Results 1 to 4 of 4
  1. #1
    Join Date
    Feb 2004
    Posts
    2

    Question Unanswered: 9iAS Memory Issue

    Iam facing a serious problem , there are some remote sessions (LOCAL=NO) which seem to occupy unusually large chunk of memory
    and not ready to leave it for hours and my SSO portal hangs as a result...

    below is a typical output of top command on my Suse machine..
    PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
    12288 oracle 15 0 333M 333M 35552 S 0.0 10.9 1:05 oracle
    1631 oracle 16 0 321M 321M 24264 S 1.1 10.6 1:23 oracle
    4221 oracle 16 0 320M 320M 23136 S 0.0 10.5 0:58 oracle
    25202 oracle 15 0 318M 318M 21320 S 0.0 10.4 2:59 oracle
    25211 oracle 16 0 319M 179M 22360 S 0.0 5.9 1:07 oracle

    if you notice the top 4 processes are taking up approx 320M of memory...

    portalpdb:~ # ps -eaf | grep 12288
    oracle 12288 1 0 13:58 ? 00:01:05 oracleiasdb (LOCAL=NO)
    root 10595 3921 0 16:00 pts/4 00:00:00 grep 12288



    Please help me if you have got any solution or idea where I can find one...

    Iam a sysadmin not a DBA so my question might sound amature..

  2. #2
    Join Date
    Jan 2004
    Posts
    370
    You are getting a problem because some remote sessions "seem" to occupy unusually large chunks of memory?

    You must be careful when looking at memory usage of oracle processes at the O/S level because Oracle uses shared memory, and this can result in much higher process memory usage being reported than is actually being used. There used to be a paper on metalink regarding this - don't know if it is still available.

    Are you sure your problem is memory related?

  3. #3
    Join Date
    Feb 2004
    Posts
    2
    I do think that it is related to memory occupied by Oracle processes as Iam having a machine with 3GB Memory and 3 GB of swap space..

    normally the swap space is low utilised but after 2-3 days when these process go on accumulating it goes full...

    below is the current status of swap space and top command

    total used free shared buffers cached
    Mem: 3105516 2807912 297604 0 152588 257212
    -/+ buffers/cache: 2398112 707404
    Swap: 3157416 1554280 1603136
    -------------------------------------------------------

    PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
    4221 oracle 15 0 328M 328M 31496 S 0.0 10.8 1:43 oracle
    3178 oracle 15 0 324M 324M 27472 S 0.0 10.7 1:24 oracle
    29418 oracle 15 0 324M 324M 27212 S 3.4 10.6 1:25 oracle
    13534 oracle 15 0 312M 312M 15304 S 0.0 10.3 1:00 oracle
    5008 oracle 15 0 321M 170M 24216 S 0.0 5.6 1:02 oracle

  4. #4
    Join Date
    Jan 2004
    Posts
    370
    I can't see any apparent memory problems from the information you've posted.

    Why do you think that the memory problem is caused by the oracle processes?

    Is it because top shows them to be using more memory than any other processes?
    How much memory do you think they are actually using, and how are you calculating it?

    Or does their size continue to increase unbounded over time?

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •