update transformation

[apache] / docs / manual / caching.xml

<?xml version="1.0" encoding="UTF-8" ?>
<!DOCTYPE manualpage SYSTEM "style/manualpage.dtd">
<?xml-stylesheet type="text/xsl" href="style/manual.en.xsl"?>
<!-- $LastChangedRevision$ -->

<!--
 Licensed to the Apache Software Foundation (ASF) under one or more
 contributor license agreements.  See the NOTICE file distributed with
 this work for additional information regarding copyright ownership.
 The ASF licenses this file to You under the Apache License, Version 2.0
 (the "License"); you may not use this file except in compliance with
 the License.  You may obtain a copy of the License at

     http://www.apache.org/licenses/LICENSE-2.0

 Unless required by applicable law or agreed to in writing, software
 distributed under the License is distributed on an "AS IS" BASIS,
 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License.
-->

<manualpage metafile="caching.xml.meta">

  <title>Caching Guide</title>

  <summary>
    <p>This document supplements the <module>mod_cache</module>,
    <module>mod_cache_disk</module>, <module>mod_file_cache</module> and <a
    href="programs/htcacheclean.html">htcacheclean</a> reference documentation.
    It describes how to use the Apache HTTP Server's caching features to accelerate web and
    proxy serving, while avoiding common problems and misconfigurations.</p>
  </summary>

  <section id="introduction">
    <title>Introduction</title>
    
    <p>The Apache HTTP server offers a range of caching features that
    are designed to improve the performance of the server in various
    ways.</p>

    <dl>
        <dt>Three-state RFC2616 HTTP caching</dt>
        <dd>
            <module>mod_cache</module>
            and its provider modules
            <module>mod_cache_disk</module>
            provide intelligent, HTTP-aware caching. The content itself is stored
            in the cache, and mod_cache aims to honor all of the various HTTP
            headers and options that control the cacheability of content
            as described in
            <a href="http://www.w3.org/Protocols/rfc2616/rfc2616-sec13.html">Section
            13 of RFC2616</a>.
            <module>mod_cache</module>
            is aimed at both simple and complex caching configurations, where
            you are dealing with proxied content, dynamic local content or
            have a need to speed up access to local files on a potentially
            slow disk.
        </dd>

        <dt>Two-state key/value shared object caching</dt>
        <dd>
            The <a href="socache.html">shared object cache API</a> (socache)
            and its provider modules provide a
            server wide key/value based shared object cache. These modules
            are designed to cache low level data such as SSL sessions and
            authentication credentials. Backends allow the data to be stored
            server wide in shared memory, or datacenter wide in a cache such
            as memcache or distcache.
        </dd>

        <dt>Specialized file caching</dt>
        <dd>
            <module>mod_file_cache</module>
            offers the ability to pre-load
            files into memory on server startup, and can improve access
            times and save file handles on files that are accessed often,
            as there is no need to go to disk on each request.
        </dd>
    </dl>

    <p>To get the most from this document, you should be familiar with
    the basics of HTTP, and have read the Users' Guides to
    <a href="urlmapping.html">Mapping URLs to the Filesystem</a> and
    <a href="content-negotiation.html">Content negotiation</a>.</p>

  </section>

  <section id="http-caching">

    <title>Three-state RFC2616 HTTP caching</title>

    <related>
      <modulelist>
        <module>mod_cache</module>
        <module>mod_cache_disk</module>
      </modulelist>
      <directivelist>
        <directive module="mod_cache">CacheEnable</directive>
        <directive module="mod_cache">CacheDisable</directive>
        <directive module="core">UseCanonicalName</directive>
        <directive module="mod_negotiation">CacheNegotiatedDocs</directive>
      </directivelist>
    </related>

    <p>The HTTP protocol contains built in support for an in-line caching
    mechanism 
    <a href="http://www.w3.org/Protocols/rfc2616/rfc2616-sec13.html">
    described by section 13 of RFC2616</a>, and the
    <module>mod_cache</module> module can be used to take advantage of
    this.</p>

