Role of the JMS server

Sorry for posting this kind of question, but after reading several times BEA documentation regarding JMS and clustering, I still do not understand the purpose of a JMS server.
          Questions:
          1. What is really a JMS server for? What is its purpose?
          2. Do I need to create and deploy one JMS server per node in a cluster, or I just need one?
          Thank you for your assistance.

While a DD is location transparent within a cluster, no matter which nodes it runs on, a DD is not required to get location transparency. Any physical destination with a JNDI name is transparently accessable from any server in the same cluster, or any remote client connected to any server in the same cluster - just by looking up the JNDI name. This works the same way other JNDI advertised resources work (EJBs, JDBC connections, etc.)
          A DD is useful for load-balancing, performance, and high-availability. It does this by providing a virtual destination that consists of multiple physical destinations. There is usually little point in a DD that only consists of a single destination, or that has all of its destinations on a single host machine/WL server. To get the full use out of a DD, configure a JMS server on more than one cluster node, and configure the DD to use a physical destination on each one.
          Whether or not you choose to use a DD is application specific. I estimate that the majority of applications don't, but performance sensitive applications, or apps that have high HA requirements typically do. Note that DD apps must typically be able to tolerate out-of-order message processing due to the very nature of DDs (WL 9.0s unique "unit-of-order" feature is designed to address this issue).
          Tom

Similar Messages

How the JMS client detect if the JMS server is still running?

I have a JMS server running on Weblogic and a JMS client running as a standalone application on my local machine. I ran into the problem that if the JMS server is down for a period of time and then get re-started, the JMS client will lose its connection to the server. Since JMS client is just passively listening to the topic it subscribes, it will have no information about the status of the JMS server. Therefore it will not be able to receive any new message if the JMS server is restarted.
I wonder if there is a way I can have the JMS client automatically detect if the connection to the server is lost or reset. I tried to use a while loop in JMS client that does a JNDI loopup every minute to check if the JMS server is up. But if the WL server is down, the JMS client will just catch an exception and crash.

Hi,
You should use an exception listener that allows a client to be notified of a problem asynchronously. As your client only consumes messages, it would have no other way to learn that its connection has failed.
Hence, you should implement the interface ExceptionListener (let says MyExceptionListener) and define your reconnection logic in the method onException(JMSException exception). MyExceptionListener has to be registered with your connection with the method:
public void setExceptionListener(ExceptionListener listener) throws JMSException
Hope it helps,
Arnaud
www.arjuna.com

Connection broker and RemoteApp manager role on the same server OK?

Hello. I may have what seems to be a strange request here. I have a currently working 3 node TS farm consisting of 2 Terminal Servers with the RemoteApp role on them and 1 server running the connection broker. All is running fine but the RemoteApps I have
published are prohibited to have the clipboard enabled for them due to PCI DSS compliance. I wanted to install some other apps and have the clipboard enabled but since it seems that the clipboard is a server setting and not a per app setting, I was hoping
to just install these apps right on the server running the connection broker vs building a separate new TS server to host them. These other apps do not need to be in the load balanced environment.
Can I just install the RemoteApp role on the CB server without any interruption to the farm?
Thanks,
Joe

Hi Joe,
Thank you for your comment.
Can you let me the OS version of RDS Server?
For better understanding, I can explain you that; RemoteApp is not a role which you need to install. But you need to install RD Web access role and then you can publish RemoteApp through that. For eg. If you are using Server 2012\R2 then you need to create
a collection and after that you can publish the RemoteApp feature which you or your user can use through RD Web Access.
As per your comment you have already TS role installed, then you can install RD Web access through role based installation, select the role and server on which you want to install it. Please refer beneath article if you have server 2008 r2.
Remote Desktop Web Access (RD Web Access)
http://technet.microsoft.com/en-us/library/cc731923.aspx
Hope it helps!
Thanks,
Dharmesh

If the JMS server is down......

If the JMS Server is down when the java application client is sending a message to the server, how can I save this message in order to send it when the server has restarted?
If you have some good solutions, please tell me and we can discuss them! Thanks......

