<?xml version="1.0" encoding="iso-8859-1"?>
<!--
The FreeBSD Documentation Project
$FreeBSD$
-->
<chapter id="config-tuning">
<chapterinfo>
<authorgroup>
<author>
<firstname>Chern</firstname>
<surname>Lee</surname>
<contrib>Written by </contrib>
</author>
</authorgroup>
<authorgroup>
<author>
<firstname>Mike</firstname>
<surname>Smith</surname>
<contrib>Based on a tutorial written by </contrib>
</author>
</authorgroup>
<authorgroup>
<author>
<firstname>Matt</firstname>
<surname>Dillon</surname>
<contrib>Also based on tuning(7) written by </contrib>
</author>
</authorgroup>
</chapterinfo>
<title>Configuration and Tuning</title>
<sect1 id="config-synopsis">
<title>Synopsis</title>
<indexterm><primary>system configuration</primary></indexterm>
<indexterm><primary>system optimization</primary></indexterm>
<para>One of the important aspects of &os; is proper system
configuration. This chapter explains much of the &os;
configuration process, including some of the parameters which
can be set to tune a &os; system.</para>
<para>After reading this chapter, you will know:</para>
<itemizedlist>
<listitem>
<para>How to efficiently work with file systems and swap
partitions.</para>
</listitem>
<listitem>
<para>The basics of <filename>rc.conf</filename> configuration
and <filename
class="directory">/usr/local/etc/rc.d</filename> startup
scripts.</para>
</listitem>
<listitem>
<para>How to configure and test a network card.</para>
</listitem>
<listitem>
<para>How to configure virtual hosts on network
devices.</para>
</listitem>
<listitem>
<para>How to use the various configuration files in <filename
class="directory">/etc</filename>.</para>
</listitem>
<listitem>
<para>How to tune &os; using &man.sysctl.8; variables.</para>
</listitem>
<listitem>
<para>How to tune disk performance and modify kernel
limitations.</para>
</listitem>
</itemizedlist>
<para>Before reading this chapter, you should:</para>
<itemizedlist>
<listitem>
<para>Understand &unix; and &os; basics (<xref
linkend="basics"/>).</para>
</listitem>
<listitem>
<para>Be familiar with the basics of kernel configuration and
compilation (<xref linkend="kernelconfig"/>).</para>
</listitem>
</itemizedlist>
</sect1>
<sect1 id="configtuning-initial">
<title>Initial Configuration</title>
<sect2>
<title>Partition Layout</title>
<indexterm><primary>partition layout</primary></indexterm>
<indexterm>
<primary><filename class="directory">/etc</filename></primary>
</indexterm>
<indexterm>
<primary><filename class="directory">/var</filename></primary>
</indexterm>
<indexterm>
<primary><filename class="directory">/usr</filename></primary>
</indexterm>
<sect3>
<title>Base Partitions</title>
<para>When laying out file systems with &man.bsdlabel.8; or
&man.sysinstall.8;, remember that hard drives transfer data
faster from the outer tracks to the inner. Thus, smaller
and heavier-accessed file systems should be closer to the
outside of the drive, while larger partitions like
<filename class="directory">/usr</filename> should be placed
toward the inner parts of the disk. It is a good idea to
create partitions in an order similar to: <filename
class="directory">/</filename>, swap,
<filename class="directory">/var</filename>, and
<filename class="directory">/usr</filename>.</para>
<para>The size of the
<filename class="directory">/var</filename> partition
reflects the intended machine's usage. This partition
is used to hold mailboxes, log files, and printer spools.
Mailboxes and log files can grow to unexpected sizes
depending on the number of users and how long log files
are kept. On average, most users rarely need more than
about a gigabyte of free disk space in <filename
class="directory">/var</filename>.</para>
<note>
<para>Sometimes, a lot of disk space is required in
<filename class="directory">/var/tmp</filename>. When
new software is installed with &man.pkg.add.1;, the
packaging tools extract a temporary copy of the packages
under <filename class="directory">/var/tmp</filename>.
Large software packages, like
<application>Firefox</application>,
<application>OpenOffice</application> or
<application>LibreOffice</application> may be tricky to
install if there is not enough disk space under <filename
class="directory">/var/tmp</filename>.</para>
</note>
<para>The <filename class="directory">/usr</filename>
partition holds many of the files which support the system,
including the &os; Ports Collection and system source code.
At least 2 gigabytes is recommended for this
partition.</para>
<para>When selecting partition sizes, keep the space
requirements in mind. Running out of space in one partition
while barely using another can be a hassle.</para>
<note>
<para>The <literal>Auto-defaults</literal> partition sizer
used by &man.sysinstall.8; will sometimes select smaller
than adequate <filename class="directory">/var</filename>
and <filename class="directory">/</filename> partitions.
Partition wisely and generously.</para>
</note>
</sect3>
<sect3 id="swap-design">
<title>Swap Partition</title>
<indexterm><primary>swap sizing</primary></indexterm>
<indexterm><primary>swap partition</primary></indexterm>
<para>As a rule of thumb, the swap partition should be about
double the size of physical memory (<acronym>RAM</acronym>)
as the kernel's virtual memory (<acronym>VM</acronym>)
paging algorithms are tuned to perform best when the swap
partition is at least two times the size of main memory.
Systems with minimal <acronym>RAM</acronym> may perform
better with more swap. Configuring too little swap can
lead to inefficiencies in the <acronym>VM</acronym> page
scanning code and might create issues later if more memory
is added.</para>
<para>On larger systems with multiple <acronym>SCSI</acronym>
disks or multiple <acronym>IDE</acronym> disks operating
on different controllers, it is recommended that swap be
configured on each drive, up to four drives. The swap
partitions should be approximately the same size. The
kernel can handle arbitrary sizes but internal data
structures scale to 4 times the largest swap partition.
Keeping the swap partitions near the same size will allow
the kernel to optimally stripe swap space across disks.
Large swap sizes are fine, even if swap is not used much.
It might be easier to recover from a runaway program before
being forced to reboot.</para>
</sect3>
<sect3>
<title>Why Partition?</title>
<para>Several users think a single large partition will be
fine, but there are several reasons why this is a bad idea.
