Serious problem with WebLogic

We have an application with servlets and tcp servers running in weblogic which receives request, then process requests and then sends the response back the clients.
This application has been fully tested and put into production. This application is mainly used as a stress test tool which means that hundreds of requests are sent to the application every second.
Just recently, we are having problems after 1400-1700 something requests. WebLogic would hang up and the application stops responding.
We did many stress tests with this application before and it was working perfectly fine until now. We used to send it hundreds of thousands of requests to the application for three days without any problems.
I have tried rebuilding the application, reinstalling WebLogic. Nothing seems to help. Memory usage is not a problem.
The thing is that no code changes was ever made, nobody touched the application.
Anybody has any ideas?
+sometimes the servlet would still receive http request, however, tcp/ip requests stop working.  And there was no problem with the logs.  Our application logs are perfectly fine too.
Edited by: user11856615 on Sep 2, 2009 4:34 PM
Edited by: user11856615 on Sep 2, 2009 4:36 PM

We are running WebLogic 9.2, Windows 2000 Advance Server, and is the info for Java:
java.version = 1.5.0_04
java.vm.info = mixed mode
java.vm.name = Java HotSpot(TM) Client VM
java.vm.specification.name = Java Virtual Machine Specification
java.vm.specification.vendor = Sun Microsystems Inc.
java.vm.specification.version = 1.0
java.vm.vendor = Sun Microsystems Inc.
java.vm.version = 1.5.0_04-b05
The thing is that this application has been used as stress test tools for months without any problems. We ran it for weeks at a time and it processed hundreds of thousands of requests without any problem. And now suddenly this problem occurs.
I have noticed that another stuck thread in the logs. I set to ignore stuck thread and will be running another test today and see what happens.

Similar Messages

  • Urgent! Serious problem with J2EE Classloader!

    Hi all,
    I'm having a serious problem with j2ee Classloading.
    I have "inherited" a J2EE application where all business classes (invoked by the EJB) are placed in the Application's Server classpath (Weblogic 8.1.4).
    Very bad because every time a class is changed the a.s. must be restarted.
    So I re-enginereed the package putting the business classes in a jar that is referenced by the EJB's MANIFEST file.
    This way all the classes would be loaded by the EJB classloader and no need to start/stop weblogic.
    Unfortunately it's not such !! I discovered that all business classes are loaded via REFLECTION
    Object objCommand = objClass.newInstance();
    Object[] param = new Object[1];
    param[0] = aDataContainer;
    objDataContainerReturn = (DataContainer) objMethod.invoke(objCommand,param);
    and the application server cannot find them using the standard J2EE Classloader mechanism.
    java.lang.reflect.InvocationTargetException
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
    at java.lang.reflect.Method.invoke(Method.java:324)
    at irma.business.Dispatcher.callService(Dispatcher.java:70)
    Caused by: java.lang.NoClassDefFoundError: irma/utility/log/LogPrintStream
    at irma.business.service.LoginService.login(LoginService.java:79)
    On the other hand they're found if I put the classes on the application server classpath.
    Has anybody got advice for this ? Should I refactor the application giving up reflection ?
    Thanks a lot in advance.

    Reflection uses the classloader as well so it should work fine.
    A common cause of a NoClassDefFoundError is something in a parent classloader that refers to something only available in a child loader.
    For instance, do you have any of your application classes which are in the $CLASSPATH but reference something that is only (now) in the business.jar in your ear?
    -- Rob
    -- Rob
    WLS Blog http://dev2dev.bea.com/blog/rwoollen/

  • Serious problem with my iPod Classic/iTunes

    I have a 6th Generation iPod classic (80 GB), and recently, I have been having some serious problems with it.
    When I connect it to iTunes, there is usually no problem, it shows up, plays and displays the music correctly, etc. However, if I try to ADD any music to it, iTunes will freeze, then close, and within 5 minutes, my entire computer will freeze completely, and be shut off. For whatever reason, happen when my iPod finally disconnects because my computer turned off, all of my music is missing from the library (This doesn't happen every time it freezes, because I try to disconnect my iPod manually before everything freezes). The music is still on the iPod, (it says that I have XX.X GB of "other" instead of music/videos), but it doesn't show up. When I plug in my iPod, iTunes says that there was a problem and I need to restore it. The first time this happened, I lost over 5,800 songs, but I forgave Apple, because these things do occasionally happen, plus I had no idea what exactly went on. The SECOND time happened within the same week, after I had put 800 songs back on the restored iPod. The first time happened when I was using iTunes 10 (which is seriously buggy on my computer, and freezes often), so I switched to iTunes 9 (still buggy, but doesn't freeze quite as much), and the same thing happened. So now I'm thinking it's my iPod's fault. My iPod does a whole host of minor problems over time such as making certain songs cut off at a certain specific point for no reason (Always the same song, always the same cut off point, and I can't fast forward past it, I also can't delete the songs because if I highlight them on iTunes, iTunes will freeze, and if I find them under the music folder in iPod control, my computer will freeze), freezing the entire iPod for seemingly no reason, and somehow messing up my album artwork (it will appear as "Colourful Static"), along with alot more.
    I know this is a troublesome problem, but if anyone could somehow shine some light on this problem, it would be tremendously appreciated.
    Thank you
    -Eric Luna

    Unfortunately, when a hard drive starts clicking that means it's going bad. The mechanism is having trouble reading the drive. Since the iPod is mostly just a harddrive and a battery I have a feeling that this is your problem.
    watch this video. http://www.youtube.com/watch?v=wMm6c3fr-fU
    about 25ss in is where the clicking starts. If it sounds like that, but softer and more muffled since it is a smaller drive that is being enclosed then I think you might be SOL. Maybe if you're iPod is still under warranty you can get a credit and go for a touch and avoid moving parts all together?

