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

Re: [Wrapper-user] Shell script won't execute on Mac

[Mark L. <mid...@ve...>]

Chris wrote:
> Are you sure you don't have to type "./RCBServer" to run a local script?
> Otherwise (without the "./"), won't it just look in the PATH for the 
> RCBServer executable?
>
> Does the first line in the script point to a real interpreter on your 
> system? i.e.
> #!/path/to/sh
>
Thanks for responding. I tried ./RCBServer just now, and still get the 
"Command not found" response. Also the first line in the script is 
#!/bin/sh. I do have sh at that location, and when I type "/bin/sh" I 
get a shell prompt. I verified that the execute bit is set for all three 
groups on /bin/sh. Also, I have many other scripts that I can run by 
typing their name. For example I have a Tomcat startup script named 
Tomcat that I can execute by typing Tomcat from the script directory. It 
has the same first line as the script that won't execute. The Wrapper 
script came with a space after !, i.e. #! /bin/sh. My Tomcat script 
doesn't have this space, i.e. #!/bin/sh, so I tried that also and still 
get the Command not found error.

If I try to launch the script via double-clicking in the Finder, the 
system asks me what application I want to use to open that document. But 
the metadata for the file says that it's a Unix executable file. I don't 
know if that's normal, or a hint to what is preventing this script from 
executing.

-Mark

Re: [Wrapper-user] Shell script won't execute on Mac

[Chris <ch...@hm...>]

Are you sure you don't have to type "./RCBServer" to run a local script?
Otherwise (without the "./"), won't it just look in the PATH for the 
RCBServer executable?

Does the first line in the script point to a real interpreter on your 
system? i.e.
#!/path/to/sh

HTH
Chris



Mark Leone wrote:
> I was able to get Wrapper running on Windows in about 10 minutes, but 3 
> hours so far trying to make it work with the Mac have been fruitless. 
> The problem is simple. I copied script.sh.in to my app's /bin directory, 
> renamed it RCBServer (my app name), set its file mode to 774, edited the 
> app name variable settings as directed in the instructions, and made 
> sure to save the script file with Unix line breaks. But when I type 
> "RCBServer" from the bin directory that contains the script file with 
> that name, I get the error message "Command not found."  I can run shell 
> scripts anywhere else on my Mac in this fashion, but this one just won't 
> run for some reason. I must be missing something really basic, but it's 
> obviously not going to occur to me without help.
> 
> -Mark
> 
> 
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Wrapper-user mailing list
> Wra...@li...
> https://lists.sourceforge.net/lists/listinfo/wrapper-user
> 
> PLEASE NOTE: THE ABOVE MESSAGE WAS RECEIVED FROM THE INTERNET.
> On entering the GSI, this email was scanned for viruses by the Government Secure Intranet (GSi) virus scanning service supplied exclusively by Cable & Wireless in partnership with MessageLabs.
> In case of problems, please call your organisational IT Helpdesk.
> The MessageLabs Anti Virus Service is the first managed service to achieve the CSIA Claims Tested Mark (CCTM Certificate Number 2006/04/0007), the UK Government quality mark initiative for information security products and services.  For more information about this please visit www.cctmark.gov.uk
> 

-- 
Chris
HMGCC

The information contained in this message (and any attachments) may
be confidential and is intended for the sole use of the named addressee.
Access, copying, alteration or re-use of the e-mail by anyone other
than the intended recipient is unauthorised. If you are not the intended
recipient please advise the sender immediately by returning the e-mail
and deleting it from your system.

This information may be exempt from disclosure under Freedom Of Information 
Act 2000 and may be subject to exemption under other UK information 
legislation. Refer disclosure requests to the Information Officer.


The original of this email was scanned for viruses by Government Secure Intranet (GSi)  virus scanning service supplied exclusively by Cable & Wireless in partnership with MessageLabs.
On leaving the GSI this email was certified virus free.
The MessageLabs Anti Virus Service is the first managed service to achieve the CSIA Claims Tested Mark (CCTM Certificate Number 2006/04/0007), the UK Government quality mark initiative for information security products and services.  For more information about this please visit www.cctmark.gov.uk

[Wrapper-user] Shell script won't execute on Mac

[Mark L. <mid...@ve...>]

I was able to get Wrapper running on Windows in about 10 minutes, but 3 
hours so far trying to make it work with the Mac have been fruitless. 
The problem is simple. I copied script.sh.in to my app's /bin directory, 
renamed it RCBServer (my app name), set its file mode to 774, edited the 
app name variable settings as directed in the instructions, and made 
sure to save the script file with Unix line breaks. But when I type 
"RCBServer" from the bin directory that contains the script file with 
that name, I get the error message "Command not found."  I can run shell 
scripts anywhere else on my Mac in this fashion, but this one just won't 
run for some reason. I must be missing something really basic, but it's 
obviously not going to occur to me without help.

-Mark

[Wrapper-user] HP-UX 3.2.3 Released

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

Hi all,
The 32 and 64-bit parisc HP-UX 3.2.3 distributions were released today.
A big thanks to Nicolas Varney who also is helping out with the AIX
releases.

Thanks also to Ben Jansen, who has been helping users with these
releases in the midterm.

Please give them a try and let me know if you encounter any
problems.

