Results 1 to 9 of 9

Thread: othersockbuf

  1. #1
    Join Date
    Aug 2007
    Location
    Toronto, Ontario
    Posts
    29

    Default othersockbuf

    I keep getting failcnt on the othersockbuf line in the /proc/user_beancounters file... how can I find out what is causing this? do i even need to worry?

    Thanks,

  2. #2
    Join Date
    Jun 2006
    Posts
    74

    Default

    Hello Marcusone,

    From OpenVZ documentation, othersockbuf is 'the total size of UNIX-domain socket buffers, UDP, and other datagram protocol send buffers'.

    There must be some kind of service on your VPS configured to send data using the UDP protocol (could be DNS, maybe VPN or some type of media stream,...), and the buffer size can't keep up with the amount of data it is trying to send.

    Once you will know what is causing it, you will know if you should worry or not...

  3. #3
    Join Date
    Aug 2007
    Location
    Toronto, Ontario
    Posts
    29

    Default

    Here is my bean readout
    othersockbuf 105632 2363192 2684354 5368709
    so most of the time, it is FAR within the limits...
    How can I find out what program is causing such a spike?

  4. #4
    Join Date
    Aug 2007
    Location
    Toronto, Ontario
    Posts
    29

    Default

    I think I might have found the problem... I have sendmail checking dnsbl lists, as well as spamassassin... so that was probably causing a lot of UDP traffic... ops
    I've disabled the one in spamassassin (as sendmail won't accept mail that is on the list anyways).

    Thanks for the "tip" on checking DNS

    QUESTION: upgrading to Link-5 (will this limit be increased) ? As if I have a busy server, I may still hit this limit!

  5. #5
    Join Date
    Aug 2006
    Posts
    1,019

    Default

    Here are the values for different plans:

    http://wiki.vpslink.com/index.php?ti...guration_files

  6. #6
    Join Date
    Aug 2007
    Location
    Toronto, Ontario
    Posts
    29

    Default

    Very nice! Thanks Jason (again)... now get on updating your monit patch for the new version

  7. #7
    Join Date
    Feb 2007
    Posts
    117

    Default

    Is there really any problem with hitting socket buffer limits? My limited understanding would make me expect that a program would just block on the socket write until it could allocate a buffer. I don't think it would cause program crashes or anything like that.

    I'm not a linux programmer, so maybe someone who knows more can correct me on this.
    intelligence is a tool to be used toward a goal, and goals are not always chosen intelligently
    Larry Niven, Protector

  8. #8
    Join Date
    Aug 2007
    Location
    Toronto, Ontario
    Posts
    29

    Default

    Good Question Steve... anyone know? I have yet to see any negative impact due to this error (and it appears my modifications didn't work )

  9. #9
    Join Date
    Feb 2007
    Posts
    117

    Default

    From http://wiki.openvz.org/Othersockbuf#othersockbuf
    Increased limit for othersockbuf is necessary for high performance of communications through local (UNIX-domain) sockets. However, similarly to tcpsndbuf, hitting othersockbuf affects the communication performance only and does not affect the functionality.
    intelligence is a tool to be used toward a goal, and goals are not always chosen intelligently
    Larry Niven, Protector

Bookmarks

Posting Permissions

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