  • Apple Have Serious Problems With 10.7.5

    Apple have very serious problems with their software and they do not have the courage to openly admit this to their customers. My company has been dedicated to Apple and all that is good. We have 2 Imacs, I macbook Pro, 1 mac Mini, 1 mac mini server, 1 ipad, 4 iphones and peripheries. I am on the verge of placing them all on a heap and putting a sledgehammer to them all because I am certain that 10.7.5 is such a corrupted release that even Apple technicians cannot help. My company has been on more and more get around fixes that are becoming more and more complicated by the minute. For the last 4 weeks I have been working 8 to 8 daily including some of the weekends to try and fix things. I have had the pleasure (genuinely) of working for two days following a knowledgeable Apple Technicians instructions to resolve a cloud issue and as of today I sit with one imac completely unusable with all sorts of errors and malfunctions - all software based. Other symptoms on other hardware include error messages saying the computer restarted due to a problem with the computer - even though we are carrying out clean shutdowns or restarts; inconsistent acceptance of passwords to i cloud on most equipment but in particular as highlighted above; nearly all applications opening on restart (I do know how to shut that off - in theory!!!!!!!); frequent and very lengthy beachballs; keychain is an absolute mess, slowdowns to the extent that I have to work on three machines to get the minimum of work done; and other random malfunctions or error messages. Apple's positive attitude to its customers used to be second to none. It seems to me that software (and hardware) releases are now too rushed and Apple do not care about software failures but rely on customers to troubleshoot the glitches for them and do so under warranty or more likely care plan support at further cost to the customer.
    Just look at all the questions being asked by customers via web searches on the above issues and more. Lots of questions, no reliable fixes, and certainly no interest shown by apple to resolve the issues - in fact Apple have adopted the say and do nothing approach.
    I love the products and have been a loyal ambassador to Apple but now my stress level, and the inability of my company to engage its business put me on the verge of taking every item my company owns with an Apple logo and at least gaining some satisfaction by placing them all in a heap and smashing the lots with a sledgehammer (deliberate repeat as this is now a realistic option)
    It used to be a pleasure to use the products, and efficient. I am not looking for help on these matters as an Apple technician is working very hard with us but I feel he is as frustrated as us. We are nowhere nearer resolving our problems and almost daily there are other glitches emerging.
    Apple are losing their way. They must listen to the directions given them by their customers frustrations. 

    What happens when you click the button?

  • My Client is having a serious problem with the Contact Form Widget

    My Client is having a serious problem with the Contact Form Widget that I can not figure out how to correct.
    I created a website for a client and inserted a Contact Form widget on their 'Contact Us' page. Whenever anyone uses the form, the e-mail my client receive shows my e-mail address as the sender. So, when they 'reply' to the e-mail, the reply is sent to me and not the person who sent the original message. This creates a major problem in that I get barraged with e-mail replies while my client's potential customers go without a response.
    When I look at  'Site Manager > System E-Mails > Workflow Information > E-mail From Address' on the site's Admin Console, it shows my e-mail information. Is this is what needs to be changed?  If it is, what needs to be placed in that field so that the e-mail my client receives shows the senders e-mail information in the 'From' field.  My client wants to be able to click reply and have the message sent to the right party. They would be very upset if they have to cut and past the senders e-mail address from the body text on every contact they receive.
    Their feeling is that if I can't find a way to make the contact form work the way it should, then it's useless to them.
    Can anyone help me figure this out? I really don't want to disappoint my first client.

    Are you on a plan that has the CRM feature that stores your customers' data? If so, the idea of the contact form is that you'll receive a notification that there's been an inquiry filled out on your contact form and there should be a link in that email notification that leads to the "Case" that was created when that customer filled out a contact form.  You can click that link and visit the case for that customer in your BC site and reply to them from there so that all of the correspondence is logged in the CRM for safe-keeping and for your records.
    If your client finds this is too much work to login to BC to reply every time, then you should check to see if you have the "Customer Service Ticketing" feature on your hosting plan.  This is a feature where you create a dedicated email account (like [email protected]) and the BC system will automatically login to that email account and pull any emails in that inbox and convert them to a customer case for the sender of the email and then it will send out a notification via email to any of the BC Admin users you delegate as "Customer Service Agents".  Since the CST integrates with the BC CRM it lets you reply directly within the email-- but when you reply it will be going to that same email address dedicated to the CST but once the CST service checks your email again and sees that you replied to this inquiry it will log your reply against the customer's case and sends your reply via email back to them so that all the correspondance gets logged on BC's CRM but you can still reply via email.
    There's no way to use the default web forms to update the "From" or "Reply-To" address.. it must come from an approved email address to avoid spam issues.  On most web services you cannot change the "from" address anyway but usually you can at least specify the "Reply-To" address so that when someone hits "reply" in their email client it will reply to whoever you setup as the "Reply-To" address.
    Here's some more information about CST: http://kb.worldsecuresystems.com/kb/customer-service-ticketing.html?bc-partner
    I don't think the CST feature is in the webMarketing BC plan-- I think it's only in the webCommerce plan so if you have less than an webCommerce plan you have to tell your client to reply through BC by clicking the link in the notification they get.  You can justify it by saying its one or two more steps but it keeps the entire convo on record in their CRM for easy referral later.