Cheers,
Leif

Re: [Wrapper-user] updated HP-UX binaries

[Santo <san...@ya...>]

I was at our customer's offices the whole week to do some more installation=
 and configuration tests of our application on HP-UX=0Aand the upgrade of t=
he wrapper executable to v3.2.3 (the missing part ;-)) solved all problems =
I was experiencing before.=0A=0AThanks again for the effort !=0A=0ASanto=0A=
=0A----- Original Message ----=0AFrom: Santo <san...@ya...>=0AT=
o: Ben Jansen <bj...@tr...>; wra...@li...=0A=
Sent: Thursday, February 15, 2007 8:59:58 PM=0ASubject: Re: [Wrapper-user] =
updated HP-UX binaries=0A=0ANo problem Ben.=0A=0AThanks for the quick respo=
nse.=0A=0A =0A=0ASanto=0A=0A=0A=0A----- Original Message ----=0AFrom: Ben J=
ansen <bj...@tr...>=0ATo: Santo <san...@ya...>; wrappe=
r-...@li...=0ASent: Thursday, February 15, 2007 7:00:26 PM=
=0ASubject: RE: [Wrapper-user] updated HP-UX binaries=0A=0A=0AHello,=0A=0AS=
orry about forgetting that.=0A=0AI've attached the binary (gzipped).=0A=0A-=
 Ben=0A=0A=0A-----Original Message-----=0AFrom: Santo [mailto:santo_portabl=
e...@ya...]=0ASent: Wed 2/14/2007 11:39 PM=0ATo: wra...@li...=
forge.net=0ACc: Ben Jansen=0ASubject: Re: [Wrapper-user] updated HP-UX bina=
ries=0A=0AHi Ben,=0A=0Ain the meantime I deployed our application on the HP=
-UX system at our customer together with the service wrapper you compiled f=
or me,=0Abut it's still not handling the signals correctly (which should be=
 fixed in v3.2.1).=0AI was wondering if this could be caused by the fact th=
at I'm using the wrapper executable from v3.2.0 because that was not=0Aavai=
lable in you tar.gz file.=0A=0ASo in fact I'm running the wrapper in the fo=
llowing configuration:=0Aeverything v3.2.0, except:=0A- libwrapper.sl -> v3=
.2.3=0A- wrapper.jar -> v3.2.3=0A=0ACould you please send me the wrapper ex=
ecutable itself (for v3.2.3) too=0A ?=0A=0AThanks in advance,=0A=0ASanto =
=0A=0A=0A----- Original Message ----=0AFrom: Santo <san...@ya...=
m>=0ATo: wra...@li...=0ASent: Friday, February 2, 200=
7 5:01:12 PM=0ASubject: Re: [Wrapper-user] updated HP-UX binaries=0A=0A=0AW=
onderful !=0A=0ASanto=0A=0A=0A----- Original Message ----=0AFrom: Ben Janse=
n <bj...@tr...>=0ATo: wra...@li...=0ASent: T=
hursday, February 1, 2007 10:47:22 PM=0ASubject: Re: [Wrapper-user] updated=
 HP-UX binaries=0A=0A=0AHello,=0A=0ANo problem :-)=0A=0AThis is a 32-bit ve=
rsion for PA-RISC.=0A=0A- Ben=0A=0A-----Original Message-----=0AFrom: wrapp=
er-...@li... on behalf of First Last=0ASent: Thu 2=
/1/2007 2:01 AM=0ATo: wra...@li...=0ASubject: Re: [Wr=
apper-user] updated HP-UX binaries=0A=0ABen,=0A=0AThis is really great, tha=
nk you very much !=0AOne question though: Is this a 32-bit version or=0A 64=
-bit ?=0AI'd like a 32-bit version, but you don't mention this and I can te=
st it ;-)=0A=0AThanks again.=0A=0A=0ASanto=0A------------------------------=
-------------------------------------------=0AUsing Tomcat but need to do m=
ore? Need to support web services, security?=0AGet stuff done quickly with =
pre-integrated technology to make your job easier.=0ADownload IBM WebSphere=
 Application Server v.1.0.1 based on Apache Geronimo=0Ahttp://sel.as-us.fal=
kag.net/sel?cmd=3Dlnk&kid=3D120709&bid=3D263057&dat=3D121642=0A____________=
___________________________________=0AWrapper-user mailing list=0AWrapper-u=
se...@li...=0Ahttps://lists.sourceforge.net/lists/listinfo/wr=
apper-user=0A=0A=0A=0A=0A=0AGet your own web address.=0AHave a HUGE year th=
rough=0A Yahoo! Small Business.=0A-----------------------------------------=
--------------------------------=0AUsing Tomcat but need to do more? Need t=
o support web services, security?=0AGet stuff done quickly with pre-integra=
ted technology to make your job easier.=0ADownload IBM WebSphere Applicatio=
n Server v.1.0.1 based on Apache Geronimo=0Ahttp://sel.as-us.falkag.net/sel=
?cmd=3Dlnk&kid=3D120709&bid=3D263057&dat=3D121642=0A_______________________=
________________________=0AWrapper-user mailing list=0AWrapper-user@lists.s=
ourceforge.net=0Ahttps://lists.sourceforge.net/lists/listinfo/wrapper-user=
=0A=0A=0A=0A_______________________________________________________________=
_____________________=0AThe fish are biting. =0AGet more visitors on your s=
ite using Yahoo! Search=0A Marketing.=0Ahttp://searchmarketing.yahoo.com/ar=
p/sponsoredsearch_v2.php=0A=0A=0A=0A=0A=0A=0A=0A=0AIt's here! Your new mess=
age!=0AGet=0A new email alerts with the free Yahoo! Toolbar.---------------=
----------------------------------------------------------=0ATake Surveys. =
Earn Cash. Influence the Future of IT=0AJoin SourceForge.net's Techsay pane=
l and you'll get the chance to share your=0Aopinions on IT & business topic=
s through brief surveys-and earn cash=0Ahttp://www.techsay.com/default.php?=
page=3Djoin.php&p=3Dsourceforge&CID=3DDEVDEV=0A____________________________=
___________________=0AWrapper-user mailing lis...@li...=
forge.net=0Ahttps://lists.sourceforge.net/lists/listinfo/wrapper-user=0A=0A=
=0A=0A=0A=0A=0A=0A =0A_____________________________________________________=
_______________________________=0AThe fish are biting. =0AGet more visitors=
 on your site using Yahoo! Search Marketing.=0Ahttp://searchmarketing.yahoo=
.com/arp/sponsoredsearch_v2.php

Re: [Wrapper-user] How to disable wrapper stopping, when killing JVM ?

