Google Git
Sign in
apache / httpd / 5422fe57879e6867ad06b52ba861ded9b7dd0b91 / . / docs / manual / mod / core.html
blob: c7225d732b192c83b0cdf94a8d2f37f8086ff432 [file] [log] [blame]
<!--%hypertext -->
<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN">
<HTML>
<HEAD>
<TITLE>Apache Core Features</TITLE>
</HEAD>
<BODY>
<IMG SRC="../images/apache_sub.gif" ALT="">
<!--/%hypertext -->
<H1>Apache Core Features</h1>
These configuration parameters control the core Apache features, and are
always available.
<!--%hypertext -->
<ul>
<li><A HREF="#accessconfig">AccessConfig</A>
<li><A HREF="#accessfilename">AccessFileName</A>
<li><A HREF="#allowoverride">AllowOverride</A>
<li><A HREF="#authname">AuthName</A>
<li><A HREF="#authtype">AuthType</A>
<li><A HREF="#bindaddress">BindAdress</A>
<li><A HREF="#defaulttype">DefaultType</A>
<li><A HREF="#directory">&lt;Directory&gt;</A>
<li><A HREF="#documentroot">DocumentRoot</A>
<li><A HREF="#errordocument">ErrorDocument</A>
<li><A HREF="#errorlog">ErrorLog</A>
<li><A HREF="#group">Group</A>
<li><A HREF="#identitycheck">IdentityCheck</A>
<li><A HREF="#keepalive">KeepAlive</A>
<li><A HREF="#keepalivetimeout">KeepAliveTimeout</A>
<li><A HREF="#limit">&lt;Limit&gt;</A>
<li><A HREF="#listen">Listen;</A>
<li><A HREF="#location">&lt;Location&gt;</A>
<li><A HREF="#maxclients">MaxClients</A>
<li><A HREF="#maxrequestsperchild">MaxRequestsPerChild</A>
<li><A HREF="#maxspareservers">MaxSpareServers</A>
<li><A HREF="#minspareservers">MinSpareServers</A>
<li><A HREF="#options">Options</A>
<li><A HREF="#pidfile">PidFile</A>
<li><A HREF="#port">Port</A>
<li><A HREF="#require">require</A>
<li><A HREF="#resourceconfig">ResourceConfig</A>
<li><A HREF="#serveradmin">ServerAdmin</A>
<li><A HREF="#serveralias">ServerAlias</A>
<li><A HREF="#servername">ServerName</A>
<li><A HREF="#serverpath">ServerPath</A>
<li><A HREF="#serverroot">ServerRoot</A>
<li><A HREF="#servertype">ServerType</A>
<li><A HREF="#startservers">StartServers</A>
<li><A HREF="#timeout">TimeOut</A>
<li><A HREF="#user">User</A>
<li><A HREF="#virtualhost">&lt;VirtualHost&gt;</A>
</ul>
<hr>
<!--/%hypertext -->
<A name="accessconfig"><h2>AccessConfig directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt AccessConfig} directive&gt; -->
<strong>Syntax:</strong> AccessConfig <em>filename</em><br>
<strong>Default:</strong> <code>AccessConfig conf/access.conf</code><br>
<strong>Context:</strong> server config, virtual host<br>
<strong>Status:</strong> core<p>
The server will read this file for more directives after reading the
<A HREF="#resourceconfig">ResourceConfig</A> file. <em>Filename</em> is
relative to the <A HREF="#serverroot">ServerRoot</A>.
This feature can be disabled using:
<blockquote><code>AccessConfig /dev/null</code></blockquote>
Historically, this file only contained
<A HREF="#directory">&lt;Directory&gt;</A> sections; in fact it can now
contain any server directive allowed in the <em>server config</em> context.
<p><hr>
<A name="accessfilename"><h2>AccessFileName directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt AccessFileName} directive&gt; -->
<strong>Syntax:</strong> AccessFileName <em>filename</em><br>
<strong>Default:</strong> <code>AccessFileName .htaccess</code><br>
<strong>Context:</strong> server config, virtual host<br>
<strong>Status:</strong> core<p>
When returning a document to the client the server looks for an
access control file with this name in every directory of the path to
the document, if access control files are enabled for that directory.
