No announcement yet.

Got error 12 from storage engine - Need help with my.cnf

  • Filter
  • Time
  • Show
Clear All
new posts

  • Got error 12 from storage engine - Need help with my.cnf


    I've been getting between 10 to 50 "Got error 12 from storage engine" errors since a week. Based on the provided information, could someone please tell me if I need to upgrade my machine or change my config? If it's config related, what can I do?

    The other weird thing is, it's always on the same query :
    user.username, (user.options & 512) AS invisible, user.usergroupid,
    session.userid, session.inforum, session.lastactivity,
    IF(displaygroupid=0, user.usergroupid, displaygroupid) AS displaygroupid
    FROM session AS session
    LEFT JOIN user AS user ON(user.userid = session.userid)
    WHERE session.lastactivity > 1182128748
    ORDER BY username ASC;

    This dedicated server runs 3 vBulletin forums and 3 websites (those 6 sites are on 3 lighttpd servers using a load balancer module).

    Machine specs: 2xINTEL XEON 3.20G/800/1M/S604, 4xSATA 36 GB RAPTOR, 2xKINGSTON 2GB KIT ECC KVR400D2S4R3K2/2G

    [[email protected] etc]# uname -a
    Linux 2.6.20-1.2319.fc5smp #1 SMP Mon May 21 20:11:15 EDT 2007 i686 i686 i386 GNU/Linux

    [[email protected] etc]# rpm -q mysql

    [[email protected] etc]# cat my.cnf

    port = 3306
    key_buffer = 2G



    # TMP Table Settings
    max_heap_table_size = 256M # Used as needed, no adverse reactions
    tmp_table_size = 256M # Used as needed, no adverse reactions
    max_join_size = 256M # used to catch bad joins and disallow them
    join_buffer_size = 64M # used for unindexed table joins (never or rarely ever)
    #max_tmp_tables = 256 # (This option does not yet do anything.)



    top - 00:32:59 up 3 days, 22:17, 1 user, load average: 0.26, 0.21, 0.18
    Tasks: 90 total, 1 running, 89 sleeping, 0 stopped, 0 zombie
    Cpu(s): 2.9%us, 1.2%sy, 0.0%ni, 95.2%id, 0.1%wa, 0.0%hi, 0.7%si, 0.0%st
    Mem: 4139744k total, 3425784k used, 713960k free, 110120k buffers
    Swap: 2031608k total, 88k used, 2031520k free, 2834564k cached

    This MySQL server has been running for 0 days, 0 hours, 54 minutes and 56 seconds. It started up on Jun 17, 2007 at 11:39 PM.

    Query statistics: Since its startup, 1,755,632 queries have been sent to the server.
    Total ø per hour ø per minute ø per second
    1,756 k 1.92 M 31.96 k 532.66

    If anybody needs any further information, just ask .

    This problem is driving me nuts. Also, I'm the person in charge of the server so saying "check with your ISP" won't help .

  • #2
    We don't offer MySQL server support. Only vBulletin. And vB is just reporting back there's an issue with the server.

    MySQL Error: Cannot allocate memory
    This is a server issue, the system is out of memory that is allowed for MySQL to allocate. MySQL may need to be restarted or the server itself.

    You will need to contact your host about this.


    • #3
      Error 12 means that the MySQL server attempted to allocate more virtual memory than is allowed by the OS. You'll have to look at both MySQL's and Linux's configuration with respect to memory usage and limits (ulimit).

      When it happens shortly after a fresh restart of mysqld, this usually means it's caused by configuration. When it only happens after mysqld has been running for a long time, your MySQL may have a memory leak.
      "The lurking suspicion that something could be simplified is the world's richest source of rewarding challenges"
      - Edsger Dijkstra


      • #4
        Originally posted by Floris View Post
        You will need to contact your host about this.
        Originally posted by Fuitad View Post
        Also, I'm the person in charge of the server so saying "check with your ISP" won't help .
        Yeah... shows you've really read what I wrote.

        The reason why I come here for help is that it's always the same query. Everything else works fine. It's just this very specific query. So you'll have to admit it's weird.


        • #5
          Yes, I did read what you write. But I rather tell you the facts and how it is, and where to go to for the support. Then give you false info and only what you want to hear.

          I don't mind going in depth, which starts with. Does this happen on 3.6.7pl1? Do you have any plugins installed, customizations, custom apache/php/mysql configurations or addons that require additional memory? Can you reproduce this on a bigger account, does it still happen when you increase your virtual memory? Does the mysql log file and other log files disclose why it runs out of that required memory? All answers will be unofficial and unsupported, because it's simply not a vBulletin product I am helping you fix, but something I know nothing about, well, very little : your hosting account.

          Here's a forum on our site where we give unofficial help with optimizing your hosting account :

          The sticky threads have optimization request threads, tips and hints for these web services, and info to read before asking questions. Hopefully any of that gets you to the next level of resolving this.


          widgetinstance 262 (Related Topics) skipped due to lack of content & hide_module_if_empty option.