  • Word experienced a serious problem with the 'acrobat pdfmaker office com addin' add-in. after Office Updates.

    I cant cant use mail merge feature in Winword in conjunction with Abobe Acrobat 10 (pdfmaker add-in) after the recent Office April Windows and Office Updates.
    Here is a quick note from the event viewer:
    The program WINWORD.EXE version 14.0.7134.5000 stopped interacting with Windows and was closed. To see if more information about the problem is available, check the problem history in the Action Center control panel. 
     Process ID: 1138 
     Start Time: 01d07dcede19c458 
     Termination Time: 15 
     Application Path: C:\Program (file:///C:/Program) Files (x86)\Microsoft Office\Office14\WINWORD.EXE 
     Report Id: 
    Microsoft Word: Accepted Safe Mode action : Word experienced a serious problem with the 'acrobat pdfmaker office com addin' add-in. If you have seen this message multiple times, you should disable
    this add-in and check to see if an update is available. Do you want to disable this add-in?.
    I do not wish to disable the pdfmaker add-in as I use it to run mail merge on WORD and send 10s and 1000s of pdfs via email to external users. I am not sure what are my options here and I wish
    someone could help me get to the root cause of the matter.
    My environment is as follows:
    Windows 7 x64 SP1
    Office 2010 v 14.0.7128.5000
    Adobe Reader MUI X and Adobe Acrobat X
    Feroze

    Hi,
    As the event log indicated, this is due to acrobat pdfmaker office com addin doesn't work well with Microsoft Word.
    I would suggest you go ahead and try to upgrade your Adobe product to the latest version and see if issue persists.
    Or contact the support of Adobe and see if there is a known compatibility issue or not.
    Regards,
    Ethan Hua
    TechNet Community Support
    It's recommended to download and install
    Configuration Analyzer Tool (OffCAT), which is developed by Microsoft Support teams. Once the tool is installed, you can run it at any time to scan for hundreds of known issues in Office
    programs.
    Please remember to mark the replies as answers if they help, and unmark the answers if they provide no help. If you have feedback for TechNet Support, contact
    [email protected]

  • PowerPoint experienced a serious problem with the 'adobe presenter powerpoint com addin' add-in.

    I've installed Adobe Presenter V7.0.7 Build 7746. When I open an existing PowerPoint file and try and use Presenter, I get the following error: PowerPoint experienced a serious problem with the 'adobe presenter powerpoint com addin' add-in. If you have seen this message multiple times, you should disable this add-in and check to see if an update is available. Do you want to disable this add-in?
    No, I don't want to disable it. When I click No, PowerPoint stops working. I get a message that says: Microsoft PowerPoint has stopped working. A problem caused the program to stop working correctly, Widows will close the program and notify you if a solution is available. My only choice is to click the Close program button.
    When I open the PPT file again, I get the same "experienced a serious problem" message. If anyone has a solution to this, I'd appreciate knowing what it is. It took me a while just to get the Presenter option to appear in the PowerPoint toolbar. Now that I have it, it's not working.
    Thank you!

    Thank you for your quick reply. Here are the answers to your questions.
    1. The file I'm having issues with was not created using any version of Presenter. It is simply a PowerPoint file.
    2. The file is a PPT not PPTX.
    3. I am running Windows7 and Office 2010.
    I think I might have found the solution to the problem. I opened the PPT file and saved it as a PPTX file. Then I clicked the Presenter option in the PowerPoint toolbar and was able to select different Presenter options with no issues. I opened other PPTX files and I did not get the error window stating 'PowerPoint experienced a serious problem with the 'adobe presenter powerpoint com addin' add-in'.
    This seems to work. So, it appears that any PPT file needs to be saved as a PPTX file before using Presenter. If you have another solution that I might be able to keep in my back pocket (in case this happens again and what I did does not work), then please answer. Otherwise, I'm good now.
    Thanks!

  • I downloaded IOS6 earlier today and my 4S has died twice, it has gone from 45 to 32% in the last 5 minutes and i have icloud, internet, all apps closed with nothing running at all!! There is a serious problem with ios6, my phone was hot at charging time.

    I downloaded IOS6 earlier today and my 4S has died twice, it has gone from 45 to 32% in the last 5 minutes and i have icloud, internet, all apps closed with nothing running at all!! There is a serious problem with ios6, my phone was working hot at chargering time. Apple need to address this asap!!!!