First, each partition has different operational
characteristics and separating them allows the file system
to tune accordingly. For example, the root and <filename
class="directory">/usr</filename> partitions are
read-mostly, with few writes, while a lot of reads and
writes could occur in <filename
class="directory">/var</filename> and <filename
class="directory">/var/tmp</filename>.</para>
<para>By properly partitioning a system, fragmentation
introduced in the smaller write heavy partitions will not
bleed over into the mostly read partitions. Keeping the
write loaded partitions closer to the disk's edge will
increase I/O performance in the partitions where it occurs
the most. While I/O performance in the larger partitions
may be needed, shifting them more toward the edge of the
disk will not lead to a significant performance
improvement over moving <filename
class="directory">/var</filename> to the edge. Finally,
there are safety concerns. A smaller, neater root
partition which is mostly read-only has a greater chance of
surviving a bad crash.</para>
</sect3>
</sect2>
</sect1>
<sect1 id="configtuning-core-configuration">
<title>Core Configuration</title>
<indexterm>
<primary>rc files</primary>
<secondary><filename>rc.conf</filename></secondary>
</indexterm>
<para>The principal location for system configuration information
is <filename>/etc/rc.conf</filename>. This file contains a
wide range of configuration information and it is read at
system startup to configure the system. It provides the
configuration information for the <filename>rc*</filename>
files.</para>
<para>The entries in <filename>/etc/rc.conf</filename> override
the default settings in
<filename>/etc/defaults/rc.conf</filename>. The file containing
the default settings should not be edited. Instead, all
system-specific changes should be made to
<filename>/etc/rc.conf</filename>.</para>
<para>A number of strategies may be applied in clustered
applications to separate site-wide configuration from
system-specific configuration in order to keep administration
overhead down. The recommended approach is to place
system-specific configuration into
<filename>/etc/rc.conf.local</filename>. For example:</para>
<itemizedlist>
<listitem>
<para><filename>/etc/rc.conf</filename>:</para>
<programlisting>sshd_enable="YES"
keyrate="fast"
defaultrouter="10.1.1.254"</programlisting>
</listitem>
<listitem>
<para><filename>/etc/rc.conf.local</filename>:</para>
<programlisting>hostname="node1.example.org"
ifconfig_fxp0="inet 10.1.1.1/8"</programlisting>
</listitem>
</itemizedlist>
<para>Distribute <filename>/etc/rc.conf</filename> to every
system using <command>rsync</command> or a similar program,
while <filename>/etc/rc.conf.local</filename> remains
unique.</para>
<para>Upgrading the system using &man.sysinstall.8; or
<command>make world</command> will not overwrite
<filename>/etc/rc.conf</filename>, so system configuration
information will not be lost.</para>
<tip>
<para>The configuration in <filename>/etc/rc.conf</filename>
is parsed by &man.sh.1;. This allows system operators to
create complex configuration scenarios. Refer to
&man.rc.conf.5; for further information on this topic.</para>
</tip>
</sect1>
<sect1 id="configtuning-appconfig">
<title>Application Configuration</title>
<para>Typically, installed applications have their own
configuration files and syntax. It is important that these
files be kept separate from the base system, so that they may be
easily located and managed by the package management
tools.</para>
<indexterm><primary>/usr/local/etc</primary></indexterm>
<para>Typically, these files are installed in <filename
class="directory">/usr/local/etc</filename>. In the case
where an application has a large number of configuration
files, a subdirectory will be created to hold them.</para>
<para>Normally, when a port or package is installed, sample
configuration files are also installed. These are usually
identified with a suffix such as <filename>.sample</filename>.
If there are no existing configuration files for the
application, they can be created by copying the sample
files.</para>
<para>For example, consider the contents of the directory
<filename
class="directory">/usr/local/etc/apache</filename>:</para>
<literallayout class="monospaced">-rw-r--r-- 1 root wheel 2184 May 20 1998 access.conf
-rw-r--r-- 1 root wheel 2184 May 20 1998 access.conf.default
-rw-r--r-- 1 root wheel 9555 May 20 1998 httpd.conf
-rw-r--r-- 1 root wheel 9555 May 20 1998 httpd.conf.default
-rw-r--r-- 1 root wheel 12205 May 20 1998 magic
-rw-r--r-- 1 root wheel 12205 May 20 1998 magic.default
-rw-r--r-- 1 root wheel 2700 May 20 1998 mime.types
-rw-r--r-- 1 root wheel 2700 May 20 1998 mime.types.default
-rw-r--r-- 1 root wheel 7980 May 20 1998 srm.conf
-rw-r--r-- 1 root wheel 7933 May 20 1998 srm.conf.default</literallayout>
<para>The file sizes show that only <filename>srm.conf</filename>
has been changed. A later update of the
<application>Apache</application> port would not overwrite
this changed file.</para>
</sect1>
<sect1 id="configtuning-starting-services">
<sect1info>
<authorgroup>
<author>
<firstname>Tom</firstname>
<surname>Rhodes</surname>
<contrib>Contributed by </contrib>
</author>
</authorgroup>
</sect1info>
<title>Starting Services</title>
<indexterm><primary>services</primary></indexterm>
<para>Many users install third party software on &os; from the
Ports Collection and require the installed services to be
started upon system initialization. Services, such as
<filename role="package">mail/postfix</filename> or
<filename role="package">www/apache22</filename> are just two
of the many software packages which may be started during system
initialization. This section explains the procedures available
for starting third party software.</para>
<para>In &os;, most included services, such as &man.cron.8;, are
started through the system start up scripts.</para>
<sect2>
<title>Extended Application Configuration</title>
<para>Now that &os; includes <filename>rc.d</filename>,
configuration of application startup is easier and provides
more features. Using the key words discussed in <xref
linkend="configtuning-rcd"/>, applications can be set to
start after certain other services and extra flags can be
passed through <filename>/etc/rc.conf</filename> in place of
hard coded flags in the start up script. A basic script may
look similar to the following:</para>
<programlisting>#!/bin/sh
#
# PROVIDE: utility
# REQUIRE: DAEMON
# KEYWORD: shutdown
. /etc/rc.subr
name=utility
rcvar=utility_enable
command="/usr/local/sbin/utility"
load_rc_config $name
#
# DO NOT CHANGE THESE DEFAULT VALUES HERE
# SET THEM IN THE /etc/rc.conf FILE
#
utility_enable=${utility_enable-"NO"}
pidfile=${utility_pidfile-"/var/run/utility.pid"}
run_rc_command "$1"</programlisting>
<para>This script will ensure that the provided
<literal>utility</literal> will be started after the
<literal>DAEMON</literal> pseudo-service. It also provides a
method for setting and tracking the process ID
(<acronym>PID</acronym>).</para>
<para>This application could then have the following line placed
in <filename>/etc/rc.conf</filename>:</para>
<programlisting>utility_enable="YES"</programlisting>
<para>This method allows for easier manipulation of command
line arguments, inclusion of the default functions provided
in <filename>/etc/rc.subr</filename>, compatibility with
&man.rcorder.8;, and provides for easier configuration via
<filename>rc.conf</filename>.</para>
</sect2>
<sect2>
<title>Using Services to Start Services</title>
<para>Other services can be started using &man.inetd.8;.
Working with &man.inetd.8; and its configuration is
described in depth in
<xref linkend="network-inetd"/>.</para>
<para>In some cases, it may make more sense to use
&man.cron.8; to start system services. This approach
has a number of advantages as &man.cron.8; runs these
processes as the owner of the &man.crontab.5;. This allows
regular users to start and maintain their own
applications.</para>
<para>The <literal>@reboot</literal> feature of &man.cron.8;,
may be used in place of the time specification. This causes
the job to run when &man.cron.8; is started, normally during
system initialization.</para>
</sect2>
</sect1>
<sect1 id="configtuning-cron">
<sect1info>
<authorgroup>
<author>
<firstname>Tom</firstname>
<surname>Rhodes</surname>
<contrib>Contributed by </contrib>
<!-- 20 May 2003 -->
</author>
</authorgroup>
</sect1info>
<title>Configuring &man.cron.8;</title>
<indexterm><primary>cron</primary>
<secondary>configuration</secondary></indexterm>
<para>One of the most useful utilities in &os; is &man.cron.8;.
