Wl60, jsp reloading, and managed servers

JSP pages don't reload on "Managed" nodes/servers in WLS6.0. Is this
          intentional or a bug? and can this behavior be modified?
          -jim

For those interested, here's the reply I got from BEA support:
 Just got some information regarding redeploying of the JSP. I got this
 from one of my colleague.
 "The only way to cause a jsp change to propagate to all the targets is
 to redeploy the whole web app. This has undesirable side effects, such
 as losing servlet session (since we have to do an undeploy/redeploy). We
 are hoping to be able to fix this soon"
 I don't know the timeframe of when this problem will be fixed. It is
 indicated that it is not fixed in 6.1 and maybe the release after 6.1
 (1Q 2002).
 Out of curiosity, how is everyone else developing with 6.0? Is everyone
 just doing their development under the Admin server? Has anyone come up
 with any other creative alternatives?
 -jim
 In article <[email protected]>, [email protected]
 says...
 > I did.
 > I would post here their answer, because I think it must disturb lots of J2EE developper
 > - especially those that have already use other J2EE servers - to have applications
 > that are in development phase to target the administration server...
 >
 > Kumar Allamraju <[email protected]> wrote:
 > >This is a known issue.
 > >I would suggest you to contact [email protected], so that they can help
 > >you with suitable workaround.
 > >
 > >--
 > >Kumar
 > >
 > >Jim Prantzalos wrote:
 > >
 > >> JSP pages don't reload on "Managed" nodes/servers in WLS6.0. Is this
 > >> intentional or a bug? and can this behavior be modified?
 > >>
 > >> -jim

Similar Messages

Command for STOPPING admin and managed servers

I am writing a script for our application, which apart from building and
deploying application, also needs to stop and start both admin and managed
servers. Is there any command available for properly stopping the weblogic
servers? ( something that "initiates the shutdown")
Thanks

Scott, could you try specifying the protocol: t3 in the url of the
servers.
eg) java weblogic.Admin -url t3://172.17.13.46:7013 -username system
-password
xxx SHUTDOWN
-mihir
Scott Jones wrote:
I have a script that shutsdown all my servers java weblogic.Admin -url
neo.conway.acxiom.com:$S4_ONLINE_PORT SHUTDOWN -username system
-password $S4_WLS_ADMIN_PW
java weblogic.Admin -url neo.conway.acxiom.com:$S4_S4MT_PORT SHUTDOWN
-username system -password $S4_WLS_ADMIN_PW
java weblogic.Admin -url neo.conway.acxiom.com:$S4_WLS_ADMIN_PORT
SHUTDOWN -username system -password $S4_WLS_ADMIN_PWThe first two
lines shut down my managed server. The last shuts down my admin
server. Scott [email protected]
"Mihir Kulkarni" <[email protected]> wrote in
message news:[email protected] here -
Solaris 2.7 with WLS 60 sp1.
I am not facing any problems, though!
Fyi, my managed-server is not in a cluster.
JD wrote:
I am running on Solaris 2.7 using WL6.0sp1.
Mihir Kulkarni <[email protected]> wrote:
<!doctype html public "-//w3c//dtd html 4.0transitional//en">
<html>
Hi,
 Could you please specify on which platform are youfacing this problem.Also,
which sp of the WLS 60 are you using ?
I tried the weblogic.Admin SHUTDOWN command on NT andSolaris and
it
runs correctly on both. These are the results:
on NT:
 on the client:
 java weblogic.Admin -url t3://172.17.13.46:7013-username system
-password
xxx SHUTDOWN
 Shutdown initiated
 The shutdown sequence has been initiated.
on the Managed server:
 <Jun 13, 2001 11:45:49 AM PDT> <Alert><WebLogicServer>
<The
disabling of server
 logins has been requested by system>
 <Jun 13, 2001 11:45:49 AM PDT> <Alert><WebLogicServer>
<Server
logins have been
 disabled.>
 <Jun 13, 2001 11:45:49 AM PDT> <Alert><WebLogicServer>
<Server
shutdown has bee
 n requested by system>
 <Jun 13, 2001 11:45:49 AM PDT> <Alert><WebLogicServer>
<The
shutdown sequence h
 as been initiated.>
 <Jun 13, 2001 11:45:49 AM PDT> <Info><WebLogicServer> <Server
shutdown is comme
 ncing NOW and is irreversible.>
 <Jun 13, 2001 11:45:49 AM PDT> <Info> <HTTP> <[HTTPstandAloneServer]
Destroying
 servlets in default servlet context>
 <Jun 13, 2001 11:45:49 AM PDT> <Info> <JMS><Undeployed
4 default connection fac
 tories.>
 <Jun 13, 2001 11:45:49 AM PDT> <Info> <JMS> <JMSshutdown
is complete.>
 <Jun 13, 2001 11:45:49 AM PDT> <Info><WebLogicServer> <Removing
"ClientContext
 - id: '#|standAloneServer|1.992457846784', bound:'false', dead:
'false'"
becaus
 e of hard disconnect timeout.>
 <Jun 13, 2001 11:45:50 AM PDT> <Info><WebLogicServer> <Requesting
all threads t
 o perform their own shutdown.>
 <Jun 13, 2001 11:45:50 AM PDT> <Info><WebLogicServer> <Shutting
down Thread[SSL
 ListenThread,5,WebLogicServer]>
 <Jun 13, 2001 11:45:50 AM PDT> <Info><WebLogicServer> <Waiting
20 seconds for t
 hreads to complete shutdown.>
 <Jun 13, 2001 11:45:50 AM PDT> <Info><WebLogicServer> <Shutting
down Thread[Lis
 tenThread,5,WebLogicServer]>
 <Jun 13, 2001 11:46:10 AM PDT> <Info><WebLogicServer> <Stopping
execute threads
 .>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '0' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '1' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '2' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '3' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '4' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '5' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '6' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '7' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '8' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '9' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '10' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '11' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '12' for queue: 'default'' stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '0' for queue: '__weblogic_admin_html_queue''stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '1' for queue: '__weblogic_admin_html_queue''stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '0' for queue: '__weblogic_admin_rmi_queue''stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Critical> <Kernel><Execute
Thread: 'ExecuteThre
 ad: '1' for queue: '__weblogic_admin_rmi_queue''stopped.>
 <Jun 13, 2001 11:46:10 AM PDT> <Info><WebLogicServer> <Shutdown
has completed.>
on Solaris platform:
 on client:
 java weblogic.Admin -url t3://willard.beasys.com:7003-username system
-xxx mihirk00 SHUTDOWN
 Shutdown initiated
 The shutdown sequence has been initiated.
on the Managed server:
 <Jun 13, 2001 11:52:26 AM PDT> <Alert><WebLogicServer>
<Server
shutdown has been requested by system>
 <Jun 13, 2001 11:52:26 AM PDT> <Alert><WebLogicServer>
<The
shutdown sequence has been initiated.>
 <Jun 13, 2001 11:52:26 AM PDT> <Emergency><WebLogicServer>
<ListenThread.run() failed: java.lang.ThreadDeath>
 <Jun 13, 2001 11:52:27 AM PDT> <Emergency><WebLogicServer>
<The WebLogic Server is no longer listening forconnections. You should
probably restart it.>
 <Jun 13, 2001 11:52:27 AM PDT> <Alert><WebLogicServer>
<Server
shutdown has been requested by system>
 <Jun 13, 2001 11:52:27 AM PDT> <Alert><WebLogicServer>
<The
shutdown sequence has been initiated.>
So, it works fine for me.
 Please let me know if I am missing something.
cheers,
 Mihir
 