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

Martin,
The version number was important.  If you encounter problems with an old 
version
in the future, please check the release notes.
http://wrapper.tanukisoftware.org/doc/english/release-notes.html

There was a bug fixed in 3.2.2 where the TERM signal handler was being 
disabled when
a CHLD signal was received.  This would happen when the JVM was 
restarted for any
reason.

There was also this particular bug fixed in 3.2.0 where the TERM signals 
were not being
trapped at all by the JVM process.

If you upgrade to 3.2.3, things should work correctly for you.   The 
Wrapper will restart
a JVM killed by KILL -9.  There is no way to trap that and prevent the 
JVM from being
killed.   It will however, completely ignore TERM signals so your JVM 
will never be
restarted.

Cheers,
Leif

Martin Gazak wrote:
> ---------- Original Message -----------
> From: Leif Mortenson <le...@ta...>
> To: wra...@li...
> Sent: Thu, 22 Feb 2007 10:39:07 +0900
> Subject: Re: [Wrapper-user] How to disable wrapper stopping, when killing 
> JVM ?
>
>   
>> Martin,
>> I just retested this and it is working correctly for me. Most likely,
>> this is a misunderstanding about when it is used. The wrapper's
>> sh script by default will only use the ignore signals setting when it
>> is being run as a daemon process. Ie when "start" is used, not
>> "console".
>> Let me know your use case as it may require some changes.
>>     
>
> Leif,
> thanks for the answer. I am sorry for not being precise - we start the 
> wrapper version 3.1.2 (and therefore our application) as Linux (SuSE 10.0)
> service by "start" parameter, not "console", therefore my problems are 
> related to daemon mode. 
> The start script includes the line
> IGNORE_SIGNALS=true
>
> Our wrapper.conf also includes line
> wrapper.ignore_signals=true
>
> We want to run wrapper as service and we want the JVM to be restarted by 
> wrapper no matter what signal is sent to JVM process (in this post I am not 
> speaking about signals sent to wrapper process, but signals sent to spawned 
> java process).
> Anyway "kill -9 " (SIGKILL) sent to JVM  results even now in desired 
> behaviour (wrapper detects unexpected termination of JVM and restarts it), 
> however "kill " (SIGTERM?) sent to JVM is recognized by wrapper as a signal 
> to be stopped. This stopping is what I need to change, the JVM should be 
> restarted.
>  
>   
>> If you want to ignore signals in console mode as well, you still need
>> to set the IGNORE_SIGNALS flag in the sh script to put the sh script
>> into that mode, but in your wrapper.conf file, add the following
>> property:
>> wrapper.ignore_signals=TRUE
>>     
>
> We speak about daeomon mode. We use also console mode, however only for 
> debugging.
>
> Best regards
> Martin
>   

Re: [Wrapper-user] How to disable wrapper stopping, when killing JVM ?

[Martin G. <ma...@mi...>]

---------- Original Message -----------
From: Leif Mortenson <le...@ta...>
To: wra...@li...
Sent: Thu, 22 Feb 2007 10:39:07 +0900
Subject: Re: [Wrapper-user] How to disable wrapper stopping, when killing 
JVM ?

> Martin,
> I just retested this and it is working correctly for me. Most likely,
> this is a misunderstanding about when it is used. The wrapper's
> sh script by default will only use the ignore signals setting when it
> is being run as a daemon process. Ie when "start" is used, not
> "console".
> Let me know your use case as it may require some changes.

Leif,
thanks for the answer. I am sorry for not being precise - we start the 
wrapper version 3.1.2 (and therefore our application) as Linux (SuSE 10.0)
service by "start" parameter, not "console", therefore my problems are 
related to daemon mode. 
The start script includes the line
IGNORE_SIGNALS=true

Our wrapper.conf also includes line
wrapper.ignore_signals=true

We want to run wrapper as service and we want the JVM to be restarted by 
wrapper no matter what signal is sent to JVM process (in this post I am not 
speaking about signals sent to wrapper process, but signals sent to spawned 
java process).
Anyway "kill -9 " (SIGKILL) sent to JVM  results even now in desired 
behaviour (wrapper detects unexpected termination of JVM and restarts it), 
however "kill " (SIGTERM?) sent to JVM is recognized by wrapper as a signal 
to be stopped. This stopping is what I need to change, the JVM should be 
restarted.
 
> 
> If you want to ignore signals in console mode as well, you still need
> to set the IGNORE_SIGNALS flag in the sh script to put the sh script
> into that mode, but in your wrapper.conf file, add the following
> property:
> wrapper.ignore_signals=TRUE

We speak about daeomon mode. We use also console mode, however only for 
debugging.

Best regards
Martin

Re: [Wrapper-user] logging: is there a way to switch the logging level at runtime

[Hubert F. <hub...@ab...>]

Wow, what a cool feature !
hub

>>> Leif Mortenson <le...@ta...> 22.02.2007 02:27 >>>
Peter,
    I added a feature to do just this a little while back.  It is done=20
using a command
file that the Wrapper periodically reads and executes.   There are a=20
number of
commands which are possible.  Take a look at this page.  It is mainly=20
meant as an
administration tool.
http://wrapper.tanukisoftware.org/doc/english/prop-commandfile.html=20

    Could you describe your use case a little bit more.  There may be=20
ways that this
could be improved.

    Be aware that the Wrapper's logging system is not meant to be used=20
to replace
other logging systems like Log4j or Excalibur Logkit (my favorite).