This utility runs in the background and regularly checks
<filename>/etc/crontab</filename> for tasks to execute and
searches <filename class="directory">/var/cron/tabs</filename>
for custom &man.crontab.5; files. These files store
information about specific functions which &man.cron.8; is
supposed to perform at certain times.</para>
<para>Two different types of configuration files are used by
&man.cron.8;: the system <filename>crontab</filename> and user
<filename>crontab</filename>s. These formats only differ in
the sixth field and later. In the system
<filename>crontab</filename>, &man.cron.8; runs the command as
the user specified in the sixth field. In a user
<filename>crontab</filename>, all commands run as the user who
created the <filename>crontab</filename>, so the sixth field
is the last field; this is an important security feature.
The final field is always the command to run.</para>
<note>
<para>User crontabs allow individual users to schedule tasks
without the need for <username>root</username> privileges.
Commands in a user's crontab run with the permissions of the
user who owns the crontab.</para>
<para>The <username>root</username> user can have a user
<filename>crontab</filename> just like any other user. The
<username>root</username> user <filename>crontab</filename>
is separate from the system <filename>crontab</filename>,
<filename>/etc/crontab</filename>. Because the system
<filename>crontab</filename> invokes the specified commands as
<username>root</username>, there is usually no need to create
a user <filename>crontab</filename> for
<username>root</username>.</para>
</note>
<para>Here is a sample entry from
<filename>/etc/crontab</filename>:</para>
<programlisting># /etc/crontab - root's crontab for FreeBSD
#
# $FreeBSD$
# <co id="co-comments"/>
#
SHELL=/bin/sh
PATH=/etc:/bin:/sbin:/usr/bin:/usr/sbin <co id="co-env"/>
#
#minute hour mday month wday who command <co id="co-field-descr"/>
#
*/5 * * * * root /usr/libexec/atrun <co id="co-main"/></programlisting>
<calloutlist>
<callout arearefs="co-comments">
<para>Like most &os; configuration files, lines that begin
with the <literal>#</literal> character are comments. A
comment can be placed in the file as a reminder of what and
why a desired action is performed. Comments cannot be on
the same line as a command or else they will be interpreted
as part of the command; they must be on a new line. Blank
lines are ignored.</para>
</callout>
<callout arearefs="co-env">
<para>The equals (<literal>=</literal>) character is used to
define any environment settings. In this example, it is
used to define the <envar>SHELL</envar> and
<envar>PATH</envar>. If the <envar>SHELL</envar> is
omitted, &man.cron.8; will use the default of &man.sh.1;.
If the <envar>PATH</envar> is omitted, no default will be
used and file locations will need to be absolute.</para>
</callout>
<callout arearefs="co-field-descr">
<para>This line defines a total of seven fields:
<literal>minute</literal>,
<literal>hour</literal>, <literal>mday</literal>,
<literal>month</literal>, <literal>wday</literal>,
<literal>who</literal>, and <literal>command</literal>.
These are almost all self explanatory.
<literal>minute</literal> is the time in minutes when the
specified command will be run. <literal>hour</literal> is
the hour when the specified command will be run.
<literal>mday</literal> stands for day of the month and
<literal>month</literal> designates the month. The
<literal>wday</literal> option stands for day of the week.
These fields must be numeric values, representing the
twenty-four hour clock, or a <literal>*</literal>,
representing all values for that field. The
<literal>who</literal> field only exists in the system
crontab. This field specifies which user the command
should be run as. The last field is the command to be
executed.</para>
</callout>
<callout arearefs="co-main">
<para>This last line defines the values discussed above.
This example has a <literal>*/5</literal> listing,followed
by several more <literal>*</literal> characters. These
<literal>*</literal> characters mean
<quote>first-last</quote>, and can be interpreted as
<emphasis>every</emphasis> time. In this example,
&man.atrun.8; is invoked by <username>root</username>
every five minutes, regardless of the day or month.</para>
<para>Commands can have any number of flags passed to them;
however, commands which extend to multiple lines need to be
broken with the backslash <quote>\</quote> continuation
character.</para>
</callout>
</calloutlist>
<para>This is the basic setup for every &man.crontab.5;.
However, field number six, which specifies the username, only
exists in the system &man.crontab.5;. This field should be
omitted for individual user &man.crontab.5; files.</para>
<sect2 id="configtuning-installcrontab">
<title>Installing a Crontab</title>
<important>
<para>Do not use the procedure described here to edit and
install the system <filename>crontab</filename>,
<filename>/etc/crontab</filename>. Instead, use an
editor and &man.cron.8; will notice that the file has
changed and immediately begin using the updated version.
See <ulink
url="&url.books.faq;/admin.html#root-not-found-cron-errors">
this FAQ entry</ulink> for more information.</para>
</important>
<para>To install a freshly written user &man.crontab.5;, use
an editor to create and save a file in the proper format.
Then, specify the file name with &man.crontab.1;:</para>
<screen>&prompt.user; <userinput>crontab crontab-file</userinput></screen>
<para>In this example, <filename>crontab-file</filename> is the
filename of a &man.crontab.5; that was previously
created.</para>
<para>To list installed &man.crontab.5; files, pass
<option>-l</option> to &man.crontab.1;.</para>
<para>Users who wish to begin their own
<filename>crontab</filename> file from scratch, without the
use of a template, can use <command>crontab -e</command>. This
will invoke the default editor with an empty file. When this
file is saved, it will be automatically installed by
&man.crontab.1;.</para>
<para>In order to remove a user &man.crontab.5; completely,
use <command>crontab -r</command>.</para>
</sect2>
</sect1>
<sect1 id="configtuning-rcd">
<sect1info>
<authorgroup>
<author>
<firstname>Tom</firstname>
<surname>Rhodes</surname>
<contrib>Contributed by </contrib>
<!-- 16 May 2003 -->
</author>
</authorgroup>
</sect1info>
<title>Using &man.rc.8; Under &os;</title>
<para>In 2002, &os; integrated the NetBSD &man.rc.8; system for
system initialization. The files listed in <filename
class="directory">/etc/rc.d</filename> provide basic services
which can be controlled with the <option>start</option>,
<option>stop</option>, and <option>restart</option> options
to &man.service.8;. For instance, &man.sshd.8; can be restarted
with the following command:</para>
<screen>&prompt.root; <userinput>service sshd restart</userinput></screen>
<para>This procedure can be used to start services on a running
system. Services will be started automatically at boot time
as specified in &man.rc.conf.5;. For example, to enable
&man.natd.8; at system startup, add the following line to
<filename>/etc/rc.conf</filename>:</para>
<programlisting>natd_enable="YES"</programlisting>
<para>If a <option>natd_enable="NO"</option> line is already
present, change the <literal>NO</literal> to
<literal>YES</literal>. The &man.rc.8; scripts will
automatically load any dependent services during the next boot,
as described below.</para>
<para>Since the &man.rc.8; system is primarily intended to start
and stop services at system startup and shutdown time, the
<option>start</option>, <option>stop</option> and
<option>restart</option> options will only perform their action
if the appropriate <filename>/etc/rc.conf</filename> variable
is set. For instance, <command>sshd restart</command> will
only work if <varname>sshd_enable</varname> is set to
<option>YES</option> in <filename>/etc/rc.conf</filename>.
