JMS Sender Channel not Polling

Hi Guys,
Scenario: BEA Weblogic -
> JMS Queue -
> Sender Channel JMS -
> SAP XI -
> SAP R3
We had initially setup this scenario using MessageSelector because the provider JMS Queue had mutiple message types on it. As Murphy's law would have it, the MessageSelector stopped polling the day before go-live. Basically, if we pointed one JMS SC to the provider JMS Queue using MessageSelector it worked fine, if we pointed multiple JMS SC's no messages were being picked up from the queue. We still haven't identified why the MessageSelector functionality stopped working but as a quick work around to MessageSelector before go-live we used blog:
[/people/robert.warde3/blog/2009/03/18/pi-basics-generic-message-interface|/people/robert.warde3/blog/2009/03/18/pi-basics-generic-message-interface]
Now we have a situation where, BEA Weblogic manages the provider JMS Queue & they perform a health check 4 times a day & if they encounter a problem they bounce the server & JMS Queue service. XI and the SC seem to re-connect fine most of the time when Weblogic & JMS Queue is back up but there are times when I see a green light in RWB for the SC but it's not polling for messages on the JMS Queue. We use Hermes JMS to monitor messages on the Queue to check if there are in fact messages to be picked up & the messages are queued.
What I'm looking for is a best practice guide for how to manage this. I don't want to keep toggling between Hermes JMS & RWB to verify if the SC is in fact polling messages altho' it shows as green in RWB. Please advise. Thanks.
Regards, Trevor

Thanks Kai & Joel.
Joel, I'm not really looking to add another layer with the JMS Connector Service. For me it's just adds another possible point of failure. At the moment the message passes thru 5 systems including SAP R3 & end system & then the response comes back the same route.
Going out from SAP, the Receiver channel to the JMS queue seems to work fine. With the responses coming back, the Sender Channel seems to re-connect ok (I'm assuming) otherwise it wouldn't be green in RWB but doesn't pick up any messages. It's also an intermittant phenomenon. Manually stopping & re-starting the Sender Comm Channel fixes the problem but we won't know if there's a problem until we check if messages are being picked up from the queue.
Kai's notes led me to some things I could possibly look at like additional parameter in JMS channel config, CONNECTION_TEST_POLL (note 948016). I'm still not sure how we can make this work for us because we essentially need to bounce the JMS channel so it starts polling again in the event of a JMS Queue Provider bounce but I need to first establish that the JMS Queue was bounced.
Note 1088675 will also be irrelevant cos the channel looks like it re-connects, just doesn't poll.
Trevor

Similar Messages

  • Doubt in JMS Sender Channel

    Dear Experts,
      I have some queries and an issue also..
    First Query : is the JMS sender channel is polling the message from JMS Queue or when ever the channel is connected to the JMS queue then the messages are pushed from JMS ? Simply I want to understand the then message is Puss or Pull. (my understanding is messages are pulled when ever the channel opens the session with JMS provider)
    Second Query : is there any provision that the sender channel will have the log When the JMS Queue is empty ??
    Issue : we are using WebSphereMQ JMS Provider, with QoS is EO . MQ claims that the messages are missing when the message were sent to PI.
    is there any way to trace the issue ? any clue ?
    -Danny

    Thanks for your help on this.. but please excuse me for asking more layman query..
    There no other JMS Client is connected with that Queue. and I am sure the messages are
    So you say, only some message are missing, but the most message run correctly?   Yes Most of the messages run correctly
    Are you sure, that the messages were sent to the queue correctly?  Since only the PI channel is connected the message should be forwarded to PI correctly.
    Do you have a log for the MS queue?  I think there should be mentioned, which KMS client has received the message. --> I don't have the log of MQ side.. But I can request them..
    can you please provide the info "what I can ask them the MQ Team?"  if I ask them about the "the JMS client details of the message" will it give the information of where that message has been sent.
    Thanks for your help
    Edited by: Dhanabal Thangavel on Sep 20, 2011 1:21 PM