    Hi Suzy Q,
    I can send and receive calls.
    I get "messaging stopped" when doing a lot of texts.
    Texts come up "ghosted" meaning they have the text on the bottom and a transparent one overlaid  on top. they then become one. But I get the ghosting most of the time when I go into a message from a contact to send a text.
    When I go to call logs or text logs, the numbers come up first, and then then the names pop up randomly, meaning not in order down the page, until they are all shown. lots of lag time.
    It runs slow.
    I have 5 apps on here. I should be able to have a  lot of apps on here, and play music, and have files, and...and....and.
    I can get online but doing a Google search, when the page with the choices comes up, it is blurred and I have to tap on the screen for it to become clear and readable.
    When online and scrolling, it hitches sometimes/lags/gets stuck then unstuck
    I get VZWAVS has stopped. (not sure if those are the right letters)
    The status bar that is supposed to be gray with the new update, goes back and forth between green--the old os--to gray. And I have found no continuity with when I get green or when I get gray.
    I get a black screen on occasion.
    sometimes the proximity sensor doesn't work.
    sometimes when I have my bluetooth on and have "hands free mode" on, I get the a text message read through the headset and also through the handset.
    I soft reboot often--turn it off wait a bit and

  • Help me please : Serious problems with collection-mapping, list-mapping and map-mappi

    Hi everybody;
    I have serious problems with list-mapping, collection-mapping and map-mapping.
    Acording to specifications and requirements in a system I am working on it is needed to
    get a "list" of values or an indivudual value.I am working with ORACLE 9i Database,
    ORACLE 9i AS and ORACLE 9i JDEVELOPER.
    I tried to map a master-detail relationship in an entity-bean, using list-mapping.
    And this was very useful in order to get a "list" of details, ...but, when I wanted
    to get a single value I have some problems with persistence, something about "saving a state"
    despite I just want to get the value of a single detail.
    I decided to change it to map-mapping and the problem related with a single detail
    worked successfully, but I can get access to the whole bunch of details.
    May anyone of you help me with that?
    I am very confused I do not know what to do.
    Have any of you a solution for that problem?
    Thank you very much.

    Have you tried a restore in iTunes?

  • Problem with Weblogic Builder SP1

    I have a problem with Weblogic Builder SP1. I basically re-used my EJB deployment
    descriptors from WL5.1 in WL8.1. Before SP1, i have no problem loading my EJB
    in WB. But with SP1, WB throws weblogic.marathon.model.LoadFailureException.
    I re-attempted by jarring my EJB with WL5.1 descriptors and used my friend's WB
    (not SP1) to export new descriptors which I used to re-jar. This new EJB is again
    not recognised by WB SP1?
    What is wrong?

    The exception is as follows:
    Opening module myproj.jar
    java.lang.NullPointerException
         at java.util.Arrays.sort(Arrays.java:1181)
         at weblogic.marathon.model.EJBJarCMBean.initRelations(EJBJarCMBean.java:1212)
         at weblogic.marathon.model.EJBJarCMBean.initBeans(EJBJarCMBean.java:1098)
         at weblogic.marathon.model.EJBJarCMBean.setup(EJBJarCMBean.java:1144)
         at weblogic.marathon.tasks.OpenModuleTask.getEJBModule(OpenModuleTask.java:156)
         at weblogic.marathon.tasks.OpenModuleTask.figureOutModule(OpenModuleTask.java:191)
         at weblogic.marathon.tasks.OpenModuleTask.figureOutModule(OpenModuleTask.java:230)
         at weblogic.marathon.tasks.OpenModuleTask.runBackground(OpenModuleTask.java:89)
         at weblogic.tools.jellybeans.core.task.TaskThread.execute(TaskThread.java:127)
         at weblogic.tools.jellybeans.core.task.TaskThread.run(TaskThread.java:64)
    Module myproj.jar descriptors failed to load
    The EJB has no problem deploying.
    Rob Woollen <[email protected]> wrote:
    You'll have to at least show us the error you get from marathon.
    Does your application deploy to the server?
    -- Rob
    Robin Tan wrote:
    I have a problem with Weblogic Builder SP1. I basically re-used myEJB deployment
    descriptors from WL5.1 in WL8.1. Before SP1, i have no problem loadingmy EJB
    in WB. But with SP1, WB throws weblogic.marathon.model.LoadFailureException.
    I re-attempted by jarring my EJB with WL5.1 descriptors and used myfriend's WB
    (not SP1) to export new descriptors which I used to re-jar. This newEJB is again
    not recognised by WB SP1?
    What is wrong?

  • Gtx860m 2gb serious Problem with gtaV fifa15 MGS ground zero

    Hi people,
    i have a serious problem with my vga (gtx86m)
    when i wanna play with gtav,fifa 15 or mgs ground zero and i start msi after burner i notice that my vga usage is 0%.why?
    i have the last nvidia drivers installed on my notebook(350.12)
    anyone have the same problem?
    do u have some solution?
    i have a msi ge60 apache pro
    i7mobile
    gtx860m 2GB
    8gb ram
    ssd samsung 120GB
    thanks anyway
    [ ERROR: SPECIFIED ATTACHMENT MISSING ]

    Reminder:
    Forum Rules: Don’t… - Bump posts. (Bumps will be deleted without further notice)
    There are many people having the same problem (according to the ROCKSTAR forum), so I think the problem is from the GTA V and the NV Graphics Driver.
    There are users sharing the temporary fix:
    Quote:"There is a temporary fix for the Nvidia cards where you disable the Nvidia cards in the device manager and reenable after launching game."
    Source: https://support.rockstargames.com/hc/communities/public/questions/203308668-GTA-V-on-Laptop-Issues-Intel-AMD
    Give it a try.
    I'm sure that the long term solution should be a game patch or the new NV driver. It's better that you report the issue to the ROCKSTAR GAMES.