Scott Simpson wrote:
<blockquote TYPE=CITE>"JD" <[email protected]> wrote
in message
 <a
href="news:[email protected]">news:[email protected]</a>...
 > I ran this problem past BEA support and as usualthey can not duplicate
 the problem.
They can't duplicate the problem! Everybody on thisnewsgroup has
the
same
 problem. How can they not duplicate it?!</blockquote>
</html>
[att1.html]

Automating start / restart of WebLogic Admin and managed servers on MS Windows

Hi All,
Please help me to do the automation of start and stop scrip of Oracle Weblogic 10.3.6 and Forms and Reports 11.1.2.2.0(Managed Servers) On Windows server 2008.
Here are two options for automating start / stop of WebLogic Admin and managed servers on MS Windows. Please suggesst which one is standrad procedure.
Use NodeManager (installed as a MS Windows Service)
or
Create Windows Service(s) for the server(s) - no NodeManager
Following Document : 1060058.1
Please suggest me the best practice. Many thanks in advance.
Kind Regards,
Mohan

Hi Borys,
Thank you for quick reply...
If we use the first option "Create Windows Service(s) for the server(s) - no NodeManager" then we will be facing issue Like "Admin Server or Managed Server may go into an unhealthy or failed state but the JVM is still running"
In the document : 1060058.1 : Mentioned as Important Note: The second option has a limitations.
Limitaion is below when we use first option : "Create Windows Service(s) for the server(s) - no NodeManager" -
Windows Services Manager only monitors the JVM process. If the JVM process fails (shuts down), then Windows Services Manager will attempt to restart it. But there are some scenarios where the Admin Server or Managed Server may go into an unhealthy or failed state but the JVM is still running. The Windows Service will not know to restart the process in these cases. If a server is started using NodeManager, NodeManager would recognize such a state and restart the process accordingly. Furthermore, if you start up managed servers via a MS Windows Service, you cannot monitor and/ or stop the managed server via the Weblogic Admin Console and/or Fusion Middleware ( Enterprise Manager ) Control.
I have implemented the second option "Use NodeManager starting admin and manged servers via WLST script" - Python script. Its working well without issue...No Limitation...
Script Contains Below :
print "================== Start scripts of Managed servers and Admin server ======================"
nmConnect(userConfigFile='D:/myFiles/myuserconfigfile.secure',userKeyFile='D:/myFiles/myuserkeyfile.secure',host='00.0.00.00',port='5556',domainName='ClassicDomain',domainDir='D:\Oracle\Middleware\user_projects\domains\ClassicDomain',nmType='ssl')
print "================== Adminserver Starting ================================================"
nmStart('AdminServer')
connect('weblogic','Password1','10.4.93.117:7001')
print "================== MANAGED SERVER (FROMS) Starting ====================================="
start('WLS_FORMS','Server')
print "================== MANAGED SERVER (REPORTS) Starting ==================================="
start('WLS_REPORTS','Server')
exit()
HOPE WE WILL MEET AGAIN AND AGAIN...
Many thanks for your comments!
Kind Regards,
Mohan

Alert setup for Admin and Manage servers in weblogic

Hi All,
We need to setup e-mail notification for health status of admin and managed servers on fusion middleware 11g. Kindly advise.
Thanks,
Vijay