  • Need help about setting up a JMS sender channel

    I just started to play with PI for a few weeks and I'm trying to setup a simple JMS sender channel to read stuff from a WebSphere MQ.
    I've someone installed JMS adaptor for me, but we never know it is correctly installed or not. First question is "Is there any way we can verify JMS adapter is correctly installed or not?"
    In the Design: integration builder, I've a very defined a very simple data type(src_dt) which is a string and a message type(src_mt) which uses resp_dt. Finally I created a message interface (test_mia_out) that contains src_mt. At this point of time, I don't need message mapping so I didn't configurate this section.
    Then I went to Configuration: Integration builder and created a new communcation channel with the following setting:
    Parameter section:
    Adapter Type: JMS
    Transport Protocol: WebSphereMQ (Non-JMS)  <- if i don't use this option, i can't enter MQ connection detail in the Source section.
    Message Protocol: JMS 1.x
    Adapter Engine: Integration Server
    Source section:
    QueueConnectionFactory Java Class: com.ibm.mq.jms.MQQueueConnectionFactory
    Queue Java Class: com.ibm.mq.jms.MQQueue
    ...MQ connection detail...(IP, Port, Queue Manager, Channel and Queue Name )
    Transport Protocol: TCP/IP  <- this is consists to server side setting
    JMS-Compliant: JMS-Compliant
    Since I only want a very simple working sample, again I didn't change the content in receiver determination, interface determaintion and sender agreement.
    Now when I active this change and I saw the following error in the channel monitor:
    Short Log - Channel has internal error
    Admin Info - Exception: Channel with Status Unknown or Not Registered
    I did stop and start the service, refresh CPA cache and even restart the entire PI server, I always see this message and cannot find any more detail description about what is causing this problem.
    My PI Server version is 7.0 and MQ version is 5.3
    There's very limited resource about how to setup to read MQ message in PI and I'm stuck now. Therefore, I need a help from some experienced people to tell me if I have did something wrong or missed some steps?
    And if possible, could people mind to post a step by step go through how you setup it up?
    Edited by: Robert Fu on Jan 10, 2008 6:51 AM

    Before you use the JMS, you need to deploy some drivers related to specific MQseries, Kindly ensure that proper driver are deployed....
    If you use MQSeries 53x.xxx, you must enter the following JAR files in aii_af_jmsproviderlib.sda.
    com.ibm.mq.jar, com.ibm.mqbind.jar, com.ibm.mqjms.jar, connector.jar (use the JAR file from the J2EE client directory)
    check this link it will give you more inofmration
    http://help.sap.com/saphelp_nw04/helpdata/en/cd/d85a9d6fab7d4dbb7ae421f710626c/content.htm
    check this to get the condifuration the JMS
    How to use conversion modules in JMS - https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/f02d12a7-0201-0010-5780-8bfc7d12f891
    Ref How To Use the Content Conversion Module in JMS Adapter - https://websmp106.sap-ag.de/~form/sapnet?_SHORTKEY=01100035870000582377&

  • EOIO JMS Sender Channel in Error

    Hi All,
    We have a error on JMS sender channels in Pre-Production environment.
    Let me explain you the problem.
    On Dev and QA there is no cluster environment. On Pre-production we have 3 clusters. We tested this Sender JMS channel on Dev and QA without giving a Cluster ID and it worked.
    Now on Pre prod, we had not mentioned the cluster ID and when messages were picked from the Queue, the channel went into a error mode stating the " Channel is locked due to EOIO".
    Then when we added the cluster ID to the channel, in RWB the channel goes into inactive mode. The status is " The EOIO channel has been made inactive on Node 206485150"
    The node mentioned in channel is 251902150. This means that the other node has been made inactive. That sounds logical as it is not supposed to process the message.
    But the node which is mentioned in the channel has also gone into error mode.
    " Error creating JMS connection. failed to create MQManager for ' Hostname:Channel name'". An MQException occured: Completion code 2, Reason 2009
    MQJE016: MQ queue managerclosed channel immediatly during connect. Closure reason= 2009, and erro code as MQJMS2005
    After this all other sender JMS channels have gone into the same error mode on this node.
    We have stopped the JMS service and started it. We also restarted the XI server but no change was encountered.
    Please let us know whether JMS sender with EOIO works with cluster ID as I'm not able to find any document which says that JMS with EOIO works succesfully.
    Please also let us know how to overcome the error on this node.
    Thank you.
    Regards,
    Sarvesh Desai

