.  So this is pretty
much a no-go.

2) Introduce an escape character to make it possible to escape the # when
it
should be included.  The obvious escape character would be the backslash.
But I unfortunately can not use it because users have been using unescaped
backslash characters in their wrapper.conf files throughout the life of the
Wrapper.   The option I am looking at right now for the escape character is
another # character.  So 2 # characters would resolve into 1 in the actual
property value.  It seems a little hackish and would not be usable in other
contexts if other characters ever need to be escaped in the future.

Currently, the ## option is the only one that is viable.  Does anyone
have any
other ideas.

Using the above example,
wrapper.java.additional.1=-Dtest=my##test#Comment
Would resolve to:
wrapper.java.additional.1=-Dtest=my#test

Cheers,
Leif




Leif Mortenson wrote:

> Sebastian,
>    Sorry, this is currently not possible to do within the wrapper.conf
> file.  An
> oversight on my part.  I did find a workaround, but am not sure if it
> will work
> for you.
>
>    The Wrapper has the ability to accept any property from the command
> line.
> If you launch the Wrapper using the following command, all on a single
> line
> then your application should work:
>
> Wrapper.exe -c ..\conf\wrapper.conf
> wrapper.java.additional.11=-Dxenon.service.args=foo#bar
>
> I opened a bug for this at the following URL.  Please monitor it to
> track its progress.
>
https://sourceforge.net/tracker/index.php?func=detail&aid=777303&group_id=39428&atid=425187

>
>
> I have to play around with my options on the different platforms.
>
> Cheers,
> Leif
>
> Sebastian Hauer wrote:
>
>> Hi,
>>
>> I am new to the java service wrapper so far it seems very promising.
>> But I am having a problem setting a system property on windows.  The
>> value of this system property unfortunately has a # character, like
>> this:
>>
>> wrapper.java.additional.11=-Dxenon.service.args=foo#bar
>>
>> Unfortunately our application receives only "foo" as the value of
>> xenon.service.args.
>> I tried escaping # using \# but this would not work and instead got
>> "foo\".
>> I assume this is a bug in how the wrapper C code parses the properties
>> file.
>>
>> Regards,
>> Sebastian
>>
>>
>> -------------------------------------------------------
>> This SF.Net email sponsored by: Free pre-built ASP.NET sites including
>> Data Reports, E-commerce, Portals, and Forums are available now.
>> Download today and enter to win an XBOX or Visual Studio .NET.
>>
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01

>>
>> _______________________________________________
>> Wrapper-user mailing list
>> Wra...@li...
>> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>>
>>
>>
>
>
>
> -------------------------------------------------------
> This SF.Net email sponsored by: Free pre-built ASP.NET sites including
> Data Reports, E-commerce, Portals, and Forums are available now.
> Download today and enter to win an XBOX or Visual Studio .NET.
>
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01

>
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>



-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01
_______________________________________________
Wrapper-user mailing list
Wra...@li...
https://lists.sourceforge.net/lists/listinfo/wrapper-user

Request for opinions. Re: [Wrapper-user] wrapper.java.additional.11 with # in value

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

I have been looking into the options for how to make it possible to 
include #
characters in a property value.
The problem is that property declarations like the following do not work 
because
The Wrapper clips everything from the # as a comment.
wrapper.java.additional.1=-Dtest=my#test
The Wrapper sees the above as:
wrapper.java.additional.1=-Dtest=my

1) One option is to ignore # characters inside of quotes, but this 
causes problems
on *NIX platforms.  It is not possible to strip the quotes because 
changing that
would cause compatibility problems with users who are already using quotes
and expecting them to remain unmodified by the Wrapper.  So this is pretty
much a no-go.

2) Introduce an escape character to make it possible to escape the # when it
should be included.  The obvious escape character would be the backslash.
But I unfortunately can not use it because users have been using unescaped
backslash characters in their wrapper.conf files throughout the life of the
Wrapper.   The option I am looking at right now for the escape character is
another # character.  So 2 # characters would resolve into 1 in the actual
property value.  It seems a little hackish and would not be usable in other
contexts if other characters ever need to be escaped in the future.

Currently, the ## option is the only one that is viable.  Does anyone 
have any
other ideas.

Using the above example,
wrapper.java.additional.1=-Dtest=my##test#Comment
Would resolve to:
wrapper.java.additional.1=-Dtest=my#test

Cheers,
Leif




Leif Mortenson wrote:

> Sebastian,
>    Sorry, this is currently not possible to do within the wrapper.conf 
> file.  An
> oversight on my part.  I did find a workaround, but am not sure if it 
> will work
> for you.
>
>    The Wrapper has the ability to accept any property from the command 
> line.
> If you launch the Wrapper using the following command, all on a single 
> line
> then your application should work:
>
> Wrapper.exe -c ..\conf\wrapper.conf 
> wrapper.java.additional.11=-Dxenon.service.args=foo#bar
>
> I opened a bug for this at the following URL.  Please monitor it to 
> track its progress.
> https://sourceforge.net/tracker/index.php?func=detail&aid=777303&group_id=39428&atid=425187 
>
>
> I have to play around with my options on the different platforms.
>
> Cheers,
> Leif
>
> Sebastian Hauer wrote:
>
>> Hi,
>>
>> I am new to the java service wrapper so far it seems very promising.
>> But I am having a problem setting a system property on windows.  The
>> value of this system property unfortunately has a # character, like
>> this:
>>
>> wrapper.java.additional.11=-Dxenon.service.args=foo#bar
>>
>> Unfortunately our application receives only "foo" as the value of
>> xenon.service.args.
>> I tried escaping # using \# but this would not work and instead got
>> "foo\".
>> I assume this is a bug in how the wrapper C code parses the properties
>> file.
>>
>> Regards,
>> Sebastian
>>
>>
>> -------------------------------------------------------
>> This SF.Net email sponsored by: Free pre-built ASP.NET sites including
>> Data Reports, E-commerce, Portals, and Forums are available now.
>> Download today and enter to win an XBOX or Visual Studio .NET.
>> http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 
>>
>> _______________________________________________
>> Wrapper-user mailing list
>> Wra...@li...
>> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>>
>>  
>>
>
>
>
> -------------------------------------------------------
> This SF.Net email sponsored by: Free pre-built ASP.NET sites including
> Data Reports, E-commerce, Portals, and Forums are available now.
> Download today and enter to win an XBOX or Visual Studio .NET.
> http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 
>
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
>

Re: [Wrapper-user] wrapper.java.additional.11 with # in value

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

Sebastian,
    Sorry, this is currently not possible to do within the wrapper.conf 
file.  An
oversight on my part.  I did find a workaround, but am not sure if it 
will work
for you.

    The Wrapper has the ability to accept any property from the command 
line.
If you launch the Wrapper using the following command, all on a single line
then your application should work:

Wrapper.exe -c ..\conf\wrapper.conf 
wrapper.java.additional.11=-Dxenon.service.args=foo#bar

I opened a bug for this at the following URL.  Please monitor it to 
track its progress.
https://sourceforge.net/tracker/index.php?func=detail&aid=777303&group_id=39428&atid=425187

I have to play around with my options on the different platforms.

Cheers,
Leif

Sebastian Hauer wrote:

>Hi,
>
>I am new to the java service wrapper so far it seems very promising.
>But I am having a problem setting a system property on windows.  The
>value of this system property unfortunately has a # character, like
>this:
>
>wrapper.java.additional.11=-Dxenon.service.args=foo#bar
>
>Unfortunately our application receives only "foo" as the value of
>xenon.service.args.
>I tried escaping # using \# but this would not work and instead got
>"foo\".
>I assume this is a bug in how the wrapper C code parses the properties
>file.
>
>Regards,
>Sebastian
>
>
>-------------------------------------------------------
>This SF.Net email sponsored by: Free pre-built ASP.NET sites including
>Data Reports, E-commerce, Portals, and Forums are available now.
>Download today and enter to win an XBOX or Visual Studio .NET.
>http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01
>_______________________________________________
>Wrapper-user mailing list
>Wra...@li...
>https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>  
>

RE: [Wrapper-user] wrapper.java.additional.11 with # in value

[Luis G. <lui...@sk...>]

Hi thomas,
Yes we did try that.  I tried 'foo#bar' "foo#bar"  and a whole variety
of different \ / @ etc, etc.  I'm out of alternatives to trying
bypassing the #.  The quotes cause the jvm to not even start. Error
below.

wrapper  | Launching a JVM...
jvm 1    | Usage: java [-options] class [args...]
jvm 1    |            (to execute a class)
jvm 1    |    or  java -jar [-options] jarfile [args...]
jvm 1    |            (to execute a jar file)
jvm 1    |

-- Luis


-----Original Message-----
From: Sebastian Hauer=20
Sent: Wednesday, July 23, 2003 6:43 PM
To: Luis Guzman
Subject: FW: [Wrapper-user] wrapper.java.additional.11 with # in value


Did we try that?

-----Original Message-----
From: Thomas Hart [mailto:de...@ha...]=20
Sent: Wednesday, July 23, 2003 6:22 PM
To: wra...@li...
Subject: Re: [Wrapper-user] wrapper.java.additional.11 with # in value


* Sebastian Hauer wrote @ 23. Jul 2003 18:13:
> wrapper.java.additional.11=3D-Dxenon.service.args=3Dfoo#bar

try wrapper.java.additional.11=3D-Dxenon.service.args=3D"foo#bar"

Thomas


-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01
/01
_______________________________________________
Wrapper-user mailing list
Wra...@li...
https://lists.sourceforge.net/lists/listinfo/wrapper-user

Re: [Wrapper-user] wrapper.java.additional.11 with # in value

[Thomas H. <de...@ha...>]

* Sebastian Hauer wrote @ 23. Jul 2003 18:13:
> wrapper.java.additional.11=-Dxenon.service.args=foo#bar

try wrapper.java.additional.11=-Dxenon.service.args="foo#bar"

Thomas

[Wrapper-user] wrapper.java.additional.11 with # in value

[Sebastian H. <seb...@sk...>]

Hi,

I am new to the java service wrapper so far it seems very promising.
But I am having a problem setting a system property on windows.  The
value of this system property unfortunately has a # character, like
this:

wrapper.java.additional.11=3D-Dxenon.service.args=3Dfoo#bar

Unfortunately our application receives only "foo" as the value of
xenon.service.args.
I tried escaping # using \# but this would not work and instead got
"foo\".
I assume this is a bug in how the wrapper C code parses the properties
file.

