Results 1 to 5 of 5

Thread: MySQL and zimbra_logger disk usage?

  1. #1
    tnelson is offline Intermediate Member
    Join Date
    May 2007
    Posts
    17
    Rep Power
    8

    Default MySQL and zimbra_logger disk usage?

    On a handful of boxes I have running Zimbra 5.0.4_GA_2101_RHEL5, I've noticed that the directory holding MySQL data for the Zimbra logger ( in /opt/zimbra/logger/db/data/zimbra_logger/ )is massive. The raw_logs.MYD and raw_logs.MYI files are nearly 30GB in some instances. If this is just log data, is there any way to purge it and set smaller log retention times?

    I've seen other posts here in the forums that reference this issue but none have a clear fix that I can see. All help is appreciated. Thank you!

    Edit:
    After looking in /opt/zimbra/logger/db/data/<hostname>.com.err I see a bunch of these:
    Code:
    081003 11:52:39 [ERROR] /opt/zimbra/logger/mysql/libexec/mysqld: Table './zimbra_logger/mta' is marked as crashed and last (automatic?) repair failed
    081003 11:52:39 [ERROR] /opt/zimbra/logger/mysql/libexec/mysqld: Table './zimbra_logger/mta' is marked as crashed and last (automatic?) repair failed
    I'm running a "check table raw_logs;" from mysql right now and will probably need to follow up with a "repair".

    BUT, that still does not explain why the database is so huge does it?

    --Tim
    Last edited by tnelson; 10-03-2008 at 10:00 AM. Reason: More info from logs...

  2. #2
    brian is offline Project Contributor
    Join Date
    Jul 2006
    Posts
    623
    Rep Power
    10

    Default

    Default is 31d for raw logs. If the mta summary table was corrupt the raw_logs will not be summarized and pruned as it would result in a lose of data.

    After fixing the corruption you can reduce the lifetime of raw_logs with the following.
    Code:
    zmprov mcf zimbraLogRawLifetime 2d
    Bugzilla - Wiki - Downloads - Before posting... Search!

  3. #3
    tnelson is offline Intermediate Member
    Join Date
    May 2007
    Posts
    17
    Rep Power
    8

    Default

    When my 'repair table raw_logs;' completes, I'll give that a shot. I'm assuming it will immediately 'prune' the database or will I need to initiate that process after making the change?

  4. #4
    brian is offline Project Contributor
    Join Date
    Jul 2006
    Posts
    623
    Rep Power
    10

    Default

    The next scheduled run of zmlogprocess will prune it. You can always run /opt/zimbra/libexec/zmlogprocess by hand if you want it to happen sooner.
    Bugzilla - Wiki - Downloads - Before posting... Search!

  5. #5
    tnelson is offline Intermediate Member
    Join Date
    May 2007
    Posts
    17
    Rep Power
    8

    Default

    I ran a repair on the database which took a few days to complete. However, zmlogprocess is still throwing errors complaining about the table being corrupt and the automatic repair process not fixing it. How do I proceed? I'm more than happy to just rm these files if mysql/logger will automatically recreate them. I just can't see keeping some 30GB of what I consider useless data sitting around...

    Suggestions and/or comments?

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

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