Hi there Vijay,
you can easily do this using the WebLogic Diagnostic Framework. If your WebLogic installation hosts products from the Fusion Middleware portfolio, e.g. Oracle SOA Suite, Oracle WebCenter, etc., you will have a preconfigured WLDF module targeted to all servers named Module-FMWDFW.
In general you will have to:
1. Configure mail session and target it to all servers. Details on the mail server are provided as properties. Use the following as a starting point:
mail.host=<your_outgoing_mail_server_host>
mail.user=<your_outgoing_mail_server_user>
mail.transport.protocol=smtp
mail.from=<address_to_send_emails_from>
mail.debug=True
2. Configure WebLogic Diagnostic Framework Module or reconfigure the Module-FMWDFW if exists
3. Configure Watches and Notifications for that module and the data/events you are interested in. By default the Module-FMWDFW is configured to generate ADR incidents in certain occasions (you can examine the configuration of Module-FMWDFW).
You can get more details from these resources:
http://docs.oracle.com/cd/E29542_01/web.1111/e13714/architecture.htm#WLDFC115
https://blogs.oracle.com/jamesbayer/entry/weblogic_notifications_watch_t
Hope this helps,
A.

Weblogic server and managed servers sudden shutdown

Hi,
when trying to restart both the admin and managed osb servers. after some time the admin and managed servers are getting shut down. This is what the information i got from logs. Appreciate your help in this regard
####<Aug 18, 2011 12:51:44 PM EDT> <Info> <Health> <weblogic.GCMonitor> <<anonymous>> <> <8beb4c4d69347bb8:6c6553af:131ddaf5db4:-7ff7-0000000000000011> <1313686304697> <BEA-310002> <38% of the total memory in the server is free>
####<Aug 18, 2011 12:52:06 PM EDT> <Error> <JDBC> <[ACTIVE] ExecuteThread: '7' for queue: 'weblogic.kernel.Default (self-tuning)'> <<WLS Kernel>> <> <8beb4c4d69347bb8:6c6553af:131ddaf5db4:-7ff7-0000000000000109> <1313686326579> <BEA-001112> <Test "SELECT 1 FROM DUAL" set up for pool "GLTOdatasource" failed with exception: "oracle.jdbc.xa.OracleXAException".>
####<Aug 18, 2011 12:52:06 PM EDT> <Info> <JDBC> <[ACTIVE] ExecuteThread: '7' for queue: 'weblogic.kernel.Default (self-tuning)'> <<WLS Kernel>> <> <8beb4c4d69347bb8:6c6553af:131ddaf5db4:-7ff7-0000000000000109> <1313686326580> <BEA-001128> <Connection for pool "GLTOdatasource" closed.>
####<Aug 18, 2011 12:52:44 PM EDT> <Info> <Health> <weblogic.GCMonitor> <<anonymous>> <> <8beb4c4d69347bb8:6c6553af:131ddaf5db4:-7ff7-0000000000000011> <1313686364766> <BEA-310002> <61% of the total memory in the server is free>
####<Aug 18, 2011 12:56:45 PM EDT> <Info> <Health> <weblogic.GCMonitor> <<anonymous>> <> <8beb4c4d69347bb8:6c6553af:131ddaf5db4:-7ff7-0000000000000011> <1313686605051> <BEA-310002> <44% of the total memory in the server is free>
####<Aug 18, 2011 12:58:18 PM EDT> <Info> <Security> <[ACTIVE] ExecuteThread: '6' for queue: 'weblogic.kernel.Default (self-tuning)'> <<WLS Kernel>> <> <8beb4c4d69347bb8:6c6553af:131ddaf5db4:-7ff7-0000000000000111> <1313686698580> <BEA-090516> <The CredentialMapper provider has preexisting LDAP data.>
####<Aug 18, 2011 1:02:23 PM EDT> <Notice> <WebLogicServer> <Thread-1> <<WLS Kernel>> <> <8beb4c4d69347bb8:6c6553af:131ddaf5db4:-7ff7-0000000000000117> <1313686943979> <BEA-000388> <JVM called WLS shutdown hook. The server will force shutdown now>
####<Aug 18, 2011 1:02:23 PM EDT> <Alert> <gifosb_admin> <Thread-1> <<WLS Kernel>> <> <8beb4c4d69347bb8:6c6553af:131ddaf5db4:-7ff7-0000000000000117> <1313686943996> <BEA-000396> <Server shutdown has been requested by <WLS Kernel>>
####<Aug 18, 2011 1:02:24 PM EDT> <Notice> <WebLogicServer> <Thread-1> <<WLS Kernel>> <> <8beb4c4d69347bb8:6c6553af:131ddaf5db4:-7ff7-0000000000000117> <1313686944005> <BEA-000365> <Server state changed to FORCE_SUSPENDING>
Thanks,
Bharath

<JVM called WLS shutdown hook. The server will force shutdown now>This is your JVM process which is getting killed and hence server is forced to shutdown. On which OS, servers are running? How are you starting the server? If you are on windows, DO NOT close the command prompt through which you started the server.
Regards,
Anuj

Creating Weblogic domain with Admin Server and managed servers on different

I am trying to create Weblogic domain where Admin Server and managed servers on different machines. However I am unable to find any steps which would allow me to do so. The config.sh script always creates an Admin Server. Please help.

