Jump to content
  • Member Statistics

    17,609
    Total Members
    7,904
    Most Online
    NH8550
    Newest Member
    NH8550
    Joined

Server Performance...feedback needed


stormtracker

Recommended Posts

Yes, caching is and has been enabled.  We're still going through all the logs to determine the cause of this error.  It seems it's only affecting the sql processing. Overall server loads on both servers have been fine, even during our high traffic events.

Link to comment
Share on other sites

  • Replies 128
  • Created
  • Last Reply

Yes, caching is and has been enabled.  We're still going through all the logs to determine the cause of this error.  It seems it's only affecting the sql processing. Overall server loads on both servers have been fine, even during our high traffic events.

 

Unless your http headers are lying, you're running php 5.2.17. That's pretty old. Not necessarily the source of the problem, but depending on the versions of MySQL(I assume) you're running, and the version of the MySQL client you've got, there could be some problems there.

 

The likelier possibility, given that you mentioned it happens during large processing runs of model data, is that some code on your webserver, either IPB or your model processing code, is using something like mysql_pconnect, spawning too many long-running queries on multiple simultaneous connections, and failing a connect call somewhere.

 

Either way, enterprise php app development (boy, that's a joke) is what I do for a living, and I'd be happy to take a look under the hood pro bono. I like this board.

Link to comment
Share on other sites

  • 2 weeks later...

Unless your http headers are lying, you're running php 5.2.17. That's pretty old. Not necessarily the source of the problem, but depending on the versions of MySQL(I assume) you're running, and the version of the MySQL client you've got, there could be some problems there.

The likelier possibility, given that you mentioned it happens during large processing runs of model data, is that some code on your webserver, either IPB or your model processing code, is using something like mysql_pconnect, spawning too many long-running queries on multiple simultaneous connections, and failing a connect call somewhere.

Either way, enterprise php app development (boy, that's a joke) is what I do for a living, and I'd be happy to take a look under the hood pro bono. I like this board.

As someone else somewhat experienced in PHP/MySQL development, and who runs a large message board myself, I cannot emphasize the above enough.

I'm not surprised to hear that it's not a problem with overall CPU/mem load. That's rarely the case. It's far more likely the maximum number of simultaneous SQL connections being exceeded. This is where careful examination of log files, and perhaps addition of some debugging/error-handling code to the forum software, would be helpful. It may be as simple as substantially increasing a software limit, or it may require some rewriting to eliminate pconnects.

Short answer - this is definitely fixable.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...