• Last night my server began to crash for no reason I can tell after I made a change to one plug in. So now I’m try to trouble shoot the issue.

    Here’s the set up

    I’m running WP 3.5.2 on an AWS EC2 c1.medium instance, the database is running off an MYSQL RDS and I’m using w3c total cache for page, database, object and CDN support. Most of the theme files and hard assest (css,js, images) are stored on an Amazon s3 bucket.

    For the combatiltiy test for w3c I’m getting this back

    Plugin Version: 0.9.2.11
    PHP Version: 5.3.6-13ubuntu3.10;
    Web Server: Apache
    FTP functions: Installed (required for Self-hosted (FTP) CDN support)
    Multibyte String support: Installed (required for Rackspace Cloud Files support)
    cURL extension: Installed (required for Amazon S3, Amazon CloudFront, Rackspace CloudFiles support)
    zlib extension: Installed (required for compression support)
    Opcode cache: Installed (APC)
    Memcache extension: Not installed
    HTML Tidy extension: Not installed (required for HTML Tidy minifier suppport)
    Mime type detection: Installed (Fileinfo) (required for CDN support)
    Hash function: Installed (hash) (required for NetDNA / MaxCDN CDN purge support)
    Safe mode: Off
    Open basedir: Off
    zlib output compression: Off
    set_time_limit: Available
    mod_deflate: Installed (required for disk enhanced Page Cache and Browser Cache)
    mod_env: Installed (required for disk enhanced Page Cache and Browser Cache)
    mod_expires: Installed (required for disk enhanced Page Cache and Browser Cache)
    mod_headers: Installed (required for disk enhanced Page Cache and Browser Cache)
    mod_mime: Installed (required for disk enhanced Page Cache and Browser Cache)
    mod_rewrite: Installed (required for disk enhanced Page Cache and Browser Cache)
    mod_setenvif: Installed (required for disk enhanced Page Cache and Browser Cache)
    Additional modules
    
    New Relic
    PHP version: Supported
    Operating System: Supported
    Web Server: Supported
    WordPress Resources
    
    /var2/www/vhosts/mysite.com/.htaccess: OK
    /var2/www/vhosts/mysite.com/wp-content: OK
    /var2/www/vhosts/mysite.com/wp-content/uploads/2013/07: OK
    Fancy permalinks: /%year%/%monthnum%/%day%/%postname%/
    WP_CACHE define: Defined (true)
    URL rewrite: Enabled
    Network mode: No

    I think this looks okay, but I have no clue.

    My PHP ini loos like this:

    apc
    
    APC Support => disabled
    Version => 3.1.7

    Which I’m not sure if that means anything.

    The error I’m most seeing in my error log is this

    [Thu Jul 18 14:56:29 2013] [error] [client 198.228.200.46] PHP Warning: apc_store(): Potential cache slam averted for key ‘w3tc_key_mysite.com_0_object_ce2b0918636aa420fc452eccd6f14ce7_0’ in /var2/www/vhosts/mysite.com/wp-content/plugins/w3-total-cache/lib/W3/Cache/Apc.php on line 55, referer: https://mysite.com/2013/07/14/pagename/3/

    Anyhow, the site was running fine until I made a change to one of themes plugins “Theia Post Slider” to have to refresh on each slide to take advantage of the page hits.

    By the way I’m using Opcode for Database Cache and Object Cache.

    Please any insight of where to start trouble shooting would be helpful.

Viewing 1 replies (of 1 total)
  • Thread Starter Philip Isaacs

    (@pisaacs)

    Hey guys,

    So here is the follow up.

    First I was able to fix the issue with the apc_store() warning message i was seeing in the logs, but making a change to the apc.ini file. This is a known bug I guess. Anyhow, if your also having this problem just change apc.slam_defense = 0 to make it go away.

    As far as the other issue of my server crashing, it had to do with the “Theia Post Slider”. The plugin which came with my theme makes it super easy to create slide shows out of a single post. Very useful. The problem I found is that when I change the settings in the plugin to refresh the page when a user clicks to the next slide, it started to choke my database to death. I thought it was related to W3 Total Cache, but it seems to have nothing to do with it. Although I guess it’s strange that it didn’t used the cached page when the page was refreshed, and that’s what I still need to figure out.

    Okay, hope this little update it useful to someone.

    Floyd

Viewing 1 replies (of 1 total)
  • The topic ‘Trouble shooting server crashing – W3C and APC issue’ is closed to new replies.