JMS uniform distributed Queue control unit, problem when a node crashes

Hello
I have the code according to which a post (with all of the control unit) to a line of uniform distribution in a cluster with two servers (Server1 and Server2) members.
UDQ - addressing a subdeployment mapped to two servers JMS pointing to each of the Member servers
-Factory connections using default targeting (I tried mapping them to deployment Sub too)

serverContext javax.naming.InitialContext = new javax.naming.InitialContext ();
javax.jms.QueueConnectionFactory qConnFactory = (javax.jms.QueueConnectionFactory) serverContext.lookup (jmsQConnFactoryName);
javax.jms.QueueConnection qConn = (javax.jms.QueueConnection) qConnFactory.createConnection ();
javax.jms.QueueSession qSession = qConn.createQueueSession (false, Session.AUTO_ACKNOWLEDGE);
javax.jms.Queue q = (javax.jms.Queue) serverContext.lookup (jmsQName);
weblogic.jms.extensions.WLMessageProducer qSender = qSession.createProducer (q) (weblogic.jms.extensions.WLMessageProducer);
qSender.setUnitOfOrder ("MyUnitOfOrder");
javax.jms.ObjectMessage message = qSession.createObjectMessage ();
< String, Object > HashMap map = new HashMap < String, Object > ();
Map.put ("something", "SomeObject" "");
message.setObject (map);
qSender.send (message);
} catch (Exception e) {}
}

Steps followed:
1 post a message of "Server1."
2. the message picked up by "Server2."
3. all fine
4. stop "Server2."
5 post a message of "Server1."
6 ERROR: ' chopped format member of MyAppJMSModule! MyDistributedQ is MyAppJMSModule! MyJMSServer-2@MyDistributedQ who is not available.

WebLogic version: 10.3.5

Is there a way (except Path Service configuration) to operate "with control unit" for a UDQ, even if some Member servers go down this code?

Thank you very much for your time.

If you want to avoid using the Service path, then the alternative is to make members of highly available destination. This will help to ensure that the host for a particular Member PUA is in place.

One approach to HA is to set up "service migration." For more information, see the white paper of the automatic Migration of Service to
http://www.Oracle.com/technology/products/WebLogic/PDF/WebLogic-automatic-Service-Migration-whitepaper.PDF

In addition, I recommend SEO best practices for JMS beginners and advanced users
http://docs.Oracle.com/CD/E17904_01/Web.1111/e13738/best_practice.htm#JMSAD455 to help WL configuration in general.

Hope this helps,

Tom

Tags: Fusion Middleware

Similar Questions

  • JMS Timestamp in a uniform distributed queue on Weblogic Console

    Dear Experts,

    I should need a clarification on how JMS Timestamp is valued in a uniform distributed queue. Let me explain better:

    I'll use the JMS timestamp to re - order the messages retrieved from the uniform distributed queue. East - reliable? How can I synchronize this JMS property to make sure it is not dependent on the physical machine where distributed uniforms are deployed a queue?

    Thank you very much
    Mike

    Hi Mike,.

    From your brief description it sounds to the AUP is a perfect fit. It is specifically designed to handle this use case and works in all conditions. The application must define a PUA on each message, in accordance with the documentation. See chapter of AUP of the programmer's guide:

    http://download.Oracle.com/docs/CD/E14571_01/Web.1111/e13727/uoo.htm#i1049853

    Tom

  • How to put on a uniform Distributed queue/topic break

    Good day to all.  I'm under WebLogic 10.3.6.0.  When I create a standard queue or a topic, I have a control tab on the settings of the queue.  In this tab control, I can pause all of good things, production, consumption.

    When I went to HA, I made a uniform distributed queue.  I don't have the control tab.  The only option I see is to pause at startup.  Pause for a direct queue was very helpful.  How can you do with a distributed queue?

    Well, actually it does:

    "You can pause or resume the insertion during execution for all destinations on a JMS server, to a group of destinations that link to the same JMS model or to different destinations. The most recent configuration change always have priority, regardless of the level at which it is made (Server JMS, JMS model or level of destination). »

    So when you use a JMS server for your distributed destination, you can pause production, insertion and consumption on the JMS server.

    the Tracking tab shows what queues distributed it is 'manage '.

    by using the tab control, you can suspend and resume operations.