Your solution depends a lot on the JMS provider you're using. Some providers support clustering, such that your client will transparently send to which ever server in the cluster is available. However, clustering in a messaging system typically has different semantics than what you might be used to (or expect). You need to carefully study your provider's documentation to see if it fits your requirements.
Your provider might also support message routing. Then you can use a local server, which resides on the same machine as your client. The local server will forward to a remote server when ever that server is available. This is basically store and forward. Some providers allow you to start the server in-process with the client, so you get store and forward where's it's guaranteed that if the client is up so is the server.
- Bjarne.

Getting the below error while using the JMS server.

<Sep 20, 2010 9:27:43 AM EDT> <Warning> <J2EE> <BEA-160064> <There was an error while making the initial connection to the JMS resource named EMSTopicFactoryNonXA from within an EJB or a servlet. The server will attempt the connection again later. The error was javax.jms.JMSException: [J2EE:160048]JNDI lookup of the JMS connection factory EMSTopicFactoryNonXA failed: javax.naming.AuthenticationException: Not permitted: invalid name or password [Root exception is javax.jms.JMSSecurityException: invalid name or password]>
<Sep 20, 2010 9:27:43 AM EDT> <Warning> <J2EE> <BEA-160082> <There was an error while making the initial connection to the JMS resource named com/lehman/fit/ice/agg2/TopicConnectionFactory from the EJB "Agg2BatchBean" inside application "Agg2". The server will attempt the connection again later. The error was javax.jms.JMSException: [J2EE:160048]JNDI lookup of the JMS connection factory com/lehman/fit/ice/agg2/TopicConnectionFactory failed: javax.naming.AuthenticationException: Not permitted: invalid name or password [Root exception is javax.jms.JMSSecurityException: invalid name or password]>
Note:-weblogic version 9.1sp0
Please suggest me how to resolve the above exception.

<Sep 20, 2010 9:27:43 AM EDT> <Warning> <J2EE> <BEA-160064> <There was an error while making the initial connection to the JMS resource named EMSTopicFactoryNonXA from within an EJB or a servlet. The server will attempt the connection again later. The error was javax.jms.JMSException: [J2EE:160048]JNDI lookup of the JMS connection factory EMSTopicFactoryNonXA failed: javax.naming.AuthenticationException: Not permitted: invalid name or password [Root exception is javax.jms.JMSSecurityException: invalid name or password]>
<Sep 20, 2010 9:27:43 AM EDT> <Warning> <J2EE> <BEA-160082> <There was an error while making the initial connection to the JMS resource named com/lehman/fit/ice/agg2/TopicConnectionFactory from the EJB "Agg2BatchBean" inside application "Agg2". The server will attempt the connection again later. The error was javax.jms.JMSException: [J2EE:160048]JNDI lookup of the JMS connection factory com/lehman/fit/ice/agg2/TopicConnectionFactory failed: javax.naming.AuthenticationException: Not permitted: invalid name or password [Root exception is javax.jms.JMSSecurityException: invalid name or password]>
Note:-weblogic version 9.1sp0
Please suggest me how to resolve the above exception.

Impact of reinstalling the SUP role at the Central server

Hi Experts
Env: 1 Central Server
15 Primary servers
1 WSUS Server
Multiple Active update sources (all primary servers)
We have identified an issue with multiple WSUS update sources in the database. This is from a previous WSUS installation where in the stale entries still prevail in the database. We have determined to delete the inactive update source from the db and reinstall
the SUP role alone. Apart from redoing the updates classification and custom site settings what impact is this going to have in the environment. e.g. on the old update lists, DMs, CIs. Am I going to lose all of that as well?
Looking forward to a response :)

Thanks Jason.
Would I have to reinstall the WSUS server again? I am indeed testing this on my lab, and my DMs and U/Ls were all in place after uninstalling the SUP role and on reinstalling I can see entries in my wsyncmgr.log such as:
"Syncronizing sms database with wsus server"
"Synchronizing update
"Incrementing CI version to '#' which I believe is pretty standard.
and I was following the blog: http://blogs.msdn.com/b/george_bethanis/archive/2013/01/31/confmgr2007-security-updates-cannot-be-downloaded-on-a-win2008-server-with-error-on-wuahandler-log-failed-to-add-update-source-for-wuagent-of-type-and-id-error-0x80040694.aspx
It nowhere said, we would have to reinstall the entire WSUS server!!
Kindly clarify.. </guid>

Windows 2012 R2 Active Directory Domain Services and Remote Desktop services Role on the same server.