For example:
<blockquote><code>AccessFileName .acl</code></blockquote>
before returning the document /usr/local/web/index.html, the
server will read /.acl, /usr/.acl, /usr/local/.acl and /usr/local/web/.acl
for directives, unless they have been disabled with
<blockquote><code>
&lt;Directory /&gt;<br>
AllowOverride None<br>
&lt;/Directory&gt;</code></blockquote><p><hr>
<A name="allowoverride"><h2>AllowOverride directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt AllowOverride} directive&gt; -->
<strong>Syntax:</strong> AllowOverride <em>override override ...</em><br>
<strong>Default:</strong> <code>AllowOverride All</code><br>
<strong>Context:</strong> directory<br>
<strong>Status:</strong> core<p>
When the server finds an .htaccess file (as specified by
<A HREF="#accessfilename">AccessFileName</A>) it needs to know which
directives declared in that file can override earlier access information.<p>
<em>Override</em> can be set to <code>None</code>, in which case the server
will not read the file, <code>All</code> in which case the server will
allow all the directives, or one or more of the following:
<dl>
<dt>AuthConfig
<dd>
<!--%plaintext &lt;?INDEX {\tt AuthConfig} override&gt; -->
Allow use of the authorization directives
(<A HREF="mod_auth_dbm.html#authdbmgroupfile">AuthDBMGroupFile</A>,
<A HREF="mod_auth_dbm.html#authdbmuserfile">AuthDBMUserFile</A>,
<A HREF="mod_auth.html#authgroupfile">AuthGroupFile</A>,
<A HREF="#authname">AuthName</A>, <A HREF="#authtype">AuthType</A>,
<A HREF="mod_auth.html#authuserfile">AuthUserFile</A> and
<A HREF="#require">require</A>).
<dt>FileInfo
<dd>
<!--%plaintext &lt;?INDEX {\tt FileInfo} override&gt; -->
Allow use of the directives controlling document types
(<A HREF="mod_mime.html#addencoding">AddEncoding</A>,
<A HREF="mod_mime.html#addlanguage">AddLanguage</A>,
<A HREF="mod_mime.html#addtype">AddType</A>,
<A HREF="#defaulttype">DefaultType</A> and
<A HREF="mod_negotiation.html#languagepriority">LanguagePriority</A>).
<dt>Indexes
<dd>
<!--%plaintext &lt;?INDEX {\tt Indexes} override&gt; -->
Allow use of the directives controlling directory indexing
(<A HREF="mod_dir.html#adddescription">AddDescription</A>,
<A HREF="mod_dir.html#addicon">AddIcon</A>,
<A HREF="mod_dir.html#addiconbyencoding">AddIconByEncoding</A>,
<A HREF="mod_dir.html#addiconbytype">AddIconByType</A>,
<A HREF="mod_dir.html#defaulticon">DefaultIcon</A>,
<A HREF="mod_dir.html#directoryindex">DirectoryIndex</A>,
<A HREF="mod_dir.html#fancyindexing">FancyIndexing</A>,
<A HREF="mod_dir.html#headername">HeaderName</A>,
<A HREF="mod_dir.html#indexignore">IndexIgnore</A>,
<A HREF="mod_dir.html#indexoptions">IndexOptions</A> and
<A HREF="mod_dir.html#readmename">ReadmeName</A>).
<dt>Limit
<dd>
<!--%plaintext &lt;?INDEX {\tt Limit} override&gt; -->
Allow use of the directives controlling host access (allow, deny and order).
<dt>Options
<dd>
<!--%plaintext &lt;?INDEX {\tt Options} override&gt; -->
Allow use of the directives controlling specific directory features
(<A HREF="#options">Options</A> and
<A HREF="mod_include.html#xbithack">XBitHack</A>).
</dl><p><hr>
<A name="authname"><h2>AuthName directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt AuthName} directive&gt; -->
<strong>Syntax:</strong> AuthName <em>auth-domain</em><br>
<strong>Context:</strong> directory, .htaccess<br>
<strong>Override:</strong> AuthConfig<br>
<strong>Status:</strong> core<p>
This directive sets the name of the authorization realm for a directory.
This realm is given to the client so that the user knows which username and
password to send.
It must be accompanied by <A HREF="#authtype">AuthType</A> and
<A HREF="#require">require</A> directives, and directives such as
<A HREF="mod_auth.html#authuserfile">AuthUserFile</A> and
<A HREF="mod_auth.html#authgroupfile">AuthGroupFile</A> to work.<p><hr>
<A name="authtype"><h2>AuthType directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt AuthType} directive&gt; -->
<strong>Syntax:</strong> AuthType <em>type</em><br>
<strong>Context:</strong> directory, .htaccess<br>
<strong>Override:</strong> AuthConfig<br>
<strong>Status:</strong> core<p>
This directive selects the type of user authentication for a directory.
Only <code>Basic</code> is currently implemented.
