Results 1 to 2 of 2
  1. #1
    Join Date
    Jun 2003
    Posts
    1

    Question Unanswered: redo log file and log_buffer size

    Hello!
    I have a 9i Rel 1 database with log_buffer = 5M and 5 redo log groups of 2 members each with 950M.
    I am experiencing lots and lots of log file sync and log buffer space events; below I added one result of a statspack report (of 10 minuts collect).
    Top 5 Wait Events
    ~~~~~~~~~~~~~~~~~ Wait % Total
    Event Waits Time (s) Wt Time
    -------------------------------------------- ------------ ----------- -------
    enqueue 4,529 12,976 41.49
    log buffer space 17,683 7,269 23.24
    library cache lock 973 2,775 8.87
    latch free 58,102 1,736 5.55
    log file sync 2,700 1,689 5.40

    I heard that increasing the log_buffer beyond 1M is not of much result.
    My log switch is from 3 to 3 minuts.

    I wouldn't like to increase much more the redo log files because it is already very large.

    So, what should I do to decrease this events?
    obs: i have already set to nologging everything that was possible.

    Thanks in advance

  2. #2
    Join Date
    Sep 2003
    Location
    china
    Posts
    3
    the log buffer space wait events caused by your inadequate log files,your log switch is too frequent.we understand during log switch oracle cant produce new redo,so present log buffer space wait.pls add more log group to maintain switch duration up to 20min+.
    the log file sync events due to your oversize log buffer,every log buffer flush need long time so subsequent redo write must wait.try to decrease your log buffer to 200k.
    good luck!

Posting Permissions

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