active redo logs
DBAsupport.com Forums - Powered by vBulletin
Results 1 to 6 of 6

Thread: active redo logs

Hybrid View

  1. #1
    Join Date
    Jun 2000
    Location
    Madrid, Spain
    Posts
    7,447
    Hi

    I have some question regarding the status of redo logs.
    The status are

    Active
    Inactive
    Current

    I understand that in archive log mode Active means the redo log is being archived. In a noarchive log mode what would Active mean? This is because I am getting

    BYTES MEMBERS ARC STATUS
    --------- ---------- --- ----------------
    1 10485760 2 NO CURRENT
    2 10485760 2 NO ACTIVE
    3 10485760 2 NO ACTIVE
    4 10485760 2 NO ACTIVE


    And it takes a while for group 2 to become inactive, what can this be?

  2. #2
    Join Date
    Jul 2000
    Location
    Amsterdam
    Posts
    234

    Hello Pando,


    The active status according to Oracle indicates the log is active but is not the current log, it is needed for crash recovery, it may be in use for block recovery or it might or might not be archived.

    It looks like your checkpoint has not yet finished up as the lgwr process starts to write to this online redo log. Try to add more log groups (with members).

    (Sure there are no mistakes in the init.ora like archive_log_start=true?)

    HTH

    Tycho

  3. #3
    Join Date
    Jun 2000
    Location
    Madrid, Spain
    Posts
    7,447
    no itŮs not in archive log mode, as you can see from the table I posted


    BYTES MEMBERS ARC STATUS
    --------- ---------- --- ----------------
    1 10485760 2 NO CURRENT
    2 10485760 2 NO ACTIVE
    3 10485760 2 NO ACTIVE
    4 10485760 2 NO ACTIVE


    ARC = ARCHIVE and is NO. What it puzzles me is how can three groups being active. Is the DISK really that slow so the redo logs are queued to be written?

  4. #4
    Join Date
    Jul 2000
    Location
    Amsterdam
    Posts
    234
    Hi Pando,

    Can you see the checkpoint processes are running with "ps -ef | grep ckpt" or is the instance running on a NT platform?

    Can it be there is some sort of IO bottleneck for the datafiles?

    Are the disks slow?

    (The ARC says they are not archived (which is normal in noarchivelog mode), It doesn't state that the archiving process is not running maybe keeping the logs in an active state (with out any reason).)

    Realy hope you can pinpoint your problem now.

    Tycho



  5. #5
    Join Date
    Jun 2000
    Location
    Madrid, Spain
    Posts
    7,447
    Yes there is I/O bottlenet still 3 groups being active seems too much.
    Acrhive process is not enabled

  6. #6
    Join Date
    Oct 2000
    Location
    Dallas:TX:USA
    Posts
    407
    what are your log sizes and the log switching frequency ?

    In one of my DSS systems (when big batch jobs are running), I have seen my 3 or 4 groups ACTIVE and logs (100MB size) were switching every 2-5 minutes ; I had to add more groups to allow proper checkpoint.

    - Rajeev

Posting Permissions

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


Click Here to Expand Forum to Full Width