    Check if there is any unprocessed message stuck in the queue:
    /people/stefan.grube/blog/2006/04/27/how-to-deal-with-stuck-eoio-messages-in-the-xi-30-adapter-framework

  • JDBC and FTP sender channels stop polling

    Hi,
    We have a great volume of sender JDBC and FTP sender channels.
    For this reason there is nothing unusual to have temporary error connections.
    Several months ago, with an XI 3.0 system, sender channels recover its functionality  automatically when temporary error connection is solved.
    But now, he have migrate to a  PI 7.1 sytem, and in the same scenario and configuration,  sender channel stop polling after a temporary error . In cc monitor this channels remains in  green status and there were no signs of stopped interfaces.
    Timeouts are correctly defined and there is no difference between channel configuration in our Xi and PI system.
    How to solve this problem?
    Thanks, in advance.
    Carme.

    Thanks for your fast response, Stefan.
    After writing this post, i ask the same question to our basis team and  we are in the latest sp.
    ¿do you know if there is any note or can you provide me some additional information?
    I'm not able to find any note about this issue...
    Regards,
    Carme.

  • JMS sender channel binary format possible?

    Hi,
    We're still having codepage problems from MQ to SAP via JMS.
    We send text messages, the QM format name is MQSTR and the message type is datagram.
    At the moment SAP development is saying:
    If the XML is encoded in anything other than UTF-8, then the
    javax.jms.Message object must be be a BinaryMessage and not a
    TextMessage. All TextMessages are interpreted to be UTF-8
    My remarks on this:
    -If we send UTF-8 the data is still corrupted in the end sap system so still the adapter seems to working incorrectly
    -I have the feeling that in the JMS sender channel the CCSID parameter has no influence whatsoever on the application data translation
    But I want to try anything, so how would I configure a JMS (MQ) sender channel for binarymessage. It seems that the JMSMessageType parameter is only valid for receiver channels. Ok, I can image that the sending side (MQ) should set the message type but then still I think the module configuration/sequence should be different than the standard one.
    Anyone a clue what this should be?
    thanks
    Tom

    Problem solved...
    We finally managed to send a binary file to XI.
    We had to do two changes to accomplish this, first make sure when we read the file from its source directory that no automatic translation is done (remove O_TEXTDATA switch from C api on iseries).
    Secondly in MQ change the message format from MQSTR meaning text message to Blanks.
    The messages in MQ are not human readable anymore, but when looking at the hex codes it is proper utf-8.
    Now the special characters are correct in the XI monitors and also in the receiving SAP systems.
    I suspect the cause of the problem to be in the iseries only partial support for the utf-8 encoding.
    a nice way to end the week..

  • JDBC Sender channel not processing the messages

    Hello,
        From yesterday onwards JDBC Sender channel not processing the messages. In CC monitoring it is showing in Green Led and status is functioning. Cahnnel is polling for messages to DB server as per polling interval. But it is not processing the msg's. Under Processing detials for cluster node it is showing only "Processing Started" for each polling interval.
    I have done the check in DB server by executing the query in the channel, whether there are any records are exisitng or not. There are 1000+ records.
    Previously the parameter "Disconnect from DB after processing each message" was not set. I have set the parameter and activated in Productiion. Still the messages are not processing by channel
    What was the problem? How to rectify with this.
    Cheers
    Veera

