htcacheclean.xml revision 78f97ce162b66a0dbfd7af4dcd9984f162569b04
<?xml version='1.0' encoding='UTF-8' ?>
<!-- $LastChangedRevision$ -->
<!--
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
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="htcacheclean.xml.meta">
<parentdocument href="./">Programs</parentdocument>
<title>htcacheclean - Clean up the disk cache</title>
<summary>
<p><code>htcacheclean</code> is used to keep the size of
<module>mod_cache_disk</module>'s storage within a given size limit, or
limit on inodes in use. This tool can run either manually or in daemon mode.
When running in daemon mode, it sleeps in the background and checks the cache
directory at regular intervals for cached content to be removed. You can stop
the daemon cleanly by sending it a TERM or INT signal. When run manually, a
once off check of the cache directory is made for cached content to be
removed. If one or more URLs are specified, each URL will be deleted from
the cache, if present.</p>
</summary>
<seealso><module>mod_cache_disk</module></seealso>
<section id="synopsis"><title>Synopsis</title>
<p><code><strong>htcacheclean</strong>
[ -<strong>D</strong> ]
[ -<strong>v</strong> ]
[ -<strong>t</strong> ]
[ -<strong>r</strong> ]
[ -<strong>n</strong> ]
[ -<strong>R</strong><var>round</var> ]
-<strong>p</strong><var>path</var>
[-<strong>l</strong><var>limit</var>|
-<strong>L</strong><var>limit</var>]</code></p>
<p><code><strong>htcacheclean</strong>
[ -<strong>n</strong> ]
[ -<strong>t</strong> ]
[ -<strong>i</strong> ]
[ -<strong>P</strong><var>pidfile</var> ]
[ -<strong>R</strong><var>round</var> ]
-<strong>d</strong><var>interval</var>
-<strong>p</strong><var>path</var>
[-<strong>l</strong><var>limit</var>|
-<strong>L</strong><var>limit</var>]</code></p>
<p><code><strong>htcacheclean</strong>
[ -<strong>v</strong> ]
[ -<strong>R</strong><var>round</var> ]
-<strong>p</strong><var>path</var>
[ -<strong>a</strong> ]
[ -<strong>A</strong> ]</code></p>
<p><code><strong>htcacheclean</strong>
[ -<strong>D</strong> ]
[ -<strong>v</strong> ]
[ -<strong>t</strong> ]
[ -<strong>R</strong><var>round</var> ]
-<strong>p</strong><var>path</var>
<var>url</var></code></p>
</section>
<section id="options"><title>Options</title>
<dl>
<dt><code>-d<var>interval</var></code></dt>
<dd>Daemonize and repeat cache cleaning every <var>interval</var> minutes.
This option is mutually exclusive with the <code>-D</code>, <code>-v</code>
and <code>-r</code> options. To shutdown the daemon cleanly, just send it
a <code>SIGTERM</code> or <code>SIGINT</code>.</dd>
<dt><code>-D</code></dt>
<dd>Do a dry run and don't delete anything. This option is mutually
exclusive with the <code>-d</code> option. When doing a dry run and
deleting directories with <code>-t</code>, the inodes reported deleted
in the stats cannot take into account the directories deleted, and will
be marked as an estimate.</dd>
<dt><code>-v</code></dt>
<dd>Be verbose and print statistics. This option is mutually exclusive
with the <code>-d</code> option.</dd>
<dt><code>-r</code></dt>
<dd>Clean thoroughly. This assumes that the Apache web server is
not running (otherwise you may get garbage in the cache). This option
is mutually exclusive with the <code>-d</code> option and implies
the <code>-t</code> option.</dd>
<dt><code>-n</code></dt>
<dd>Be nice. This causes slower processing in favour of other
processes. <code>htcacheclean</code> will sleep from time to time
so that (a) the disk IO will be delayed and (b) the kernel can schedule
other processes in the meantime.</dd>
<dt><code>-t</code></dt>
<dd>Delete all empty directories. By default only cache files are
removed, however with some configurations the large number of
directories created may require attention. If your configuration
requires a very large number of directories, to the point that
