Skip to Content
author's profile photo
Former Member

Reprocessing Messages stuck in AE/IE

Hi ,

If a message gets stuck in the Adapter Engine due to an error, say, due to, the target application is down. After the target application is up , when the next message comes in , i want the current message plus the previous message(s) also to be processed. rite now i dont see this happening , a manual restart from RWB is what i do . Now Is there a way to auto restart these messages stuck ?

Thanks in advance

Saravana

Add comment
10|10000 characters needed characters exceeded

  • Follow
  • Get RSS Feed

2 Answers

  • Aug 25, 2005 at 12:01 PM

    Hi Saravana:)

    have you tried scheduling report RSXMB_RESTART_MESSAGES

    Regards,

    michal

    Message was edited by: Michal Krawczyk

    Add comment
    10|10000 characters needed characters exceeded

    • Former Member

      Hello Michal,

      I have not tried scheduling this report, one clarification , is this report goin to restart messages stuck in adapter framework also or only in the Integration Engine ?

      I was thinking , it will help if somehow we can trigger autorestart based on the next message. Say, the target appln is down, messages will get stuck undelivered. Now i will troubleshoot the issue , bring the appln up again, then only send the next message(assumin i will stop furthur publishing of messages from source until then), now all the messages get picked up. Is this possible ?.

      Scheduling the report will mean it will run at a regular time intervals picking all errored out messages and push them again rite!!.

      cheers

      Saravana

  • author's profile photo
    Former Member
    Aug 25, 2005 at 12:04 PM

    First I would reactivate the specific adapter.

    Then try restarting that specific adapter in Visual Admin (Ex: Visual Admin --> Server --> Services --> SAP XI: File).

    Add comment
    10|10000 characters needed characters exceeded

    • Former Member Former Member

      Hi Roberto , i ve answered based on my understanding , hope it helps

      > 1) When this kind of situation occurs; are the

      > messages automatically stored/queued on XI?

      > 1a) Does this apply for both asynch/synchr

      > communication?

      If its EO/EOIO , they are definitely stored. For Synch scenarios , my understanding is that your source application will timeout without getting a response from middleware, now you chk this, if timedout, then send the request again

      >

      > 2) Is this an automatically procedure built-in on XI,

      > or do you need to configure it somewhere in the

      > integration engine?

      The QOS(Quality of Service) you choose will determine the behaviour in XI.

      > 3) When the receiver system is again available, are

      > the queued messages then automatically sent to the

      > receiver system or do you nedd to start this process

      > manually?

      No , you need to schedule RSXMB_RESTART_MESSAGES , refer note 813029 . For messages stuck in Adapter Engine, select the errored out messages in MDT , reprocess them(this was a suggestion for a note we raised to SAP)

      cheers

      Saravana