  • Reading Messages from a uniform distributed queue without an MDB

    WebLogic Server Version: 10.3.4.0

    I am currently facing a problem trying to read messages from a queue of distributed uniform with a spring JMSTemplate Weblogic.
    We have a cluster created with 2 servers.
    Each server in the cluster has it's own server JMS (JMSServer-1, JMSServer-2), each has its own persistent store
    We have a JMSModule. This JMSModule contains a factory connections (Subdeployment = sub1 Target = mycluster) and also contains a uniform tail (Subdeployment = sub1 Target = MyCluster) distributed

    A peak at our Spring JNDITemplate configuration looks like:
    < bean id = "messagingJndiTemplate" class = "org.springframework.jndi.JndiTemplate" >
    < property name = "environment" >
    < Accessories >
    < prop key = "java.naming.factory.initial" > weblogic.jndi.WLInitialContextFactory < / prop >
    < prop key = "java.naming.provider.url" > t3://some.server.name1,some.server.name2:8001 < / prop >
    < / Accessories >
    < / property >
    < / bean >

    It seems that every time that I call the JMSTemplate.browse function in my client, I'll read the messages in the server 1. Then the next bed request messages that are found on server 2. Each request is sent around in circles.

    I then tried to change the JMSModule. The new JMSModule contains a factory connections (Subdeployment = sub1 Target = mycluster) and also contains a uniform distributed Queue (Subdeployment = sub2 Target = JMSServer-1, JMSServer-2)

    After this change, it seems that every time that I call the JMSTemplate.browse function, I'll read the messages in any server, I have a session with my client. If I'm logged on server 1, I can only see messages in Server 1 and vice versa.


    My question is, how to configure Weblogic to pass messages from TWO servers in my cluster? I guess since it was a cluster installation, I would get some sort of uniform distributed queue unique view to access via JNDI, but appears not to be the case.
    Thanks in advance.

    Browsers and receivers always attach to a single member of a destination distributed.

    WebLogic BMD, on the other hand, automatically handle the task of fixing the receivers to each Member and are fairly simple to code and use these days. If you have the possibility to use WL BMD, I recommend their use. (There is no equivalent for the browsers).

    Spring does not have the same STANDARD, but it seems not be a workaround for the spring receiver number (but not for browsers - receivers only). Here is a sample spring impl that attaches a subscription for each Member of a subject is distributed:
    http://sleeplessinslc.blogspot.com/2011/12/WebLogic-JMS-partitioned-distributed.html.

    If the foregoing is not useful and that you must scroll through each message on each server in the cluster, then you must write special code case to check each separately. There are two options to list the destinations and work with each of them - 'messages management' JMX mbean API (WLST Jython scripting or Java-based) and the availability of destination weblogic.jms.extensions API.

    HTH,

    Tom

  • BAM consume does not correctly uniform distributed queue?

    Hello

    We see that the EMS uses not correctly uniform Distributed queue, this is a limitation or're missing us some configurations?

    Thank you.

    Finally understood... looks like that BAM EMS does not create many consumers connecting to a UDQ

    Technical support article: BAM doesn't have several consumer for JMS (Doc ID 1493533.1) queues

    Open enhancement, delivery possible in 12 c

  • Installation of the Photoshot 13 elements. error message: "this is a problem when extracting the files. Check the amount of space available on your computer and the rights to write to the destination folder. I have 1.38 TB of free space and total control

    I try to install photoshop elements 13, that I bought and downloaded from Amazon. When I try to install it I get the following error message: "this is a problem when extracting the files. Check the amount of space available on your computer and the rights to write to the destination folder. I administrar 1.38 TB of free space and total control.

    Charles please do not remove the installation files in the Program Files directory.  I recommend to extract the installation to your downloads folder or your desktop files.  Once installation files are extracted, you can then run the installation program to install Photoshop elements 13.

  • compare the control unit when obtained from scriptures Sub

    Dear,

    I'm working on oracle apps R12, and I two business units when do two bills of different operating units and transfer to the GL, how do I compare journals it is for a specific operating unit in the database is there any ID in the database that I can compare different operating unit transactions that come from the sup to the GL book.


    Thanks in advance,

    Hello

    Ledger doesn't store information directly on the control unit which is normally captured at the transaction level.
    However, there is a table called GL_IMPORT_REFERENCES that keeps track of each worksheet that is passed from subsidiary ledgers to the General Ledger. This table has a column GL_SL_LINK_ID... This can be compared with a few tables over to reach the transaction source and pick up the reference of operating unit...