  • Why Apple don't fix half year serious problem with HD3000, Lion and Adobe products? Doesn't Apple really care about advanced users?

    Why Apple don't fix half year serious problem with HD3000 drivers, Lion and Adobe products? Doesn't Apple really care about advanced users?
    I have Mac mini and Macbook Pro 13". Bouth with HD3000. And I can't work at all. With Snow Lepard I had no problems. It's tiny problem in code. And such pain for lots of users!

    If you have a problem with Adobe products, voice your issues with them.
    It is the software developers responsibility to maintain compatability
    with any operating system, beit OSX or Windows.

  • Having serious problems with calendar duplications, one birthday now showing 174 times on one day! Help!

    Having serious problems with calendar duplications, one birthday now showing 174 times on one day! Help!

    Having serious problems with calendar duplications, one birthday now showing 174 times on one day! Help!

  • Some serious problems with Mountain Lion

    Hi everyone,
    I upgraded my mid 2010 15" MBP to Mountain Lion 2 days ago, and I have some serious problems with my machine now.
    - Wifi problem: like many members 's machines, mine has problem to connect to the wifi after sleep or restarting, I had to wait 10 to 15 minutes for it to connected to my home network. Anh I tried to remove all of my settings in System Preferences/Network and added new connection. And now it cant connect to wifi anymore. I'm writing this article while sitting on the floor with my Ethernet cable plugged in my MBP. I connected to internet through a DSL modem and a Airport Extreme as a bridge and Access Point. I also tried to set up all of my network device but my MBP can't connect to the wifi still.
    - Keyboard problem with Pages: I was very happy with the new iCloud syncing for Pages documents but after 5 minutes, I met a big problem with Mountain Lion. Whenever I type in Pages with Vietnames Keyboard( Mac OS X built in or from another developer: http://www.trankynam.com/gotv/ ), Pages stops working immediately. I can still drag the Pages's window around but can't type or do any editing. After that I have to force quit pages. It is really stupid that I'm writing a book in Pages and now I have to stop my working till apple fixes this ridiculous bug. I tried German Keyboard as well and Pages worked flawlessly. Do you hate Vietnamese, Apple?
    - Messages problem: Since upgrading to Mountain Lion I couln't send or receive any messages. It worked perfectly in Messages beta.
    Does anyone have the same problems like me? Any solution such as how to reinstall Messages(I googled but found nothing)
    Apple, please test your software more before release it. I feel like I have no computer at the moment because I can't work, can't surfing the web wirelessly and can't communicate with my friends.

    HI,
    - Messages problem: Since upgrading to Mountain Lion I couldn't send or receive any messages. It worked perfectly in Messages beta.
    Does anyone have the same problems like me? Any solution such as how to reinstall Messages(I googled but found nothing)
    What Messages to what type of Buddies or Contacts ?
    This
    I found out that the problem with my wifi was my MBP received automatically a ridiculous IP such as 169.254.170.0 instead of 192.168.1.4 as it should be.
    WIll be part of the Issue as will be DNS issues.
    mDNSresponder crashed with Handsoff! v1.5.3. After upgrading Handsoff! to v2.0.2, it works again.
    In this case mDNSresponder also being stopped will stop DNS working.
    I would guess if you have resolved these that you can Chat to people again.
    4:33 PM      Friday; July 27, 2012
    Please, if posting Logs, do not post any Log info after the line "Binary Images for iChat"
      iMac 2.5Ghz 5i 2011 (Lion 10.7.2)
     G4/1GhzDual MDD (Leopard 10.5.8)
     MacBookPro 2Gb (Snow Leopard 10.6.8)
     Mac OS X (10.6.8),
    "Limit the Logs to the Bits above Binary Images."  No, Seriously