Whilst the method that Lawrence gives can result in a domain that is perfectly functional, it is a method that can result in errors as you are having to enter the same information twice. Also you will create two admin servers, as the second domain creation will have the console added to it on the renamed admin server and this will give issues.
The best method to create a domain spamming multiple physical machines, is to create the domain with all the components on the first machine. This means planning the domain and making sure all components are added once; so having the information about the domain written out is wise. Ensure that the WLS software is installed in the same path on each machine before starting.
Then copy the domain that you have created from one machine to the other, keeping the same paths. You can simply copy (I've used this method multiple times without issue) but the supported way would be to use the WLST PACK and UNPACK methods. These create and extract domains from templates. Some more information on these is given here -> http://docs.oracle.com/cd/E13179_01/common/docs100/pack/intro.html#wp1069056

Commands for STOPPING admin and managed servers

I am writing a script for our application, which apart from building and
deploying application, also needs to stop and start both admin and managed
servers. Is there any command available for properly stopping the weblogic
servers? ( something that "initiates the shutdown")
Thanks

sorry i haven't moved up to 6.0 yet, so i don't know much about it :(
hope someone else can help you there
Manmari wrote:
thanks bhavin. I will try this on my 6.0 However what is the command for
stopping any of my "managed servers"? Is it the same or something like
java weblogic.Managed "t3://managedserver:7003" SHUTDOWN
Have you tried it. I founf a note on other newsgroup that this does not work
for maged servers. Do you know the solution? Does it work for you?
youe help is much appreciated
Thanks
"bhavin patel" <[email protected]> wrote in message
news:[email protected]..
i don't know if this is what you are looking for, but to stop wl 5.10
you just need to run the weblogic.Admin class
like this
java weblogic.Admin "t3://yourserver:7001" SHUTDOWN <username> <password>
hth
Manmari wrote:
I am writing a script for our application, which apart from building and
deploying application, also needs to stop and start both admin and
managed
servers. Is there any command available for properly stopping theweblogic
servers? ( something that "initiates the shutdown")
Thanks

Re: Precompiling JSP with admin/managed servers

Thanks, but I'm not doing any copying.
 The admin/managed-server communication copies things to the managed server,
 which then always recompiles the pages when hit.
 -Greg
 Check out my WebLogic 6.1 Workbook for O'Reilly EJB Third Edition
 www.oreilly.com/catalog/entjbeans3 or www.titan-books.com
 "Robert Coonrad" <[email protected]> wrote in message
 news:[email protected]...
 >
 > check out post 8366...i found that i was not preserving
 > the lastmodified date on my jsps and this was causing
 > unnecessary re-compilation.
 >
 > hope it helps...
 > bobc
 >
 > "Greg Nyberg" <greg.nyberg.at.objectpartners.com> wrote:
 > >I believe I have exhausted all permutations of EARing/notEARing,
 > >WARing/notWARing, placing precompiled jsp class files in WEB-INF/classes,
 > >placing them in a static location and setting workingDir to that
 location,
 > >combinations of the above.
 > >
 > >No matter what, the managed server re-compiles pages the first time they
 > >are
 > >hit. Non admin/managed-server I have no problems.
 > >
 > >Can anyone from BEA comment on this problem? Or give me a workaround
 > >for
 > >getting a cluster working with precompiled jsps?
 > >
 > >-Greg
 > >
 > >"Greg Nyberg" <greg.nyberg.at.objectpartners.com> wrote in message
 > >news:[email protected]...
 > >> Grrr... The JSP engine is extremely frustrating! I've spent many hours
 > >> fighting the "staleness" checker in WL. I've been through all of the
 > >> newsgroup messages pertaining to pre-compiling, etc., and I've gotten
 > >> pre-compilation working on single-server deployments, but admin/managed
 > >> server deployments have me beat.
 > >>
 > >> WL6.1, SP1, Solaris
 > >>
 > >> I've done the pageCheckSeconds=-1 and the workingDir is set to a fixed
 > >> place. The fixed place contains pre-compiled versions of all jsps
 > >made on
 > >> that machine using jspc not 20 minutes earlier using the JSP files
 > >in the
 > >> exploded EAR file used by the admin server as the model for managed
 > >> servers.. The managed servers are on the same machine.
 > >>
 > >> When the admin server gives an application to a managed server, the
 > >managed
 > >> server creates a temporary directory containing all of the webapp
 > >> components, etc. The file timestamps on these files is the set by
 > >the
 > >> copying process to the time of the managed server boot (why?!?!????!?),
 > >so
 > >> the staleness check always thinks they are new and could care less
 > >what
 > >> precompiled jsps I have in my workingDir, the WEB-INF/classes
 directory,
 > >or
 > >> anywhere else. The pageCheckSeconds=-1 seems to be completely ignored
 > >in
 > >> this scenario.
 > >>
 > >> If I tell the managed server to precompile everything on boot (about
 > >45
 > >> minutes for this app) it will create versions of the classes that match
 > >th
 > >e
 > >> new JSP file timestamps, but this does not even survive a reboot of
 > >the
 > >> managed server because it AGAIN creates a new temp version of
 everything
 > >on
 > >> the next reboot with new timestamps.
 > >>
 > >> If I wait for the managed server to boot and find the directory like
 > >> .../applications/.wlnotdelete_man1/wlap7336/webapp/... and physically
 > >copy
 > >> (via cp -pr to retain timestamps) all of the original webapp components
 > >on
 > >> top of the temp versions, the staleness checker is happy and the
 > >> pre-compiled versions work fine.
 > >>
 > >> There HAS to be a way to package pre-compiled versions of the JSPs
 > >in the
 > >> "model" application in the admin server and keep from having to
 precompile
 > >> the JSPs on every managed server every time managed server is booted..
 > >>
 > >> It would help if we had a way to bypass the staleness checking
 > >completely..
 > >> Or you guys should make the timestamps on the files copied by the
 > >> admin/managed deployment process match properly so the staleness
 checker
 > >> doesn't think the JSP is different.
 > >>
 > >> It would also help if the engineer who wrote this could explain the
 > >rules
 > >> being implemented by the staleness checker. So far all the messages
 > >in
 > >the
 > >> newsgroup have amounted to point solutions for problems without a good
 > >> understanding of what the engine is checking for and/or doing under
 > >the
 > >> covers. Looking at the generated .java files for the JSP pages helps,
 > >but
 > >> it is not good enough...
 > >>
 > >> Anyone out there have a working admin/managed server JSP application?
 > >> -Greg
 > >>
 > >> -----------------------------------------------------------
 > >> Check out my WebLogic 6.1 Workbook for O'Reilly EJB Third Edition
 > >> www.oreilly.com/catalog/entjbeans3 or www.titan-books.com
 > >>
 > >>
 > >>
 > >
 > >
 >


