Remote JMS Queue in Non-Clustered Environment

          We have two hosts, both running WLS 8.1, that are not clustered. We would like
          to create a queue on the first and be able to read it from the second. Is this
          possible? How does one go about setting this up?

          Paul wrote:
          > We have two hosts, both running WLS 8.1, that are not clustered. We would like
          > to create a queue on the first and be able to read it from the second. Is this
          > possible? How does one go about setting this up?
          Hi Paul,
          This a common use case. Even user transactions
          can include operations from both servers.
          The remote server just looks a regular
          client as far as the JMS server is concerned.
          I recommend using MDBs - which can be be configured with a URL
          and connection factory, messaging bridges, and/or foreign destination
          wrappers. In addition, you can use the JMS resource reference
          pooling feature to pool the "foreign" JMS server's resources.
          For information you can start here:
          http://edocs.bea.com/wls/docs81/jms/intro.html#jms_features
          and here:
          http://dev2dev.bea.com/technologies/jms/index.jsp
          Important: As documented, for various reasons, some not
          having to do with JMS, you will need to ensure the following are
          true to enable "non-clustered" communication:
          - all domains should be named uniquely
          - all wl servers must be named uniquely
          - all jms servers must be named uniquely
          - all jms stores must be named uniquely
          Tom

Similar Messages

How to config JMS queues for non-AQueue products? Only config files?

Assume I have the requirement NOT to use Oracles built-in Aqueue JMS provider but use another (external)product like:
Apache ActiveMQ, JBossMQ, IBM MQseries,... instead.
Can I config these JMS products through OC4Js web browser config web pages as well or am I forced in such cases to use configuration through XML config files?
How do I tell OC4J NOT to use internal JMS Aqueue queues but use other JMS product?
Peter

"lilian zhu" <[email protected]> wrote:
>
Hi,
i am having a admin server running on port 7001 for servlet1 and managed
server
running on port 8001 for servlet2. how can config the httpd.conf to make
the apache
web server recongize both uri path?
i tried <IfModule mod_weblogic.c> WebLogicHost localhost WebLogicPort
7001</IfModule>
and <IfModule mod_weblogic.c> WebLogicHost localhost WebLogicPort 8001</IfModule>,
only can recongize port 8001.
i also tried WebLogicCluster localhost:7001,localhost:8001, doesn't work.
i would like to know, for my case, will the clustered server work, how
can i make
this working for non-clustered servers?
Thanks.
Lilian
Hi,
we have the same problem. We are running two non-clustered weblogic servers and
we want to configure only a failover
with the plugin.
We contacte the bea support but we haven't find a solution.
Greetings
Hans.

Process not writing to JMS queue with non-xa data source

I have a process reading from a JMS-AQ with non XA, but it does not seem to be performing a commit and writing to the queue. If I use XA all the way ut works, but I don't get the desired error handling which I have built. Anything obvious I am missing? 11.1.1.5 SOA

Hi,
Under Weblogic JDBC Datasources select the datasource and then select Transaction tab there you can find Use XA datasource interface.
and In DB adapter ... go to Configuration Tab ---Outbound Connection Pool--- Unhide the pool group -Select JNDI name ( e.g eis/DB/local ) ---enter the JDBC datasource created earlier in weblogic JDBC datasource against xADataSourceName..
Now Click on Transaction tab & select Transaction Support: as XA Transaction .
Regards,
Abhinav Gupta

OSM v7.2 - More that 1 instance in a non clustered environment

I am just wondering...
1) Can i install OSM separately on 2 servers while using the same database instance in a non-clustered manner?
This is for circumstances where I need a dedicated server resource for bulk provisioning for example.
2) If I am not using AIA in my implementation, how does COM calls a SOM cartridge and how does SOM reply COM on the completed status?
Is it via some default provided webservice, or do i need to find a way to implement it myself.
Thanks :)

Hi,
1. Though technically it is feasible to use two OSM instances with same database, it is not suggested as it may lead to locking issues and not sure how you are planning to acheive bulk provisioning by this way
2. The calling of SOM cartridge by COM and status updates of SOM to COM has to be implemented by yourself if you are not using AIA. You can make use of default webservices proivded by OSM
Regards,
JP

TimerManager in non-clusterer environment

Hi all,
Is it possible to lookup TimerManager object in a POJO? Since it's not EJB I don't know the JNDI name of TimerManager, and I couldn't find it in JNDI tree.
@Resource injection will not work in this case.
Any help would be greatly appreciated.