Regards,
Sebastian

[Wrapper-user] building the win32 native sources

[Patrick W. <wo...@em...>]

Is there a way to build the win32 native sources from
the command-line?  I'm relatively inexperienced with
MSVC6 (normally using mingw).  

I'd even be satisfied with building from inside Visual
Studio but if I open the Wrapper.dsw project file it
seems to be an empty project.  I'm using the sources
pulled from CVS BTW.

Any suggestions?

--Woody

[Wrapper-user] 64bit JVMs

[<da...@ix...>]

Has anyone used wrapper with a 64bit JVM yet?   Interested in experience on
Windows, Linux and Solaris.

Does it build just fine?  And does everything work as expected?

Thanks,
mrc

-- 
     Mike Castle      da...@ix...      www.netcom.com/~dalgoda/
    We are all of us living in the shadow of Manhattan.  -- Watchmen
fatal ("You are in a maze of twisty compiler features, all different"); -- gcc

Re: [Wrapper-user] Wrapper & FreeBSD

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

Please send the modified files along with the version to which you made the
changes to be directly and I will get them added to the next release.

Thanks,
Leif

Alphonse Bendt wrote:

>hi,
>
>if anyone is interested i have successfully compiled and executed
>the java wrapper on my freebsd box:
>freebsd 4.7 with gcc2.95 and gmake.
>
>i had to change the files property.c, wrapper.h, wrapper.c,
>wrapper_unix.c, build.xml and add Makefile.freebsd and a additional c
>file.
>
>regards,
>	alphonse
>
>
>
>
>
>
>-------------------------------------------------------
>This SF.net email is sponsored by: VM Ware
>With VMware you can run multiple operating systems on a single machine.
>WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines at the
>same time. Free trial click here: http://www.vmware.com/wl/offer/345/0
>_______________________________________________
>Wrapper-user mailing list
>Wra...@li...
>https://lists.sourceforge.net/lists/listinfo/wrapper-user
>
>  
>

RE: [Wrapper-user] Wrapper & FreeBSD

[Max S. <MSt...@li...>]

Forgot my e-mail: mst...@li...

-----Original Message-----
From: Alphonse Bendt [mailto:be...@in...] 
Sent: Friday, July 18, 2003 6:41 PM
To: wra...@li...
Subject: [Wrapper-user] Wrapper & FreeBSD

hi,

if anyone is interested i have successfully compiled and executed
the java wrapper on my freebsd box:
freebsd 4.7 with gcc2.95 and gmake.

i had to change the files property.c, wrapper.h, wrapper.c,
wrapper_unix.c, build.xml and add Makefile.freebsd and a additional c
file.

regards,
	alphonse






-------------------------------------------------------
This SF.net email is sponsored by: VM Ware
With VMware you can run multiple operating systems on a single machine.
WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines at the
same time. Free trial click here: http://www.vmware.com/wl/offer/345/0
_______________________________________________
Wrapper-user mailing list
Wra...@li...
https://lists.sourceforge.net/lists/listinfo/wrapper-user

RE: [Wrapper-user] Wrapper & FreeBSD

[Max S. <MSt...@li...>]

Hello Alphonse,

  I would be very interested in your changes. Can you send then to me?
Thanks in advance. 

Max

-----Original Message-----
From: Alphonse Bendt [mailto:be...@in...] 
Sent: Friday, July 18, 2003 6:41 PM
To: wra...@li...
Subject: [Wrapper-user] Wrapper & FreeBSD

hi,

if anyone is interested i have successfully compiled and executed
the java wrapper on my freebsd box:
freebsd 4.7 with gcc2.95 and gmake.

i had to change the files property.c, wrapper.h, wrapper.c,
wrapper_unix.c, build.xml and add Makefile.freebsd and a additional c
file.

regards,
	alphonse






-------------------------------------------------------
This SF.net email is sponsored by: VM Ware
With VMware you can run multiple operating systems on a single machine.
WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines at the
same time. Free trial click here: http://www.vmware.com/wl/offer/345/0
_______________________________________________
Wrapper-user mailing list
Wra...@li...
https://lists.sourceforge.net/lists/listinfo/wrapper-user

[Wrapper-user] Wrapper & FreeBSD

[Alphonse B. <be...@in...>]

hi,

if anyone is interested i have successfully compiled and executed
the java wrapper on my freebsd box:
freebsd 4.7 with gcc2.95 and gmake.

i had to change the files property.c, wrapper.h, wrapper.c,
wrapper_unix.c, build.xml and add Makefile.freebsd and a additional c
file.

regards,
	alphonse

Re: [Wrapper-user] Garbage collection and timeouts

[<da...@ix...>]

In article <3F1...@ta...>,
Leif Mortenson  <wra...@li...> wrote:
>Mike,
>Ouch. You are right about avoiding calling System.gc() Still that is one
>long lock. You are using quite a bit of memory though.