  • Problems with weblogic clustering in 6.1 sp3

    We have spent a lot of time trying to get our application deployed to
              a cluster using weblogic 6.1 sp3 and we consistently receive a failure
              when we attempt to start the managed server. This was not a problem
              with weblogic sp1--we got our application to deploy to the cluster
              successfully; although there was another weblogic bug there with
              clients accessing EJB clusters--we won't go into that here...
              We have tried this on both a Windows 2000 machine and an HP machine
              running weblogic sp3. The same error occurs on both platforms.
              The error in deploying our application to the cluster against weblogic
              sp3 looks to have to do with our custom security realm. Inside our
              custom realm we make use of a configurable providerUrl which we set to
              the cluster address/port. The custom realm makes a call where it
              passes in the providerUrl to:
                   weblogic.management.Helper.getMBeanHome(..., providerUrl,...)
              When we have our providerUrl set to the cluster address/port--e.g.,
                   t3://clustermember1:7001
              and attempt to start the managed server we get the error:
              Starting WebLogic Server ....
              Connecting to http://adminserver:7117...
              The WebLogic Server did not start up properly.
              Exception raised:
              weblogic.management.configuration.ConfigurationException:
              clustermember1 not found
                   at weblogic.management.Admin.getBootstrapLocalServer(Admin.java:1084)
                   at weblogic.management.Admin.initialize(Admin.java:340)
                   at weblogic.t3.srvr.T3Srvr.initialize(T3Srvr.java:359)
                   at weblogic.t3.srvr.T3Srvr.run(T3Srvr.java:206)
                   at weblogic.Server.main(Server.java:35)
              Reason: Fatal initialization exception
              When we have our providerUrl set to the admin server address--e.g.,
                   t3://adminserver:7117
              everything starts up fine.
              Does anyone know why this would work on sp1 and not sp3 of weblogic
              6.1?
              We verified that all passwords are correct and everything else we
              could determine--any ideas would be helpful.
              We don't want the providerUrl to point at our admin server, we want it
              to point at the cluster address/port.
              When we get the managed server error, we received this error on the
              AdminServer:
              2002-08-15 16:52:23,019 ERROR [ExecuteThread: '11' for queue:
              'default'] (com.msa.gabriel.share.security.wlrealm.GabrielRealm) -
              Caught naming exception null; throwing RuntimeException.
              javax.naming.CommunicationException. Root exception is
              java.net.ConnectException: t3://tomtate.msais.com:7119: Destination
              unreachable; nested exception is:
                   java.net.ConnectException: Connection refused; No available router to
              destination
                   at weblogic.rjvm.RJVMFinder.findOrCreate(RJVMFinder.java:155)
                   at weblogic.rjvm.ServerURL.findOrCreateRJVM(ServerURL.java:207)
                   at weblogic.jndi.WLInitialContextFactoryDelegate.getInitialContext(WLInitialContextFactoryDelegate.java:307)
                   at weblogic.jndi.WLInitialContextFactoryDelegate.getInitialContext(WLInitialContextFactoryDelegate.java:211)
                   at weblogic.jndi.WLInitialContextFactory.getInitialContext(WLInitialContextFactory.java:149)
                   at javax.naming.spi.NamingManager.getInitialContext(NamingManager.java:665)
                   at javax.naming.InitialContext.getDefaultInitCtx(InitialContext.java:246)
                   at javax.naming.InitialContext.init(InitialContext.java:222)
                   at javax.naming.InitialContext.<init>(InitialContext.java:198)
                   at com.msa.gabriel.share.security.wlrealm.GabrielRealm.getNamingContext(GabrielRealm.java:416)
                   at com.msa.gabriel.share.security.wlrealm.GabrielRealm.getConnection(GabrielRealm.java:347)
                   at com.msa.gabriel.share.security.wlrealm.GabrielRealm.access$000(GabrielRealm.java:51)
                   at com.msa.gabriel.share.security.wlrealm.GabrielRealm$2.run(GabrielRealm.java:225)
                   at weblogic.security.acl.Security.doAsPrivileged(Security.java:489)
                   at com.msa.gabriel.share.security.wlrealm.GabrielRealm.myDoAsPrivileged(GabrielRealm.java:578)
                   at com.msa.gabriel.share.security.wlrealm.GabrielRealm.getUser(GabrielRealm.java:221)
                   at weblogic.security.acl.CachingRealm.getUserEntry(CachingRealm.java:832)
                   at weblogic.security.acl.CachingRealm.getUser(CachingRealm.java:696)
                   at weblogic.security.acl.Security.getCurrentUser(Security.java:250)
                   at weblogic.servlet.security.internal.SecurityModule.auditPerm(SecurityModule.java:356)
                   at weblogic.servlet.security.internal.ServletSecurityManager.checkAccess(ServletSecurityManager.java:205)
                   at weblogic.servlet.internal.WebAppServletContext.invokeServlet(WebAppServletContext.java:2518)
                   at weblogic.servlet.internal.ServletRequestImpl.execute(ServletRequestImpl.java:2260)
                   at weblogic.kernel.ExecuteThread.execute(ExecuteThread.java:139)
                   at weblogic.kernel.ExecuteThread.run(ExecuteThread.java:120)
              <Aug 15, 2002 4:52:23 PM EDT> <Error> <HTTP>
              <[WebAppServletContext(8091823,wl_management_internal2,/wl_management_internal2)]
              Servlet failed with Exception
              java.lang.RuntimeException
                   at com.msa.gabriel.share.security.wlrealm.GabrielRealm.getUser(GabrielRealm.java:260)
                   at weblogic.security.acl.CachingRealm.getUserEntry(CachingRealm.java:832)
                   at weblogic.security.acl.CachingRealm.getUser(CachingRealm.java:696)
                   at weblogic.security.acl.Security.getCurrentUser(Security.java:250)
                   at weblogic.servlet.security.internal.SecurityModule.auditPerm(SecurityModule.java:356)
                   at weblogic.servlet.security.internal.ServletSecurityManager.checkAccess(ServletSecurityManager.java:205)
                   at weblogic.servlet.internal.WebAppServletContext.invokeServlet(WebAppServletContext.java:2518)
                   at weblogic.servlet.internal.ServletRequestImpl.execute(ServletRequestImpl.java:2260)
                   at weblogic.kernel.ExecuteThread.execute(ExecuteThread.java:139)
                   at weblogic.kernel.ExecuteThread.run(ExecuteThread.java:120)
              Thanks for any help.
              Rich
              