<!--%plaintext &lt;?INDEX {\tt Basic} authentication scheme&gt; -->
It must be accompanied by <A HREF="#authname">AuthName</A> and
<A HREF="#require">require</A> directives, and directives such as
<A HREF="mod_auth.html#authuserfile">AuthUserFile</A> and
<A HREF="mod_auth.html#authgroupfile">AuthGroupFile</A> to work.<p><hr>
<A name="bindaddress"><h2>BindAddress directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt BindAddress} directive&gt; -->
<strong>Syntax:</strong> BindAddress <em>saddr</em><br>
<strong>Default:</strong> <code>BindAddress *</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
A Unix&#174; http server can either listen on for connections to every
IP address of the server machine, or just one IP address of the server
machine. <em>Saddr</em> can be
<menu>
<li>*
<li>An IP address
<li>A fully-qualified internet domain name
</menu>
If the value is *, then the server will listen for connections on
every IP address, otherwise it will only listen on the IP address
specified. <p>
This option can be used as an alternative method for supporting
<A HREF="virtual-host.html">virtual hosts</A> instead of using
<A HREF="#virtualhost">&lt;VirtualHost&gt;</A> sections.<p><hr>
<p><strong>See Also:</strong>
<a href="bind.html">Setting which addresses and ports Apache uses</a></p>
<A name="defaulttype"><h2>DefaultType directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt DefaultType} directive&gt; -->
<strong>Syntax:</strong> DefaultType <em>mime-type</em><br>
<strong>Default:</strong> <code>DefaultType text/html</code><br>
<strong>Context:</strong> server config, virtual host, directory, .htaccess<br>
<strong>Override:</strong> FileInfo<br>
<strong>Status:</strong> core<p>
There will be times when the server is asked to provide a document
whose type cannot be determined by its MIME types mappings.<p>
The server must inform the client of the content-type of the document, so in
the event of an unknown type it uses the <CODE>DefaultType</CODE>. For
example:
<blockquote><code>DefaultType image/gif</code></blockquote>
would be appropriate for a directory which contained many gif images
with filenames missing the .gif extension.<p><hr>
<A name="directory"><h2>&lt;Directory&gt; directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt Directory} section directive&gt; -->
<strong>Syntax:</strong> &lt;Directory <em>directory</em>&gt; ...
&lt;/Directory&gt; <br>
<strong>Context:</strong> server config, virtual host<br>
<strong>Status:</strong> Core. <p>
&lt;Directory&gt; and &lt;/Directory&gt; are used to enclose a group of
directives which will apply only to the named directory and sub-directories
of that directory. Any directive which is allowed in a directory
context may be used. <em>Directory</em> is either the full path to a directory,
or a wildcard string. In a wildcard string, `?' matches any single character,
and `*' matches any sequences of characters. Example:
<blockquote>
<code>
&lt;Directory /usr/local/httpd/htdocs&gt;<br>
Options Indexes FollowSymLinks<br>
&lt;/Directory&gt;</code></blockquote>
If multiple directory sections match the directory (or its parents) containing
a document, then the directives are applied in the order of shortest match
first, interspersed with the directives from the
<A HREF="#accessfilename">.htaccess</A> files. For example, with
<blockquote><code>
&lt;Directory /&gt;<br>
AllowOverride None<br>
&lt;/Directoy&gt;<br><br>
&lt;Directory /home/*&gt;<br>
AllowOverride FileInfo<br>
&lt;/Directory&gt;</code></blockquote>
The for access to the document <code>/home/web/dir/doc.html</code> the
steps are:
<menu>
<li>Apply directive <code>AllowOverride None</code> (disabling
<code>.htaccess</code> files).
<li>Apply directive <code>AllowOverride FileInfo</code> (for directory
<code>/home/web</code>).
<li>Apply any FileInfo directives in <code>/home/web/.htaccess</code>
</menu>
The directory sections typically occur in the access.conf file, but they
may appear in any configuration file. &lt;Directory&gt; directives cannot
nest, and cannot appear in a <A HREF="#limit">&lt;Limit&gt;</A> section.
<p><hr>
<A NAME="documentroot"><h2>DocumentRoot directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt DocumentRoot} directive&gt; -->
<strong>Syntax:</strong> DocumentRoot <em>directory-filename</em><br>
<strong>Default:</strong> <code>DocumentRoot
/usr/local/etc/httpd/htdocs</code><br>
<strong>Context:</strong> server config, virtual host<br>
<strong>Status:</strong> core<p>
This directive sets the directory from which httpd will serve files.
Unless matched by a directive like Alias, the server appends the path
from the requested URL to the document root to make the path to the
document. Example:
<blockquote><code>DocumentRoot /usr/web</code></blockquote>
then an access to <code>http://www.my.host.com/index.html</code> refers
to <code>/usr/web/index.html</code>.
<P>There appears to be a bug in mod_dir which causes problems when the
DocumentRoot has a trailing slash (i.e. "DocumentRoot /usr/web/") so
please avoid that.
<p><hr>
<A name="errordocument"><h2>ErrorDocument directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt ErrorDocument} directive&gt; -->
<strong>Syntax:</strong> ErrorDocument <em>error-code document</em><br>
<strong>Context</strong> server config, virtual host, directory, .htaccess<br>
<strong>Status:</strong> core<br>
<strong>Compatibility:</strong> The directory and .htaccess contexts
are only available in Apache 1.1 and later.<p>
In the event of a problem or error, Apache can be configured to do
one of four things,
<OL>
<LI>behave like NCSA httpd 1.3
<LI>output a customized message
<LI>redirect to a local URL to handle the problem/error
<LI>redirect to an external URL to handle the problem/error
</OL>
<P>2-4 are configured using <CODE>ErrorDocument</CODE>, which
is followed by the HTTP response code and a message or URL.
<P><em>Messages</em> in this context, begin with a single quote
(<code>"</code>), which does not form part of the message itself. Apache will
sometime offer additional information regarding the problem/error. This can be
embedded into the message using <code>%s</code>
<P>URLs will begin with a slash (/) for local URLs, or will be a full
URL which the client can resolve. Examples:
<blockquote><code>
ErrorDocument 500 /cgi-bin/tester<br>
ErrorDocument 404 /cgi-bin/bad_urls.pl<br>
ErrorDocument 401 http://www2.foo.bar/subscription_info.html<br>
ErrorDocument 403 "Sorry can't allow you access today
</code></blockquote>
See Also: <A HREF="custom-error.html">documentation of customizable
responses.</A><p><hr>
<A name="errorlog"><h2>ErrorLog directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt ErrorLog} directive&gt; -->
<strong>Syntax:</strong> ErrorLog <em>filename</em><br>
<strong>Default:</strong> <code>ErrorLog logs/error_log</code><br>
<strong>Context:</strong> server config, virtual host<br>
<strong>Status:</strong> core<p>
The error log directive sets the name of the file to which the server will log
any errors it encounters. If the filename does not begin with a slash (/)
then it is assumed to be relative to the <A HREF="#serverroot">ServerRoot</A>.
Example:
<blockquote><code>ErrorLog /dev/null</code></blockquote>
This effectively turns off error logging.<p><hr>
<A name="group"><h2>Group directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt Group} directive&gt; -->
<strong>Syntax:</strong> Group <em>unix-group</em><br>
<strong>Default:</strong> <code>Group #-1</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The Group directive sets the group under which the server will answer requests.
In order to use this directive, the standalone server must be run initially
as root. <em>Unix-group</em> is one of:
<dl>
<dt>A group name
<dd>Refers to the given group by name.
<dt># followed by a group number.
<dd>Refers to a group by its number.
</dl>
It is recommended that you set up a new group specifically for running the
server. Some admins use user <code>nobody</code>, but this is not always
possible or desirable.<p>
Note: if you start the server as a non-root user, it will fail to change
to the specified, and will instead continue to run as the group of the
original user. <p>
SECURITY: See <A HREF="#user">User</A> for a discussion of the security
considerations.<p><hr>
<A name="identitycheck"><h2>IdentityCheck directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt IdentityCheck} directive&gt; -->
<strong>Syntax:</strong> IdentityCheck <em>boolean</em><br>
<strong>Default:</strong> <code>IdentityCheck off</code><br>
<strong>Context:</strong> server config, virtual host, directory, .htaccess<br>
<strong>Status:</strong> core<p>
This directive enables RFC931-compliant logging of the remote user name
for each connection, where the client machine runs identd or something similar.
This information is logged in the access log. <em>Boolean</em> is either
<code>on</code> or <code>off</code>.<p>
The information should not trusted in any way except for rudimentary usage
tracking.<p><hr>
<h2><a name="keepalive">KeepAlive</a></h2>
<strong>Syntax:</strong> KeepAlive <em>max-requests</em><br>
<strong>Default:</strong> <code>KeepAlive 5</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> Core<br>
<strong>Compatibility:</strong> KeepAlive is only available in Apache
1.1 and later.<p>
This directive enables
<a href="keepalive.html">Keep-Alive</a>
support. Set <em>max-requests</em>
to the maximum number of requests you want Apache to entertain per
request. A limit is imposed to prevent a client from hogging your
server resources. Set this to <code>0</code> to disable support.
<h2><a name="keepalivetimeout">KeepAliveTimeout</a></h2>
<strong>Syntax:</strong> KeepAliveTimeout <em>seconds</em><br>
<strong>Default:</strong> <code>KeepAliveTimeout 15</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> Core<br>
<strong>Compatibility:</strong> KeepAliveTimeout is only available in Apache
1.1 and later.<p>
The number of seconds Apache will wait for a subsequent request before
closing the connection. Once a request has been received, the timeout
value specified by the <a
href="../core.html#timeout"><code>Timeout</code></a> directive
applies.
<A name="listen"><h2>Listen</h2></A>
<strong>Syntax:</strong>
Listen [<em>IP address</em>:]<em>port number</em><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<br>
<strong>Compatibility:</strong> Listen is only available in Apache
1.1 and later.<p>
<p>The Listen directive instructs Apache to listen to more than one IP
address or port; by default it responds to requests on all IP
interfaces, but only on the port given by the <a href="#port">Port</a>
directive.</p>
<p><strong>See Also</strong>:
<a href="bind.html">Setting which addresses and ports Apache uses</a></p>
<A name="limit"><h2>&lt;Limit&gt; directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt Limit} section directive&gt; -->
<strong>Syntax:</strong>
&lt;Limit <em>method method</em> ... &gt; ... &lt;/Limit&gt;<br>
<strong>Context:</strong> any<br>
<strong>Status:</strong> core<p>
&lt;Limit&gt; and &lt;/Limit&gt; are used to enclose a group of
access control directives which will then apply only to the specified
access methods, where <em>method</em> is any valid HTTP method.
Any directive except another &lt;Limit&gt; or
<A HREF="#directory">&lt;Directory&gt;</A> may be used; the majority will be
unaffected by the &lt;Limit&gt;. Example:
<blockquote><code>
&lt;Limit GET POST&gt;<br>
require valid-user<br>
&lt;/Limit&gt;</code></blockquote>
If an access control directive appears outside a &lt;Limit&gt; directive,
then it applies to all access methods.<p><hr>
<h2><a name="location">&lt;Location&gt;</a></h2>
<strong>Syntax:</strong> &lt;Location <em>URL prefix</em>&gt;<br>
<strong>Context:</strong> server config, virtual host<br>
<strong>Status:</strong> core<br>
<strong>Compatibility:</strong> Location is only available in Apache
1.1 and later.<p>
<p>The &lt;Location&gt; directive provides for access control by
URL. It is comprable to the <a
href="#directory">&lt;Directory&gt;</a> directive, and
should be matched with a &lt;/Location&gt; directive. Directives that
apply to the URL given should be listen
within. <code>&lt;Location&gt;</code> sections are processed in the
order they appear in the configuration file, after the
&lt;Directory&gt; sections and <code>.htaccess</code> files are
read.</p>
<p>Note that, due to the way HTTP functions, <em>URL prefix</em>
should, save for proxy requests, be of the form <code>/path/</code>,
and should not include the <code>http://servername</code>. It doesn't
neccessarily have to protect a directory (it can be an individual
file, or a number of files), and can include wildcards. In a wildcard
string, `?' matches any single character, and `*' matches any
sequences of characters.
<p>This functionality is especially useful when combined with the
<code><a href="mod_mime.html#sethandler">SetHandler</a></code>
directive. For example, to enable status requests, but allow them only
from browsers at foo.com, you might use:
<pre>
&lt;Location /status&gt;
SetHandler server-status
<Limit GET>
order deny,allow
deny from all
allow from .foo.com
</Limit>
&lt;/Location&gt;
</pre>
<A name="maxclients"><h2>MaxClients directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt MaxClients} directive&gt; -->
<strong>Syntax:</strong> MaxClients <em>number</em><br>
<strong>Default:</strong> <code>MaxClients 150</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The MaxClients directive sets the limit on the number of simultaneous
requests that can be supported; not more than this number of child server
processes will be created.<p><hr>
<A name="maxrequestsperchild"><h2>MaxRequestsPerChild directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt MaxRequestsPerChild} directive&gt; -->
<strong>Syntax:</strong> MaxRequestsPerChild <em>number</em><br>
<strong>Default:</strong> <code>MaxRequestsPerChild 0</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The MaxRequestsPerChild directive sets the limit on the number of requests
that an individual child server process will handle. After MaxRequestsPerChild
requests, the child process will die. If MaxRequestsPerChild is 0, then
the process will never expire.<p>
Setting MaxRequestsPerChild to a non-zero limit has two beneficial effects:
<ul>
<li>it limits the amount of memory that process can consume by (accidental)
memory leakage;
<li> by giving processes a finite lifetime, it helps reduce the
number of processes when the server load reduces.
</ul><p><hr>
<A name="maxspareservers"><h2>MaxSpareServers directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt MaxSpareServers} directive&gt; -->
<strong>Syntax:</strong> MaxSpareServers <em>number</em><br>
<strong>Default:</strong> <code>MaxSpareServers 10</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The MaxSpareServers directive sets the desired maximum number of <em>idle</em>
child server processes. An idle process is one which is not handling
a request. If there are more than MaxSpareServers idle, then the parent
process will kill off the excess processes.<p>
Tuning of this parameter should only be necessary on very busy sites.
Setting this parameter to a large number is almost always a bad idea.<p>
See also <A HREF="#minspareservers">MinSpareServers</A> and
<A HREF="#startservers">StartServers</A>.<p><hr>
<A name="minspareservers"><h2>MinSpareServers directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt MinSpareServers} directive&gt; -->
<strong>Syntax:</strong> MinSpareServers <em>number</em><br>
<strong>Default:</strong> <code>MinSpareServers 5</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The MinSpareServers directive sets the desired minimum number of <em>idle</em>
child server processes. An idle process is one which is not handling
a request. If there are fewer than MinSpareServers idle, then the parent
process creates new children at a maximum rate of 1 per second.<p>
Tuning of this parameter should only be necessary on very busy sites.
Setting this parameter to a large number is almost always a bad idea.<p>
See also <A HREF="#maxspareservers">MaxSpareServers</A> and
<A HREF="#startservers">StartServers</A>.<p><hr>
<A name="options"><h2>Options directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt Options} directive&gt; -->
<strong>Syntax:</strong> Options <em>option option ...</em><br>
<strong>Context:</strong> server config, virtual host, directory, .htaccess<br>
<strong>Override:</strong> Options<br>
<strong>Status:</strong> core<p>
The Options directive controls which server features are available in
a particular directory.<p>
<em>option</em> can be set to <code>None</code>, in which case none of
the extra features are enabled, or one or more of the following:
<dl>
<dt>All
<dd>All options except for MultiViews.
<dt>ExecCGI
<dd>
<!--%plaintext &lt;?INDEX {\tt ExecCGI} option&gt; -->
Execution of CGI scripts is permitted.
<dt>FollowSymLinks
<dd>
<!--%plaintext &lt;?INDEX {\tt FollowSymLinks} option&gt; -->
The server will follow symbolic links in this directory.
<dt>Includes
<dd>
<!--%plaintext &lt;?INDEX {\tt Includes} option&gt; -->
Server-side includes are permitted.
<dt>IncludesNOEXEC
<dd>
<!--%plaintext &lt;?INDEX {\tt IncludesNOEXEC} option&gt; -->
Server-side includes are permitted, but the #exec command and
#include of CGI scripts are disabled.
<dt>Indexes
<dd>
<!--%plaintext &lt;?INDEX {\tt Indexes} option&gt; -->
If a URL which maps to a directory is requested, and the there is no
DirectoryIndex (e.g. index.html) in that directory, then the server will
return a formatted listing of the directory.
<dt>MultiViews
<dd>
<!--%plaintext &lt;?INDEX {\tt MultiViews} option&gt; -->
<A HREF="content-negotiation.html">Ccontent negotiatad</A> MultiViews are
allowed.
<dt>SymLinksIfOwnerMatch
<dd>
<!--%plaintext &lt;?INDEX {\tt SymLinksIfOwnerMatch} option&gt; -->
The server will only follow symbolic links for which the target
file or directory is owned by the same user id as the link.
</dl>
If multiple Options could apply to a directory, then the most specific one is
taken complete; the options are not merged. For example:
<blockquote><code>
&lt;Directory /web/docs&gt; <br>
Options Indexes FollowSymLinks<br>
&lt;/Directory&gt;<br>
&lt;Directory /web/docs/spec&gt; <br>
Options Includes<br>
&lt;/Directory&gt;
</code></blockquote>
then only <code>Includes</code> will be set for the /web/docs/spec
directory.<p><hr>
<A name="pidfile"><h2>PidFile directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt PidFile} directive&gt; -->
<strong>Syntax:</strong> PidFile <em>filename</em><br>
<strong>Default:</strong> <code>PidFile logs/httpd.pid</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The PidFile directive sets the file to which the server records the
process id of the daemon. If the filename does not begin with a slash (/)
then it is assumed to be relative to the <A HREF="#serverroot">ServerRoot</A>.
The PidFile is only used in <A HREF="#servertype">standalone</A> mode.<p>
It is often useful to be able to send the server a signal, so that it closes
and then reopens its <A HREF="#errorlog">ErrorLog</A> and TransferLog, and
re-reads its configuration files. This is done by sending a SIGHUP (kill -1)
signal to the process id listed in the PidFile.<p><hr>
<A name="port"><h2>Port directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt Port} directive&gt; -->
<strong>Syntax:</strong> Port <em>number</em><br>
<strong>Default:</strong> <code>Port 80</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The Port directive sets the network port on which the server listens.
<em>Num</em> is a number from 0 to 65535; some port numbers (especially below
1024) are reserved for particular protocols. See <code>/etc/services</code>
for a list of some defined ports; the standard port for the http protocol
is 80.<p>
Port 80 is one of Unix's special ports. All ports numbered
below 1024 are reserved for system use, i.e. regular (non-root) users cannot
make use of them; instead they can only use higher port numbers.<p>
To use port 80, you must start the server from the root account.
After binding to the port and before accepting requests, Apache will change
to a low privileged user as set by the <A HREF="#user">User directive</A>.<p>
If you cannot use port 80, choose any other unused port. Non-root users
will have to choose a port number higher than 1023, such as 8000.<p>
SECURITY: if you do start the server as root, be sure
not to set <A HREF="#user">User</A> to root. If you run the server as
root whilst handling connections, your site may be open to a major security
attack.<p><hr>
<A name="require"><h2>require directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt require} directive&gt; -->
<strong>Syntax:</strong> require <em>entity-name entity entity...</em><br>
<strong>Context:</strong> directory, .htaccess<br>
<strong>Override:</strong> AuthConfig<br>
<strong>Status:</strong> core<p>
This directive selects which authenticated users can access a directory.
The allowed syntaxes are:
<ul>
<li>require user <em>userid userid ...</em><p>
Only the named users can access the directory.<p>
<li>require group <em>group-name group-name ...</em><p>
Only users in the named groups can access the directory.<p>
<li>require valid-user<p>
All valid users can access the directory.