For role-based vlan mapping, the vlan is switched to the final vlan and then agent requests an ip refresh since if you had the access vlan IP address, and you're now in a new vlan, network access wont be there.
I don't think we have any white papers discussing setting up NAC with static IPs, but I'll look.
HTH,
Faisal

HowTo:Publish Msg to JMS queue deployed in a Weblogic Cluster from a client

Hello All-
I am new to Weblogic and need help from this forum. Forgive me for my ignorance. I am learning.
Below is a brief about the context and the issues that we are facing:
Background
=======
1. We have Oracle Service Bus (OSB) installed on top of Oracle Weblogic Server (ver 10.3). In OSB we have developed a proxy service that listens to a JMS queue for an incoming message
2. The Weblogic environment is a cluster having 2 managed servers listening to different ports and it is front-ended by a Apache Load Balancer.
3. Our requirement is to post an XML message from a Java client to the JMS queue which will then initiate the proxy service deployed. We are using t3 protocol to post the message from the client to the JMS queue
Issue
====
1. To post an XML message from the client, how do we give the address of the JMS queue in the clustered environment to enable failover/loadbalancing if one managed server goes down. I understand from your earlier posts in this forum that directly hitting the JMS queue on a particular managed server will not enable failover/loadbalancing in case one of them goes down
2. Do we need to modify any config file in the Apache Proxy Plugin?
3. Do we need to provide the address of the Apache proxy in the client to send the message?
Kindly clarify and let me know me know if you need any additional information.
Greatly appreciate your help!!!
Regards,
Dibya

Hi Dibya,
Please refer to : http://weblogic-wonders.com/weblogic/2010/05/13/apache-proxy-as-static-content-repository/#comment-1379
Thanks
Jay SenSharma

Load balancing, failover and fallback in Non-Clustered WebLogic environment

hi,
Has anyone implemented WebLogic 10.3.3 (or 10.3.4) in a Non-Clustered environment, but also got load balancing, failover and fallback work?
We were successful in getting failover working using t3://server1:7001,server2:7002 provider URL, but not load balancing or fallback.
The fallback is working when it was connecting to server2 and if we kill server2, then it switches to server1, but not when server2 is still running while server1 comes back.
All we need to find a way to enforce fallback to primary site, even if secondary which the client connected is still up and running and primary site comes back.
Any help appreciated.
Thanks.
Best regards,
Bala

hi,
Has anyone implemented WebLogic 10.3.3 (or 10.3.4) in a Non-Clustered environment, but also got load balancing, failover and fallback work?
We were successful in getting failover working using t3://server1:7001,server2:7002 provider URL, but not load balancing or fallback.
The fallback is working when it was connecting to server2 and if we kill server2, then it switches to server1, but not when server2 is still running while server1 comes back.
All we need to find a way to enforce fallback to primary site, even if secondary which the client connected is still up and running and primary site comes back.
Any help appreciated.
Thanks.
Best regards,
Bala

How to create a JMS queue linked to external (remote) JMS server?

Hello,
I'm trying to receive messages from a remote JMS queue. I found some information on the Internet on how to create an external JNDI resource (my remote JMS queue connection factory is bound to LDAP context on remote box). I was able to configure JNDI resource that points to remote queue connection factory, but I don't know what to do next.
How can I create a local JMS destination (queue) that will be the image of the remote queue? At least I'm guessing this is what needs to be done in order to deploy MDBs.
I would appreciate it if someone could tell me how to get this working. Maybe the steps I took were not in the right direction - the bottom line is that JMS queue and queue connection factory are deployed on a remote server, and I need to get messages sent to them on the local server. I'm using Sun App Server from J2EE 1.4.
Thank you,
Mark

Please refer to http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6377822, particularly the evaluation. Could your situation be the same too (i.e server ip address changing dynamically)? Also, which OS are you using and do you have any firewalls setup that prevents access to port 3700?