Findings:
Currently, Windows 2012 R2 AD DS role and RDS With Broker services can only seem to coexist properly in a new domain not an existing domain. Any attempt to add to an existing domain causes internal database user access denied issues and any attempt to
adjust rights and circumvent is dubious at best.
The escalation technician said it best. Out of 50 clients that want to do this, they end up not being able to help 5 right off the bat for whatever reason. As for the other 40 they might be able to help by running reports, adjusting rights and trying to add
the roles until it works. This can end up being a 20 day process. Basically they are playing whack-a-mole with user rights and permissions until something sticks.
We tried creating an OU where any other domain policies would not be inherited to see if that was the issue, a fresh install with different sequence of adding the Roles, no effect.
Given the errors I witnessed when running procmon and then trying to add the roles, the NT System and the Windows Internal database user had access denied issues on 100+ registry keys when trying to add the roles. After that the system is not behaving normally.
The errors displayed almost mirror the errors that would occur on Windows 2012 when those two roles would be added which of course is officially NOT supported on that system.
This blog needs serious revision:
http://blogs.msdn.com/b/rds/archive/2013/07/09/what-s-new-in-remote-desktop-services-for-windows-server-2012-r2.aspx
This is the excerpt from that blog: Single server RDS deployment including Active Directory. We now support running our RD Connection Broker role service on the same physical instance as an Active Directory Domain Controller. In addition, we published
guidelines for how RD Session Host could be used without the RD Connection Broker.
Microsoft Support was curteous and helpful and they were the ones who advised cutting our losses, which mirrored my hunch after seeing what was transpiring in the system. They refunded my money for the support call.
For me, it was an opportunity to find out if there was any way to configure Windows 2012 R2 in the Same manner that it was setup as Windows 2008 R2 and lay that to rest. The coexistence is poorly implemented. It is as if there was a reaction from all the deprecation
of bread and butter features such as shadowing in TS and the coexistence of AD DS and RDS to where those features were re-added haphazardly. (I have no complaints on shadowing on Windows 2012 R2 it works, just do not like having to go to server manager to
use it).
I opted for virtualizing the Domain controller to eliminate the incompatibility issues and that is what I will be doing from now on. I found free solutions for backing up and reporting for virtual machines as well as the suggested procedures for configruing
a Domain controller as a virtual machine on a Hyper-V environment and I will be sticking to those. Thus far the setup has been operational.
I am not allergic to virtualization, but for really small setups it adds additional time and considerations but if that is how it has to be done, so be it. Windows 2008 R2 days are numbered and since we can usually squeeze 5-7 years on quality server equipment,
buying a Windows 2008 R2 setup now is a borderline disservice in my opinion.
Hopefully someone finds this useful and saves some time.

Hi,
Thank you for posting in Windows Server Forum.
Do you need any other assistance?
Based on your description, you are describing your story of successfully implementing RDS server with AD role and more regarding all RDS related scenario. For shadowing feature, you can use with command also. Below is the syntax to shadow a session.
mstsc /v:<ServerName> /shadow:<SessionID>
Hope it helps!
Thanks.
Dharmesh Solanki
TechNet Community Support

SAP MII 14.0 SP5 Patch 11 - Error has occurred while processing data stream Dynamic Query role is not assigned to the Data Server

Hello All,
We are using a two tier architecture.
Our Corp server calls the refinery server.
Our CORP MII server uses user id abc_user to connect to the refinery data server.
The user id abc_user has the SAP_xMII_Dynamic_Query role.
The data server also has the checkbox for allow dynamic query enabled.
But we are still getting the following error
Error has occurred while processing data stream
Dynamic Query role is not assigned to the Data Server; Use query template
Once we add the SAP_xMII_Dynamic_Query role to the data server everything works fine. Is this feature by design ?
Thanks,
Kiran