</ul>
<p>
If <code>require</code> appears in a <A HREF="#limit">&lt;Limit&gt;</A>
section, then it restricts access to the named methods, otherwise
it restricts access for all methods. Example:
<blockquote><code>
AuthType Basic<br>
AuthName somedomain<br>
AuthUserFile /web/users<br>
AuthGroupFile /web/groups<br>
Limit &lt;GET POST&gt;<br>
require group admin<br>
&lt;/Limit&gt;
</code></blockquote>
Require must be accompanied by <A HREF="#authname">AuthName</A> and
<A HREF="#authtype">AuthType</A> directives, and directives such as
<A HREF="mod_auth.html#authuserfile">AuthUserFile</A> and
<A HREF="mod_auth.html#authgroupfile">AuthGroupFile</A> (to define users and
groups) in order to work correctly.<p><hr>
<A name="resourceconfig"><h2>ResourceConfig directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt ResourceConfig} directive&gt; -->
<strong>Syntax:</strong> ResourceConfig <em>filename</em><br>
<strong>Default:</strong> <code>ResourceConfig conf/srm.conf</code><br>
<strong>Context:</strong> server config, virtual host<br>
<strong>Status:</strong> core<p>
The server will read this file for more directives after reading the
httpd.conf file. <em>Filename</em> is relative to the
<A HREF="#serverroot">ServerRoot</A>.
This feature can be disabled using:
<blockquote><code>ResourceConfig /dev/null</code></blockquote>
Historically, this file contained most directives except for server
configuration directives and <A HREF="#directory">&lt;Directory&gt;</A>.
sections; in fact it can now contain any server directive allowed in the
<em>server config</em> context.<p>
See also <A HREF="#accessconfig">AccessConfig</A>.<p><hr>
<A name="serveradmin"><h2>ServerAdmin directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt ServerAdmin} directive&gt; -->
<strong>Syntax:</strong> ServerAdmin <em>email-address</em><br>
<strong>Context:</strong> server config, virtual host<br>
<strong>Status:</strong> core<p>
The ServerAdmin sets the e-mail address that the server includes in any
error messages it returns to the client.<p>
It may be worth setting up a dedicated address for this, e.g.
<blockquote><code>ServerAdmin www-admin@foo.bar.com</code></blockquote>
as users do not always mention that they are talking about the server!<p><hr>
<A name="serveralias"><h2>ServerAlias directive</h2></A>
<strong>Syntax:</strong> ServerAlias <em>host1 host2 ...</em><br>
<strong>Context:</strong> virtual host<br>
<strong>Status:</strong> core<br>
<strong>Compatibility:</strong> ServerAlias is only available in Apache
1.1 and later.<p>
The ServerAlias directive sets the alternate names for a host, for use
with
<a href="host.html">Host-header based virtual hosts</a>.
<p><hr>
<A name="servername"><h2>ServerName directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt ServerName} directive&gt; -->
<strong>Syntax:</strong> ServerName <em>fully-qualified domain name</em><br>
<strong>Context:</strong> server config, virtual host<br>
<strong>Status:</strong> core<p>
The ServerName directive sets the hostname of the server; this is only
used when creating redirection URLs. If it is not specified, then the
server attempts to deduce it from its own IP address; however this may
not work reliably, or may not return the preferred hostname. For example:
<blockquote><code>ServerName www.wibble.com</code></blockquote>
would be used if the canonical (main) name of the actual machine
were <code>monster.wibble.com</code>.<p><hr>
<A name="serverpath"><h2>ServerPath directive</h2></A>
<strong>Syntax:</strong> ServerPath <em>pathname</em><br>
<strong>Context:</strong> virtual host<br>
<strong>Status:</strong> core<br>
<strong>Compatibility:</strong> ServerPath is only available in Apache
1.1 and later.<p>
The ServerAlias directive sets the legacy URL pathname for a host, for
use with <a href="host.html">Host-header based virtual hosts</a>.
<A name="serverroot"><h2>ServerRoot directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt ServerRoot} directive&gt; -->
<strong>Syntax:</strong> ServerRoot <em>directory-filename</em><br>
<strong>Default:</strong> <code>ServerRoot /usr/local/etc/httpd</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The ServerRoot directive sets the directory in which the server lives.
Typically it will contain the subdirectories <code>conf/</code> and
<code>logs/</code>. Relative paths for other configuration files are taken
as relative to this directory.<p><hr>
<A name="servertype"><h2>ServerType directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt ServerType} directive&gt; -->
<strong>Syntax:</strong> ServerType <em>type</em><br>
<strong>Default:</strong> <code>ServerType standalone</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The ServerType directive sets how the server is executed by the system.
<em>Type</em> is one of
<dl>
<dt>inetd
<dd>The server will be run from the system process inetd; the command to start
the server is added to <code>/etc/inetd.conf</code>
<dt>standalone
<dd>The server will run as a daemon process; the command to start the server
is added to the system startup scripts. (<code>/etc/rc.local</code> or
<code>/etc/rc3.d/...</code>.)
</dl>
Inetd is the lesser used of the two options. For each http
connection received, a new copy of the server is started from scratch;
after the connection is complete, this program exits. There is a high price to
pay per connection, but for security reasons, some admins prefer this option.