Oracle AQ to JMS Queue which is on Oracle Application server

Hi,
I am using Oracle Streams AQ Oracle database 10g
and i have a JMS Queue on oracle Application server.
how i can Propagate message from Oracle AQ (AQ$_JMS_TEXT_MESSAGE)Queue to JMS Queue. ?
Thanks,
Satnani

Hi Justin,
Thanks for Reply, but in my case my JMS queue is on Oracle Application Server.
and i want to propagate message from AQ to JMS queue.
and none of the document is available to describe how to send message from oracle database AQ to JMS queue which is on application server.
please help
Thanks,
Satnani

Reading Message from Foreign JMS Queue in SOA Suite 11g.

Hi,
I am trying to read a JMS message that has been put into a JMS Queue on a remote weblogic Server.
The steps that I know is that I need to create a foreign JMS Server and then probably use JMS Adapter to subscribe to the message that is lying in the remote queue.
Has anybody been able to achieve this kind of scenario working for a remote JMS queue? Also, is this supported by JMS Adapter in SOA11g?
Any help will be appreciated.
Regards,
Varun Maheshwari

Hi Atheek1,
Thanks for your reply.
The confusion that is pondering me is that do I need to create a Foreign JMS Server or should I create a JMS adapter and override the properties.
I have tried the steps for creating a Foreign Server from the below link
Unable to access remote JMS Queue through JMS Adapter
But I am not able to find the Foreign Server created and any connection factories and Destinations related to foreign Server in my JMS Destination while configuring JMS Adapter.
I am not sure how to do step2 as entailed by you.
Could you please explain me in little detail on how to do this or redirect me to some blog or forum which will be really helpful.
Cheers,
Varun

OCFS2 in a non-RAC environment

Here's a question I've been batting around with a sysadmin. I thought I'd run it past this group.
We're spec'ing out the hardware for a very large Oracle database (3+ TB) running on Linux. This will be a single, non-RAC instance. This particular SA has very bad feelings about the ext3 filesystem (I am not exactly sure what his objections are). He has heard about ocfs2 and wonders if it would be superior to ext3.
What do you think? Would ocfs2 be appropriate in a non-clustered environment? Can you suggest any arguments for or against this approach?
Thanks, I appreciate any thoughts anyone might have.

Note that paper was written in 2004, and their tests used Pentium III processors. The OS was RHAS 2.1 and their kernel was 2.4.9
I'm not saying that invalidates the paper's conclusions, but I'm certainly saying you'd have to be very careful indeed in assuming that conclusions drawn from that combination of hardware and software had anything much to say about today's quad cores, multi-gigabyte RAM, kernel 2.6.x setups.
I'm concerned when I see their graphs on page 8: no axis scales in either x or y directions. If they're going to come to the conclusion that ext2/3 performance tails off, it would be nice to know whether it does so when the number of users hits 10, 50, 100, 500 or whatever, wouldn't it? But you can't tell anything from their graphs on that score.
And I don't see any mention of them adjusting the buffer cache when using raw, which (having just abolished the file system cache) is something you should sensibly do when making such comparisons.
And, of course, they were testing OCFS not OCFS2, so again... the applicability of anything concluded in that paper to today's situation is highly questionable, at least in my mind. But I do note that OCFS with "high numbers" of users ended up clobbering the CPU to death, whereas ext2/3 didn't...
And they seem to claim that transactions per second levelled out with ext2/3 because they hit a lot of free buffer waits -and there are ways of dealing with them that don't involve getting rid of the file system! The fact that they didn't report a lot of write complete waits as well as free buffer waits might indicate that they just need more DBWRs, and not a switch of file system, for example.
Anyway... short answer is, I wouldn't read too much into that report, which looks to me to be poorly devised and written and very much 'of its time and place' with little general applicability outside of that.

Synchronization of JMS queue with backup instance

Hello,
          I have the following problem: there are two machines in active - passive
          configuration. Both those machines serve as JMS servers in two environments.
          What I want to do is to keep a backup of JMS queue from the productive
          environment in the backup one, so in case of a disaster in the productive
          environment the backup one will stand up and will start processing the waiting
          messages. In order to do this I need to keep an up-to-date (or almous up to
          date, I will propably have to accept that some of the messages will be lost)
          copy of JMS queue.
          Have you ever worked on such a problem? Maybe you can suggest some solution?
          Best regards,
          Dawid Duda

Hi,
          Thanks for informations, but it is not exactly what I need.
          In my case I need to keep a backup of the queue in a separate location, to be
          able to start processing the data in case if the first one was i.e. burned.
          Therefore I was thinking about something like duplicating the database store
          under the queue (using database tools) in on-line mode to be able to start
          another instance in case of problems with the first one. Do you think such
          aproach would work?
          Best regards,
          dave
          Tom Barnes wrote:
          > Hi,
          >
          > Search the newsgroup and docs for keyword "migration". Automated "whole server" migration is available in 9.0, and automated "service" level migration is slated to be available in 9.5. (8.1 provides manual/scriptable service-level migration.) Also, depending on your use case, it might be that distributed destinations, bridges, server-side-store-and-forward (9.0+), or client side store-and-forward (9.2+) would fit your needs.
          >
          > http://edocs.bea.com/wls/docs81/jms/intro.html#jms_features
          >
          > http://edocs.bea.com/wls/docs92/jms_admin/intro.html#jms_features
          >
          > http://dev2dev.bea.com/pub/a/2004/05/ClusteredJMS.html (service-level migration paper)
          >
          > Tom

2 non-clustered weblogic servers, 1 MQ queue. Messaging bridge fairness

Hey all,
          More of a general question. Has anyone ever had 2 non-clustered
          servers, each with a messaging bridge, pulling from one queue? If so, how is
          the fairness of the machines pulling.
          Assuming same machines / same configurations for the weblogic servers,
          should it be
          approximately fair?
          Just want to verify that if the pulling machine sees more messages after
          pulling it's batch,
          it doesn't keep the channel locked or antyhing, before we pursue this
          option.
          Thanks much,
          Greg


Hi Greg,
          The consumer fairness is up to the JMS provider, which sees the bridge
          as a regular consumer no different from any other client consumer.
          If the queue being pulled from is a WebLogic 6.0 or greater queue then
          fairness will be observed. (I'm not sure about earlier versions.) Of
          course, this "fairness" is proportional to the rates at which the
          consumers consume.
          There is no such guarantee for other providers. And in fact, we (BEA)
          have noticed that some of the other well-known JMS vendors do
          not exhibit fairness.
          Tom, BEA
          Greg Kaestle wrote:
          > Hey all,
          >
          > More of a general question. Has anyone ever had 2 non-clustered
          > servers, each with a messaging bridge, pulling from one queue? If so, how is
          > the fairness of the machines pulling.
          > Assuming same machines / same configurations for the weblogic servers,
          > should it be
          > approximately fair?
          >
          > Just want to verify that if the pulling machine sees more messages after
          > pulling it's batch,
          > it doesn't keep the channel locked or antyhing, before we pursue this
          > option.
          >
          > Thanks much,
          > Greg
          >
          >

Data retrieval from JMS queue using single SOA composite in a clustered env

Hi,
I have following situation:
1) A requester SOA composite (Composite1)is reading data from a File
2) Mediator is routing data received from Composite1 and writing in a JMS queue.
3) A provider composite (Composite2) is reading from the JMS queue.
Both Composite1 and Composite2 is deployed in a clustered environment.
Problem is Composite2 is not able to read data sequentially from JMS Queue sometime (for example out of 30000 transactions its failing once). Here it's a queue and only one producer and only one consumer is there.
What may cause this issue? I thought a single producer and single consumer on a queue will guarantee FIFO retrieval . Could you please advise where may be the gap?
Thanks
Edited by: user1054373 on Sep 18, 2012 11:29 PM

Hi,
I thought a single producer and single consumer on a queue will guarantee FIFO retrieval Java Message Service Specification does not guarantee ordered message delivery when applications use distributed queues...
Using Weblogic Message Unit-of-Order may solve your issue...
http://docs.oracle.com/cd/E23943_01/web.1111/e13727/uoo.htm#JMSPG389
Hope this helps...
Cheers,
Vlad

JMS in Clustered Environment

Hi All,
We have a clustered environment ; instance A and B of the same application server is running. Both the servers are configured for a JMS queue. But when I send message to the queue to refreh certain data, only one server is getting refreshed.
What can I do for this? Please suggest?
Regards
Sajitha

Hi,
You should tell us which product you are using. However, I think that you should use a topic rather than a queue because a message that is sent to a clustered queue will be delivered to only one MDB instance.
Hope it helps
Arnaud
www.arjuna.com

Remote JMS Queue in Non-Clustered Environment

Similar Messages

Maybe you are looking for