The admin/managed-server communication copies things to the managed server, which then always recompiles the pages when hit.
 This is a known issue and is fixed. The timestamps of the compiled classes was not being preserved when extracted from the war file used to distribute to the managed servers. This will be available in WLS6.1 Service Pack 3 - and there is a temporary patch available for SP2. Please ask your friendly BEA support person for it (you can refer to CR058946)
 I'd give you the patch myself, but they like to keep track of these things...
 Regards,
 Alex
 "Girish" <[email protected]> wrote in message news:[email protected]...
 >
 > "Aditya Kiran Gavvala" <[email protected]> wrote:
 > >Greg,
 > >
 > >I have been following your posts, because our application deployment
 > >ran
 > >into exact same problem you ran into. I had spent a full two days
 > >researching into the problem. And I figured the solution. Hope this
 > >helps.
 > >
 > >Here are my discoveries:
 > >
 > >The following applies only to the following environment:
 > >OS: Linux (perhaps for Win/Unix/Solaris etc)
 > >WLS 6.0 SP2 ( no rolling patches): I found Rolling Patch2 (RP2) not useful
 > >for this problem.
 > >Clustered environment with Admin/Managed servers
 > >
 > >- When you compile JSP using weblogic.jspc compiler it puts the JSP file
 > >timestamp into the compiled class. You can see it in the generated java
 > >file
 > >(you need to supply -keepgenerated switch to jspc)
 > >
 > >- When a request is made to a JSP page after the application is deployed,
 > >it
 > >seems to be retrieving this timestamp from the compiled class file and
 > >comparing it with the JSP file timestamp. If they dont match a compile
 > >command gets run by the server. Thereby you see a compile happening at
 > >run
 > >time.
 > >
 > >- If you have exploded directory deployment, when you start the managed
 > >servers they create a ".war" file (under some temp dir) with all the
 > >JSP
 > >source files going into the file. You can notice this by looking into
 > >the
 > >server log file. Therefore all JSP source files get a brand new timestamp
 > >in
 > >the archive (a timestamp later than what was put class files by
 > >weblogic.jspc). So, the server at run time sees that the timestamp in
 > >the
 > >class file is older than the JSP source file and runs a recompile. So
 > >DONT
 > >DO EXPLODED directory deployment if your environment is as described
 > >in this
 > >post.
 > >
 > >- If you have ".war" file deployment, you will not have a problem. At
 > >the
 > >start up time managed server still creates "".war" file under a temp
 > >directory however it seems to be copying the content of the your ".war"
 > >file. So, the timestamps of JSP remain the same as they were before.
 > >SO NO
 > >RE-COMPILATION.
 > >
 > >- Another important thing to remember is to make sure you specify the
 > >workingDir in the weblogic.xml file. That is where the precompiled class
 > >files should reside. This should be any directory the server uses as
 > >scratch
 > >pad to compile classes or find (pre)compiled classes. This is not a
 > >directory inside your .war file is what I am trying to get at.
 > >
 > >Hope this helps,
 > >Aditya
 > >
 > >"Greg Nyberg" <greg.nyberg.at.objectpartners.com> wrote in message
 > >news:[email protected]...
 > >> Thanks, but I'm not doing any copying.
 > >>
 > >> The admin/managed-server communication copies things to the managed
 > >server,
 > >> which then always recompiles the pages when hit.
 > >>
 > >> -Greg
 > >>
 > >> -----------------------------------------------------------
 > >> Check out my WebLogic 6.1 Workbook for O'Reilly EJB Third Edition
 > >> www.oreilly.com/catalog/entjbeans3 or www.titan-books.com
 > >>
 > >> "Robert Coonrad" <[email protected]> wrote in message
 > >> news:[email protected]...
 > >> >
 > >> > check out post 8366...i found that i was not preserving
 > >> > the lastmodified date on my jsps and this was causing
 > >> > unnecessary re-compilation.
 > >> >
 > >> > hope it helps...
 > >> > bobc
 > >> >
 > >> > "Greg Nyberg" <greg.nyberg.at.objectpartners.com> wrote:
 > >> > >I believe I have exhausted all permutations of EARing/notEARing,
 > >> > >WARing/notWARing, placing precompiled jsp class files in
 > >WEB-INF/classes,
 > >> > >placing them in a static location and setting workingDir to that
 > >> location,
 > >> > >combinations of the above.
 > >> > >
 > >> > >No matter what, the managed server re-compiles pages the first time
 > >they
 > >> > >are
 > >> > >hit. Non admin/managed-server I have no problems.
 > >> > >
 > >> > >Can anyone from BEA comment on this problem? Or give me a workaround
 > >> > >for
 > >> > >getting a cluster working with precompiled jsps?
 > >> > >
 > >> > >-Greg
 > >> > >
 > >> > >"Greg Nyberg" <greg.nyberg.at.objectpartners.com> wrote in message
 > >> > >news:[email protected]...
 > >> > >> Grrr... The JSP engine is extremely frustrating! I've spent many
 > >hours
 > >> > >> fighting the "staleness" checker in WL. I've been through all
 > >of the
 > >> > >> newsgroup messages pertaining to pre-compiling, etc., and I've
 > >gotten
 > >> > >> pre-compilation working on single-server deployments, but
 > >admin/managed
 > >> > >> server deployments have me beat.
 > >> > >>
 > >> > >> WL6.1, SP1, Solaris
 > >> > >>
 > >> > >> I've done the pageCheckSeconds=-1 and the workingDir is set to
 > >a
 > >fixed
 > >> > >> place. The fixed place contains pre-compiled versions of all
 > >jsps
 > >> > >made on
 > >> > >> that machine using jspc not 20 minutes earlier using the JSP files
 > >> > >in the
 > >> > >> exploded EAR file used by the admin server as the model for managed
 > >> > >> servers.. The managed servers are on the same machine.
 > >> > >>
 > >> > >> When the admin server gives an application to a managed server,
 > >the
 > >> > >managed
 > >> > >> server creates a temporary directory containing all of the webapp
 > >> > >> components, etc. The file timestamps on these files is the set
 > >by
 > >> > >the
 > >> > >> copying process to the time of the managed server boot
 > >(why?!?!????!?),
 > >> > >so
 > >> > >> the staleness check always thinks they are new and could care
 > >less
 > >> > >what
 > >> > >> precompiled jsps I have in my workingDir, the WEB-INF/classes
 > >> directory,
 > >> > >or
 > >> > >> anywhere else. The pageCheckSeconds=-1 seems to be completely
 > >ignored
 > >> > >in
 > >> > >> this scenario.
 > >> > >>
 > >> > >> If I tell the managed server to precompile everything on boot
 > >(about
 > >> > >45
 > >> > >> minutes for this app) it will create versions of the classes that
 > >match
 > >> > >th
 > >> > >e
 > >> > >> new JSP file timestamps, but this does not even survive a reboot
 > >of
 > >> > >the
 > >> > >> managed server because it AGAIN creates a new temp version of
 > >> everything
 > >> > >on
 > >> > >> the next reboot with new timestamps.
 > >> > >>
 > >> > >> If I wait for the managed server to boot and find the directory
 > >like
 > >> > >> .../applications/.wlnotdelete_man1/wlap7336/webapp/... and physically
 > >> > >copy
 > >> > >> (via cp -pr to retain timestamps) all of the original webapp
 > >components
 > >> > >on
 > >> > >> top of the temp versions, the staleness checker is happy and the
 > >> > >> pre-compiled versions work fine.
 > >> > >>
 > >> > >> There HAS to be a way to package pre-compiled versions of the
 > >JSPs
 > >> > >in the
 > >> > >> "model" application in the admin server and keep from having to
 > >> precompile
 > >> > >> the JSPs on every managed server every time managed server is
 > >booted..
 > >> > >>
 > >> > >> It would help if we had a way to bypass the staleness checking
 > >> > >completely..
 > >> > >> Or you guys should make the timestamps on the files copied by
 > >the
 > >> > >> admin/managed deployment process match properly so the staleness
 > >> checker
 > >> > >> doesn't think the JSP is different.
 > >> > >>
 > >> > >> It would also help if the engineer who wrote this could explain
 > >the
 > >> > >rules
 > >> > >> being implemented by the staleness checker. So far all the messages
 > >> > >in
 > >> > >the
 > >> > >> newsgroup have amounted to point solutions for problems without
 > >a
 > >good
 > >> > >> understanding of what the engine is checking for and/or doing
 > >under
 > >> > >the
 > >> > >> covers. Looking at the generated .java files for the JSP pages
 > >helps,
 > >> > >but
 > >> > >> it is not good enough...
 > >> > >>
 > >> > >> Anyone out there have a working admin/managed server JSP application?
 > >> > >> -Greg
 > >> > >>
 > >> > >> -----------------------------------------------------------
 > >> > >> Check out my WebLogic 6.1 Workbook for O'Reilly EJB Third Edition
 > >> > >> www.oreilly.com/catalog/entjbeans3 or www.titan-books.com
 > >> > >>
 > >> > >>
 > >> > >>
 > >> > >
 > >> > >
 > >> >
 > >>
 > >>
 > >
 > >
 >
 [att1.html]

