Clovertech

java process on the host server consuming resources

› Clovertech Forums › Read Only Archives › Cloverleaf › Cloverleaf › java process on the host server consuming resources

  • Creator
    Topic
  • August 4, 2010 at 5:28 pm #51914
    Highland Dave
    Participant

      Currently on version 5.7, REV 2.

      Initially, over a few weeks, processing and GUI response really slows down.  

      Rebooting the server is the initial fix. What is occurring more recently,

      this is becoming an every other day problem.

      Really effects any GUI work, i.e.: netconfig, xlate

      There was a post on June 22, that a user upgraded to 5.7 and were getting java core/heap dumps.   The reply for this was possibly memory allocation.  I do not believe this is my situation.  I believe our issue is more and more resources are used, not released.

      example of top review

      last pid: 11902;  load avg:  2.04,  2.26,  2.29;       up 70+23:45:33

      228 processes: 2 running, 226 sleeping

      CPU states:  4.3% user,  0.0% nice, 51.7% system, 44.0% idle,  0.0% iowait

      Memory: 1842M used, 185M free, 293M buffers, 810M cached

      Swap: 272M used, 2800M free, 17M cached

       PID USERNAME PRI NICE  SIZE   RES STATE   TIME   WCPU    CPU COMMAND

      19171 hci       25    0 1240M  288M sleep 832:46 71.00% 72.26% java

      11901 hci       19    0 5016K 1048K run     0:00  0.05%  0.57% hcilicstatus

      13177 root      16    0 8592K 1476K sleep 151:51  0.23%  0.38% conky

      11393 root      15    0   22M 4108K sleep 127:19  0.20%  0.38% Xorg

      10632 hci       15    0   22M 9068K sleep   2:56  0.26%  0.38% hcimonitord

      7996 hci       18    0   21M 6216K sleep  75:02  0.09%  0.19% hcimonitord

      20476 hci       18    0   20M 5208K sleep  41:18  0.06%  0.19% hcimonitord

      13207 root      15    0 1868K  580K sleep  24:20  0.05%  0.19% pam_timestamp_c

      2626 hci       15    0   20M 5136K sleep  15:57  0.04%  0.19% hcimonitord

      6262 hci       15    0   20M 6688K sleep   9:58  0.04%  0.19% hcimonitord

      22538 hci       18    0   20M 4976K sleep   7:42  0.03%  0.19% hcimonitord

      24487 hci       15    0 2620K  704K sleep 341:18  0.70%  0.00% lm

      1289 root      10   -5    0K    0K sleep 122:00  0.10%  0.00% kjournald

      23524 hci       15    0 2492K  588K sleep 108:58  0.09%  0.00% lm

      26541 hci       18    0   20M 5584K sleep  65:51  0.07%  0.00% hcimonitord

    • Creator
      Topic
    Viewing 2 reply threads
    • Author
      Replies
      • August 5, 2010 at 6:14 pm #72261
        Connie Kraska
        Participant

          We have also are experiencing the issue of java core dumps and heap dumps on 5.7 Rev 2.  We have to stop and start the client nearly daily.  It’s a big hassle.  Are ulimits on AIX 6.1 are generous.  We also do not think this is really a memory issue but rather processes that do not shut down.  Occassionally we get the error “too many files open”.

          We would like to know if there is a fix out there.  Charlie?

        • August 5, 2010 at 7:22 pm #72262
          Highland Dave
          Participant

            I forgot to include we are running Cloverleaf in a VM environment with

            RedHat Linux 5.1

            Also, I wanted to let you know that a Lawson person is connecting into our system this afternoon to check/try about our problem.  It could be something with a OS setting, or a combination of OS and GUI.  If we find out anything, I’ll update the post

          • August 5, 2010 at 7:27 pm #72263
            Connie Kraska
            Participant

              We too are in a VM environment hosted by an IBM P6.  We have to have this configuration in order to fail over to our disaster recovery site.  Our storage is EMC’s Clarion.

          • Author
            Replies
          Viewing 2 reply threads
          • The forum ‘Cloverleaf’ is closed to new topics and replies.