    Bottom line:
              In our custom realm we were not handling user guest correctly. Even if you
              have guest user disabled, Weblogic seems to have hard-coded guest to
              send messages to the cluster every-so-often. Not sure there--couldn't get
              an answer out of bea as to exactly why we see guest still being used...
              It seems that with sp3, the user guest interaction started happening earlier
              than it did with sp1, consequently making our realm code fail when trying to
              retrieve the guest user like someone we knew about in our system.
              Hence, our getUser and authUserPassword methods now return null for both
              users guest and system, making the secondary realm (file realm) be used to
              authenticate and resolve guest & system. BEA helped get our code fixed;
              however, we still don't have a lot of depth in understanding Weblogic
              server's use of guest...
              Apparently, in 7.x the guest & system user behavior and configuration is
              different also... We'll see when we start digging into that migration.
              Rich
              [email protected] (Rich Koch) wrote in message news:<[email protected]>...
              > Thanks for the responses--we're working with weblogic support now. We
              > think that the custom realm that we have [the developer that wrote it
              > left the company] is the problem.
              >
              > The original developer was told by someone to check:
              > weblogic.security.acl.internal.ClusterRealm.THE_ONE != null
              >
              > In order to determine if the JNDI was available/ready for the whole
              > cluster.
              >
              > It seems that the meaning/setting of THE_ONE changed with this respect
              > from
              > sp1 to sp3--i.e., this is no longer a valid test to tell us if the
              > JNDI is
              > ready for the cluster.
              >
              > We believe this was used because the 'system' user has to be
              > authenticated before the custom realm is up. Originally, before this
              > check was in place, an exception was received when authenticating
              > 'system'. This appears to be a weblogic limitation/issue. Support
              > has told us that this is different in weblogic 7.0. Unfortunately, we
              > can't upgrade from 6.1 yet.
              >
              > We'll post the solution when this gets figured out.
              >
              > Rak
              >
              > "Sabha" <[email protected]> wrote in message news:<[email protected]>...
              > > There was a security restriction enforced from sp2/sp3 onwards in terms of
              > > looking up mbeans from admin server.
              > >
              > > This might cause things to fail if you are attempting to lookup Mbeans with
              > > guest priviliges from admin server. Also, can you try doing the following:
              > >
              > > Run " java weblogic.Admin -url adminServer -username system -password
              > > .... -GET -pretty -type Server" and check whether the named clustermember1
              > > is available in the list or not.
              > >
              > > Also you seem to be getting some security exception - can you check that.
              > >
              > > t3://tomtate.msais.com:7119: Destination
              > > unreachable; nested exception is:
              > >
              > > --- Try running weblogic.Admin PING on this one and see whether you are
              > > able to reach this server upon the error message.
              > >
              > > --Sabha
              > >
              > > "Rich Koch" <[email protected]> wrote in message
              > > news:[email protected]...
              > > > We have spent a lot of time trying to get our application deployed to
              > > > a cluster using weblogic 6.1 sp3 and we consistently receive a failure
              > > > when we attempt to start the managed server. This was not a problem
              > > > with weblogic sp1--we got our application to deploy to the cluster
              > > > successfully; although there was another weblogic bug there with
              > > > clients accessing EJB clusters--we won't go into that here...
              > > >
              > > > We have tried this on both a Windows 2000 machine and an HP machine
              > > > running weblogic sp3. The same error occurs on both platforms.
              > > >
              > > > The error in deploying our application to the cluster against weblogic
              > > > sp3 looks to have to do with our custom security realm. Inside our
              > > > custom realm we make use of a configurable providerUrl which we set to
              > > > the cluster address/port. The custom realm makes a call where it
              > > > passes in the providerUrl to:
              > > > weblogic.management.Helper.getMBeanHome(..., providerUrl,...)
              > > >
              > > > When we have our providerUrl set to the cluster address/port--e.g.,
              > > > t3://clustermember1:7001
              > > >
              > > > and attempt to start the managed server we get the error:
              > > >
              > > > Starting WebLogic Server ....
              > > > Connecting to http://adminserver:7117...
              > > >
              > ***************************************************************************
              > > > The WebLogic Server did not start up properly.
              > > > Exception raised:
              > > > weblogic.management.configuration.ConfigurationException:
              > > > clustermember1 not found
              > > > at weblogic.management.Admin.getBootstrapLocalServer(Admin.java:1084)
              > > > at weblogic.management.Admin.initialize(Admin.java:340)
              > > > at weblogic.t3.srvr.T3Srvr.initialize(T3Srvr.java:359)
              > > > at weblogic.t3.srvr.T3Srvr.run(T3Srvr.java:206)
              > > > at weblogic.Server.main(Server.java:35)
