2008-01-11-ahhh-that-s-better.html

---
layout: post
title: "Ahhh...  That's Better"
permalink: ahhh-that-s-better.html
categories: [perpubplat, typo, reliability]
---

<p>I <a href="http://blog.prb.io/administrivia.html">cut over</a> to
perpubplat about a week ago, so it's worth a quick check to see if I
met my goal of getting a few more nines.</p>

<p>On the left, we have an every-30-minutes response time chart for
the Typo version of this site, and on the right, the perpubplat
version. (The charts are from the free monitoring service from <a
href="http://mon.itor.us">mon.itor.us</a>.)  The Typo configuration
used Apache2 with <code>mod_fcgid</code> (see my earlier <a
href="http://blog.prb.io/apache2-mod_fcgid-rails.html">post</a> on the
subject) configured for seven processes with a maximum 60-minute
lifetime; the perpubplat configuration uses Apache2 with
<code>mod_fastcgi</code> configured for a single process that runs on
50 lightweight Haskell threads internally.  (N.B.: The graphs use different
scales on the y-axis.)</p>

<p style="align: center;"><img src="http://blog.prb.io/files/typo_reliability.png" alt="mon.itor.us graph for Typo install of blog.prb.io" />&#160;<img src="http://blog.prb.io/files/perpubplat_reliability.png" alt="mon.itor.us graph for perpubplat install of blog.prb.io" /></p>

<p>The graph hints that things are better, but some quick text
processing on server logs make the difference more explicit.  Here's
an analysis of the response codes from the most recent log file for
perpubplat:</p>

<pre class="code"><b>$</b> head -n 1 multifarious-combined.log | awk '{print $4}'
[07/Jan/2008:00:32:50
<b>$</b> ^head^tail
[11/Jan/2008:15:03:50
<b>$</b> awk '{print $9}' multifarious-combined.log | sort | uniq -c | sort
      1 400
      2 206
     73 304
   1572 302
   2976 404
   5574 301
  14772 200</pre>

<p>No <code>500</code>s.  (Most of the <code>404</code>'s are comment
spammers trying to hit old URLs for comments.)  Here's the same
analysis for Typo from a week back in December of last year
(2007):</p>

<pre class="code"><b>$</b> zcat mult.ifario.us-access.log.3.gz | head -n 1 | awk '{print $4}'
[10/Dec/2007:19:49:06
<b>$</b> ^head^tail
[19/Dec/2007:14:28:17
<b>$</b> zcat mult.ifario.us-access.log.3.gz | awk '{print $9}' | sort | uniq -c | sort
      5 206
      5 400
     <span style="color: red;">23 503</span>
     30 302
    237 404
   1098 301
   <span style="color: red'">2259 500</span>
   9319 304
  20885 200</pre>

<p>A change from one-ish nines (-log<sub><small>10</small></sub>
(33861 - 2282)/33861 ~ 1.17) to 100% uptime is a positive change, and
the CPU trace for the virtual server suggests that the perpubplat
configuration uses a tiny fraction of the machine resources of the
Typo configuration.</p>