    (1) for the sheet, you want to know the GL_SL_LINK_ID of table GL_IMPORT_REFERENCES...
    (2) once GL_SL_LINK_ID is obtained, use the value in the XLA_AE_LINES table and collect the AE_HEADER_ID value...
    (3) once the AE_HEADER_ID value is obtained, use the value in the XLA_AE_HEADERS table and capture the value EVENT_ID...
    (4) once the event_ID value is obtained, you can check it with the ACCOUNTING_EVENT_ID of the base transaction... table column for accounts payable invoice it would be AP_INVOICE_DISTRIBUTIONS_ALL and for the payment of accounts payable AP_CHECKS_ALL... Similarly for other modules, the basic transaction tables would hold the reference... .Similarly each transaction table at the level of the subledger database allows to store the ORG_ID which is your value of operating unit...

    In addition, if your users provide a name for the JOURNAL by operating unit during the transfer of auxiliary accounting journal entries in the log book, you can identify by the same...

    Kind regards
    Ivruksha

  • The OSB distributed queue proxy configuration

    I implemented a distributed WeblogicServer queue and I wonder how can I configure proxy OSB to consume her JMS messages. I have read the documentation on the Oracle website, and he told a group of distributed queues a bunch of local JMS queues in different servers which can be accesed through a JNDI name. This failover and load to provide balance to put in the queue of messages.
    The BSO a JMS destination proxy is configured with the following URL: jms://server:port / destination. So if I have a distributed queue that maps to two JMS, what server and port servers do I set proxy URL? Do I need to configure two different agents and one for JMS server1 to server2 JMS?

    Kind regards

    CCI

    For weblogic JMS, producer of message load balancing happens in 3 different places.

    1. to search for initial context - it depends on the URI that you specify in the call for Initial context. for example if you specify t3: / / localhost:7002, localhost:7003, all research context occur on port 7002. Port 7003 is used only if 7002 is out of service. So this supports the failover instead of the load balancing. If you want the load balancer for executive search, then you must use a dns based address dns cluster where the name to a resolution address different every time.

    2 - when a jms connection is created. The jms connection can be created with any managed server to which the connection factory caters to the. So, you can have your search for context on 7002, but real jms connection can get created on 7003. If the server affinity is enabled for the connection factory and then the jms connection will be created to the same server instance managed that in search of context that's happened.

    3 - when a message send a producer is run. Send message can land on any member of the cluster dd if NLB is enabled. If the server affinity is enabled, then the message will end up on dd members on the same instance of the server to which the jms connection has been made. For example suppose you have your jms client application has a connection to jms on ms1 and product message ends up on top of ms2, then the following path would have been taken.

    JMS client app - via jms---> ms1 - in-house connection transmits---> ms2 - message puts---> dd2

    I recommend you read the chapter JMS the book professional Oracle Weblogic Server where it is explained clearly.

  • Message Listener listens only 1 queue in the configuration of distributed queues!

    Hi guys,.

    This may be the wrong place for this post.

    We have created a system of e-mail with the OSB proxies come and messages between distributed queues. After all this messages autour brewing, the messages themselves in a distributed queue where a java client is waiting to pick up the results. Our problem is that it is only reading one of the queues in this queue distributed.

    The following is the code segment that wants the message with a specific id:
    public Message (String queueName, String messageId) receiveMessage throws {FrameworkException}
    BytesMessage, bytesMessage = null;
    If (null == queueSession) {}
    init();
    }
    try {}
    Queue receiveQueue = lookupQueue (queueName);
    QueueReceiver queueReceiver = null;
    Channel selector = null;
    If (null == messageId) {}
    queueReceiver = queueSession.createReceiver (receiveQueue);
    } else {}
    selector = "JMSCorrelationID = '" + messageId + "'";
    queueReceiver = queueSession.createReceiver (receiveQueue, selector);
    }

    bytesMessage (BytesMessage) = queueReceiver.receive ();
    queueReceiver.close ();
    } catch {...


    QueueReceiver.receive () will listen to all members in a distributed queue (spread over 2 servers managed in a cluster)?

    Thank you very much

    No.. Since you have a single consumer, he can listen to only one destination member.

    Is your java client to run in Server or an external client? If its operation in Server convert the client to an mdb.

    Since this one issue of jms, try to post in the forum of jms:
    WebLogic Server - JMS

  • Why firefox is always problem when I put cache 0 MB Web site?

    Hello

    I develop Web sites and about 1 month, I think that to the top of the Version 30 or 31 have problems with the cache.