Thanks Anushree !!
I thought that just adding the role to the user and enabling the dynamic query checkbox on the data server should work.
But we even needed to add the role to the data server.
Thanks,
Kiran

What is the right behaviour when jms server is down

Hi,
several jms implementation behave differently while trying to connect to the jms server which is not available or when the jms server is suddenly down in the middle.
For example:
iBus Mobile - handles reconnections internally, doesn't throw exceptions. I can only know that i am in the disconnected mode from ExceptionListener.
ActiveMQ and openJMS throw exceptions when they can not connect to the jms provider or when the provider is lost in the middle
Sun MQ has "number of reconnection" parameter which internally tries to reconnect "number" times and after that he throws exception.
While the initial connection failure i can handle in while loop , busy waiting for successful connection, i can not handle the later case, when connection is lost in the middle.
I thought that seemless reconnection is part of jms implementation.
Our application requires smooth working even when there is no connection to jms.
How to handle this issue? I think the behaviour of the most jms described is not so logical in terms of jms requirement.
Slava.

SwiftMq is a commercial product, also it also throws
exception when the broker is not running. Yes, it is a commercial product and high quality. But transparent reconnect is disabled by default in SwiftMQ Router Standard (it is enabled by default in SwiftMQ High Availability Router), hence you get an exception. So always check the docs before you suggest wrong behavior.
To enable transparent reconnect, check these links:
http://www.swiftmq.com/products/router/swiftlets/sys_jndi/introduction/index.html
("JNDI Provider URL" to enabled transparent reconnect for JNDI)
http://www.swiftmq.com/products/router/swiftlets/sys_jms/clients/remote/reconnect/index.html
Since this is a general JMS forum, you might use the SwiftMQ forum if you have specific questions:
http://www.nabble.com/SwiftMQ-JMS-Forum-f15660.html
But keep in mind that SwiftMQ costs money. It is not for free.
I think only iBus is the one who implements
transparent connection and reconnection for 100%Haven't heard from them a couple of years. Are their business still operating? At least their message server wasn't updated since a couple of years.
-- Andreas

What is the association between JMS Server and JMS Module in weblogic 10?

Hi,
I have a created a ConnectionFactory, Queue, Topic under JMS Modules in Weblogic Server 10 using admin console.
And I have created a JMS Server and Targetted to the current running server (Admin Server).
But when I run the Queue Sender program it is unable to LookUP the jndi which I gave for Queue.
queue = (Queue) initialContext.lookup("jndi_queue");
Exception:
javax.naming.NameNotFoundException: Unable to resolve 'jndi_queue'.
But I configured this jndi name when I create the Queue in JMS Module in Weblogic Server 10.
How JMS Module and JMS server are associated.
My Doubt is if i have several JMS Modules and only one JMS Server created then how does the JMS Server identifies which Queue/Topic belongs to which JMS Module????
Please advice.
Thanks....

Hi,
Once you have JMS servers configured:
Step 1: Target the JMS configuration module to the cluster that will host the destination, or, if not using clustering, target the module to the individual WL server that will host the JMS server.
Step 2: Configure a "subdeployment" for the module and set it to reference the exact JMS server(s) that will host the destination. If you're using a distributed desitnation, you can specify multiple JMS servers, if you're using a regular destination, only specify a single destination.
Step 3: Create a destination resource in the module, and set its subdeployment to the subdeployment you specified in step 1 (the console calls this "advanced targeting"). As a best practice avoid the "default targeting" option for the resource - always choose subdeployment (advanced) targeting.
For more information, reference the "best practices" chapter of the JMS configuration guide:
http://download.oracle.com/docs/cd/E14571_01/web.1111/e13738/best_practice.htm#CACJCGHG
I also I highly recommend the JMS chapter of the book "Professional Oracle WebLogic".
Hope this helps,
Tom

Adapter-specific message attributes for the JMS Receiver adapter

Dear SAP Community,
We are trying to activate some parameters at message level when SAP XI sent an interface (XML data) to IBM Websphere MQ. Programmaticaly, this is done the following way:
       TextMessage outMessage = session.createTextMessage();
       MQQueue replyToQ = new MQQueue("TESTQM01","SYSTEM.ADMIN.ACTIVITY.QUEUE");
       Destination replyTo = (Destination) replyToQ;
       outMessage.setJMSReplyTo(replyTo);
       outMessage.setIntProperty("JMS_IBM_Report_Exception", MQC.MQRO_EXCEPTION);
       outMessage.setIntProperty("JMS_IBM_Report_Expiration", MQC.MQRO_EXPIRATION);;
       outMessage.setIntProperty("JMS_IBM_Report_COA", MQC.MQRO_COA);
       outMessage.setIntProperty("JMS_IBM_Report_COD", MQC.MQRO_COD);
       outMessage.setIntProperty("JMS_IBM_Report_Discard_Msg", MQC.MQRO_DISCARD_MSG);
This code is working fine.
Since the documentation about this feature (Adapter-specific message attributes) is very light, We am not sure about the interpretation we made.
First, we defined the following attributes: MS_IBM_Report_Exception, JMS_IBM_Report_Expiration, JMS_IBM_Report_COA, JMS_IBM_Report_COD, JMS_IBM_Report_Discard_Msg. All of them are with type int.
Then we changed the mapping (XSL) to pass the value to the dynamic configuration, using the technical names (DCJMSMessageProperty0, ..., DCJMSMessageProperty4) and the value of the constants. We also specified a value in the Reply To Queue parameter of the adapter. As a test, We also change another parameter via the mapping (the correlation ID, to use the idoc number) and according to the log of the processed message, every thing is finde. According to the trace of the JMS adapter, everything is fine to. However, this has no impact on the message, only the correlation ID is changed.
The JMS server is IBM Websphere MQ (v6.0.2.1). It is already in production and working fine. This is a change we have been asked to implement.
Has anyone experience with the JMS receiver adapter-specific message attrbitutes ? What have we done wrong ?
Any help is greatly appreciated.
Regards,
Gregory PFISTER
SAP XI Analyst - Holcim

Hi Neha,
pls do chk the below links for SOAP adapter message attributes
http://help.sap.com/saphelp_nwpi71/helpdata/en/fc/5ad93f130f9215e10000000a155106/frameset.htm
http://help.sap.com/saphelp_nw04/helpdata/en/43/0a7d1be4e622f3e10000000a1553f7/content.htm
http://publib.boulder.ibm.com/infocenter/wbihelp/v6rxmx/index.jsp?topic=/com.ibm.wbia_adapters.doc/doc/sap_xi/sapximst38.htm
http://publib.boulder.ibm.com/infocenter/wbihelp/v6rxmx/index.jsp?topic=/com.ibm.wbia_adapters.doc/doc/webservices/webservices16.htm
Regards
Sampath

Unable to select jms destination in the jms adapter configuration

hi,
I am trying to use JMS queues for consuming messages but after creating the JMS queues(destination) in the application server, i am unable to find the destination while configuring the jms adapter in jdeveloper.
any help would be appreciated

i am using soa 10g. are you saying there is an option for starting the jms server in the application sever or do i have to create a seprate jms erver for creating the jms server etc.
it will be really helpful if you can explain the how to create/start jms server,jms queue etc or provide link to any article that is available. also please note that i am using SOA 10g
thanks in advance

Error in JMS server in OSB

Hi All,
I have been using JMS queues for few of my OSB projects.
From what I have noticed, the JMS server fails to queue messages often.
This is solved once the weblogic server is restarted.
I come across this problem atlest thrice a week. Everytime I restart the weblogic server to overcome this issue.
Is there any configuration setting within the JMS to avoid this?
Or is there any other way to overcome this?
Thanks,
Mani

weblogic.messaging.kernel.KernelException: Error persisting message.
But sometimes i dont get any errors. The transaction seems successul.
But the message has not been queued.
Thanks,
Mani