inode or file allocation table exhaustion may become an issue, use
of this option is advised.</dd>
<dt><code>-p<var>path</var></code></dt>
<dd>Specify <var>path</var> as the root directory of the disk cache. This
should be the same value as specified with the <directive
module="mod_cache_disk">CacheRoot</directive> directive.</dd>
<dt><code>-P<var>pidfile</var></code></dt>
<dd>Specify <var>pidfile</var> as the name of the file to write the
process ID to when daemonized.</dd>
<dt><code>-R<var>round</var></code></dt>
<dd>Specify <var>round</var> as the amount to round sizes up to, to
compensate for disk block sizes. Set to the block size of the cache
partition.</dd>
<dt><code>-l<var>limit</var></code></dt>
<dd>Specify <var>limit</var> as the total disk cache size limit. The value
is expressed in bytes by default (or attaching <code>B</code> to the
number). Attach <code>K</code> for Kbytes or <code>M</code> for
MBytes.</dd>
<dt><code>-L<var>limit</var></code></dt>
<dd>Specify <var>limit</var> as the total disk cache inode limit.</dd>
<dt><code>-i</code></dt>
<dd>Be intelligent and run only when there was a modification of the disk
cache. This option is only possible together with the <code>-d</code>
option.</dd>
<dt><code>-a</code></dt>
<dd>List the URLs currently stored in the cache. Variants of the same URL
will be listed once for each variant.</dd>
<dt><code>-A</code></dt>
<dd>List the URLs currently stored in the cache, along with their
attributes in the following order: url, header size, body size, status,
entity version, date, expiry, request time, response time, body present,
head request.</dd>
</dl>
</section>
<section id="delete"><title>Deleting a specific URL</title>
<p>If <code>htcacheclean</code> is passed one or more URLs, each URL will
be deleted from the cache. If multiple variants of an URL exists, all
variants would be deleted.</p>
<p>When a reverse proxied URL is to be deleted, the effective URL is
constructed from the <strong>Host</strong> header, the
<strong>port</strong>, the <strong>path</strong> and the
<strong>query</strong>. Note the '?' in the URL must always be specified
explicitly, whether a query string is present or not. For example, an
attempt to delete the path <strong>/</strong> from the server
<strong>localhost</strong>, the URL to delete would be
</section>
<section id="list"><title>Listing URLs in the Cache</title>
<p>By passing the <code>-a</code> or <code>-A</code> options to
<code>htcacheclean</code>, the URLs within the cache will be listed
as they are found, one URL per line. The <code>-A</code> option
dumps the full cache entry after the URL, with fields in the
following order:</p>
<dl>
<dt>url</dt><dd>The URL of the entry.</dd>
<dt>header size</dt><dd>The size of the header in bytes.</dd>
<dt>body size</dt><dd>The size of the body in bytes.</dd>
<dt>status</dt><dd>Status of the cached response.</dd>
<dt>entity version</dt><dd>The number of times this entry has been
revalidated without being deleted.</dd>
<dt>date</dt><dd>Date of the response.</dd>
<dt>expiry</dt><dd>Expiry date of the response.</dd>
<dt>request time</dt><dd>Time of the start of the request.</dd>
<dt>response time</dt><dd>Time of the end of the request.</dd>
<dt>body present</dt><dd>If 0, no body is stored with this request,
1 otherwise.</dd>
<dt>head request</dt><dd>If 1, the entry contains a cached HEAD
request with no body, 0 otherwise.</dd>
</dl>
</section>
<section id="exit"><title>Exit Status</title>
<p><code>htcacheclean</code> returns a zero status ("true") if all
operations were successful, <code>1</code> otherwise. If an URL is
specified, and the URL was cached and successfully removed,
<code>0</code> is returned, <code>2</code> otherwise. If an error
occurred during URL removal, <code>1</code> is returned.</p>
</section>
</manualpage>