    <p>Unlike a simple two state key/value cache where the content
    disappears completely when no longer fresh, an HTTP cache includes
    a mechanism to retain stale content, and to ask the origin server
    whether this stale content has changed and if not, make it fresh
    again.</p>

    <p>An entry in an HTTP cache exists in one of three states:</p>

    <dl>
    <dt>Fresh</dt>
    <dd>
        If the content is new enough (younger than its <strong>freshness
        lifetime</strong>), it is considered <strong>fresh</strong>. An
        HTTP cache is free to serve fresh content without making any
        calls to the origin server at all.
    </dd>
    <dt>Stale</dt>
    <dd>
        <p>If the content is too old (older than its <strong>freshness
        lifetime</strong>), it is considered <strong>stale</strong>. An
        HTTP cache should contact the origin server and check whether
        the content is still fresh before serving stale content to a
        client. The origin server will either respond with replacement
        content if not still valid, or ideally, the origin server will
        respond with a code to tell the cache the content is still
        fresh, without the need to generate or send the content again.
        The content becomes fresh again and the cycle continues.</p>

        <p>The HTTP protocol does allow the cache to serve stale data
        under certain circumstances, such as when an attempt to freshen
        the data with an origin server has failed with a 5xx error, or
        when another request is already in the process of freshening
        the given entry. In these cases a <code>Warning</code> header
        is added to the response.</p>
    </dd>
    <dt>Non Existent</dt>
    <dd>
        If the cache gets full, it reserves the option to delete content
        from the cache to make space. Content can be deleted at any time,
        and can be stale or fresh. The <a
        href="programs/htcacheclean.html">htcacheclean</a> tool can be
        run on a once off basis, or deployed as a daemon to keep the size
        of the cache within the given size, or the given number of inodes.
        The tool attempts to delete stale content before attempting to
        delete fresh content.
    </dd>
    </dl>

    <p>Full details of how HTTP caching works can be found in
    <a href="http://www.w3.org/Protocols/rfc2616/rfc2616-sec13.html">
    Section 13 of RFC2616</a>.</p>

    <section>
      <title>Interaction with the Server</title>

      <p>The <module>mod_cache</module> module hooks into the server in two
      possible places depending on the value of the
      <directive module="mod_cache">CacheQuickHandler</directive> directive:
      </p>

      <dl>
        <dt>Quick handler phase</dt>
        <dd>
          <p>This phase happens very early on during the request processing,
              just after the request has been parsed. If the content is
              found within the cache, it is served immediately and almost
              all request processing is bypassed.</p>

              <p>In this scenario, the cache behaves as if it has been "bolted
              on" to the front of the server.</p>
              
              <p>This mode offers the best performance, as the majority of
              server processing is bypassed. This mode however also bypasses the
              authentication and authorization phases of server processing, so
              this mode should be chosen with care when this is important.</p>
          </dd>
          <dt>Normal handler phase</dt>
          <dd>
              <p>This phase happens late in the request processing, after all
              the request phases have completed.</p>

              <p>In this scenario, the cache behaves as if it has been "bolted
              on" to the back of the server.</p>

              <p>This mode offers the most flexibility, as the potential exists
              for caching to occur at a precisely controlled point in the filter
              chain, and cached content can be filtered or personalized before
              being sent to the client.</p>
          </dd>
        </dl>

        <p>If the URL is not found within the cache, <module>mod_cache</module>
        will add a <a href="filter.html">filter</a> to the filter stack in order
        to record the response to the cache, and then stand down, allowing normal
        request processing to continue. If the content is determined to be
        cacheable, the content will be saved to the cache for future serving,
        otherwise the content will be ignored.</p>

        <p>If the content found within the cache is stale, the
        <module>mod_cache</module> module converts the request into a
        <strong>conditional request</strong>. If the origin server responds with
        a normal response, the normal response is cached, replacing the content
        already cached. If the origin server responds with a 304 Not Modified
        response, the content is marked as fresh again, and the cached content
        is served by the filter instead of saving it.</p>
    </section>