To <option>start</option>, <option>stop</option> or
<option>restart</option> a service regardless of the settings
in <filename>/etc/rc.conf</filename>, these commands should be
prefixed with <quote>one</quote>. For instance, to restart
&man.sshd.8; regardless of the current
<filename>/etc/rc.conf</filename> setting, execute the following
command:</para>
<screen>&prompt.root; <userinput>service sshd onerestart</userinput></screen>
<para>To check if a service is enabled in
<filename>/etc/rc.conf</filename>, run the appropriate
&man.rc.8; script with <option>rcvar</option>. This example
checks to see if &man.sshd.8; is enabled in
<filename>/etc/rc.conf</filename>:</para>
<screen>&prompt.root; <userinput>service sshd rcvar</userinput>
# sshd
$sshd_enable=YES</screen>
<note>
<para>The <literal># sshd</literal> line is output from the
above command, not a <username>root</username> console.</para>
</note>
<para>To determine whether or not a service is running, use
<option>status</option>. For instance, to verify that
&man.sshd.8; is running:</para>
<screen>&prompt.root; <userinput>service sshd status</userinput>
sshd is running as pid 433.</screen>
<para>In some cases, it is also possible to
<option>reload</option> a service. This attempts to send a
signal to an individual service, forcing the service to reload
its configuration files. In most cases, this means sending
the service a <literal>SIGHUP</literal> signal. Support for
this feature is not included for every service.</para>
<para>The &man.rc.8; system is used for network services and it
also contributes to most of the system initialization. For
instance, when the
<filename>/etc/rc.d/bgfsck</filename> script is executed, it
prints out the following message:</para>
<screen>Starting background file system checks in 60 seconds.</screen>
<para>This script is used for background file system checks,
which occur only during system initialization.</para>
<para>Many system services depend on other services to function
properly. For example, &man.yp.8; and other
<acronym>RPC</acronym>-based services may fail to start until
after the &man.rpcbind.8; service has started. To resolve this
issue, information about dependencies and other meta-data is
included in the comments at the top of each startup script.
The &man.rcorder.8; program is used to parse these comments
during system initialization to determine the order in which
system services should be invoked to satisfy the
dependencies.</para>
<para>The following key word must be included in all startup
scripts as it is required by &man.rc.subr.8; to
<quote>enable</quote> the startup script:</para>
<itemizedlist>
<listitem>
<para><literal>PROVIDE</literal>: Specifies the services this
file provides.</para>
</listitem>
</itemizedlist>
<para>The following key words may be included at the top of each
startup script. They are not strictly necessary, but are
useful as hints to &man.rcorder.8;:</para>
<itemizedlist>
<listitem>
<para><literal>REQUIRE</literal>: Lists services which are
required for this service. The script containing this key
word will run <emphasis>after</emphasis> the specified
services.</para>
</listitem>
<listitem>
<para><literal>BEFORE</literal>: Lists services which depend
on this service. The script containing this key word will
run <emphasis>before</emphasis> the specified
services.</para>
</listitem>
</itemizedlist>
<para>By carefully setting these keywords for each startup script,
an administrator has a fine-grained level of control of the
startup order of the scripts, without the need for
<quote>runlevels</quote> used by some &unix; operating
systems.</para>
<para>Additional information can be found in &man.rc.8; and
&man.rc.subr.8;. Refer to <ulink
url="&url.articles.rc-scripting;">this article</ulink> for
instructions on how to create custom &man.rc.8;
scripts.</para>
</sect1>
<sect1 id="config-network-setup">
<sect1info>
<authorgroup>
<author>
<firstname>Marc</firstname>
<surname>Fonvieille</surname>
<contrib>Contributed by </contrib>
<!-- 6 October 2002 -->
</author>
</authorgroup>
</sect1info>
<title>Setting Up Network Interface Cards</title>
<indexterm>
<primary>network cards</primary>
<secondary>configuration</secondary>
</indexterm>
<para>Adding and configuring a network interface card
(<acronym>NIC</acronym>) is a common task for any &os;
administrator.</para>
<sect2>
<title>Locating the Correct Driver</title>
<indexterm>
<primary>network cards</primary>
<secondary>driver</secondary>
</indexterm>
<para>First, determine the model of the <acronym>NIC</acronym>
and the chip it uses. &os; supports a wide variety of
<acronym>NIC</acronym>s. Check the Hardware Compatibility
List for the &os; release to see if the <acronym>NIC</acronym>
is supported.</para>
<para>If the <acronym>NIC</acronym> is supported, determine
the name of the &os; driver for the <acronym>NIC</acronym>.
Refer to <filename>/usr/src/sys/conf/NOTES</filename> and
<filename>/usr/src/sys/<replaceable>arch</replaceable>/conf/NOTES</filename>
for the list of <acronym>NIC</acronym> drivers with some
information about the supported chipsets. When in doubt, read
the manual page of the driver as it will provide more
information about the supported hardware and any known
limitations of the driver.</para>
<para>The drivers for common <acronym>NIC</acronym>s are
already present in the <filename>GENERIC</filename> kernel,
meaning the <acronym>NIC</acronym> should show up during boot.
In this example, two <acronym>NIC</acronym>s using the
&man.dc.4; driver are present on the system:</para>
<screen>dc0: <82c169 PNIC 10/100BaseTX> port 0xa000-0xa0ff mem 0xd3800000-0xd38
000ff irq 15 at device 11.0 on pci0
miibus0: <MII bus> on dc0
bmtphy0: <BCM5201 10/100baseTX PHY> PHY 1 on miibus0
bmtphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
dc0: Ethernet address: 00:a0:cc:da:da:da
dc0: [ITHREAD]
dc1: <82c169 PNIC 10/100BaseTX> port 0x9800-0x98ff mem 0xd3000000-0xd30
000ff irq 11 at device 12.0 on pci0
miibus1: <MII bus> on dc1
bmtphy1: <BCM5201 10/100baseTX PHY> PHY 1 on miibus1
bmtphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
dc1: Ethernet address: 00:a0:cc:da:da:db
dc1: [ITHREAD]</screen>
<para>If the driver for the <acronym>NIC</acronym> is not
present in <filename>GENERIC</filename>, but a driver is
available, the driver will need to be loaded before the
<acronym>NIC</acronym> can be configured and used. This may
be accomplished in one of two ways:</para>
<itemizedlist>
<listitem>
<para>The easiest way is to load a kernel module for the
<acronym>NIC</acronym> using &man.kldload.8;. To also
automatically load the driver at boot time, add the
appropriate line to
<filename>/boot/loader.conf</filename>. Not all
<acronym>NIC</acronym> drivers are available as
modules.</para>
</listitem>
<listitem>
<para>Alternatively, statically compile support for the
<acronym>NIC</acronym> into a custom kernel. Refer to
<filename>/usr/src/sys/conf/NOTES</filename>,
<filename>/usr/src/sys/<replaceable>arch</replaceable>/conf/NOTES</filename>
and the manual page of the driver to determine which line
to add to the custom kernel configuration file. For more
information about recompiling the kernel, refer to <xref
linkend="kernelconfig"/>. If the
<acronym>NIC</acronym> was detected at boot, the kernel
does not need to be recompiled.</para>
</listitem>
</itemizedlist>
<sect3 id="config-network-ndis">
<title>Using &windows; <acronym>NDIS</acronym> Drivers</title>
<indexterm><primary><acronym>NDIS</acronym></primary></indexterm>
<indexterm><primary>NDISulator</primary></indexterm>
<indexterm><primary>&windows; drivers</primary></indexterm>
<indexterm><primary>µsoft.windows;</primary>
<secondary>device drivers</secondary>
</indexterm>
<indexterm>
<primary><acronym>KLD</acronym> (kernel loadable
object)</primary>
</indexterm>
<!-- We should probably omit the expanded name, and add a <see> entry
for it. Whatever is done must also be done to the same indexterm in
linuxemu/chapter.xml -->
<para>Unfortunately, there are still many vendors that do not
provide schematics for their drivers to the open source
community because they regard such information as trade
secrets. Consequently, the developers of &os; and other
operating systems are left with two choices: develop the
drivers by a long and pain-staking process of reverse
engineering or using the existing driver binaries available
for µsoft.windows; platforms.</para>
<para>&os; provides <quote>native</quote> support for the
Network Driver Interface Specification
(<acronym>NDIS</acronym>). It includes &man.ndisgen.8;
which can be used to convert a &windowsxp; driver into a
format that can be used on &os;. Because the &man.ndis.4;
driver uses a &windowsxp; binary, it only runs on &i386;
and amd64 systems. <acronym>PCI</acronym>, CardBus,
<acronym>PCMCIA</acronym>, and <acronym>USB</acronym>
devices are supported.</para>
<para>To use &man.ndisgen.8;, three things are needed:</para>
<orderedlist>
<listitem>
<para>&os; kernel sources.</para>
</listitem>
<listitem>
<para>A &windowsxp; driver binary with a
<filename>.SYS</filename> extension.</para>
</listitem>
<listitem>
<para>A &windowsxp; driver configuration file with a
<filename>.INF</filename> extension.</para>
</listitem>
</orderedlist>
<para>Download the <filename>.SYS</filename> and
<filename>.INF</filename> files for the specific
<acronym>NIC</acronym>. Generally, these can be found on
the driver CD or at the vendor's website. The following
examples use <filename>W32DRIVER.SYS</filename> and
<filename>W32DRIVER.INF</filename>.</para>
<para>The driver bit width must match the version of &os;.
For &os;/i386, use a &windows; 32-bit driver. For
&os;/amd64, a &windows; 64-bit driver is needed.</para>
<para>The next step is to compile the driver binary into a
loadable kernel module. As <username>root</username>, use
&man.ndisgen.8;:</para>
<screen>&prompt.root; <userinput>ndisgen <replaceable>/path/to/W32DRIVER.INF</replaceable> <replaceable>/path/to/W32DRIVER.SYS</replaceable></userinput></screen>
<para>This command is interactive and prompts for any extra
information it requires. A new kernel module will be
generated in the current directory. Use &man.kldload.8;
to load the new module:</para>
<screen>&prompt.root; <userinput>kldload <replaceable>./W32DRIVER_SYS.ko</replaceable></userinput></screen>
<para>In addition to the generated kernel module, the
<filename>ndis.ko</filename> and
<filename>if_ndis.ko</filename> modules must be loaded.
This should happen automatically when any module that
depends on &man.ndis.4; is loaded. If not, load them
manually, using the following commands:</para>
<screen>&prompt.root; <userinput>kldload ndis</userinput>
&prompt.root; <userinput>kldload if_ndis</userinput></screen>
<para>The first command loads the &man.ndis.4; miniport driver
wrapper and the second loads the generated
<acronym>NIC</acronym> driver.</para>
<para>Check &man.dmesg.8; to see if there were any load
errors. If all went well, the output should be similar to
the following:</para>
<screen>ndis0: <Wireless-G PCI Adapter> mem 0xf4100000-0xf4101fff irq 3 at device 8.0 on pci1
ndis0: NDIS API version: 5.0
ndis0: Ethernet address: 0a:b1:2c:d3:4e:f5
ndis0: 11b rates: 1Mbps 2Mbps 5.5Mbps 11Mbps
ndis0: 11g rates: 6Mbps 9Mbps 12Mbps 18Mbps 36Mbps 48Mbps 54Mbps</screen>
<para>From here, <devicename>ndis0</devicename> can be
configured like any other <acronym>NIC</acronym>.</para>
<para>To configure the system to load the &man.ndis.4; modules
at boot time, copy the generated module,
<filename>W32DRIVER_SYS.ko</filename>, to <filename
class="directory">/boot/modules</filename>. Then, add the
following line to
<filename>/boot/loader.conf</filename>:</para>
<programlisting>W32DRIVER_SYS_load="YES"</programlisting>
</sect3>
</sect2>
<sect2>
<title>Configuring the Network Card</title>
<indexterm>
<primary>network cards</primary>
<secondary>configuration</secondary>
</indexterm>
<para>Once the right driver is loaded for the
<acronym>NIC</acronym>, the card needs to be configured. It
may have been configured at installation time by
&man.sysinstall.8;.</para>
<para>To display the <acronym>NIC</acronym> configuration,
enter the following command:</para>
<screen>&prompt.user; <userinput>ifconfig</userinput>
dc0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=80008<VLAN_MTU,LINKSTATE>
ether 00:a0:cc:da:da:da
inet 192.168.1.3 netmask 0xffffff00 broadcast 192.168.1.255
media: Ethernet autoselect (100baseTX <full-duplex>)
status: active
dc1: flags=8802<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=80008<VLAN_MTU,LINKSTATE>
ether 00:a0:cc:da:da:db
inet 10.0.0.1 netmask 0xffffff00 broadcast 10.0.0.255
media: Ethernet 10baseT/UTP
status: no carrier
lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> metric 0 mtu 16384
options=3<RXCSUM,TXCSUM>
inet6 fe80::1%lo0 prefixlen 64 scopeid 0x4
inet6 ::1 prefixlen 128
inet 127.0.0.1 netmask 0xff000000
nd6 options=3<PERFORMNUD,ACCEPT_RTADV></screen>
<para>In this example, the following devices were
displayed:</para>
<itemizedlist>
<listitem>
<para><devicename>dc0</devicename>: The first Ethernet
interface.</para>
</listitem>
<listitem>
<para><devicename>dc1</devicename>: The second Ethernet
interface.</para>
</listitem>
<listitem>
<para><devicename>lo0</devicename>: The loopback
device.</para>
</listitem>
</itemizedlist>
<para>&os; uses the driver name followed by the order in which
the card is detected at boot to name the
<acronym>NIC</acronym>. For example,
<devicename>sis2</devicename> is the third
<acronym>NIC</acronym> on the system using the &man.sis.4;
driver.</para>
<para>In this example, <devicename>dc0</devicename> is up and
running. The key indicators are:</para>
<orderedlist>
<listitem>
<para><literal>UP</literal> means that the card is
configured and ready.</para>
</listitem>
<listitem>
<para>The card has an Internet (<literal>inet</literal>)
address, <hostid
role="ipaddr">192.168.1.3</hostid>.</para>
</listitem>
<listitem>
<para>It has a valid subnet mask
(<literal>netmask</literal>), where <hostid
role="netmask">0xffffff00</hostid> is the same as
<hostid role="netmask">255.255.255.0</hostid>.</para>
</listitem>
<listitem>
<para>It has a valid broadcast address, <hostid
role="ipaddr">192.168.1.255</hostid>.</para>
</listitem>
<listitem>
<para>The <acronym>MAC</acronym> address of the card
(<literal>ether</literal>) is <hostid
role="mac">00:a0:cc:da:da:da</hostid>.</para>
</listitem>
<listitem>
<para>The physical media selection is on autoselection mode
(<literal>media: Ethernet autoselect (100baseTX
<full-duplex>)</literal>). In this example,
<devicename>dc1</devicename> is configured to run with
<literal>10baseT/UTP</literal> media. For more
information on available media types for a driver, refer
to its manual page.</para>
</listitem>
<listitem>
<para>The status of the link (<literal>status</literal>) is
<literal>active</literal>, indicating that the carrier
signal is detected. For <devicename>dc1</devicename>, the
<literal>status: no carrier</literal> status is normal
when an Ethernet cable is not plugged into the
card.</para>
</listitem>
</orderedlist>
<para>If the &man.ifconfig.8; output had shown something similar
to:</para>
<screen>dc0: flags=8843<BROADCAST,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=80008<VLAN_MTU,LINKSTATE>
ether 00:a0:cc:da:da:da
media: Ethernet autoselect (100baseTX <full-duplex>)
status: active</screen>
<para>it would indicate the card has not been configured.</para>
<para>The card must be configured as <username>root</username>.
The <acronym>NIC</acronym> configuration can be performed
from the command line with &man.ifconfig.8; but will not
persist after a reboot unless the configuration is also added
to <filename>/etc/rc.conf</filename>. Add a line for each
<acronym>NIC</acronym> present on the system, as seen in
this example:</para>
<programlisting>ifconfig_dc0="inet 192.168.1.3 netmask 255.255.255.0"
ifconfig_dc1="inet 10.0.0.1 netmask 255.255.255.0 media 10baseT/UTP"</programlisting>
<para>Replace <devicename>dc0</devicename> and
<devicename>dc1</devicename> and the <acronym>IP</acronym>
address information with the correct values for the system.
Refer to the man page for the driver, &man.ifconfig.8;, and
&man.rc.conf.5; for more details about the allowed options and
the syntax of <filename>/etc/rc.conf</filename>.</para>
<para>If the network was configured during installation, some
entries for the <acronym>NIC</acronym>(s) may be already
present. Double check <filename>/etc/rc.conf</filename>
before adding any lines.</para>
<para>If the network is not using <acronym>DNS</acronym>, edit
<filename>/etc/hosts</filename> to add the names and
<acronym>IP</acronym> addresses of of the hosts on the
<acronym>LAN</acronym>, if they are not already there. For
more information, refer to &man.hosts.5; and to
<filename>/usr/share/examples/etc/hosts</filename>.</para>
<note>
<para>If there is no <acronym>DHCP</acronym> server and
access to the Internet is needed, manually configure the
default gateway and the nameserver:</para>
<screen>&prompt.root; <userinput>echo 'defaultrouter="<replaceable>your_default_router</replaceable>"' >> /etc/rc.conf</userinput>
&prompt.root; <userinput>echo 'nameserver <replaceable>your_DNS_server</replaceable>' >> /etc/resolv.conf</userinput></screen>
</note>
</sect2>
<sect2>
<title>Testing and Troubleshooting</title>
<para>Once the necessary changes to
<filename>/etc/rc.conf</filename> are saved, a reboot can be
used to test the network configuration and to verify that the
system restarts without any configuration errors.
Alternatively, apply the settings to the networking system
with this command:</para>
<screen>&prompt.root; <userinput>service netif restart</userinput></screen>
<note>
<para>If a default gateway has been set in
<filename>/etc/rc.conf</filename>, also issue this
command:</para>
<screen>&prompt.root; <userinput>service routing restart</userinput></screen>
</note>
<para>Once the networking system has been relaunched, test the
<acronym>NIC</acronym>s.</para>
<sect3>
<title>Testing the Ethernet Card</title>
<indexterm>
<primary>network cards</primary>
<secondary>testing</secondary>
</indexterm>
<para>To verify that an Ethernet card is configured correctly,
&man.ping.8; the interface itself, and then &man.ping.8;
another machine on the <acronym>LAN</acronym>:</para>
<screen>&prompt.user; <userinput>ping -c5 192.168.1.3</userinput>
PING 192.168.1.3 (192.168.1.3): 56 data bytes
64 bytes from 192.168.1.3: icmp_seq=0 ttl=64 time=0.082 ms
64 bytes from 192.168.1.3: icmp_seq=1 ttl=64 time=0.074 ms
64 bytes from 192.168.1.3: icmp_seq=2 ttl=64 time=0.076 ms
64 bytes from 192.168.1.3: icmp_seq=3 ttl=64 time=0.108 ms
64 bytes from 192.168.1.3: icmp_seq=4 ttl=64 time=0.076 ms
--- 192.168.1.3 ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max/stddev = 0.074/0.083/0.108/0.013 ms</screen>
<screen>&prompt.user; <userinput>ping -c5 192.168.1.2</userinput>
PING 192.168.1.2 (192.168.1.2): 56 data bytes
64 bytes from 192.168.1.2: icmp_seq=0 ttl=64 time=0.726 ms
64 bytes from 192.168.1.2: icmp_seq=1 ttl=64 time=0.766 ms
64 bytes from 192.168.1.2: icmp_seq=2 ttl=64 time=0.700 ms
64 bytes from 192.168.1.2: icmp_seq=3 ttl=64 time=0.747 ms
64 bytes from 192.168.1.2: icmp_seq=4 ttl=64 time=0.704 ms
--- 192.168.1.2 ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max/stddev = 0.700/0.729/0.766/0.025 ms</screen>
<para>To test network resolution, use the host name instead
of the <acronym>IP</acronym> address. If there is no
<acronym>DNS</acronym> server on the network,
<filename>/etc/hosts</filename> must first be
configured.</para>
</sect3>
<sect3>
<title>Troubleshooting</title>
<indexterm>
<primary>network cards</primary>
<secondary>troubleshooting</secondary>
</indexterm>
<para>When troubleshooting hardware and software
configurations, check the simple things first. Is the
network cable plugged in? Are the network services properly
configured? Is the firewall configured correctly? Is the
<acronym>NIC</acronym> supported by &os;? Before sending
a bug report, always check the Hardware Notes, update the
version of &os; to the latest STABLE version, check the
mailing list archives, and search the Internet.</para>
<para>If the card works, yet performance is poor, read
through &man.tuning.7;. Also, check the network
configuration as incorrect network settings can cause slow
connections.</para>
<para>Some users experience one or two
<errorname>device timeout</errorname> messages, which is
normal for some cards. If they continue, or are bothersome,
determine if the device is conflicting with another device.
Double check the cable connections. Consider trying another
card.</para>
<para>To resolve <errorname>watchdog timeout</errorname>
errors, first check the network cable. Many cards
require a <acronym>PCI</acronym> slot which supports bus
mastering. On some old motherboards, only one
<acronym>PCI</acronym> slot allows it, usually slot 0.
Check the <acronym>NIC</acronym> and the motherboard
documentation to determine if that may be the
problem.</para>
<para><errorname>No route to host</errorname> messages occur
if the system is unable to route a packet to the destination
host. This can happen if no default route is specified or
if a cable is unplugged. Check the output of
<command>netstat -rn</command> and make sure there is a
valid route to the host. If there is not, read <xref
linkend="advanced-networking"/>.</para>
<para><errorname>ping: sendto: Permission denied</errorname>
error messages are often caused by a misconfigured firewall.
If a firewall is enabled on &os; but no rules have been
defined, the default policy is to deny all traffic, even
&man.ping.8;. Refer to <xref
linkend="firewalls"/> for more information.</para>
<para>Sometimes performance of the card is poor or below
average. In these cases, try setting the media
selection mode from <literal>autoselect</literal> to the
correct media selection. While this works for most
hardware, it may or may not resolve the issue. Again,
check all the network settings, and refer to
&man.tuning.7;.</para>
</sect3>
</sect2>
</sect1>
<sect1 id="configtuning-virtual-hosts">
<title>Virtual Hosts</title>
<indexterm><primary>virtual hosts</primary></indexterm>
<indexterm><primary><acronym>IP</acronym>
aliases</primary></indexterm>
<para>A common use of &os; is virtual site hosting, where one
server appears to the network as many servers. This is achieved
by assigning multiple network addresses to a single
interface.</para>
<para>A given network interface has one <quote>real</quote>
address, and may have any number of <quote>alias</quote>
addresses. These aliases are normally added by placing alias
entries in <filename>/etc/rc.conf</filename>, as seen in this
example:</para>
<programlisting>ifconfig_fxp0_alias0="inet xxx.xxx.xxx.xxx netmask xxx.xxx.xxx.xxx"</programlisting>
<para>Alias entries must start with
<literal>alias<replaceable>0</replaceable></literal> using a
sequential number such as
<literal>alias0</literal>, <literal>alias1</literal>,
and so on. The configuration process will stop at the first
missing number.</para>
<para>The calculation of alias netmasks is important. For a
given interface, there must be one address which correctly
represents the network's netmask. Any other addresses which
fall within this network must have a netmask of all
<literal>1</literal>s, expressed as either <hostid
role="netmask">255.255.255.255</hostid> or <hostid
role="netmask">0xffffffff</hostid>.</para>
<para>For example, consider the case where the
<devicename>fxp0</devicename> interface is connected to two
networks: <hostid role="ipaddr">10.1.1.0</hostid> with a
netmask of <hostid role="netmask">255.255.255.0</hostid> and
<hostid role="ipaddr">202.0.75.16</hostid> with a netmask of
<hostid role="netmask">255.255.255.240</hostid>. The system
is to be configured to appear in the ranges <hostid
role="ipaddr">10.1.1.1</hostid> through <hostid
role="ipaddr">10.1.1.5</hostid> and <hostid
role="ipaddr">202.0.75.17</hostid> through <hostid
role="ipaddr">202.0.75.20</hostid>. Only the first address
in a given network range should have a real netmask. All the
rest (<hostid role="ipaddr">10.1.1.2</hostid> through <hostid
role="ipaddr">10.1.1.5</hostid> and <hostid
role="ipaddr">202.0.75.18</hostid> through <hostid
role="ipaddr">202.0.75.20</hostid>) must be configured with
a netmask of <hostid
role="netmask">255.255.255.255</hostid>.</para>
<para>The following <filename>/etc/rc.conf</filename> entries
configure the adapter correctly for this scenario:</para>
<programlisting>ifconfig_fxp0="inet 10.1.1.1 netmask 255.255.255.0"
ifconfig_fxp0_alias0="inet 10.1.1.2 netmask 255.255.255.255"
ifconfig_fxp0_alias1="inet 10.1.1.3 netmask 255.255.255.255"
ifconfig_fxp0_alias2="inet 10.1.1.4 netmask 255.255.255.255"
ifconfig_fxp0_alias3="inet 10.1.1.5 netmask 255.255.255.255"
ifconfig_fxp0_alias4="inet 202.0.75.17 netmask 255.255.255.240"
ifconfig_fxp0_alias5="inet 202.0.75.18 netmask 255.255.255.255"
ifconfig_fxp0_alias6="inet 202.0.75.19 netmask 255.255.255.255"
ifconfig_fxp0_alias7="inet 202.0.75.20 netmask 255.255.255.255"</programlisting>
</sect1>
<sect1 id="configtuning-syslog">
<sect1info>
<authorgroup>
<author>
<firstname>Niclas</firstname>
<surname>Zeising</surname>
<contrib>Contributed by </contrib>
</author>
</authorgroup>
</sect1info>
<title>Configuring the System Logger,
<command>syslogd</command></title>
<indexterm><primary>system logging</primary></indexterm>
<indexterm><primary>syslog</primary></indexterm>
<indexterm><primary>&man.syslogd.8;</primary></indexterm>
<para>System logging is an important aspect of system
administration. It is used to detect hardware and software
issues and errors in the system. It plays an important role
in security auditing and incident response. System daemons
without a controlling terminal usually log information to a
system logging facility or other log file.</para>
<para>This section describes how to configure and use the &os;
system logger, &man.syslogd.8;, and how to perform log rotation
and log management using &man.newsyslog.8;. Focus will be on
setting up and using &man.syslogd.8; on a local machine. For
more advanced setups using a separate loghost, see <xref
linkend="network-syslogd"/>.</para>
<sect2>
<title>Using <command>syslogd</command></title>
<para>In the default &os; configuration, &man.syslogd.8; is
started at boot. This is controlled by the variable
<literal>syslogd_enable</literal> in
<filename>/etc/rc.conf</filename>. There are numerous
application arguments that affect the behavior of
&man.syslogd.8;. To change them, use
<literal>syslogd_flags</literal> in
<filename>/etc/rc.conf</filename>. Refer to &man.syslogd.8;
for more information on the arguments, and &man.rc.conf.5;,
<xref linkend="configtuning-core-configuration"/> and <xref
linkend="configtuning-rcd"/> for more information about
<filename>/etc/rc.conf</filename> and the &man.rc.8;