    I activated the manual cache size and set it to 0MB (Web sites), but I still have problems that a site is cached.

    In development, it's a big problem, when the browser still shows you an old page. I have change something in my code, but it does not apear when I reload the page. When I opened another browser chrome for example I see my changes.

    BTW, I don't like chrome and I was really happy when I stay in firefox. But the described problem could couse me to change browser.

    Best regards and continue to have fun in dev
    Michael Topor-Grabowski

    Hello Michael,
    There are actually quite a few ways to accomplish what you're trying to do. To refresh the page without querying the cache, the simplest method is to type ctrl and f5 instead of simply typing f5 or hit the button reload. The second would be to set up your development web server to transmit the "Cache-Control: non-cache ' directive cache so that any browser you test with, (as long as it is in accordance with w3c) does not cache the hosted Web pages. The last method should be considered as a last resort, but you could also disable the cache of Firefox by activating / disabling the network.http.use option - Hide all: config to false. Remember, once you decide on one of these methods to clear your browser cache before testing again.

  • two other programs to lose control of cursor when loading firefox. Nitro pdf &amp; on display at take-off from the software Center. Help, please

    two other programs to lose control of cursor when loading firefox. Nitro pdf & on screen takeoff on center software. Help, please.

    When I use the programs listed above and firefox is loaded, scrolling through these programs takes me at the edge of the window. I can't scroll 3 lines I can when FF is not loaded. I hate to say it, but FF is certainly the culprit.

    OK, if I may ask, what also happened when Firefox is in Mode safe? That will do this troubleshooting step, is whether this is caused by an add on that you have in your profile. Questions to troubleshoot Firefox in Safe Mode

    Moreover, the fact happen on the touchpad, as well as with an external mouse (if it's a laptop, if not, please ignore)Reset Firefox preferences to troubleshoot and solve problems

    Does this happen with accelerated scrolling on? This setting is in preferences/Options > Advanced Options > general

    Please post back with your results, thank you!

  • How can I solve very slow scrolling with the mouse when you press the command key? (MacBook Pro, Lion, no problem when using trackpad, no problem with other web browsers, other mice have the same problem)

    Scrolling speed is fine, but when you press the command key and then it becomes very slow.

    He does it with a mouse, but not with the trackpad.
    It does when the modules are disabled.
    No problem on Safari or Chrome.
    No problem on PC.

    I think it all started when I went form a MacBook with Snow Leopard on a MacBook Pro with Lion.

    Any help would be much appreciated.

    Thanks in advance!

    You use the mouse wheel to scroll?

    There are some prefs mousewheel, which control the behavior when you press the modifier keys.

    You can set the key to the action of each to 0 for normal scroll.

    See http://kb.mozillazine.org/About:config_entries #Mousewheel.

  • Problems when you try to move the 3 stepper motors

    Hello

    I have problems when I Isaiah to compile a 3 stepper motors control program.

    I have 3 steepest motors and controller (3 SMCP33 + SMCP33-EVA) nanotec.com, they also provide an example of Vi to control a motor that works successfully, but problems happen when I try to adapt the program to try to start the 3 Motors. The final application for this is going to be a Cartesian robot XYZ so finally I need to program coordinates to move the 3 Motors.

    I am new to LabVIEW, so what I do is try to adapt the nanotec example to order 3 engine as follows.

    -J' I assign each engine a different address, doing so that I can run each engine separately. But when I try to run 3 programs at the same time, they work but all moving them the engine (the first that has been run).

    -If I try to make a sort of sub - VI is the same thing, only one motor is driven.

    -Moreover, I get a warning when I try this, I have attached a screenshot of this.

    I have attached the Vi and texture it to improve my description.

    I woul be grateful if someona can help aport.

    Kind regards.

    Thanks for posting the original.  By comparing the two screws immediately gives an indication of what might happen.

    In the picture you posted you circled two places where you changed the address of the reader. In the Example.vi - Nanotec address of the player is connected to about 14 seats.  So when you try to run the other engines in your modified program, the 12 places, you have not changed the address of the player are always preset to motor 1.

    What you need to do is to divide the example VI in at least three parts.  The first part initialization. The next part moves the engine.  The last part ensures that the engine is stopped and made any required another stop.  Each party becomse a separate Subvi. Each Subvi has the address of the reader as input.  The initialization and shutdown parts are placed before and after a while loop.  The movement parts are inside the loop.  You can use three of each sub - VI for the three engines or a Subvi combined with subVIs move three to manage all three engines.