    <section>
      <title>Improving Cache Hits</title>

      <p>When a virtual host is known by one of many different server aliases,
      ensuring that <directive module="core">UseCanonicalName</directive> is
      set to <code>On</code> can dramatically improve the ratio of cache hits.
      This is because the hostname of the virtual-host serving the content is
      used within the cache key. With the setting set to <code>On</code>
      virtual-hosts with multiple server names or aliases will not produce
      differently cached entities, and instead content will be cached as
      per the canonical hostname.</p>

    </section>

    <section>
      <title>Freshness Lifetime</title>

      <p>Well formed content that is intended to be cached should declare an
      explicit freshness lifetime with the <code>Cache-Control</code>
      header's <code>max-age</code> or <code>s-maxage</code> fields, or
      by including an <code>Expires</code> header.</p>
      
      <p>At the same time, the origin server defined freshness lifetime can
      be overridden by a client when the client presents their own
      <code>Cache-Control</code> header within the request. In this case,
      the lowest freshness lifetime between request and response wins.</p>

      <p>When this freshness lifetime is missing from the request or the
      response, a default freshness lifetime is applied. The default
      freshness lifetime for cached entities is one hour, however
      this can be easily over-ridden by using the <directive
      module="mod_cache">CacheDefaultExpire</directive> directive.</p>

      <p>If a response does not include an <code>Expires</code> header but does
      include a <code>Last-Modified</code> header, <module>mod_cache</module>
      can infer a freshness lifetime based on a heuristic, which can be
      controlled through the use of the <directive
      module="mod_cache">CacheLastModifiedFactor</directive> directive.</p>

      <p>For local content, or for remote content that does not define its own
      <code>Expires</code> header, <module>mod_expires</module> may be used to
      fine-tune the freshness lifetime by adding <code>max-age</code> and
      <code>Expires</code>.</p>

      <p>The maximum freshness lifetime may also be controlled by using the
      <directive module="mod_cache">CacheMaxExpire</directive>.</p>

    </section>

    <section>
      <title>A Brief Guide to Conditional Requests</title>

      <p>When content expires from the cache and becomes stale, rather than
      pass on the original request, httpd will modify the request to make
      it conditional instead.</p>

      <p>When an <code>ETag</code> header exists in the original cached
      response, <module>mod_cache</module> will add an
      <code>If-None-Match</code> header to the request to the origin server.
      When a <code>Last-Modified</code> header exists in the original
      cached response, <module>mod_cache</module> will add an
      <code>If-Modified-Since</code> header to the request to the origin
      server. Performing either of these actions makes the request
      <strong>conditional</strong>.</p>

      <p>When a conditional request is received by an origin server, the
      origin server should check whether the ETag or the Last-Modified
      parameter has changed, as appropriate for the request. If not, the
      origin should respond with a terse "304 Not Modified" response. This
      signals to the cache that the stale content is still fresh should be
      used for subsequent requests until the content's new freshness lifetime
      is reached again.</p>

      <p>If the content has changed, then the content is served as if the
      request were not conditional to begin with.</p>

      <p>Conditional requests offer two benefits. Firstly, when making such
      a request to the origin server, if the content from the origin
      matches the content in the cache, this can be determined easily and
      without the overhead of transferring the entire resource.</p>

      <p>Secondly, a well designed origin server will be designed in such
      a way that conditional requests will be significantly cheaper to
      produce than a full response. For static files, typically all that is
      involved is a call to <code>stat()</code> or similar system call, to
      see if the file has changed in size or modification time. As such, even
      local content may still be served faster from the cache if it has not
      changed.</p>
      
      <p>Origin servers should make every effort to support conditional
      requests as is practical, however if conditional requests are not
      supported, the origin will respond as if the request was not
      conditional, and the cache will respond as if the content had changed
      and save the new content to the cache. In this case, the cache will
      behave like a simple two state cache, where content is effectively
      either fresh or deleted.</p>
    </section>

    <section>
      <title>What Can be Cached?</title>

      <p>The full definition of which responses can be cached by an HTTP
      cache is defined in
      <a href="http://www.w3.org/Protocols/rfc2616/rfc2616-sec13.html#sec13.4">
      RFC2616 Section 13.4 Response Cacheability</a>, and can be summed up as
      follows:</p>

      <ol>
        <li>Caching must be enabled for this URL. See the <directive
        module="mod_cache">CacheEnable</directive> and <directive
        module="mod_cache">CacheDisable</directive> directives.</li>

        <li>The response must have a HTTP status code of 200, 203, 300, 301 or
        410.</li>

        <li>The request must be a HTTP GET request.</li>

        <li>If the response contains an "Authorization:" header, it must
        also contain an "s-maxage", "must-revalidate" or "public" option
        in the "Cache-Control:" header, or it won't be cached.</li>

        <li>If the URL included a query string (e.g. from a HTML form GET
        method) it will not be cached unless the response specifies an
        explicit expiration by including an "Expires:" header or the max-age
        or s-maxage directive of the "Cache-Control:" header, as per RFC2616
        sections 13.9 and 13.2.1.</li>

        <li>If the response has a status of 200 (OK), the response must
        also include at least one of the "Etag", "Last-Modified" or
        the "Expires" headers, or the max-age or s-maxage directive of
        the "Cache-Control:" header, unless the
        <directive module="mod_cache">CacheIgnoreNoLastMod</directive>
        directive has been used to require otherwise.</li>

        <li>If the response includes the "private" option in a "Cache-Control:"
        header, it will not be stored unless the
        <directive module="mod_cache">CacheStorePrivate</directive> has been
        used to require otherwise.</li>

        <li>Likewise, if the response includes the "no-store" option in a
        "Cache-Control:" header, it will not be stored unless the
        <directive module="mod_cache">CacheStoreNoStore</directive> has been
        used.</li>

        <li>A response will not be stored if it includes a "Vary:" header
        containing the match-all "*".</li>
      </ol>
    </section>

    <section>
      <title>What Should Not be Cached?</title>

      <p>It should be up to the client creating the request, or the origin
      server constructing the response to decide whether or not the content
      should be cacheable or not by correctly setting the
      <code>Cache-Control</code> header, and <module>mod_cache</module> should
      be left alone to honor the wishes of the client or server as appropriate.
      </p>

      <p>Content that is time sensitive, or which varies depending on the
      particulars of the request that are not covered by HTTP negotiation,
      should not be cached. This content should declare itself uncacheable
      using the <code>Cache-Control</code> header.</p>
      
      <p>If content changes often, expressed by a freshness lifetime of minutes
      or seconds, the content can still be cached, however it is highly
      desirable that the origin server supports
      <strong>conditional requests</strong> correctly to ensure that
      full responses do not have to be generated on a regular basis.</p>

      <p>Content that varies based on client provided request headers can be
      cached through intelligent use of the <code>Vary</code> response
      header.</p>

    </section>

    <section>
      <title>Variable/Negotiated Content</title>

      <p>When the origin server is designed to respond with different content
      based on the value of headers in the request, for example to serve
      multiple languages at the same URL, HTTP's caching mechanism makes it
      possible to cache multiple variants of the same page at the same URL.</p>
      
      <p>This is done by the origin server adding a <code>Vary</code> header
      to indicate which headers must be taken into account by a cache when
      determining whether two variants are different from one another.</p>

      <p>If for example, a response is received with a vary header such as;</p>

      <example>
Vary: negotiate,accept-language,accept-charset
      </example>

      <p><module>mod_cache</module> will only serve the cached content to
      requesters with accept-language and accept-charset headers
      matching those of the original request.</p>
      
      <p>Multiple variants of the content can be cached side by side,
      <module>mod_cache</module> uses the <code>Vary</code> header and the
      corresponding values of the request headers listed by <code>Vary</code>
      to decide on which of many variants to return to the client.</p>
    </section>

    <section id="disk">
      <title>Caching to Disk</title>

      <p>The <module>mod_cache</module> module relies on specific backend store
      implementations in order to manage the cache, and for caching to disk
      <module>mod_cache_disk</module> is provided to support this.</p>

      <p>Typically the module will be configured as so;</p>

      <highlight language="config">
CacheRoot   "/var/cache/apache/"
CacheEnable disk /
CacheDirLevels 2
CacheDirLength 1
      </highlight>

      <p>Importantly, as the cached files are locally stored, operating system
      in-memory caching will typically be applied to their access also. So
      although the files are stored on disk, if they are frequently accessed
      it is likely the operating system will ensure that they are actually
      served from memory.</p>

    </section>

    <section>
      <title>Understanding the Cache-Store</title>

      <p>To store items in the cache, <module>mod_cache_disk</module> creates
      a 22 character hash of the URL being requested. This hash incorporates
      the hostname, protocol, port, path and any CGI arguments to the URL,
      as well as elements defined by the Vary header to ensure that multiple
      URLs do not collide with one another.</p>

      <p>Each character may be any one of 64-different characters, which mean
      that overall there are 64^22 possible hashes. For example, a URL might
      be hashed to <code>xyTGxSMO2b68mBCykqkp1w</code>. This hash is used
      as a prefix for the naming of the files specific to that URL within
      the cache, however first it is split up into directories as per
      the <directive module="mod_cache_disk">CacheDirLevels</directive> and
      <directive module="mod_cache_disk">CacheDirLength</directive>
      directives.</p>

      <p><directive module="mod_cache_disk">CacheDirLevels</directive>
      specifies how many levels of subdirectory there should be, and
      <directive module="mod_cache_disk">CacheDirLength</directive>
      specifies how many characters should be in each directory. With
      the example settings given above, the hash would be turned into
      a filename prefix as
      <code>/var/cache/apache/x/y/TGxSMO2b68mBCykqkp1w</code>.</p>

      <p>The overall aim of this technique is to reduce the number of
      subdirectories or files that may be in a particular directory,
      as most file-systems slow down as this number increases. With
      setting of "1" for
      <directive module="mod_cache_disk">CacheDirLength</directive>
      there can at most be 64 subdirectories at any particular level.
      With a setting of 2 there can be 64 * 64 subdirectories, and so on.
      Unless you have a good reason not to, using a setting of "1"
      for <directive module="mod_cache_disk">CacheDirLength</directive>
      is recommended.</p>

      <p>Setting
      <directive module="mod_cache_disk">CacheDirLevels</directive>
      depends on how many files you anticipate to store in the cache.
      With the setting of "2" used in the above example, a grand
      total of 4096 subdirectories can ultimately be created. With
      1 million files cached, this works out at roughly 245 cached
      URLs per directory.</p>

      <p>Each URL uses at least two files in the cache-store. Typically
      there is a ".header" file, which includes meta-information about
      the URL, such as when it is due to expire and a ".data" file
      which is a verbatim copy of the content to be served.</p>

      <p>In the case of a content negotiated via the "Vary" header, a
      ".vary" directory will be created for the URL in question. This
      directory will have multiple ".data" files corresponding to the
      differently negotiated content.</p>
    </section>

    <section>
      <title>Maintaining the Disk Cache</title>

      <p>The <module>mod_cache_disk</module> module makes no attempt to
      regulate the amount of disk space used by the cache, although it
      will gracefully stand down on any disk error and behave as if the
      cache was never present.</p>

      <p>Instead, provided with httpd is the <a
      href="programs/htcacheclean.html">htcacheclean</a> tool which allows you
      to clean the cache periodically. Determining how frequently to run <a
      href="programs/htcacheclean.html">htcacheclean</a> and what target size to
      use for the cache is somewhat complex and trial and error may be needed to
      select optimal values.</p>

      <p><a href="programs/htcacheclean.html">htcacheclean</a> has two modes of
      operation. It can be run as persistent daemon, or periodically from
      cron. <a
      href="programs/htcacheclean.html">htcacheclean</a> can take up to an hour
      or more to process very large (tens of gigabytes) caches and if you are
      running it from cron it is recommended that you determine how long a typical
      run takes, to avoid running more than one instance at a time.</p>
 
      <p>It is also recommended that an appropriate "nice" level is chosen for
      htcacheclean so that the tool does not cause excessive disk io while the
      server is running.</p>

      <p class="figure">
      <img src="images/caching_fig1.gif" alt="" width="600"
          height="406" /><br />
      <a id="figure1" name="figure1"><dfn>Figure 1</dfn></a>: Typical
      cache growth / clean sequence.</p>