    Check in the Visual Admin >Cluster> "LOCKING ADAPTER"
    we get a option of Display Locks .Check for an entry with
    NAME : $XIDBAD.JDBC2XI
    reset the locks and restart the CC ,now it works fine.
    Note 1083488 - XI FTP/JDBC sender channel stop polling indefinitely(04/04S)

  • Jms message id in jms sender channel

    Hi
    Is it working for you remove id from jmsmessageid in a jms sender channel?
    I can see still in the logs that my message id is starting with ID:
    Florin

    Hi Florin !
    You mean this:
    "If you have selected JMSMessageID, JMSCorrelationID, or JMSProperty and entered a JMS message ID in these fields, you should see the Remove 'ID:' Marker from JMSMessageIDindicator. Set the indicator.
    This is necessary because the JMS specification stipulates that each JMS message ID starts with 'ID'. However, this conflicts with ISO-11578."
    Maybe you need to reactivate your sender comm channel to force the update of cache and try again. To reactivate, just edit your adapter, make any dummy changes (e.g. change description) and reactivate again.
    Then try scenario once more and check if it is working.
    Regards,
    Matias.

  • JMS sender channel error

    hi,
    i am using pi 731 single stack.
    i have a jms sender channel. The jms sender channel is connecting to Webshere MQ version 6.0.2.10.
    the error is -
    Error creating JMS connection : The JMS provider gave the error message as MQJMS2005:failed to create
    MQQueueManager for x.y.com:QMgr1 MQJE001 completion code 2,Reason 2059,and the error code as MQJMS2005
    What could be the issue. The JMS drivers are successfully installed.

    Hi,
    As Indrajit said seems to be a JMS configuration problem. Check this thread in stackoverflow with the same error: java - Always receive MQJE001: Completion Code &amp;#39;2&amp;#39;, Reason &amp;#39;2059&amp;#39;. ERROR - Stack Overflo…
    Regards.

  • File Sender channel stopped polling on FTP site

    Dear all,
    We have many file to Idoc interfaces running in production and all are they good.
    Yesterday we faced one issue, one of the interface's sender communication channel stopped polling and didn't picked the files from FTP.
    There was no error on communication channel monitoring and it is green.
    We stopped & started the channel then started picking the files. But this is not usual behavior of file adapter.
    We found one SAP Note 1083488, according to this note, we are in the latest SP level( PI 7.0 SP 14). So not required to apply this patch my case.
    What should i do to not to face next time
    Regards

    Hi Bhavana,
    Check the below Q & A.....
    >> Q: After configuring a large amount of File Adapter sender channels, the J2EE Engine becomes very slow and some services start to block. How can I solve this issue?
    A: Up to and including XI 3.0 SP13 each File Adapter sender channel permanently consumes a J2EE application thread. To solve this issue, increase the number of configured J2EE application threads using the SAP J2EE Engine Config Tool ("cluster-data" -> "Global server configuration" -> "managers" -> "ApplicationThreadManager" -> "MaxThreadCount").
    Starting with XI 3.0 SP14 / PI 7.0 application threads are allocated on demand by the File Adapter and returned to the thread pool after it has finished the polling sequence, so thread shortage situations will typically occur much more rarely than with earlier SPs.
    Also as said by Other SDN'ers check the locks.
    Thanks,

  • SP15, JMS Sender, "Messages not fetched"

    JMS Sender Communication Channel, Transport Protocol "Access JMS Provider with JNDI" and Name of JNDI initial ContextFactory "com.sonicsw.jndi.mfcontext.MFContextFactory". 
    Error in Adapter Monitor "Sender channel. Details: Last Exception during send at Sat Mar 11 07:55:45 EET 2006: Value with name enableDynConfigSender not available."
    There is messages in SonicQ Queue but they are not fetched by XI. I've start/stop SAP XI Adapter: JMS, no help.
    What does that error actually mean and what should I check next?