Precompiling JSP with admin/managed servers

Grrr... The JSP engine is extremely frustrating! I've spent many hours
 fighting the "staleness" checker in WL. I've been through all of the
 newsgroup messages pertaining to pre-compiling, etc., and I've gotten
 pre-compilation working on single-server deployments, but admin/managed
 server deployments have me beat.
 WL6.1, SP1, Solaris
 I've done the pageCheckSeconds=-1 and the workingDir is set to a fixed
 place. The fixed place contains pre-compiled versions of all jsps made on
 that machine using jspc not 20 minutes earlier using the JSP files in the
 exploded EAR file used by the admin server as the model for managed
 servers.. The managed servers are on the same machine.
 When the admin server gives an application to a managed server, the managed
 server creates a temporary directory containing all of the webapp
 components, etc. The file timestamps on these files is the set by the
 copying process to the time of the managed server boot (why?!?!????!?), so
 the staleness check always thinks they are new and could care less what
 precompiled jsps I have in my workingDir, the WEB-INF/classes directory, or
 anywhere else. The pageCheckSeconds=-1 seems to be completely ignored in
 this scenario.
 If I tell the managed server to precompile everything on boot (about 45
 minutes for this app) it will create versions of the classes that match the
 new JSP file timestamps, but this does not even survive a reboot of the
 managed server because it AGAIN creates a new temp version of everything on
 the next reboot with new timestamps.
 If I wait for the managed server to boot and find the directory like
 .../applications/.wlnotdelete_man1/wlap7336/webapp/... and physically copy
 (via cp -pr to retain timestamps) all of the original webapp components on
 top of the temp versions, the staleness checker is happy and the
 pre-compiled versions work fine.
 There HAS to be a way to package pre-compiled versions of the JSPs in the
 "model" application in the admin server and keep from having to precompile
 the JSPs on every managed server every time managed server is booted..
 It would help if we had a way to bypass the staleness checking completely..
 Or you guys should make the timestamps on the files copied by the
 admin/managed deployment process match properly so the staleness checker
 doesn't think the JSP is different.
 It would also help if the engineer who wrote this could explain the rules
 being implemented by the staleness checker. So far all the messages in the
 newsgroup have amounted to point solutions for problems without a good
 understanding of what the engine is checking for and/or doing under the
 covers. Looking at the generated .java files for the JSP pages helps, but
 it is not good enough...
 Anyone out there have a working admin/managed server JSP application?
 -Greg
 Check out my WebLogic 6.1 Workbook for O'Reilly EJB Third Edition
 www.oreilly.com/catalog/entjbeans3 or www.titan-books.com