Cheers,
Leif

peter ellis wrote:
> It is possible to change the logging level within my server.conf file
> post runtime so that i can switch from'INFO' logging level to say
> DEBUG or ALL levels?? How can i achieve this easily using the
> wrapper!?
>
>  =20


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share =
your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge&CID=3DDE=
VDEV=20
_______________________________________________
Wrapper-user mailing list
Wra...@li...=20
https://lists.sourceforge.net/lists/listinfo/wrapper-user

Re: [Wrapper-user] configuring mulitple processes

[Donal M. <dom...@ya...>]

Hi,=0A=0AThanks for the response Leif. I stumbled across the #include optio=
n shortly after I sent the message below.=0A=0ARegarding the way the 4 proc=
esses are set up, all the .conf files *are* in the same directory. I notice=
d that the .sh script (on Solaris) has the .conf file path and APP_NAME wit=
hin the script itself, so I've modified this scripts to accept these as par=
ameters.=0A=0AIn the .conf files for the processes themselves (rather than =
in the common .conf file that they #include), I've specified a different wr=
apper log file for each process. If I also add a unique wrapper.java.pidfil=
e property for each process, then are there any other changes I need to mak=
e in order to safely execute all 4 processes simultaneously?=0A=0ABTW, I ex=
pect you already know this, but the "su" command in the script does not wor=
k on Solaris as it does not accept a "-m" option. I'll look into a workarou=
nd for this tomorrow and can let you know what I come up with if you're int=
erested?=0A=0AThanks for providing such a great tool!=0A- Donal=0A=0A----- =
Original Message ----=0AFrom: Leif Mortenson <le...@ta...>=0ATo=
: wra...@li...=0ASent: Wednesday, 21 February, 2007 8=
:22:08 PM=0ASubject: Re: [Wrapper-user] configuring mulitple processes=0A=
=0ADonal,=0A    Each of the 4 java processes are independent applications. =
 Are they =0Aall installed=0Ain the same directory as well?  How are things=
 set up?=0A=0A    If they are different installs, then you could generate y=
our =0Awrapper.conf files=0Aat build time my making the main class name a t=
oken that gets replaced.=0A=0A    If all 4 applications will be using the s=
ame install directory or =0Athe above does not=0Awork, another option is to=
 move everything other than the main class, =0Aservice name,=0Aand log file=
 into a file called conf/wrapper-common.conf=0A    Then modify your wrapper=
.conf to look like this:=0A---=0A#include ../conf/wrapper-common.conf=0Awra=
pper.app.parameter.1=3Dfoo.bar.MyProcessMainClass=0Awrapper.logfile=3D../lo=
gs/wrapper-app1.log=0Awrapper.console.title=3DApp1=0Awrapper.ntservice.name=
=3DApp1=0Awrapper.ntservice.displayname=3DApp1=0Awrapper.ntservice.descript=
ion=3DApp1Desc=0A---=0ASee this page for details.=0Ahttp://wrapper.tanukiso=
ftware.org/doc/english/props-cascading.html=0A=0ANote that if all 4 apps ar=
e in the same directory, you are also going to =0Aneed to make sure=0Athey =
all have their own wrapper.log file.  Any use of pid, state, or =0Alock fil=
es will also need=0Ato avoid conflicts.=0A=0AYou could also keep a single w=
rapper.conf and then set the above =0Aproperties from=0Athe wrapper command=
 line.  If you are using a bat or sh file, this would =0Abe feasible.=0Ahtt=
p://wrapper.tanukisoftware.org/doc/english/props-command-line.html=0A=0AUsi=
ng environment variables and then dereferencing them in the =0Awrapper.conf=
 is another=0Aoption.=0Ahttp://wrapper.tanukisoftware.org/doc/english/props=
-envvars.html=0A=0AIn general, the Wrapper is not designed to have more tha=
n one instance =0Arunning out=0Aof the same directory at the same time.  It=
 is possible, but you will =0Ahave some work=0Ato make sure there are no fi=
le conflicts etc.  The sh script on linux =0Awill work if you set=0Ause set=
 unique APP_NAME and WRAPPER_CONF values in the sh script.=0A=0ACheers,=0AL=
eif=0A=0ADonal Murtagh wrote:=0A> Hi,=0A>=0A> I have 4 java processes that =
I wish to wrap. Each one has a different main class, but apart from that th=
ey are identical (in terms of JVM parameters, application parameters, etc.)=
.=0A>=0A> Therefore, I currently  have 4 wrapper.conf files - one for each =
process - which are identical apart from=0A>=0A> wrapper.app.parameter.1=3D=
foo.bar.MyProcessMainClass=0A>=0A> Is there any way I can avoid the duplica=
ting the JVM parameters, application parameters for each process?=0A>=0A> T=
hanks,=0A> TD=0A>   =0A=0A=0A----------------------------------------------=
---------------------------=0ATake Surveys. Earn Cash. Influence the Future=
 of IT=0AJoin SourceForge.net's Techsay panel and you'll get the chance to =
share your=0Aopinions on IT & business topics through brief surveys-and ear=
n cash=0Ahttp://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge=
&CID=3DDEVDEV=0A_______________________________________________=0AWrapper-u=
ser mailing lis...@li...=0Ahttps://lists.sourc=
eforge.net/lists/listinfo/wrapper-user=0A=0A=0A=0A=0A=0A=09=09=0A__________=
_________________________________________________ =0ANow you can scan email=
s quickly with a reading pane. Get the new Yahoo! Mail. http://uk.docs.yaho=
o.com/nowyoucan.html

Re: [Wrapper-user] How to disable wrapper stopping, when killing JVM ?

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

Martin,
I just retested this and it is working correctly for me. Most likely,
this is a misunderstanding about when it is used. The wrapper's
sh script by default will only use the ignore signals setting when it
is being run as a daemon process. Ie when "start" is used, not
"console".

The thinking here is that daemon processes are the ones that you
really want to be protecting from stray signals. "console" mode is
mainly used for development as the wrapper would be killed when the
shell was closed by default anyway.

Let me know your use case as it may require some changes.

If you want to ignore signals in console mode as well, you still need
to set the IGNORE_SIGNALS flag in the sh script to put the sh script
into that mode, but in your wrapper.conf file, add the following
property:
wrapper.ignore_signals=TRUE

When IGNORE_SIGNALS is set, the sh script also causes the wrapper to
create an anchor file. It then uses that to tell the wrapper to
shut down. When IGNORE_SIGNALS is not set, the sh script attempts
to shut down the wrapper by sending a TERM signal.