Turns out we aren't calling gc() directly.

One thing I'd noticed was that we were thrashing objects left and right.
Something along the lines of:

while (true) {
  try {
   Object o[];
   o[0] = this;
   do stuff with o...
   }
   maybe break;
}

And the "do stuff with o" wasn't blocking like it was supposed to.  Whoops.

Still, yeah, we do have one processing mode where we allocate several
megs of a bunch of small objects that all go out of scope at once.  Still
working on that one.

>What is the timeout that you are extending? Setting the
>wrapper.ping.timeout=300

I've upped it pretty high for the time being so we can run our other tests
without rebooting in the middle of it.  :->  Correctness first, then speed.

>In general though I have found that newer JVMs do a very good job of
>handling GC
>on their own. Forcing a GC just serves to slow things down. It is
>usually better to
>let the JVM decide when it is best to do a GC. The GC that the JVM does is
>generally must more light weight. It is still taking over a second on
>your app
>though.

One of the problems is that we're running the RMI Server in process, so
that automatically kicks in System.gc() once a minute, regardless.  And
we're not to the point of tweaking the properties that will control that
rate.  Soon, though I think.

We're running 1.4.1_01 right now, but holding off switching to 1.4.2 until
next week.  Still, we do have 3 other GC mechanisms we can look at that
might help our situation.

Thanks!
mrc

-- 
     Mike Castle      da...@ix...      www.netcom.com/~dalgoda/
    We are all of us living in the shadow of Manhattan.  -- Watchmen
fatal ("You are in a maze of twisty compiler features, all different"); -- gcc

Re: [Wrapper-user] Garbage collection and timeouts

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

Mike,
Ouch. You are right about avoiding calling System.gc() Still that is one
long lock. You are using quite a bit of memory though.
What is the timeout that you are extending? Setting the
wrapper.ping.timeout=300
would cause the Wrapper to allow up to 5 minutes for the GC to complete.
If that
is still a problem then you could set it to 600 (10 minutes) But I would
say that
you have other problems long before that. You could also set it to 0,
but as is
described in the docs, I would strongly advise against that as it would
disable the
Wrapper's ability to detect a frozen JVM.
http://wrapper.tanukisoftware.org/doc/english/prop-ping-timeout.html

If you don't already know, you might want to play with the profiler to
try and figure
out what is being GCed. Maybe pooling resources etc could help you to
avoid the
need to perform garbage collection.

In general though I have found that newer JVMs do a very good job of
handling GC
on their own. Forcing a GC just serves to slow things down. It is
usually better to
let the JVM decide when it is best to do a GC. The GC that the JVM does is
generally must more light weight. It is still taking over a second on
your app
though.

Doing GCs during development can be useful to get the minimum amount of
memory that an application needs to run. But it seems to slow things down
because the JVM will usually immediately reallocate the memory just freed so
that it can run.

Cheers,
Leif


Mike Castle wrote:

>Ugh.  It looks like GC takes too long sometimes and Wrapper things the JVM
>has hung:
>
>INFO   | jvm 1    | 2003/07/15 20:36:49 | [GC
>INFO   | jvm 1    | 2003/07/15 20:36:50 |  451745K->409600K(506368K), 1.1287680 secs]
>INFO   | jvm 1    | 2003/07/15 20:36:51 | [Full GC
>DEBUG  | wrapperp | 2003/07/15 20:36:52 | send a packet 103 : ping
>DEBUG  | wrapperp | 2003/07/15 20:36:58 | send a packet 103 : ping
>DEBUG  | wrapperp | 2003/07/15 20:37:04 | send a packet 103 : ping
>DEBUG  | wrapperp | 2003/07/15 20:37:10 | send a packet 103 : ping
>DEBUG  | wrapperp | 2003/07/15 20:37:16 | send a packet 103 : ping
>ERROR  | wrapper  | 2003/07/15 20:37:22 | JVM appears hung: Timed out waiting for signal from JVM.
>ERROR  | wrapper  | 2003/07/15 20:37:22 | JVM did not exit on request, terminated
>
>
>Yeah, I am trying to figure out who is calling System.gc() once a second.
>
>Anyway, since System.gc() causes all other threads to pause until it's
>done... and no matter what timeout you set in wrapper's config file, there
>will eventually be a case where gc takes just a bit longer... how to handle
>this in the general case?  Is there anyway?
>
>Meanwhile, time to figure out who's banging away on gc, read up on GC
>tuning, and yell at the developer to not allocate so much shit.  :-/
>Any recommendations?
>
>mrc
>  
>

[Wrapper-user] Garbage collection and timeouts

[<da...@ix...>]

Ugh.  It looks like GC takes too long sometimes and Wrapper things the JVM
has hung:

INFO   | jvm 1    | 2003/07/15 20:36:49 | [GC
INFO   | jvm 1    | 2003/07/15 20:36:50 |  451745K->409600K(506368K), 1.1287680 secs]
INFO   | jvm 1    | 2003/07/15 20:36:51 | [Full GC
DEBUG  | wrapperp | 2003/07/15 20:36:52 | send a packet 103 : ping
DEBUG  | wrapperp | 2003/07/15 20:36:58 | send a packet 103 : ping
DEBUG  | wrapperp | 2003/07/15 20:37:04 | send a packet 103 : ping
DEBUG  | wrapperp | 2003/07/15 20:37:10 | send a packet 103 : ping
DEBUG  | wrapperp | 2003/07/15 20:37:16 | send a packet 103 : ping
ERROR  | wrapper  | 2003/07/15 20:37:22 | JVM appears hung: Timed out waiting for signal from JVM.
ERROR  | wrapper  | 2003/07/15 20:37:22 | JVM did not exit on request, terminated


Yeah, I am trying to figure out who is calling System.gc() once a second.

Anyway, since System.gc() causes all other threads to pause until it's
done... and no matter what timeout you set in wrapper's config file, there
will eventually be a case where gc takes just a bit longer... how to handle
this in the general case?  Is there anyway?

Meanwhile, time to figure out who's banging away on gc, read up on GC
tuning, and yell at the developer to not allocate so much shit.  :-/
Any recommendations?

mrc
-- 
     Mike Castle      da...@ix...      www.netcom.com/~dalgoda/
    We are all of us living in the shadow of Manhattan.  -- Watchmen
fatal ("You are in a maze of twisty compiler features, all different"); -- gcc

[Wrapper-user] Startup failed: Timed out waiting for signal from JVM

[<lar...@is...>]

hi,

i have tried to get the wrapper 3.0.3 working, but failed. it always =
times out at start.

os: winnt 4.0
jdk: sun 1.4 (also tried sun jdk13, no difference)
cpu-usage by other programs: 1%

that is the program to wrapp, it just writes every secound the date into =
a file.

###########################################################
public class ServiceWrapperTest {
=20
 static File file =3D new File("j://temp//bla");
 static FileWriter out;=20
=20
 private static void writeFile () throws Exception {
  out =3D new FileWriter(file);
  out.write((new Date()).toString());
  out.close();
 }

 public static void main(String[] args) throws Exception {
  file.createNewFile();
  while (true) {
   writeFile();
   Thread.currentThread().sleep(1000);
  }
 }
}
###########################################################

nothing special i think. works fine under jdk14.

now my wrapper.conf:

###########################################################
wrapper.java.command=3D%JAVA_HOME%/bin/java
wrapper.java.mainclass=3Dorg.tanukisoftware.wrapper.WrapperSimpleApp
wrapper.java.classpath.1=3D../lib/wrapper.jar
wrapper.java.library.path.1=3D../lib
wrapper.java.additional.1=3DServiceWrapperTest
wrapper.java.initmemory=3D3
wrapper.java.maxmemory=3D64
wrapper.port=3D1777
wrapper.console.format=3DPM
wrapper.console.loglevel=3DINFO
wrapper.logfile=3D../logs/wrapper.log
wrapper.logfile.format=3DLPTM
wrapper.logfile.loglevel=3DDEBUG
wrapper.logfile.maxsize=3D0
wrapper.logfile.maxfiles=3D0
wrapper.syslog.loglevel=3DNONE
wrapper.ntservice.name=3DIsquare
wrapper.ntservice.displayname=3DIsquare Wrapper
wrapper.ntservice.description=3DIsquare Service Wrapper for Java Classes
wrapper.ntservice.dependency.1=3D
wrapper.ntservice.starttype=3DAUTO_START
wrapper.ntservice.interactive=3Dfalse
###########################################################

after calling bin\iquare.bat, the log says:

STATUS | wrapper  | 2003/07/11 15:53:23 | --> Wrapper Started as Console
DEBUG  | wrapperp | 2003/07/11 15:53:23 | server listening on port 1777.
STATUS | wrapper  | 2003/07/11 15:53:24 | Launching a JVM...
DEBUG  | wrapper  | 2003/07/11 15:53:24 | command: =
"D:\doehling\j2sdk1.4.0_01\bin\java" ServiceWrapperTest -Xms3m -Xmx64m =
-Djava.library.path=3D"../lib" -classpath "../lib/wrapper.jar" =
-Dwrapper.key=3D"m0wGoAWWBv6P1ZVM" -Dwrapper.port=3D1777 =
-Dwrapper.debug=3D"TRUE" -Dwrapper.cpu.timeout=3D"10" =
-Dwrapper.jvmid=3D1 org.tanukisoftware.wrapper.WrapperSimpleApp
DEBUG  | wrapper  | 2003/07/11 15:53:24 | Java Virtual Machine started =
(PID=3D295)
ERROR  | wrapper  | 2003/07/11 15:53:55 | Startup failed: Timed out =
waiting for signal from JVM.
ERROR  | wrapper  | 2003/07/11 15:53:55 | Java Virtual Machine did not =
exit on request, terminated
DEBUG  | wrapper  | 2003/07/11 15:53:55 | JVM was only running for 31 =
seconds leading to a failed restart count of 1.
STATUS | wrapper  | 2003/07/11 15:54:01 | Launching a JVM...
DEBUG  | wrapper  | 2003/07/11 15:54:01 | command: =
"D:\doehling\j2sdk1.4.0_01\bin\java" ServiceWrapperTest -Xms3m -Xmx64m =
-Djava.library.path=3D"../lib" -classpath "../lib/wrapper.jar" =
-Dwrapper.key=3D"9_P5eXGvMFlukTWu" -Dwrapper.port=3D1777 =
-Dwrapper.debug=3D"TRUE" -Dwrapper.cpu.timeout=3D"10" =
-Dwrapper.jvmid=3D2 org.tanukisoftware.wrapper.WrapperSimpleApp
DEBUG  | wrapper  | 2003/07/11 15:54:01 | Java Virtual Machine started =
(PID=3D277)
ERROR  | wrapper  | 2003/07/11 15:54:32 | Startup failed: Timed out =
waiting for signal from JVM.
ERROR  | wrapper  | 2003/07/11 15:54:32 | Java Virtual Machine did not =
exit on request, terminated
DEBUG  | wrapper  | 2003/07/11 15:54:32 | JVM was only running for 31 =
seconds leading to a failed restart count of 2.
STATUS | wrapper  | 2003/07/11 15:54:38 | Launching a JVM...
DEBUG  | wrapper  | 2003/07/11 15:54:38 | command: =
"D:\doehling\j2sdk1.4.0_01\bin\java" ServiceWrapperTest -Xms3m -Xmx64m =
-Djava.library.path=3D"../lib" -classpath "../lib/wrapper.jar" =
-Dwrapper.key=3D"2io0rv3nflNghUlw" -Dwrapper.port=3D1777 =
-Dwrapper.debug=3D"TRUE" -Dwrapper.cpu.timeout=3D"10" =
-Dwrapper.jvmid=3D3 org.tanukisoftware.wrapper.WrapperSimpleApp
DEBUG  | wrapper  | 2003/07/11 15:54:38 | Java Virtual Machine started =
(PID=3D84)
STATUS | wrapper  | 2003/07/11 15:54:48 | CTRL-C trapped.  Shutting =
down.
DEBUG  | wrapper  | 2003/07/11 15:54:48 | wrapperStopProcess(0) called.
DEBUG  | wrapper  | 2003/07/11 15:54:48 | Sending stop signal to JVM
ERROR  | wrapper  | 2003/07/11 15:55:24 | Shutdown failed: Timed out =
waiting for signal from JVM.
ERROR  | wrapper  | 2003/07/11 15:55:24 | Java Virtual Machine did not =
exit on request, terminated
STATUS | wrapper  | 2003/07/11 15:55:24 | <-- Wrapper Stopped

the program is running fine while the jvm is alive, but the version =
string is not printed.
i tried to manipulate wrapper.startup.timeout, but nothing changed.

what am i doing wrong?

Re: [Wrapper-user] Where can I get libwrapper.so library?

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

Matija,
    It is conveniently included with the binary release of the Wrapper.  
Go to do the
download page and download the binary distribution for your platform.
https://sourceforge.net/project/showfiles.php?group_id=39428

    Note that the Windows version has a Wrapper.DLL file.  All other 
platforms are
Unix based and thus use the libwrapper.so file.

Cheers,
Leif

Matija Krajnc wrote:

>Hy
>
>I don't now, where can I get libwrapper.so library. Must I compile original source or can I download it from somewhere?
>
>Thanks, Matija
>  
>

[Wrapper-user] Where can I get libwrapper.so library?

[Matija K. <ma...@po...>]

Hy

I don't now, where can I get libwrapper.so library. Must I compile original source or can I download it from somewhere?

Thanks, Matija
-- 
__________________________________________________________
Sign-up for your own FREE Personalized E-mail at Mail.com
http://www.mail.com/?sr=signup

CareerBuilder.com has over 400,000 jobs. Be smarter about your job search
http://corp.mail.com/careers

[Wrapper-user] Re: Dependance of the wrapper on jre installation

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

Max,
I do this myself all the time. (I am ccing the user list as this is
useful to
others as well) There are several ways to do this:

1) If your application is installed at C:\MyApp\, with its bin directory at
C:\MyApp\bin, then install the jre into a C:\MyApp\jre directory. Then
assuming that your Wrapper.exe file is located inside the C:\MyApp\bin,
you can simply set the following property:
wrapper.java.command=C:/MyApp/jre/bin/java
I would advise using relative paths in this case as it makes it possible for
the user to move the application around. Or at the least, simplifies your
installer:
wrapper.java.command=../jre/bin/java

2) Use an environment variable:
Define an environment variable, MY_JRE, then access the JRE's location
using the following property:
wrapper.java.command=%MY_JRE%/bin/java

In either case, I would define a template wrapper.conf file which contains
the following:
wrapper.java.command=${jre.dir}/bin/java
Then have your installer replace the "jre.dir" token with the actual
location of the JRE. The path can contain a mixture of forward and
backward slashes and the Wrapper will handle it correctly, so
wrapper.java.command=C:\MyApp\jre/bin/java
is ok.

Another option to modifying the wrapper.conf file is to declare the property
from the command line. So you would do the following when launching the
Wrapper.
Wrapper.exe -c ../conf/wrapper.conf
wrapper.java.command=C:/MyApp/jre/bin/java
If the path contains spaces, then include it in quotes:
Wrapper.exe -c ../conf/wrapper.conf
"wrapper.java.command=C:/MyApp/jre/bin/java"

Hope this helps,
Cheers,
Leif

Max Stolyarov wrote:

> Leif,
>
> I have an interesting situation here. I am trying to determine the
> dependence of Java Service Wrapper on the jre installed on the local
> machine. During the installation process my installer searches for a
> JRE package and if it cannot find it it installs a local version of
> jre that it uses by itself and which is not available to other
> programs to use because it does not modify registry settings or path
> or classpath on the local machine. So, my question is this: Is it
> possible to pass to a wrapper parameters that specify where my
> java.exe is located so that the wrapper can use it or do I have to
> have a full and proper JRE installation on the local machine in order
> for the wrapper to work? Thanks in advance.
>
> Max
>

Re: [Wrapper-user] RE: Re: Invoking another Java service from another Java

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

Chris,
I thought I covered both of your questions in my previous post. 

>So the two main questions I have is 
>- Can I make one service spark off another service?
>
I covered two possible ways to do this below.

>- Can I do this programmatically without having to rely on configuration
>files?
>
Yes and no.  As I said before. The Wrapper requires a configuration 
file, but the
file can be empty.  You can define all properties by using a very long 
command line.
For example

Wrapper.exe -c empty.conf wrapper.java.command=java wrapper.debug=true <etc>

See:
http://www.tanukisoftware.org/doc/english/props-command-line.html

Let me know what you end up using or if you have any more questions.

Cheers,
Leif

>Chris,
>    Not sure exactly what you are wanting to do.  But individual NT 
>Services are
>independent of each other.  If you wanted to start a second service when
>
>the first
>one was started, your would want to install them both normally.  Then 
>you would
>trigger the startup of the second service from within the first using
>the java.lang.Runtime.exec methods and the "net" command with "start",
>"{service name}" arguments.  Have not tested this, but it should work.
>In
>Wrapper v3.0.4, you will also be able to execute Wrapper.exe with a "-t"
>parameter to start the service.
>
>    If you simply want to launch a second JVM from a Java application 
>running as
>a service, then you can do so by, once again, using the 
>java.lang.Runtime.exec
>method.  Only this time, execute Java.exe with all the arguments that 
>apply.  This
>second JVM will be a child process of the first so it will be a part of 
>the original
>service.  The problem here is that you have to be careful about how you 
>shutdown
>the service to make sure that the child JVM is always killed.  Failure 
>to do so will
>result in a zombie process that I believe you will have to reboot to get
>
>rid of.
>
>    If you want to execute a copy of the Wrapper from the exec method, 
>you should
>be able to specify any property in the wrapper.conf file from the 
>command line.
>The Wrapper always requires that a wrapper.conf exists.  But if 
>everything is on
>the command line, then the conf file can be empty.  This would result in
>
>a very long
>command line however, so I would only put the parameters that need to be
>dynamic on the command line.
>
>Let me know if none of these answers actually answers your question.
>
>Cheers,
>Leif
>  
>
>  
>

[Wrapper-user] RE: Re: Invoking another Java service from another Java

[Chris K. <chr...@en...>]

Thanks for the response.

The problem I'm trying to solve is to launch a Java program remotely. I
have an invoker that runs as a service but invokes other programs but
would be neat if I could invoke these programs as services also so that
I gain the recoverability functionality gained by using the Java Service
Wrapper. 

So the two main questions I have is 
- Can I make one service spark off another service?
- Can I do this programmatically without having to rely on configuration
files?

Thanks for any guidance in advance.

-----Original Message-----
From: wra...@li...
[mailto:wra...@li...] On Behalf Of
wra...@li...
Sent: 10 July 2003 04:30
To: wra...@li...
Subject: Wrapper-user digest, Vol 1 #132 - 4 msgs


Send Wrapper-user mailing list submissions to
	wra...@li...

To subscribe or unsubscribe via the World Wide Web, visit
	https://lists.sourceforge.net/lists/listinfo/wrapper-user

Message: 1
Date: Wed, 09 Jul 2003 13:03:33 +0900
From: Leif Mortenson <le...@ta...>
To:  wra...@li...
Subject: Re: [Wrapper-user] Invoking another Java service from another
Java  service wrapper
Reply-To: wra...@li...

This is a multi-part message in MIME format.
--------------060200040702020606050204
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit

Chris,
    Not sure exactly what you are wanting to do.  But individual NT 
Services are
independent of each other.  If you wanted to start a second service when

the first
one was started, your would want to install them both normally.  Then 
you would
trigger the startup of the second service from within the first using
the java.lang.Runtime.exec methods and the "net" command with "start",
"{service name}" arguments.  Have not tested this, but it should work.
In
Wrapper v3.0.4, you will also be able to execute Wrapper.exe with a "-t"
parameter to start the service.