 check out post 8366...i found that i was not preserving
 the lastmodified date on my jsps and this was causing
 unnecessary re-compilation.
 hope it helps...
 bobc
 "Greg Nyberg" <greg.nyberg.at.objectpartners.com> wrote:
 >I believe I have exhausted all permutations of EARing/notEARing,
 >WARing/notWARing, placing precompiled jsp class files in WEB-INF/classes,
 >placing them in a static location and setting workingDir to that location,
 >combinations of the above.
 >
 >No matter what, the managed server re-compiles pages the first time they
 >are
 >hit. Non admin/managed-server I have no problems.
 >
 >Can anyone from BEA comment on this problem? Or give me a workaround
 >for
 >getting a cluster working with precompiled jsps?
 >
 >-Greg
 >
 >"Greg Nyberg" <greg.nyberg.at.objectpartners.com> wrote in message
 >news:[email protected]...
 >> Grrr... The JSP engine is extremely frustrating! I've spent many hours
 >> fighting the "staleness" checker in WL. I've been through all of the
 >> newsgroup messages pertaining to pre-compiling, etc., and I've gotten
 >> pre-compilation working on single-server deployments, but admin/managed
 >> server deployments have me beat.
 >>
 >> WL6.1, SP1, Solaris
 >>
 >> I've done the pageCheckSeconds=-1 and the workingDir is set to a fixed
 >> place. The fixed place contains pre-compiled versions of all jsps
 >made on
 >> that machine using jspc not 20 minutes earlier using the JSP files
 >in the
 >> exploded EAR file used by the admin server as the model for managed
 >> servers.. The managed servers are on the same machine.
 >>
 >> When the admin server gives an application to a managed server, the
 >managed
 >> server creates a temporary directory containing all of the webapp
 >> components, etc. The file timestamps on these files is the set by
 >the
 >> copying process to the time of the managed server boot (why?!?!????!?),
 >so
 >> the staleness check always thinks they are new and could care less
 >what
 >> precompiled jsps I have in my workingDir, the WEB-INF/classes directory,
 >or
 >> anywhere else. The pageCheckSeconds=-1 seems to be completely ignored
 >in
 >> this scenario.
 >>
 >> If I tell the managed server to precompile everything on boot (about
 >45
 >> minutes for this app) it will create versions of the classes that match
 >th
 >e
 >> new JSP file timestamps, but this does not even survive a reboot of
 >the
 >> managed server because it AGAIN creates a new temp version of everything
 >on
 >> the next reboot with new timestamps.
 >>
 >> If I wait for the managed server to boot and find the directory like
 >> .../applications/.wlnotdelete_man1/wlap7336/webapp/... and physically
 >copy
 >> (via cp -pr to retain timestamps) all of the original webapp components
 >on
 >> top of the temp versions, the staleness checker is happy and the
 >> pre-compiled versions work fine.
 >>
 >> There HAS to be a way to package pre-compiled versions of the JSPs
 >in the
 >> "model" application in the admin server and keep from having to precompile
 >> the JSPs on every managed server every time managed server is booted..
 >>
 >> It would help if we had a way to bypass the staleness checking
 >completely..
 >> Or you guys should make the timestamps on the files copied by the
 >> admin/managed deployment process match properly so the staleness checker
 >> doesn't think the JSP is different.
 >>
 >> It would also help if the engineer who wrote this could explain the
 >rules
 >> being implemented by the staleness checker. So far all the messages
 >in
 >the
 >> newsgroup have amounted to point solutions for problems without a good
 >> understanding of what the engine is checking for and/or doing under
 >the
 >> covers. Looking at the generated .java files for the JSP pages helps,
 >but
 >> it is not good enough...
 >>
 >> Anyone out there have a working admin/managed server JSP application?
 >> -Greg
 >>
 >> -----------------------------------------------------------
 >> Check out my WebLogic 6.1 Workbook for O'Reilly EJB Third Edition
 >> www.oreilly.com/catalog/entjbeans3 or www.titan-books.com
 >>
 >>
 >>
 >
 >

Log4j and managed servers

This seems somewhat obvious, but...
 I have a Solaris based WLS 8.1.6 cluster with two managed servers.
 In the log4j.xml for BOTH managed servers, the logs for output from both go to the same file.
 What happens is one managed server writes to dctm_api.log and one writes to the previous log file.
 Is this as simple as Log4J in one of the Managed Servers has the file open and Log4J from two process can't have the same file open and write to it at the same time?
 Thanks,
 Michael Geiser