      <p>Because <module>mod_cache_disk</module> does not itself pay attention
      to how much space is used you should ensure that
      <a href="programs/htcacheclean.html">htcacheclean</a> is configured to
      leave enough "grow room" following a clean.</p>
    </section>

  </section>

  <section id="socache-caching">

    <title>Two-state Key/Value Shared Object Caching</title>

    <related>
      <modulelist>
        <module>mod_authn_socache</module>
        <module>mod_socache_dbm</module>
        <module>mod_socache_dc</module>
        <module>mod_socache_memcache</module>
        <module>mod_socache_shmcb</module>
        <module>mod_ssl</module>
      </modulelist>
        <directivelist>
          <directive module="mod_authn_socache">AuthnCacheSOCache</directive>
          <directive module="mod_ssl">SSLSessionCache</directive>
          <directive module="mod_ssl">SSLStaplingCache</directive>
        </directivelist>
    </related>
    
    <p>The Apache HTTP server offers a low level shared object cache for
    caching information such as SSL sessions, or authentication credentials,
    within the <a href="socache.html">socache</a> interface.</p>

    <p>Additional modules are provided for each implementation, offering the
    following backends:</p>

    <dl>
    <dt><module>mod_socache_dbm</module></dt>
    <dd>DBM based shared object cache.</dd>
    <dt><module>mod_socache_dc</module></dt>
    <dd>Distcache based shared object cache.</dd>
    <dt><module>mod_socache_memcache</module></dt>
    <dd>Memcache based shared object cache.</dd>
    <dt><module>mod_socache_shmcb</module></dt>
    <dd>Shared memory based shared object cache.</dd>
    </dl>

    <section id="mod_authn_socache-caching">
      <title>Caching Authentication Credentials</title>

      <related>
        <modulelist>
          <module>mod_authn_socache</module>
        </modulelist>
        <directivelist>
          <directive module="mod_authn_socache">AuthnCacheSOCache</directive>
        </directivelist>
      </related>

      <p>The <module>mod_authn_socache</module> module allows the result of
      authentication to be cached, relieving load on authentication backends.</p>

    </section>

    <section id="mod_ssl-caching">
      <title>Caching SSL Sessions</title>

      <related>
        <modulelist>
          <module>mod_ssl</module>
        </modulelist>
        <directivelist>
          <directive module="mod_ssl">SSLSessionCache</directive>
          <directive module="mod_ssl">SSLStaplingCache</directive>
        </directivelist>
      </related>

      <p>The <module>mod_ssl</module> module uses the <code>socache</code> interface
      to provide a session cache and a stapling cache.</p>

    </section>

  </section>

  <section id="file-caching">

    <title>Specialized File Caching</title>

    <related>
      <modulelist>
        <module>mod_file_cache</module>
      </modulelist>
      <directivelist>
        <directive module="mod_file_cache">CacheFile</directive>
        <directive module="mod_file_cache">MMapFile</directive>
      </directivelist>
    </related>

    <p>On platforms where a filesystem might be slow, or where file
    handles are expensive, the option exists to pre-load files into
    memory on startup.</p>

    <p>On systems where opening files is slow, the option exists to
    open the file on startup and cache the file handle. These
    options can help on systems where access to static files is
    slow.</p>

    <section id="filehandle">
      <title>File-Handle Caching</title>

      <p>The act of opening a file can itself be a source of delay, particularly
      on network filesystems. By maintaining a cache of open file descriptors
      for commonly served files, httpd can avoid this delay. Currently httpd
      provides one implementation of File-Handle Caching.</p>

      <section>
        <title>CacheFile</title>

        <p>The most basic form of caching present in httpd is the file-handle
        caching provided by <module>mod_file_cache</module>. Rather than caching
        file-contents, this cache maintains a table of open file descriptors. Files
        to be cached in this manner are specified in the configuration file using
        the <directive module="mod_file_cache">CacheFile</directive>
        directive.</p>

        <p>The
        <directive module="mod_file_cache">CacheFile</directive> directive
        instructs httpd to open the file when it is started and to re-use
        this file-handle for all subsequent access to this file.</p>

