Queue Limits

Clovertech Forums Read Only Archives Cloverleaf Cloverleaf Queue Limits

  • Creator
    Topic
  • #49181
    Jared Miller
    Participant

      We had a cable cut accidently by a building demolisher and it has caused a downtime in one of our interface connections (to our state dept for ER visit research).  It is not known how long it will be until the connection will be restored, but it is believed to be an extended period.

      My question is, how many messages can be queued up as ‘Pending’ until the amount begins to effect the performance of other threads?  At what point should I begin to remove them from the recovery DB (and what’s the best way to do that)?

      Thanks in advance.

    Viewing 0 reply threads
    • Author
      Replies
      • #61002
        Russ Ross
        Participant

          Here is a URL that describes how I handle extended downtime because the cloverleaf process will eventually crash when a threshold of too many messages are queued up as a result of a foriegn system being down for too long.

          <a href="https://usspvlclovertch2.infor.com/viewtopic.php?t=1621&highlight=&#8221; class=”bbcode_url”>https://usspvlclovertch2.infor.com/viewtopic.php?t=1621&highlight=

          I’ve had an interface in down time mode in one extreme case for over 3 weeks using the method described in the URL successfully.

          How many messages you can que up before the process will crash is dependant on your servers resources (ex: real memory) and how well you have utilized multiple sites and divided up your processes.

          In other words, I do not have a definitive answer for you but I can share my thresholds with you.

          On our previous smaller server that had less real memory with fewer sites and I had less understanding of how to effectively divide our sites and processes, the process would crash when around 8,000 messages backed up.

          Generally the process could not be restarted until the messages had been removed from the recovery database.

          Our current server that is more powerfull and has more granualar sites and processes has gone over 120,000 messages queued up without any issue, which has been very helpful because that gives us a whole day before we have to worry about going into down time mode instead of an hour like before.

          Russ Ross
          RussRoss318@gmail.com

      Viewing 0 reply threads
      • The forum ‘Cloverleaf’ is closed to new topics and replies.