Results 1 to 4 of 4
  1. #1
    Join Date
    Aug 2002
    Location
    Shanghai , China
    Posts
    63

    Unanswered: what's the relation between log_buffer and redo log file size

    Hi , Gurus ,

    what's the relation between log_buffer and redo log file size .

    My DB log_buffer is around 10MB but redo log file size is only 1 MB . I am worried that which log_buffer flush , the redo log file is not able to hold it
    Oracle is an ocean . I am just a little fish

  2. #2
    Join Date
    Aug 2002
    Posts
    2
    hi

    yes thre is lot of relation with the log_buffer with redo log files

    the log_buffer flushes the data into redo log files in following situations

    1.) 1/3 of log_buffer si filled up
    2.)ckpt i.e for every 3 sec
    3.)when it is 1mb filled
    4.) for every commit

    so chek out the reason why you r getting problem.

    if this is not ur expected answer please mail to me back with appropiate question

    regards
    jayanth

  3. #3
    Join Date
    Aug 2002
    Location
    Shanghai , China
    Posts
    63
    Actually I didn't face problem on redo log buffer . I want to size the redo log file properly . Do you think I use following parameters :
    log_buffer = 3MB
    redo log file = 1 MB
    Oracle is an ocean . I am just a little fish

  4. #4
    Join Date
    Sep 2002
    Location
    ITALY
    Posts
    53
    Hi.
    There is no relation between the log buffer dim and the size of the redo log file. It's simply a matter of tuning.

    They are triggering two separate actions/events:
    1) the size of logbuffer limits the quantity of redo records that can be inserted in the buffer before you have to wait for ... more room be made by logwriter (burst inserting from many transactions);
    2) the size of redo log file limits the quantity of redo records that can be inserted in the file before you have to checkpoint (and open another redo).
    The size you have to keep in mind is the joint size of all online logs,
    because if you do not have space for burst of redo inserts the db hangs waiting the dbwriter to complete the checkpoint. This size must be quite larger than redo log buffer if you do not want to wait for checkpoint completion.

    More than this: you should evalute the time between checkpoints (log switches) and this time should be at least a few minutes. This drive your choice of the size of a single redo log file.

    see you soon.
    Franco Ceotto
    SIPTI srl
    OnSite Services
    9i OCP DBA, Performance Engineer

Posting Permissions

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