        <example>
        CacheFile /usr/local/apache2/htdocs/index.html
        </example>

        <p>If you intend to cache a large number of files in this manner, you
        must ensure that your operating system's limit for the number of open
        files is set appropriately.</p>

        <p>Although using <directive module="mod_file_cache">CacheFile</directive>
        does not cause the file-contents to be cached per-se, it does mean
        that if the file changes while httpd is running these changes will
        not be picked up. The file will be consistently served as it was
        when httpd was started.</p>

        <p>If the file is removed while httpd is running, it will continue
        to maintain an open file descriptor and serve the file as it was when
        httpd was started. This usually also means that although the file
        will have been deleted, and not show up on the filesystem, extra free
        space will not be recovered until httpd is stopped and the file
        descriptor closed.</p>
      </section>

    </section>

    <section id="inmemory">
      <title>In-Memory Caching</title>

      <p>Serving directly from system memory is universally the fastest method
      of serving content. Reading files from a disk controller or, even worse,
      from a remote network is orders of magnitude slower. Disk controllers
      usually involve physical processes, and network access is limited by
      your available bandwidth. Memory access on the other hand can take mere
      nano-seconds.</p>

      <p>System memory isn't cheap though, byte for byte it's by far the most
      expensive type of storage and it's important to ensure that it is used
      efficiently. By caching files in memory you decrease the amount of
      memory available on the system. As we'll see, in the case of operating
      system caching, this is not so much of an issue, but when using
      httpd's own in-memory caching it is important to make sure that you
      do not allocate too much memory to a cache. Otherwise the system
      will be forced to swap out memory, which will likely degrade
      performance.</p>

      <section>
        <title>Operating System Caching</title>

        <p>Almost all modern operating systems cache file-data in memory managed
        directly by the kernel. This is a powerful feature, and for the most
        part operating systems get it right. For example, on Linux, let's look at
        the difference in the time it takes to read a file for the first time
        and the second time;</p>

        <example><pre>
colm@coroebus:~$ time cat testfile &gt; /dev/null
real    0m0.065s
user    0m0.000s
sys     0m0.001s
colm@coroebus:~$ time cat testfile &gt; /dev/null
real    0m0.003s
user    0m0.003s
sys     0m0.000s</pre>
        </example>

        <p>Even for this small file, there is a huge difference in the amount
        of time it takes to read the file. This is because the kernel has cached
        the file contents in memory.</p>

        <p>By ensuring there is "spare" memory on your system, you can ensure
        that more and more file-contents will be stored in this cache. This
        can be a very efficient means of in-memory caching, and involves no
        extra configuration of httpd at all.</p>

        <p>Additionally, because the operating system knows when files are
        deleted or modified, it can automatically remove file contents from the
        cache when necessary. This is a big advantage over httpd's in-memory
        caching which has no way of knowing when a file has changed.</p>
      </section>

      <p>Despite the performance and advantages of automatic operating system
      caching there are some circumstances in which in-memory caching may be
      better performed by httpd.</p>

      <section>
        <title>MMapFile Caching</title>

        <p><module>mod_file_cache</module> provides the
        <directive module="mod_file_cache">MMapFile</directive> directive, which
        allows you to have httpd map a static file's contents into memory at
        start time (using the mmap system call). httpd will use the in-memory
        contents for all subsequent accesses to this file.</p>

        <example>
        MMapFile /usr/local/apache2/htdocs/index.html
        </example>

        <p>As with the
        <directive module="mod_file_cache">CacheFile</directive> directive, any
        changes in these files will not be picked up by httpd after it has
        started.</p>

        <p> The <directive module="mod_file_cache">MMapFile</directive>
        directive does not keep track of how much memory it allocates, so
        you must ensure not to over-use the directive. Each httpd child
        process will replicate this memory, so it is critically important
        to ensure that the files mapped are not so large as to cause the
        system to swap memory.</p>
      </section>
    </section>

