Page 1 of 2 12 LastLast
Results 1 to 10 of 18

Thread: How can I truncate raw_logs?

  1. #1
    Join Date
    Oct 2005
    Posts
    18
    Rep Power
    10

    Default How can I truncate raw_logs?

    Hi, At our installation of 4.5.10, we continue to have problems with the raw_logs completely filling our disk and taking down Zimbra. I've found that the raw_logs table is 96GB, plus its index is an additional 27GB.

    I know I can set zimbraLogRawLifetime from its default of 31d down to, say, 7d, but that doesn't seem to actually go delete the old records to get the table small again. How am I supposed to do that?

    thanks,
    -Nick

  2. #2
    phoenix is offline Zimbra Consultant & Moderator
    Join Date
    Sep 2005
    Location
    Vannes, France
    Posts
    23,587
    Rep Power
    58

    Default

    Have you checked to see if there is any corruption in the table: Repair Logger Data Corruption - Zimbra :: Wiki
    Regards


    Bill


    Acompli: A new adventure for Co-Founder KevinH.

  3. #3
    Join Date
    Oct 2005
    Posts
    18
    Rep Power
    10

    Default

    Thanks for your quick reply. I have, and it doesn't appear to be corrupt. There are no concerning messages in /opt/zimbra/logger/data/<myhost>.err

    I'd like to sift through the table and search for anything that's broken, but I don't know how to effectively sift through 90GB.

    I have noticed stats don't show up in the administrator pages, and I do get this error when I attempt to go to Server Status:

    Message: Csfe service error
    Error code: service.FAILURE
    Method: ZmCsfeCommand.invoke
    Details:system failure: getting database logger connection

    I've just inherited the machine, so I don't know much history about it.

    thanks again.

  4. #4
    phoenix is offline Zimbra Consultant & Moderator
    Join Date
    Sep 2005
    Location
    Vannes, France
    Posts
    23,587
    Rep Power
    58

    Default

    Is this an NE or OSS installation? You could re-initialise the logger database but it would wipe out any historical data.
    Regards


    Bill


    Acompli: A new adventure for Co-Founder KevinH.

  5. #5
    phoenix is offline Zimbra Consultant & Moderator
    Join Date
    Sep 2005
    Location
    Vannes, France
    Posts
    23,587
    Rep Power
    58

    Default

    I forgot to to add, could you post the /etc/hosts file?
    Regards


    Bill


    Acompli: A new adventure for Co-Founder KevinH.

  6. #6
    Join Date
    Oct 2005
    Posts
    18
    Rep Power
    10

    Default

    Oh, sorry I didn't mention that before... This is network edition.

    Is there a way to wipe out the current data and retain the summary data? It's far from critical, but probably a nice-to-have.

    In either case, can you please point me to how to wipe out the data? (or do I literally just run logmysql and delete all rows?

    thanks!

  7. #7
    Join Date
    Oct 2005
    Posts
    18
    Rep Power
    10

    Default

    127.0.0.1 localhost.localdomain localhost
    <this machine's IP> mailbox.<mydomain>.com mailbox

    Our server is called 'mailbox'

  8. #8
    phoenix is offline Zimbra Consultant & Moderator
    Join Date
    Sep 2005
    Location
    Vannes, France
    Posts
    23,587
    Rep Power
    58

    Default

    Quote Originally Posted by nick View Post
    127.0.0.1 localhost.localdomain localhost
    <this machine's IP> mailbox.<mydomain>.com mailbox
    The error message you posted above is usually symptomatic of an error in the hosts file but this one is OK.

    Were there any error messages at all in the log /opt/zimbra/logger/db/data/<myhost>.err?

    Which operating system is this and is your firewall and/or SElinux disabled?
    Last edited by phoenix; 05-08-2008 at 10:17 AM.
    Regards


    Bill


    Acompli: A new adventure for Co-Founder KevinH.

  9. #9
    Join Date
    Oct 2005
    Posts
    18
    Rep Power
    10

    Default

    We aren't running a firewall on this machine.

    The only messages I found in our /opt/zimbra/logger/data/<myhost>.err were normal stop / start messages when the service would begin and end.

    We're running on RHEL (well actually, on CentOS 4.5).

    thanks again.

  10. #10
    phoenix is offline Zimbra Consultant & Moderator
    Join Date
    Sep 2005
    Location
    Vannes, France
    Posts
    23,587
    Rep Power
    58

    Default

    Did you check the lifetime setting for the log, does the following display the correct 31d?

    Code:
    zmprov gacf | grep zimbraLogRawLifetime
    Regards


    Bill


    Acompli: A new adventure for Co-Founder KevinH.

Similar Threads

  1. Replies: 5
    Last Post: 11-18-2006, 01:11 PM

Posting Permissions

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