    If you simply want to launch a second JVM from a Java application 
running as
a service, then you can do so by, once again, using the 
java.lang.Runtime.exec
method.  Only this time, execute Java.exe with all the arguments that 
apply.  This
second JVM will be a child process of the first so it will be a part of 
the original
service.  The problem here is that you have to be careful about how you 
shutdown
the service to make sure that the child JVM is always killed.  Failure 
to do so will
result in a zombie process that I believe you will have to reboot to get

rid of.

    If you want to execute a copy of the Wrapper from the exec method, 
you should
be able to specify any property in the wrapper.conf file from the 
command line.
The Wrapper always requires that a wrapper.conf exists.  But if 
everything is on
the command line, then the conf file can be empty.  This would result in

a very long
command line however, so I would only put the parameters that need to be
dynamic on the command line.

Let me know if none of these answers actually answers your question.

Cheers,
Leif

Chris Kelly wrote:

> Is there a way of programmatically invoking a Java service from
> another Java service.
> What I'm looking for is to invoke a Java service without use of 
> explicit configuration files. I could supply this as a
> list of properties instead?
>  








This email is confidential and may be protected by legal privilege. If you are not the intended recipient,  please do not copy or disclose its content but  delete the email and contact the sender immediately. Whilst we run antivirus software on all internet emails we are not liable for any loss or damage. The recipient is advised to run their own antivirus software.

Re: [Wrapper-user] Hanging terminal window after wrapper starts

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

Mirek,
I have seen this before as well.  I believe it is being caused by the 
way the Wrapper
is launching the Wrapper and JVM as detached processes.  I think the hang is
being caused by the shell thinking that there are are still processes 
running which
were launched by the current shell and it is waiting for them to 
complete.  Just a
guess though...

Even if you kill the shell, the Wrapper and its JVM continue to run in 
the background
without any problems...

I'll take a look at it again and see if I can figure out a way to avoid 
this..

If anyone has any ideas, I'd appreciate the input.

Cheers,
Leif

ms...@em... wrote:

>Hello,
>I have this little problem. I'm logged to remote server(Red Hat Linux 9)
>via ssh from Windows(Putty) or Linux(Xterm,Konsole).
>On the remote server I start wrapper (v.3.0.3), all is OK.
>But when I logged out then black screen appers and I have kill terminal
>window. In other session window user seems still log in.
>Does anobody know where the problem should be?
>
>Mirek
>
>  
>

[Wrapper-user] Hanging terminal window after wrapper starts

[<ms...@em...>]

Hello,
I have this little problem. I'm logged to remote server(Red Hat Linux 9)
via ssh from Windows(Putty) or Linux(Xterm,Konsole).
On the remote server I start wrapper (v.3.0.3), all is OK.
But when I logged out then black screen appers and I have kill terminal
window. In other session window user seems still log in.
Does anobody know where the problem should be?

Mirek

_________________________________________________________________________=
_______
ZNA=C8KOV=C9 PO=C8=CDTA=C8E DEXX
- v=EDt=ECz testu hern=EDch PC v Computeru 11/2003, dr=BEitel Chip TIP a =
TIP
Po=E8=EDta=E8 pro ka=BEd=E9ho
- cenov=FD trh=E1k - PC s Pentium 4 2,66 GHz, CDRW, DVD, 17" a tisk=E1rno=
u za
22.990 s DPH!
http://www.email.cz/dexx

Re: [Wrapper-user] Invoking another Java service from another Java service wrapper

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

Chris,
    Not sure exactly what you are wanting to do.  But individual NT 
Services are
independent of each other.  If you wanted to start a second service when 
the first
one was started, your would want to install them both normally.  Then 
you would
trigger the startup of the second service from within the first using the
java.lang.Runtime.exec methods and the "net" command with "start",
"{service name}" arguments.  Have not tested this, but it should work.   In
Wrapper v3.0.4, you will also be able to execute Wrapper.exe with a "-t"
parameter to start the service.

    If you simply want to launch a second JVM from a Java application 
running as
a service, then you can do so by, once again, using the 
java.lang.Runtime.exec
method.  Only this time, execute Java.exe with all the arguments that 
apply.  This
second JVM will be a child process of the first so it will be a part of 
the original
service.  The problem here is that you have to be careful about how you 
shutdown
the service to make sure that the child JVM is always killed.  Failure 
to do so will
result in a zombie process that I believe you will have to reboot to get 
rid of.

    If you want to execute a copy of the Wrapper from the exec method, 
you should
be able to specify any property in the wrapper.conf file from the 
command line.
The Wrapper always requires that a wrapper.conf exists.  But if 
everything is on
the command line, then the conf file can be empty.  This would result in 
a very long
command line however, so I would only put the parameters that need to be
dynamic on the command line.

Let me know if none of these answers actually answers your question.

Cheers,
Leif

Chris Kelly wrote:

> Is there a way of programmatically invoking a Java service from 
> another Java service.
> What I'm looking for is to invoke a Java service without use of 
> explicit configuration files. I could supply this as a
> list of properties instead?
>