  </section>

  <section id="security">
    <title>Security Considerations</title>

    <section>
      <title>Authorization and Access Control</title>

      <p>Using <module>mod_cache</module> in its default state where
      <directive module="mod_cache">CacheQuickHandler</directive> is set to
      <code>On</code> is very much like having a caching reverse-proxy bolted
      to the front of the server. Requests will be served by the caching module
      unless it determines that the origin server should be queried just as an
      external cache would, and this drastically changes the security model of
      httpd.</p>

      <p>As traversing a filesystem hierarchy to examine potential
      <code>.htaccess</code> files would be a very expensive operation,
      partially defeating the point of caching (to speed up requests),
      <module>mod_cache</module> makes no decision about whether a cached
      entity is authorised for serving. In other words; if
      <module>mod_cache</module> has cached some content, it will be served
      from the cache as long as that content has not expired.</p>

      <p>If, for example, your configuration permits access to a resource by IP
      address you should ensure that this content is not cached. You can do this
      by using the <directive module="mod_cache">CacheDisable</directive>
      directive, or <module>mod_expires</module>. Left unchecked,
      <module>mod_cache</module> - very much like a reverse proxy - would cache
      the content when served and then serve it to any client, on any IP
      address.</p>

      <p>When the <directive module="mod_cache">CacheQuickHandler</directive>
      directive is set to <code>Off</code>, the full set of request processing
      phases are executed and the security model remains unchanged.</p>
    </section>

    <section>
      <title>Local exploits</title>

      <p>As requests to end-users can be served from the cache, the cache
      itself can become a target for those wishing to deface or interfere with
      content. It is important to bear in mind that the cache must at all
      times be writable by the user which httpd is running as. This is in
      stark contrast to the usually recommended situation of maintaining
      all content unwritable by the Apache user.</p>

      <p>If the Apache user is compromised, for example through a flaw in
      a CGI process, it is possible that the cache may be targeted. When
      using <module>mod_cache_disk</module>, it is relatively easy to
      insert or modify a cached entity.</p>

      <p>This presents a somewhat elevated risk in comparison to the other
      types of attack it is possible to make as the Apache user. If you are
      using <module>mod_cache_disk</module> you should bear this in mind -
      ensure you upgrade httpd when security upgrades are announced and
      run CGI processes as a non-Apache user using <a
      href="suexec.html">suEXEC</a> if possible.</p>

    </section>

    <section>
      <title>Cache Poisoning</title>

      <p>When running httpd as a caching proxy server, there is also the
      potential for so-called cache poisoning. Cache Poisoning is a broad
      term for attacks in which an attacker causes the proxy server to
      retrieve incorrect (and usually undesirable) content from the origin
      server.</p>

      <p>For example if the DNS servers used by your system running httpd
      are vulnerable to DNS cache poisoning, an attacker may be able to control
      where httpd connects to when requesting content from the origin server.
      Another example is so-called HTTP request-smuggling attacks.</p>

      <p>This document is not the correct place for an in-depth discussion
      of HTTP request smuggling (instead, try your favourite search engine)
      however it is important to be aware that it is possible to make
      a series of requests, and to exploit a vulnerability on an origin
      webserver such that the attacker can entirely control the content
      retrieved by the proxy.</p>
    </section>

    <section>
      <title>Denial of Service / Cachebusting</title>

      <p>The Vary mechanism allows multiple variants of the same URL to be
      cached side by side. Depending on header values provided by the client,
      the cache will select the correct variant to return to the client. This
      mechanism can become a problem when an attempt is made to vary on a
      header that is known to contain a wide range of possible values under
      normal use, for example the <code>User-Agent</code> header. Depending
      on the popularity of the particular web site thousands or millions of
      duplicate cache entries could be created for the same URL, crowding
      out other entries in the cache.</p>
      
      <p>In other cases, there may be a need to change the URL of a particular
      resource on every request, usually by adding a "cachebuster" string to
      the URL. If this content is declared cacheable by a server for a
      significant freshness lifetime, these entries can crowd out
      legitimate entries in a cache. While <module>mod_cache</module>
      provides a
      <directive module="mod_cache">CacheIgnoreURLSessionIdentifiers</directive>
      directive, this directive should be used with care to ensure that
      downstream proxy or browser caches aren't subjected to the same denial
      of service issue.</p>
    </section>
  </section>

</manualpage>