    I can't say what approach might be better without knowing more about how you plan to order the engines - one at a time or all three at the same time, what kind of feedback is used, how management mistakes and other topics.

    Lynn

  • I have problems when sending emails on Outlook Express (Windows XP Home) told in the Outbox, even if they were sent. What can I do?

    I have problems when sending emails on Outlook Express (Windows XP Home) told in the Outbox, even if they were sent.  What can I do?

    You have apparent dbx file corruption.

    Spend most of your messages out of the Inbox and then create new folders to send and sent items box after having moved the messages you want to save to a local folder that you create.

    Tools | Options | Maintenance | Store folder will reveal the location of your Outlook Express files. Note the location and navigate on it in Explorer Windows or, copy and paste in start | Run.

    In Windows XP, Win2K & Win2K3 the OE user files (DBX and WAB) are by default marked as hidden. To view these files in Windows Explorer, you must enable Show hidden files and folders under start | Control Panel | Folder Options icon | Opinion, or in Windows Explorer. Tools | Folder options | View.

    With OE closed, find the DBX files for the items in the Outbox and sent and delete them.  New ones will be created automatically when you open OE.

    After you're done, followed by compacting your folders manually while working * off * and do it often.

    Click Outlook Express at the top of the the folder tree so no folders are open. Then: File | Work offline (or double-click on work online in the status bar). File | Folder | Compact all folders. Don't touch anything until the compacting is completed.

    General precautions for Outlook Express:

    Do not archive mail in the receipt or sent items box. Create your own user-defined folders and move messages you want to put in them. Empty the deleted items folder daily. Although the dbx files have a theoretical capacity of 2 GB, I recommend all a 300 MB max for less risk of corruption.

    Information on the maximum size of the .dbx files that are used by Outlook Express:
    http://support.Microsoft.com/?kbid=903095

    Disable analysis in your e-mail anti-virus program. It is a redundant layer of protection that devours the CPUs, slows down sending and receiving and causes a multitude of problems such as time-outs, account setting changes and has even been responsible for the loss of messages. Your up-to-date A / V program will continue to protect you sufficiently. For more information, see:
    http://www.oehelp.com/OETips.aspx#3

    Why you don't need your anti-virus to scan your email
    http://thundercloud.NET/infoave/tutorials/email-scanning/index.htm

    Note that for some AV programs, it may be necessary to uninstall the program and reinstall in custom Mode and uncheck analysis when the option is the result of e-mail messages.

    Compact often as specified above.

    And backup often.

    Outlook Express Quick Backup (OEQB Freeware)
    http://www.oehelp.com/OEBackup/default.aspx

  • "Windows Media Player has encountered a problem when burning the file to disk"?

    "I ENCOUNTERD CE WHEN BURNING YEAR MP3 AUDIO CD AND NOW I wanted TO KNOW WHY IT WON'T WORK THAT IS THE CODE POPPING up" Windows Media Player has encountered a problem when burning the file to disk ' WHAT CAN BE THE PROBLEM

    Hello MactabardX,

    I try and remove Windows Media Player and then reinstall it.
    Uninstalling and reinstalling Windows Media Player:
    Step 1.
    Uninstalling Windows Media Player:
    1. go to start and in the search type "Turn Windows willing or not.
    2. click on "Turn Windows features on or off".
    3. search for multimedia and uncheck the brand in the face of Windows Media Player.
    4 restart the computer
     
    Step 2.
    Reinstall Windows Media Player:
    1. go to start and in the search type "Turn Windows willing or not.
    2. click on "Turn Windows features on or off".
    3. find the multimedia functions and place a check mark in front of the Windows Media Player.
    4 restart the computer.

    If you encounter problems with Windows Media Player, try to use the Troubleshoot utility to see if that fixes the problem.

    Open the troubleshooting Windows Media Player settings Troubleshooter by clicking the Start button, then Control Panel. In the search box, type troubleshooting, and then click Troubleshooting. Click View all, and then click the Windows Media Player settings.

    If the size of the .wmv files is too large, it can cause Windows Media Player to stop working.

    I try and remove videos from Windows Media Player library and test to see if the size of the files could be a problem.

    If please reply back and let us know if these steps helps.

    Thank you

    Marilyn
    Microsoft Answers Support Engineer
    Visit our Microsoft answers feedback Forum and let us know what you think

Maybe you are looking for