<p>
Standalone is the most common setting for ServerType since
it is far more efficient. The server is started once, and services all
subsequent connections. If you intend running Apache to serve a busy site,
standalone will probably be your only option.<p>
SECURITY: if you are paranoid about security, run in inetd mode. Security
cannot be guaranteed in either, but whilst most people are happy to use
standalone, inetd is probably least prone to attack.<p><hr>
<A name="startservers"><h2>StartServers directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt StartServers} directive&gt; -->
<strong>Syntax:</strong> StartServers <em>number</em><br>
<strong>Default:</strong> <code>StartServers 5</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The StartServers directive sets the number of child server processes created
on startup. As the number of processes is dynamically controlled depending
on the load, there is usually little reason to adjust this parameter.<p>
See also <A HREF="#minspareservers">MinSpareServers</A> and
<A HREF="#maxspareservers">MaxSpareServers</A>.<p><hr>
<A name="timeout"><h2>TimeOut directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt TimeOut} directive&gt; -->
<strong>Syntax:</strong> TimeOut <em>number</em><br>
<strong>Default:</strong> <code>TimeOut 1200</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The TimeOut directive sets the maximum time that the server will wait
for the receipt of a request and the completion of a request, in seconds.
So if it takes more than TimeOut seconds for a client to send a request or
receive a response, the server will break off the connection. Thus TimeOut
limits the maximum a transfer can take; for large files, and slow networks
transfer times can be large.
<p><hr>
<A name="user"><h2>User directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt User} directive&gt; -->
<strong>Syntax:</strong> User <em>unix-userid</em><br>
<strong>Default:</strong> <code>User #-1</code><br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> core<p>
The User directive sets the userid as which the server will answer requests.
In order to use this directive, the standalone server must be run initially
as root. <em>Unix-userid</em> is one of:
<dl>
<dt>A username
<dd>Refers to the given user by name.
<dt># followed by a user number.
<dd>Refers to a user by their number.
</dl>
The user should have no privileges which result in it being able to access
files which are not intended to be visible to the outside world, and
similarly, the user should not be able to execute code which is not
meant for httpd requests. It is recommended that you set up a new user and
group specifically for running the server. Some admins use user
<code>nobody</code>, but this is not always possible or desirable.<p>
Notes: If you start the server as a non-root user, it will fail to change
to the lesser privileged user, and will instead continue to run as
that original user. If you do start the server as root, then it is normal
for the parent process to remain running as root.<p>
SECURITY: Don't set User (or <A HREF="#group">Group</A>) to
<code>root</code> unless you know exactly what you are doing, and what the
dangers are.<p><hr>
<A name="virtualhost"><h2>&lt;VirtualHost&gt; directive</h2></A>
<!--%plaintext &lt;?INDEX {\tt VirtualHost} section directive&gt; -->
<strong>Syntax:</strong> &lt;VirtualHost <em>addr</em>[:<em>port</em>]&gt; ...
&lt;/VirtualHost&gt; <br>
<strong>Context:</strong> server config<br>
<strong>Status:</strong> Core. <p>
&lt;VirtualHost&gt; and &lt;/VirtualHost&gt; are used to enclose a group of
directives which will apply only to a particular virtual host.
Any directive which is allowed in a virtual host context may be used.
When the server receives a request for a document on a particular virtual
host, it uses the configuration directives enclosed in the &lt;VirtualHost&gt;
section. <em>Addr</em> can be
<menu>
<li>The IP address of the virtual host
<li>A fully qualified domain name for the IP address of the virtual host.
</menu> Example:
<blockquote>
<code>
&lt;VirtualHost host.foo.com&gt; <br>
ServerAdmin webmaster@host.foo.com <br>
DocumentRoot /www/docs/host.foo.com <br>
ServerName host.foo.com <br>
ErrorLog logs/host.foo.com-error_log <br>
TransferLog logs/host.foo.com-access_log <br>
&lt;/VirtualHost&gt;
</code></blockquote>
Currently, each VirtualHost must correspond to a different IP address for
the server, so the server machine must be configured to accept IP packets for
multiple addresses. If the machine does not have multiple network interfaces,
then this can be accomplished with the <code>ifconfig alias</code> command
(if your OS supports it), or with kernel patches like
<A HREF="vif.info.txt">VIF</A> (for SunOS(TM) 4.1.x).<p>
<!--%hypertext -->
<p><strong>See also:</strong>
<A HREF="virtual-host.html">Information on Virtual Hosts.
(multihome)</A><br>
<strong>See also:</strong>
<a href="host.html">Non-IP address-based Virtual Hosts</a>
<!--/%hypertext --></p>
<hr>
<!--%hypertext -->
<A HREF="../"><IMG SRC="../images/apache_home.gif" ALT="Home"></A>
<A HREF="./"><IMG SRC="../images/apache_index.gif" ALT="Index"></A>
</BODY>
</HTML>
<!--/%hypertext -->