              > > > Reason: Fatal initialization exception
              > > >
              > ***************************************************************************
              > > >
              > > > When we have our providerUrl set to the admin server address--e.g.,
              > > > t3://adminserver:7117
              > > >
              > > > everything starts up fine.
              > > >
              > > > Does anyone know why this would work on sp1 and not sp3 of weblogic
              > > > 6.1?
              > > > We verified that all passwords are correct and everything else we
              > > > could determine--any ideas would be helpful.
              > > >
              > > > We don't want the providerUrl to point at our admin server, we want it
              > > > to point at the cluster address/port.
              > > >
              > > > When we get the managed server error, we received this error on the
              > > > AdminServer:
              > > >
              > > > 2002-08-15 16:52:23,019 ERROR [ExecuteThread: '11' for queue:
              > > > 'default'] (com.msa.gabriel.share.security.wlrealm.GabrielRealm) -
              > > > Caught naming exception null; throwing RuntimeException.
              > > > javax.naming.CommunicationException. Root exception is
              > > > java.net.ConnectException: t3://tomtate.msais.com:7119: Destination
              > > > unreachable; nested exception is:
              > > > java.net.ConnectException: Connection refused; No available router to
              > > > destination
              > > > at weblogic.rjvm.RJVMFinder.findOrCreate(RJVMFinder.java:155)
              > > > at weblogic.rjvm.ServerURL.findOrCreateRJVM(ServerURL.java:207)
              > > > at
              > > weblogic.jndi.WLInitialContextFactoryDelegate.getInitialContext(WLInitialCon
              > > textFactoryDelegate.java:307)
              > > > at
              > > weblogic.jndi.WLInitialContextFactoryDelegate.getInitialContext(WLInitialCon
              > > textFactoryDelegate.java:211)
              > > > at
              > > weblogic.jndi.WLInitialContextFactory.getInitialContext(WLInitialContextFact
              > > ory.java:149)
              > > > at
              > javax.naming.spi.NamingManager.getInitialContext(NamingManager.java:665)
              > > > at javax.naming.InitialContext.getDefaultInitCtx(InitialContext.java:246)
              > > > at javax.naming.InitialContext.init(InitialContext.java:222)
              > > > at javax.naming.InitialContext.<init>(InitialContext.java:198)
              > > > at
              > > com.msa.gabriel.share.security.wlrealm.GabrielRealm.getNamingContext(Gabriel
              > > Realm.java:416)
              > > > at
              > > com.msa.gabriel.share.security.wlrealm.GabrielRealm.getConnection(GabrielRea
              > > lm.java:347)
              > > > at
              > > com.msa.gabriel.share.security.wlrealm.GabrielRealm.access$000(GabrielRealm.
              > > java:51)
              > > > at
              > > com.msa.gabriel.share.security.wlrealm.GabrielRealm$2.run(GabrielRealm.java:
              > > 225)
              > > > at weblogic.security.acl.Security.doAsPrivileged(Security.java:489)
              > > > at
              > > com.msa.gabriel.share.security.wlrealm.GabrielRealm.myDoAsPrivileged(Gabriel
              > > Realm.java:578)
              > > > at
              > > com.msa.gabriel.share.security.wlrealm.GabrielRealm.getUser(GabrielRealm.jav
              > > a:221)
              > > > at weblogic.security.acl.CachingRealm.getUserEntry(CachingRealm.java:832)
              > > > at weblogic.security.acl.CachingRealm.getUser(CachingRealm.java:696)
              > > > at weblogic.security.acl.Security.getCurrentUser(Security.java:250)
              > > > at
              > > weblogic.servlet.security.internal.SecurityModule.auditPerm(SecurityModule.j
              > > ava:356)
              > > > at
              > > weblogic.servlet.security.internal.ServletSecurityManager.checkAccess(Servle
              > > tSecurityManager.java:205)
              > > > at
              > > weblogic.servlet.internal.WebAppServletContext.invokeServlet(WebAppServletCo
              > > ntext.java:2518)
              > > > at
              > weblogic.servlet.internal.ServletRequestImpl.execute(ServletRequestImpl.java
              > > :2260)
              > > > at weblogic.kernel.ExecuteThread.execute(ExecuteThread.java:139)
              > > > at weblogic.kernel.ExecuteThread.run(ExecuteThread.java:120)
              > > > <Aug 15, 2002 4:52:23 PM EDT> <Error> <HTTP>
              > > >
              > > <[WebAppServletContext(8091823,wl_management_internal2,/wl_management_intern
              > > al2)]
              > > > Servlet failed with Exception
              > > > java.lang.RuntimeException
              > > > at
              > > com.msa.gabriel.share.security.wlrealm.GabrielRealm.getUser(GabrielRealm.jav
              > > a:260)
              > > > at weblogic.security.acl.CachingRealm.getUserEntry(CachingRealm.java:832)
              > > > at weblogic.security.acl.CachingRealm.getUser(CachingRealm.java:696)
              > > > at weblogic.security.acl.Security.getCurrentUser(Security.java:250)
              > > > at
              > > weblogic.servlet.security.internal.SecurityModule.auditPerm(SecurityModule.j
              > > ava:356)
              > > > at
              > > weblogic.servlet.security.internal.ServletSecurityManager.checkAccess(Servle
              > > tSecurityManager.java:205)
              > > > at
              > > weblogic.servlet.internal.WebAppServletContext.invokeServlet(WebAppServletCo
              > > ntext.java:2518)
              > > > at
              > weblogic.servlet.internal.ServletRequestImpl.execute(ServletRequestImpl.java
              > > :2260)
              > > > at weblogic.kernel.ExecuteThread.execute(ExecuteThread.java:139)
              > > > at weblogic.kernel.ExecuteThread.run(ExecuteThread.java:120)
              > > >
              > > >
              > > > Thanks for any help.
              > > >
              > > > Rich
              

Maybe you are looking for