    Hi Tapio,
    If you are on SP14, SAP has introduced 'dynamic headers' with this support Pack. Your problem may be related to this. See the related section in SAP Note#856346 (JMS Adapter - FAQ)
    To fix your problem try deactivating your sender communication channel, make a small change to your description and activate it again. Test your scenario to see if it works.
    Hope this helps.
    Regards,
    Sam Raju

  • Error in JMS Communication Channel not raising alert

    Processing Errors in JMS Communication Channel are not raising alerts.
    The settings on the communication channel retires are set at 3 and 5min interval.
    In RWB the JMS communication channel shows processing errors but an alert is not raised.
    The alert are configured in our system and we even have a catch all alert rule with * (wildcard). Errors in PI message processing are raising alerts but errors in Comm Channel are not raising alerts.
    The message in SXI_MONITOR shows as processed successfully but when we check the RWB and Comm. Channel Monitoring we see processing errors "Channel error occured; detailed error description: JMS error: Connection to the server has been terminated; Linked error: Connection reset, Error Code:null"
    Are we missing any configuration? How do we get alerts/notifications if there are errors in Comm. Channel without having to login into RWB and watch it periodically?
    Any help is greatly appreciated.
    RM

    Hi,
    What is the staus of Communication channel in RWB when it fails, RED ??if yes then it should genarate alert.
    create one more alert for specific to JMS interface and check it out.
    Regards,
    Raj

  • Sender Channel not working(RFCSenderR3)

    Hello All,
    In my XI 3.0 QA env...there is one service "RFCSenderR3" which is not showing any messages in message monitoring ,but if i check in Communication channel monitoring for that service it shows as :
    Sender channel 'RFCSenderR3' for party ",service'RQ1'
    ERROR can not instantiate RfcPoolException:error initializing
    RfcServerPool:com.sap.aii.af.rfc.core.repository.RfcRepositoryException:can not connect to destination system due to :com.sap.mw.jco.JCO$Exception:(103) RFC_ERROR_LOGON_FAILURE:Mandante 101
    Here RQ1 is the R3 reciever system.
    And i have checked rfc connection from R3 end which shows it is fine..
    Can any one suggest me ,exactly wat need to be checked to solve this problem .And I am not sure what to check from R3end also....
    can any one guide to me ..
    Best Regards
    raki

    Hello All,
    Thanks for helping yaar,the problem got resolved .
    After verifying the user Id & password , I restarted that particular channel from Communication channel monitoring.Then it got worked with successful execution.
    Thanks one and all for helping me out.
    Best Regards
    Rakesh

  • JMS Sender Channel - reconnect problem

    Hi,
    I've got a sender channel which picks up messages from MQSeries and sends them to XI.
    If I set the MQSeries Queue to "Get Inhibited" to simulate a network problem, the channel goes down with a standard error message
    "Sender channel. Details: MQJMS2002: failed to get message from MQ queue Code: MQJMS2002".
    A receiver channel that looses connection gives me the same error.
    So I've got 2 problems :
    - how can I get the MQSerier error message (2016, MQRC_GET_INHIBITED )
    - how come the adapter doesn't reconnect after correction of the problem ?
    Tx in advance
    Tom

    Hi Tom,
      I get the same error , i tryed looking at the IBM and eocnte at the following links, but not able to solve the problem yet.
    <a href="http://www-1.ibm.com/support/docview.wss?rs=180&uid=swg1PK02976">http://www-1.ibm.com/support/docview.wss?rs=180&uid=swg1PK02976</a>

  • JMS Sender channel for WebSphere fails with java.lang.ArrayIndexOutOfBounds

    The adapter monitor shows
    Sender channel. Details: Last Exception during send at Mon May 22 12:09:17 EDT 2006: ArrayIndexOutOfBoundsException
    If the message is 529 byte it works but the one we are testing is 4659 bytes.
    Any ideas on what we should be checking or setting.
    thank you.

    Can you try and get the detailed log for the error?
    Also if you can throw light on any adapter specific processing you are trying with this then may be it would be easy to trace out the root cause.

Maybe you are looking for