Ant to delete JMS Server

          I am trying to use an Ant script to configure a new domain for weblogic server
          8.1
          Ideally I want to run the script and have it configure all resources I will need
          (jdbc, jms, etc.). Currently I am having issues with the configuring of the JMS
          Server. I am using the following line of code to delete the JMS Server instance
          if it currently exist.
          <delete mbean="${domain.name}:Name=cgJMSServer,Type=JMSServer"/>
          This seems to be ignored by Ant because when I call the create task I get an InstanceAlreadyExist
          error. I assume my delete task is failing but giving no error. I'd like to be
          able to delete the current JMS server if one exist and then recreate it. I was
          wondering what the correct task or syntax would be for deleting the JMS server
          through Ant.
          Additionally, I'd like to use wlserver to create a new Integration Domain. I havent
          seen any documentation on how to specify a template to use when using the wlserver
          task, so I am not sure if this is possible or not.


Hi Allen,
          I think you may need to set the target of the JMS server
          to null first.
          I think the "JMSML" utility on dev2dev does this sort of
          task. You might want to take a look at its source to
          divine what steps it goes through.
          Tom
          Allan wrote:
          > I am trying to use an Ant script to configure a new domain for weblogic server
          > 8.1
          >
          > Ideally I want to run the script and have it configure all resources I will need
          > (jdbc, jms, etc.). Currently I am having issues with the configuring of the JMS
          > Server. I am using the following line of code to delete the JMS Server instance
          > if it currently exist.
          >
          > <delete mbean="${domain.name}:Name=cgJMSServer,Type=JMSServer"/>
          >
          > This seems to be ignored by Ant because when I call the create task I get an InstanceAlreadyExist
          > error. I assume my delete task is failing but giving no error. I'd like to be
          > able to delete the current JMS server if one exist and then recreate it. I was
          > wondering what the correct task or syntax would be for deleting the JMS server
          > through Ant.
          >
          > Additionally, I'd like to use wlserver to create a new Integration Domain. I havent
          > seen any documentation on how to specify a template to use when using the wlserver
          > task, so I am not sure if this is possible or not.

Union All with Linked Servers - Works until loaded on to the report server then fails.

Hi,
On our production server I have 2 linked servers. One that leads to ServiceNow via ODBC and one that leads to an HP Openview database.
Testing these linked servers works fine. I have a query that obtains info from each source and 'union all' together. When i run this query in SQL Server management studio it works fine and I get info from both data sources union-ed together perfectly.
I transfer this into Visual Studio and create a report, which again runs perfectly.
I upload this report to the report server and try to run it and get the error:
An error has occurred during report processing. (rsProcessingAborted)
Query execution failed for dataset 'Dataset1'. (rsErrorExecutingCommand)
For more information about this error navigate to the report server on the local server machine, or enable remote errors
When I rummage through the log files for the report server, I find very little helpful errors, basically this:
ERROR: Throwing Microsoft.ReportingServices.ReportProcessing.ReportProcessingException: , Microsoft.ReportingServices.ReportProcessing.ReportProcessingException: Query execution failed for dataset 'Dataset1'. ---> System.Data.SqlClient.SqlException: Cannot
execute the query.
I did a test where I created a copy of the report and ran only he Servicenow section of the report and it works, then ran only the HP section of the report and it works. It seems the UNION ALL is the problem somehow.
Anyone have any ideas??
Thanks
Kirsty

Hi Kirsty,
As you posted, this issue is caused by the security configuration of Linked Server.
Generally, in a domain environment, we can specify a domain account as the stored credentials for the report, and then configure the Linked Server to "Be made using the login’s current security context".
However,if we specify a SQL Server login as the stored credentials for the report, please set the Linked Server security to "By using this security context", and then providing the necessary credentials to authenticate at the linked server.
Please also add the Reporting Services Security role to the Linked Server Remote Server Login Mappings.
For more information about Creating Linked Servers, please refer to
http://msdn.microsoft.com/en-us/library/ff772782.aspx
About Security for Linked Servers, please refer to
http://msdn.microsoft.com/en-us/library/ms175537.aspx
Regards,
Swallow

Role of the JMS server

Similar Messages

Maybe you are looking for