Cheers,
Leif

Martin Gazak wrote:
>  Good morning,
>  I am using Java Service Wrapper under Linux and I am facing following
>  problem:
>  When I kill the JVM process (not wrapper, but JVM) with signal "kill
>  -9", wrapper logs:
>
>  ERROR  | wrapper  | 2007/02/21 08:36:48 | JVM exited unexpectedly.
>  STATUS | wrapper  | 2007/02/21 08:36:53 | Launching a JVM...
>
>  and restarts the JVM which is what I want.
>
>  When I kill the JVM process (not wrapper, but JVM) with just "kill",
>  wrapper logs:
>
>  STATUS | wrapper  | 2007/02/21 08:32:25 | <-- Wrapper Stopped
>
>  and stops without restarting JVM.
>  Is there any way how to configure wrapper (or to change application ?)
>  to achieve, that wrapper restarts JVM no matter what signal JVM received ?
>
>  I played with wrapper.ignore_signal property, but it seems to work on
>  wrapper, not on the spawned JVM and therefore it did not help me.
>
>  Thanks.
>
>  With best regards
>
>  Martin Gazak
>  MicroStep-MIS
>   

Re: [Wrapper-user] logging: is there a way to switch the logging level at runtime

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

Peter,
    I added a feature to do just this a little while back.  It is done 
using a command
file that the Wrapper periodically reads and executes.   There are a 
number of
commands which are possible.  Take a look at this page.  It is mainly 
meant as an
administration tool.
http://wrapper.tanukisoftware.org/doc/english/prop-commandfile.html

    Could you describe your use case a little bit more.  There may be 
ways that this
could be improved.

    Be aware that the Wrapper's logging system is not meant to be used 
to replace
other logging systems like Log4j or Excalibur Logkit (my favorite).

Cheers,
Leif

peter ellis wrote:
> It is possible to change the logging level within my server.conf file
> post runtime so that i can switch from'INFO' logging level to say
> DEBUG or ALL levels?? How can i achieve this easily using the
> wrapper!?
>
>   

Re: [Wrapper-user] configuring mulitple processes

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

Donal,
    Each of the 4 java processes are independent applications.  Are they 
all installed
in the same directory as well?  How are things set up?

    If they are different installs, then you could generate your 
wrapper.conf files
at build time my making the main class name a token that gets replaced.

    If all 4 applications will be using the same install directory or 
the above does not
work, another option is to move everything other than the main class, 
service name,
and log file into a file called conf/wrapper-common.conf
    Then modify your wrapper.conf to look like this:
---
#include ../conf/wrapper-common.conf
wrapper.app.parameter.1=foo.bar.MyProcessMainClass
wrapper.logfile=../logs/wrapper-app1.log
wrapper.console.title=App1
wrapper.ntservice.name=App1
wrapper.ntservice.displayname=App1
wrapper.ntservice.description=App1Desc
---
See this page for details.
http://wrapper.tanukisoftware.org/doc/english/props-cascading.html

Note that if all 4 apps are in the same directory, you are also going to 
need to make sure
they all have their own wrapper.log file.  Any use of pid, state, or 
lock files will also need
to avoid conflicts.

You could also keep a single wrapper.conf and then set the above 
properties from
the wrapper command line.  If you are using a bat or sh file, this would 
be feasible.
http://wrapper.tanukisoftware.org/doc/english/props-command-line.html

Using environment variables and then dereferencing them in the 
wrapper.conf is another
option.
http://wrapper.tanukisoftware.org/doc/english/props-envvars.html

In general, the Wrapper is not designed to have more than one instance 
running out
of the same directory at the same time.  It is possible, but you will 
have some work
to make sure there are no file conflicts etc.  The sh script on linux 
will work if you set
use set unique APP_NAME and WRAPPER_CONF values in the sh script.

Cheers,
Leif

Donal Murtagh wrote:
> Hi,
>
> I have 4 java processes that I wish to wrap. Each one has a different main class, but apart from that they are identical (in terms of JVM parameters, application parameters, etc.).
>
> Therefore, I currently  have 4 wrapper.conf files - one for each process - which are identical apart from
>
> wrapper.app.parameter.1=foo.bar.MyProcessMainClass
>
> Is there any way I can avoid the duplicating the JVM parameters, application parameters for each process?
>
> Thanks,
> TD
>   

Re: [Wrapper-user] Application won't start

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

Justin,
You were close, but as the error message states, the WrapperSimpleApp 
needs to
receive your application's real main class name.   It looks like you 
"did" this.  But
the property is commented out.
> # Application parameters.  Add parameters as needed starting from 1
> #wrapper.app.parameter.1=myMainClass
>   
The rest of your configuration file looks fine.

Cheers,
Leif

[Wrapper-user] Application won't start

[Justin S. <ju...@se...>]

I can't get the wrapper to start via the App.bat. It starts the wrapper and 
then fails/ends:

STATUS | wrapper  | 2007/02/21 14:41:54 | --> Wrapper Started as Console
STATUS | wrapper  | 2007/02/21 14:41:55 | Launching a JVM...
INFO   | jvm 1    | 2007/02/21 14:41:55 | Wrapper (Version 3.2.3) 
http://wrapper.tanukisoftware.org
INFO   | jvm 1    | 2007/02/21 14:41:55 |   Copyright 1999-2006 Tanuki 
Software, Inc.  All Rights Reserved.
INFO   | jvm 1    | 2007/02/21 14:41:55 |
INFO   | jvm 1    | 2007/02/21 14:41:55 |
INFO   | jvm 1    | 2007/02/21 14:41:55 | WrapperSimpleApp Usage:
INFO   | jvm 1    | 2007/02/21 14:41:55 |   java 
org.tanukisoftware.wrapper.WrapperSimpleApp {app_class} [app_arguments]
INFO   | jvm 1    | 2007/02/21 14:41:55 |
INFO   | jvm 1    | 2007/02/21 14:41:55 | Where:
INFO   | jvm 1    | 2007/02/21 14:41:55 |   app_class:      The fully 
qualified class name of the application to run.
INFO   | jvm 1    | 2007/02/21 14:41:55 |   app_arguments:  The arguments 
that would normally be passed to the
INFO   | jvm 1    | 2007/02/21 14:41:55 |                   application.
STATUS | wrapper  | 2007/02/21 14:41:57 | <-- Wrapper Stopped


attached is my conf file  

[Wrapper-user] configuring mulitple processes

[Donal M. <dom...@ya...>]

Hi,=0A=0AI have 4 java processes that I wish to wrap. Each one has a differ=
ent main class, but apart from that they are identical (in terms of JVM par=
ameters, application parameters, etc.).=0A=0ATherefore, I currently  have 4=
 wrapper.conf files - one for each process - which are identical apart from=
=0A=0Awrapper.app.parameter.1=3Dfoo.bar.MyProcessMainClass=0A=0AIs there an=
y way I can avoid the duplicating the JVM parameters, application parameter=
s for each process?=0A=0AThanks,=0ATD=0A=0A=0A=0A=0A=0A=09=09=0A___________=
________________________________________________ =0AWhat kind of emailer ar=
e you? Find out today - get a free analysis of your email personality. Take=
 the quiz at the Yahoo! Mail Championship. =0Ahttp://uk.rd.yahoo.com/evt=3D=
44106/*http://mail.yahoo.net/uk 

[Wrapper-user] logging: is there a way to switch the logging level at runtime

[peter e. <pet...@go...>]

It is possible to change the logging level within my server.conf file
post runtime so that i can switch from'INFO' logging level to say
DEBUG or ALL levels?? How can i achieve this easily using the
wrapper!?

-- 
Peter Ellis
Java Developer

[Wrapper-user] How to disable wrapper stopping, when killing JVM ?

[Martin G. <mar...@mi...>]

 Good morning,
 I am using Java Service Wrapper under Linux and I am facing following
 problem:
 When I kill the JVM process (not wrapper, but JVM) with signal "kill
 -9", wrapper logs:

 ERROR  | wrapper  | 2007/02/21 08:36:48 | JVM exited unexpectedly.
 STATUS | wrapper  | 2007/02/21 08:36:53 | Launching a JVM...

 and restarts the JVM which is what I want.

 When I kill the JVM process (not wrapper, but JVM) with just "kill",
 wrapper logs:

 STATUS | wrapper  | 2007/02/21 08:32:25 | <-- Wrapper Stopped

 and stops without restarting JVM.
 Is there any way how to configure wrapper (or to change application ?)
 to achieve, that wrapper restarts JVM no matter what signal JVM received ?

 I played with wrapper.ignore_signal property, but it seems to work on
 wrapper, not on the spawned JVM and therefore it did not help me.

 Thanks.

 With best regards

 Martin Gazak
 MicroStep-MIS



-- 
RNDr. Martin Gazak, MicroStep-MIS
Cavojskeho 1, 841 04 Bratislava, Slovakia
Tel: ++421 2 602 00 128, 111
Fax: ++421 2 602 00 180
e-mail: mar...@mi...
www: http://www.microstep-mis.com

Re: [Wrapper-user] Performance issues

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

Vinod,
Great.  Thanks for the update on this. :-)

Cheers,
Leif

Vinod Panicker wrote:
> A final update on this issue.  I got this reply on the Yourkit forums -
>
> "If a native method performs waiting directly using OS calls, Java
> obviously cannot know what exactly the method does, and will report it
> as running. If OS cannot provide actual thread CPU time, the time when
> that native method runs and when it waits cannot be separated."
>
> So rather than this being an issue with wrapper, its actually a
> limitation of the platform.
>
> Regards,
> Vinod.
>   

Re: [Wrapper-user] Performance issues

[Vinod P. <vi...@gm...>]

A final update on this issue.  I got this reply on the Yourkit forums -

"If a native method performs waiting directly using OS calls, Java
obviously cannot know what exactly the method does, and will report it
as running. If OS cannot provide actual thread CPU time, the time when
that native method runs and when it waits cannot be separated."

So rather than this being an issue with wrapper, its actually a
limitation of the platform.

Regards,
Vinod.

On 2/16/07, Vinod Panicker <vi...@gm...> wrote:
> Hi,
>
> There are no calls reported to DataInputStream.readByte() when CPU
> tracing is enabled.  That makes sense, since the wrapper ping interval
> is kept to the max amount and the profling snapshot is being taken
> after the application has been running for a while.
>
> I'll revert back to the old default values for the wrapper settings
> and will use the current values only for profling.  I'll also get in
> touch with the Yourkit guys and get some more info on this behaviour.
>
> Thanks for the help.
>
> Regards,
> Vinod.
>
> On 2/15/07, Leif Mortenson <le...@ta...> wrote:
> > Vinod,
> >     That is actually correct.  The handleSocket method gets called once
> > when the
> > socket is opened.  It is called with a dedicated thread called
> > "Wrapper-Connection".
> > The handleSocket method then loops until the socket is closed or the JVM
> > shuts
> > down.  In that loop, the DataInputStream.readByte() method will be
> > called at least
> > once for each command received from the Wrapper process.
> >     If any commands are send, they will show up in the log file with
> > wrapper.debug
> > enabled.
> >
> >     I wonder if the profiler is getting confused by the native call
> > inside the read
> > method.  That method is blocking on the socket at a low level.  I'm not
> > sure how
> > often the OS is spending actual CPU cycles in that call.  It should be
> > pretty much
> > idle unless any data is coming in however.
> >
> >     Are you able to see a call count on the DataInputStream.readByte
> > method as it
> > is called from within the handleSocket method?
> >
> > Cheers,
> > Leif
> >
> > Vinod Panicker wrote:
> > > OK, I ran the application with cpu sampling and tracing and with debug enabled.
> > >
> > > According to the code, if there is communication, there should be a
> > > log message in wrapper.log.  There's nothing of that sort, which means
> > > that handleSocket() is not getting invoked.
> > >
> > > During sampling, the profiler is reporting cpu usage in the
> > > handleSocket() method call.  During tracing, it shows that the method
> > > has been invoked just once.
> > >
> > > Does this make sense to anyone?  Does it look like an issue with the profiler?
> > >
> > > Regards,
> > > Vinod.
> > >
> >
> >
> > -------------------------------------------------------------------------
> > Take Surveys. Earn Cash. Influence the Future of IT
> > Join SourceForge.net's Techsay panel and you'll get the chance to share your
> > opinions on IT & business topics through brief surveys-and earn cash
> > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> > _______________________________________________
> > Wrapper-user mailing list
> > Wra...@li...
> > https://lists.sourceforge.net/lists/listinfo/wrapper-user
> >
>

[Wrapper-user] Compilation in HP-UX IA64

[Diego R. <dru...@be...>]

Hi, I'm trying to compile the sources in a HP-UX IA64 and I'm getting
the following error:

=20

Wrapper Build System

--------------------

Buildfile: build.xml

=20

init:

OS=3D'HP-UX'

=20

update-info:

Copying 1 file to
/home/cce/wrapper/wrapper_3.1.1_src/src/java/org/tanukisoftware/wrapper

Copying 1 file to /home/cce/wrapper/wrapper_3.1.1_src/src/c

=20

compile-java:

Compiling 1 source file to
/home/cce/wrapper/wrapper_3.1.1_src/build/classes

=20

compile-c-unix:

if test ! -d .deps; then mkdir .deps; fi

gcc -O3 -Wall --pedantic -DHPUX -D_XOPEN_SOURCE_EXTENDED wrapper.c
wrapperinfo.c wrappereventloop.c wrapper_unix.c property.c logger.c -lm
-o ../../bin/wrapper

wrapper.c: In function 'wrapperProtocolOpen':

wrapper.c:558: warning: dereferencing type-punned pointer will break
strict-aliasing rules

logger.c:230: warning: missing braces around initializer

logger.c:230: warning: (near initialization for
'log_printfMutex.m_short')

ld: Unsatisfied symbol "rk_globfree" in file /var/tmp//cc025oOs.o

ld: Unsatisfied symbol "rk_glob" in file /var/tmp//cc025oOs.o

2 errors.

collect2: ld returned 1 exit status

gmake: *** [wrapper] Error 1

=20

BUILD FAILED

/home/cce/wrapper/wrapper_3.1.1_src/build.xml:316: exec returned: 2

=20

=20

Any Ideas?=20

=20

Diego.

_______________________________________________________________________
Notice:  This email message, together with any attachments, may contain
information  of  BEA Systems,  Inc.,  its subsidiaries  and  affiliated
entities,  that may be confidential,  proprietary,  copyrighted  and/or
legally privileged, and is intended solely for the use of the individual
or entity named in this message. If you are not the intended recipient,
and have received this message in error, please immediately return this
by email and then delete it.

[Wrapper-user] Socket in TIME_WAIT after shutdown

[<jan...@ub...>]

Hi all,=20
=20
I recently updated my wrapper from 3.1 to 3.2.3. Since then the wrapper
leaves its socket in TIME_WAIT status after shutdown. That means when I
am trying to restart the application, the wrapper gives a
java.net.SocketException: Address already in use. =20
I have to wait for the timeout which is around 4 minutes before I can
restart the application. How can the wrapper be forced to close the
socket properly?=20
=20
Java Version is 1.3.1_16-b06 on a Solaris 8. =20
=20
Any help would be much appreciated.=20
=20
Thanks,=20
Jan

                                                 =20

Re: [Wrapper-user] Can't start service error

[Nicolas V. <ni...@va...>]

Le 17 f=E9vr. 07 =E0 03:34, Leif Mortenson a =E9crit :

> Dave,
>> # Java Classpath (include wrapper.jar)  Add class path elements as
>> #  needed starting from 1
>> wrapper.java.classpath.1=3D../*.jar
>>
> This is actually perfectly fine.  Maintaining a full list of jar files
> in the wrapper.conf
> makes the wrapper.conf as difficult to maintain as a standard batch
> file.   To help
> the user with this, the Wrapper has a feature which allows =20
> wildcards in
> this location.
> It then expands those wildcards when generating the classpath used to
> launch the
> actual JVM.

Hum, that's really nice. I don't know that point, Il will be able to =20
simplify my wrapper.conf files now :)

Nicolas

Re: [Wrapper-user] Wrapper on AIX and HP-UX

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

Ben,
If you download the "prerelease" distribution and copy in your makefile 
changes,
you should be able to build an actual release version by running the 
following:
./build32.sh release

The prerelease dist is here:
http://sourceforge.net/project/showfiles.php?group_id=39428&package_id=183647

Once that is done, I would prefer to have you send me the resulting 
distribution
release so I can validate it and post it on the official download page.  
This makes
it much easier for other users to find and avoids having several versions of
a particular version floating around.

I have been working with Nicolas this week to get the HP-UX and AIX
versions available for everyone to use.  He has been successful with the
32 and 64 bit AIX versions, those have been uploaded.  He has also gotten
the 32-bit HP-UX version build but is still working on the 64 bit build.

The whole community really appreciates both you and Nicolas's effort
to fill this need.

Cheers,
Leif

Ben Jansen wrote:
> FYI to the list: The full set of wrapper-3.2.3 binaries I built for HP-UX PA-RISC 32bit are at <http://w007.org/~aogail/wrapper-3.2.3-hpux-parisc-32bit.tgz>. The archive contains all binaries built by the normal wrapper build. It's not a "dist" build since the machine I built on only has a 1.4 JDK, but AFAIK this only affects the jars.
>
> - Ben
>
>
> -----Original Message-----
> From: wra...@li... on behalf of Nicolas Varney
> Sent: Thu 2/15/2007 11:45 PM
> To: wra...@li...
> Subject: [Wrapper-user] Wrapper on AIX and HP-UX
>  
> Hi Wrapper users !
>
> As I'm a system architect and administrator, I have access to HP-UX  
> (PA-RISC) and AIX platforms. (And also linux platforms, but as many  
> people)
>
> I have successfully build the wrapper 3.2.3 under AIX 5.3 (32 and 64  
> bits). So I can help some users to build under different versions of  
> AIX OS.
>
> I have successfully build the wrapper 3.2.3 under HP-UX (32 bits),  
> but not 64 bits. If some of you have already build a wrapper for HP- 
> UX 64 bits, help will be appreciated.
>
> Regards
>
> Nicolas Varney
>   

