Site disruption

Yesterday we got an Instalanche (thank you, Glenn!) the traffic from which unfortunately had the effect of crashing our server. The site was down for much of the day, as was email service to and (if you sent us mail and we didn’t respond, please resend). While the front page was restored within a few hours, most of the rest of the site remained down until this morning. Thanks for your patience.

P.S. Welcome readers from the second Instalanche, which we’re happy to say our servers have succeeded in accommodating. Our new slogan, courtesy of Glenn: Overlawyered, the site “you should probably be visiting regularly anyway”.


  • It might help to note that the reason it was murdering the server was due to your site set to dynamically publish the archive pages. Each request for the permalink in question was spawning a new process, which a hundreds of requests per second, rapidly ate up all available resources. At any rate, large traffic sites should emphatically NOT publish dynamically. Maybe this tidbit will help others in the same boat.

  • There’s a caching plugin for WP that would help.

  • Interesting thought, but that’s not necessarily the truth.

    A: Even a site that publishes dynamically can override the process for a single page by simply putting up a static page. If the page is found, the process is short-circuited.

    B: The use of caching will not prevent the spawning of which you speak, but it will lower it dramatically. Rather than a request for each page individually, you get a request for a page – if it’s been built recently, you get that. If not, you get a new one.

    Nothing is wrong with dynamic publishing if implemented correctly.

  • “My Complication Had Complications.”

    And my Instalanche had an Instalanche. (Via . . . good old whatshisface.)…

  • I had one insta hit in 2004 on my site and blogspot held its own nicely. I’m simply too small to pay for my own domain at the current time.