Static fields won't get serialized.
 Peace,
 Cameron Purdy
 Tangosol, Inc.
 http://www.tangosol.com/coherence.jsp
 Tangosol Coherence: Clustered Replicated Cache for Weblogic
 "rohit" <[email protected]> wrote in message
 news:[email protected]..
 >
 > Hi, can someone please advice if using log4j in the clustered environment
 would
 > in any way result in the failure of in memory session replication. We are
 using
 > WL 7.0 and log4j is installed on each managed servers. The log4j class
 "org.apache.log4j.Logger"
 > is used as a static reference within java classes which are put in the
 session.
 > Any response greatly appreciated. Thanks. Rohit

Best practice to restart weblogic and managed servers after changing pass

hi
i need to know if have an installation with the following structer what is the best practice to restart the servers after we change weblogic password ?and i know that i should change boot.properties with the new password ,is there anything else to change ? please advise .
oracle access manager &oid (dose it affected by weblogic password change ?)
thank you
Edited by: hsweiss on Jul 31, 2012 10:45 AM

the order is to restart:All managed servers and followed by Admin server
If your weblogic server is integrated with oid, you need to change the credentilas in both the places ie weblogic server(boot.properties) and idm server
if your weblogic server is using your internal ldap for the authentication, your idm and oam wil not effect with this change
if you change the boot.properties file that is enough, you don't need to change anywhere

JNDI, EARs and managed servers

I have an EAR file containing a WAR and two EJB JAR files. When I deploy this EAR
onto a server (admin server), assign a target to the 3 modules and run, evertying
is fine.
I now have a scenario where I have added a managed server (so there are two targets
to pick from). I deploy to the admin server as before but I change the target
for the Web module to the new server. This is a typical deployment scenario where
the web container runs on a different server (perhaps in a DMZ) to the EJB container.
Now when I try and run the web app, the servlet fails with a naming exception
- it can't find the EJB in the JNDI tree. Sure enough when I go to the console
and look, the EJBs are only in the JNDI tree on the first server (those they are
targeted to). So the servlet is quite right to complain.
Is then JNDI tree not shared between managed servers? Shouldn't the EJBs be seen
even though they are on separate servers?
If I make the new server a target on the EJBs then it is obviously resolved but
I don't want these remote EJBs on my Web Application server. These are not, and
shouldn't be, clustered servers.
How do I get around this? Do I have to specify the URL_PREFIX of the first server
in my InitialContext properties? This is not very good if I want to move my EJBs
to a different target, I would then have to change the URL_PREFIX again.
I'm using WLS 6.1 sp1

I think renaming them in the config.xml IS the easy way. You cannot do it through the console. Truthfully, that is a planning step and you should probably have the correct names before building in order to avoid other issues. Might be easier to just recreate the managed servers.

Can I run the admin and managed servers as different users?

in unix, is it possible to run my admin and manged servers as different users with
7.1?

As long as users are in admin group, there could be different users starting different
servers.
"Frank" <[email protected]> wrote:
>
"Erik Johnson" <[email protected]> wrote:
Hi, Frank.
Could you be a little more specific about what it is you want to do?
Thanks.
erikUh, I'm on unix, I have Weblogic 7.0 service pack4. I have 1 managed
server and
I have one admin server...Can I run the admin server as user 'hithere'
and run
the managed server as user, 'byethere'......
"Frank" <[email protected]> wrote:
in unix, is it possible to run my admin and manged servers as different
users with
7.1?

Silent Install w/Admin and Managed Servers

I noticed on versions 11 and 12 of Weblogic, it seems the parameters available for a silent install have been greatly simplified to where it just does a base install. How would I go about getting it to do a silent install with an Admin server and 2 Managed servers? Do I need to do an install and update-install or is there a way I can do all this in 1 shot?
Thanks In Advance
Nic

An example is given here: Middleware Snippets: Automate WebLogic 12.1.2 Deployment.

Development environment and managed servers

Hi,
I am on a project with a bunch of developers each running a local copy of
weblogic for development. We were thinking it would be nice if we could make
each development weblogic server a managed server and administer connection
pools, startup classes, etc from a centralized location. This way we
wouldn't have to duplicate configuration efforts.
The problem is each person needs to develop and compile their own build of
the application to their local hard drive. My understanding is that in order
for an application to be run on a managed server, the application needs to
be deployed to that server from the admin server. Upon startup any ears,
jars, wars, or exploded applications are transferred to the managed server
machine from the admin server machine. So each developer would not be able
to compile and deploy locally as they code.
Has anyone attempted anything like this?
Thanks in advance,
Daniel

Hi Daniel,
Were you able to acomplish this task? Please let me know. I am planning to do
the same for our development team also. Can you please let me know.
Thanks
"Daniel Han" <[email protected]> wrote:
Hi,
I am on a project with a bunch of developers each running a local copy
of
weblogic for development. We were thinking it would be nice if we could
make
each development weblogic server a managed server and administer connection
pools, startup classes, etc from a centralized location. This way we
wouldn't have to duplicate configuration efforts.
The problem is each person needs to develop and compile their own build
of
the application to their local hard drive. My understanding is that in
order
for an application to be run on a managed server, the application needs
to
be deployed to that server from the admin server. Upon startup any ears,
jars, wars, or exploded applications are transferred to the managed server
machine from the admin server machine. So each developer would not be
able
to compile and deploy locally as they code.
Has anyone attempted anything like this?
Thanks in advance,
Daniel

Wl60, jsp reloading, and managed servers

Similar Messages

Maybe you are looking for