Re: [Wrapper-user] Can't start service error

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

Dave,
> # Java Classpath (include wrapper.jar)  Add class path elements as
> #  needed starting from 1
> wrapper.java.classpath.1=../*.jar
>   
This is actually perfectly fine.  Maintaining a full list of jar files 
in the wrapper.conf
makes the wrapper.conf as difficult to maintain as a standard batch 
file.   To help
the user with this, the Wrapper has a feature which allows wildcards in 
this location.
It then expands those wildcards when generating the classpath used to 
launch the
actual JVM.
> # Java Library Path (location of Wrapper.DLL or libwrapper.so)
> wrapper.java.library.path.1=../bin
> wrapper.java.library.path.2=%PATH%
>   
The Wrapper is capable of doing environment variable expansions.  So in that
respect this is correct.  I do have some comments on this however.

By default, if a specific java.library.path is not specified when 
launching a JVM,
Java defaults to using the system PATH to locate its native library DLLs.
This is similar to the way it defaults to the CLASSPATH environment variable
if a classpath is not specified on the java command line.

Both of these methods of locating libraries is considered to be bad 
practice by
much of the java community.   The problem is that you are placing your
application ability to start up in he faith that nobody or no other 
application will
ever modify the system environment in a way that affects your application.

In both the PATH and CLASSPATH cases, they are system wide variables
which can and will be modified by others.  In the case of java, these system
wide variables also do not support the ability to have more than one version
of an application installed at once as both versions would have to be 
located
on the same PATH/CLASSPATH.

In experience, developers have run into many problems where an unrelated
program happens to have a DLL or jar files with the same name that ends
up breaking their program.

In the case of the Wrapper, it requires its own native library, this it must
define a java.library.path to avoid the above problems.  In doing so, it
also prevents Java from looking on the PATH for native libraries as was
done without the Wrapper.   You can get the original behavior back by
doing the following:

wrapper.java.library.path.1=../bin
wrapper.java.library.path.append_system_path=true

See the explanation of this property here:
http://wrapper.tanukisoftware.org/doc/english/prop-java-library-path-append-system-path.html

BUT. If you do this, you will open yourself up to the same problems that I
described above.  A much better way to handle this is to locate the DLL(s)
that your Java application requires and add those specific locations to the
library path directly.  Personally, I like to keep applications as 
compartmentalized
as possible by placing the required native libraries into the lib 
directory with
the wrapper.dll file.

In cases where the native library DLL requires additional DLLs to be loaded,
the nested DLL load is outside the scope of Java.  Such DLLs will still be
located using the PATH on Windows and LD_LIBRARY_PATH on UNIX
systems.   To make that work correctly after all DLLs have been moved to
the lib directory, you also need to set up those environment variables 
correctly
as follows.  These changes are done when the wrapper is launched and are
local to the wrapper and its child process environments. (ie Java).  
This keeps
your system as a whole nice and clean, is easy to maintain and support, and
will help make sure your application runs reliably over time.

set.PATH=..%WRAPPER_FILE_SEPARATOR%lib%WRAPPER_PATH_SEPARATOR%%PATH%
set.LD_LIBRARY_PATH=..%WRAPPER_FILE_SEPARATOR%lib%WRAPPER_PATH_SEPARATOR%%LD_LIBRARY_PATH%
wrapper.java.library.path.1=../bin
wrapper.java.library.path.2=../lib

The WRAPPER_* environment variables below are there to make sure the
same wrapper.conf file can be used as is on both Windows and UNIX
systems.  The java.library.path itself is used within java, so the forward
slashes are always converted correctly before use.


Now on to your original problem:
>> STATUS | wrapper  | 2007/02/16 09:20:16 | Launching a JVM...
>> INFO   | jvm 5    | 2007/02/16 09:20:16 |
>> java.lang.NoClassDefFoundError: Files\Sybase\ASA
>>     
Try setting the following property and then relaunching your JVM.
wrapper.java.command.loglevel=INFO

You will now see the generated java command line in the wrapper.log file 
before
the failure.  From your configuration file, I see only one place that 
could be a
problem.  Your system PATH most likely contains quotes around one or more
of its entries as follows:
PATH=c:\myfirstapp;"C:\Program Files\Sybase\ASA";c:\Mysecondapp.

When the Wrapper builds the command line, it encloses the entire library 
path
in a set of quotes.  This works perfectly in all cases whether the path 
contains
spaces or not.   The problem is that with invalid paths like the one 
above, this
will break as the generated command line goes in and out of the quoted state
at inappropriate times.

I will look at ways to at least warn the user about the this problem to help
make it clearer and easier to avoid.

Cheers,
Leif

Re: [Wrapper-user] Can't start service error

[Dave H. <DH...@xr...>]

Thanks much, you are right.  The problem was due to =
wrapper.java.library.path.2=3D%PATH%

-dh

-----Original Message-----
From: wra...@li... =
[mailto:wra...@li...] On Behalf Of Nicolas =
Varney
Sent: Friday, February 16, 2007 2:59 PM
To: wra...@li...
Subject: Re: [Wrapper-user] Can't start service error


Le 16 f=E9vr. 07 =E0 20:26, Dave Hoffer a =E9crit :

> wrapper.java.classpath.1=3D../*.jar

I'm not sure that this class path is valid.
I thnk you should declare all jar file completly

> wrapper.java.library.path.2=3D%PATH%

I don't know if this is a valid library path.
I don't think that environment variable substitution can be done in =20
this configuration file

Regards

Nicolas Varney
-------------------------------------------------------------------------=

Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share =
your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge&CID=3D=
DEVDEV
_______________________________________________
Wrapper-user mailing list
Wra...@li...
https://lists.sourceforge.net/lists/listinfo/wrapper-user