subsystem.</para>
</sect2>
<sect2>
<title>Configuring <command>syslogd</command></title>
<indexterm><primary>syslog.conf</primary></indexterm>
<para>The configuration file, by default
<filename>/etc/syslog.conf</filename>, controls what
&man.syslogd.8; does with the log entries once they are
received. There are several parameters to control the
handling of incoming events, of which the most basic are
<firstterm>facility</firstterm> and
<firstterm>level</firstterm>. The facility describes
which subsystem generated the message, such as the kernel or a
daemon, and the level describes the severity of the event that
occurred. This makes it possible to log the message to
different log files, or discard it, depending on the facility
and level. It is also possible to take action depending on
the application that sent the message, and in the case of
remote logging, the hostname of the machine generating
the logging event.</para>
<para>The configuration file for &man.syslogd.8; contains one
line per action, and the syntax for each line is a selector
field followed by an action field. The syntax of the selector
field is <replaceable>facility.level</replaceable> which will
match log messages from <replaceable>facility</replaceable>
at level <replaceable>level</replaceable> or higher. It is
also possible to add an optional comparison flag before the
level to specify more precisely what is logged. Multiple
selector fields can be used for the same action, and are
separated with a semicolon (<literal>;</literal>). Using
<literal>*</literal> will match everything. The action field
denotes where to send the log message, such as to a file or
remote log host. As an example, here is the default
<filename>syslog.conf</filename> from &os;:</para>
<programlisting># $&os;$
#
# Spaces ARE valid field separators in this file. However,
# other *nix-like systems still insist on using tabs as field
# separators. If you are sharing this file between systems, you
# may want to use only tabs as field separators here.
# Consult the syslog.conf(5) manpage.
*.err;kern.warning;auth.notice;mail.crit /dev/console <co id="co-syslog-many-match"/>
*.notice;authpriv.none;kern.debug;lpr.info;mail.crit;news.err /var/log/messages
security.* /var/log/security
auth.info;authpriv.info /var/log/auth.log
mail.info /var/log/maillog <co id="co-syslog-one-match"/>
lpr.info /var/log/lpd-errs
ftp.info /var/log/xferlog
cron.* /var/log/cron
*.=debug /var/log/debug.log <co id="co-syslog-comparison"/>
*.emerg *
# uncomment this to log all writes to /dev/console to /var/log/console.log
#console.info /var/log/console.log
# uncomment this to enable logging of all log messages to /var/log/all.log
# touch /var/log/all.log and chmod it to mode 600 before it will work
#*.* /var/log/all.log
# uncomment this to enable logging to a remote loghost named loghost
#*.* @loghost
# uncomment these if you're running inn
# news.crit /var/log/news/news.crit
# news.err /var/log/news/news.err
# news.notice /var/log/news/news.notice
!ppp <co id="co-syslog-prog-spec"/>
*.* /var/log/ppp.log
!*</programlisting>
<calloutlist>
<callout arearefs="co-syslog-many-match">
<para>Match all messages with a level of
<literal>err</literal> or higher, as well as
<literal>kern.warning</literal>,
<literal>auth.notice</literal> and
<literal>mail.crit</literal>, and send these log messages
to the console
(<devicename>/dev/console</devicename>).</para>
</callout>
<callout arearefs="co-syslog-one-match">
<para>Match all messages from the <literal>mail</literal>
facility at level <literal>info</literal> or above, and
log the messages to
<filename>/var/log/maillog</filename>.</para>
</callout>
<callout arearefs="co-syslog-comparison">
<para>This line uses a comparison flag, <literal>=</literal>
to only match messages at level <literal>debug</literal>,
and log them in
<filename>/var/log/debug.log</filename>.</para>
</callout>
<callout arearefs="co-syslog-prog-spec">
<para>Here is an example usage of a <emphasis>program
specification</emphasis>. This makes the rules
following it only valid for the program in the program
specification. In this case, this and the following
lines log all messages from &man.ppp.8;, but no other
programs, to
<filename>/var/log/ppp.log</filename>.</para>
</callout>
</calloutlist>
<para>This example shows that there are plenty of levels and
subsystems. The levels are, in order from most to least
critical: <literal>emerg</literal>, <literal>alert</literal>,
<literal>crit</literal>, <literal>err</literal>,
<literal>warning</literal>, <literal>notice</literal>,
<literal>info</literal>, and <literal>debug</literal>.</para>
<para>The facilities are, in no particular order:
<literal>auth</literal>, <literal>authpriv</literal>,
<literal>console</literal>, <literal>cron</literal>,
<literal>daemon</literal>, <literal>ftp</literal>,
<literal>kern</literal>, <literal>lpr</literal>,
<literal>mail</literal>, <literal>mark</literal>,
<literal>news</literal>, <literal>security</literal>,
<literal>syslog</literal>, <literal>user</literal>,
<literal>uucp</literal>, and <literal>local0</literal> through
<literal>local7</literal>. Be aware that other operating
systems might have different facilities.</para>
<para>With this knowledge, it is easy to add a new line to
<filename>/etc/syslog.conf</filename> to log everything from
the different daemons on level <literal>notice</literal> and
higher to <filename>/var/log/daemon.log</filename>. Just add
the following:</para>
<programlisting>daemon.notice /var/log/daemon.log</programlisting>
<para>For more information about the different levels and
facilities, refer to &man.syslog.3; and &man.syslogd.8;.
For more information about
<filename>/etc/syslog.conf</filename>, its syntax, and more
advanced usage examples, see &man.syslog.conf.5; and <xref
linkend="network-syslogd"/>.</para>
</sect2>
<sect2>
<title>Log Management and Rotation with
<command>newsyslog</command></title>
<indexterm><primary>newsyslog</primary></indexterm>
<indexterm><primary>newsyslog.conf</primary></indexterm>
<indexterm><primary>log rotation</primary></indexterm>
<indexterm><primary>log management</primary></indexterm>
<para>Log files tend to grow quickly and accumulate steadily.
This leads to the files being full of less immediately useful
information while filling up the hard drive. Log management
attempts to mitigate this. In &os;, &man.newsyslog.8; is used
to manage log files. This program periodically rotates and
compresses log files, and optionally creates missing log files
and signals programs when log files are moved. The log files
are not necessarily generated by &man.syslogd.8; as
&man.newsyslog.8; works with any logs written from any
program. While &man.newsyslog.8; is normally run from
&man.cron.8;, it is not a system daemon. In the default
configuration, it is run every hour.</para>
<sect3>
<title>Configuring
<command>newsyslog</command></title>
<para>To know which actions to take, &man.newsyslog.8; reads
its configuration file, by default
<filename>/etc/newsyslog.conf</filename>. This
configuration file contains one line for each file that
&man.newsyslog.8; manages. Each line states the file
owner, permissions, when to rotate that file, optional flags
that affect log rotation, such as compression, and programs
to signal when the log is rotated. Here is the default
configuration in &os;:</para>
<programlisting># configuration file for newsyslog
# $FreeBSD$
#
# Entries which do not specify the '/pid_file' field will cause the
# syslogd process to be signalled when that log file is rotated. This
# action is only appropriate for log files which are written to by the
# syslogd process (ie, files listed in /etc/syslog.conf). If there
# is no process which needs to be signalled when a given log file is
# rotated, then the entry for that file should include the 'N' flag.
#
# The 'flags' field is one or more of the letters: BCDGJNUXZ or a '-'.
#
# Note: some sites will want to select more restrictive protections than the
# defaults. In particular, it may be desirable to switch many of the 644
# entries to 640 or 600. For example, some sites will consider the
# contents of maillog, messages, and lpd-errs to be confidential. In the
# future, these defaults may change to more conservative ones.
#
# logfilename [owner:group] mode count size when flags [/pid_file] [sig_num]
/var/log/all.log 600 7 * @T00 J
/var/log/amd.log 644 7 100 * J
/var/log/auth.log 600 7 100 @0101T JC
/var/log/console.log 600 5 100 * J
/var/log/cron 600 3 100 * JC
/var/log/daily.log 640 7 * @T00 JN
/var/log/debug.log 600 7 100 * JC
/var/log/kerberos.log 600 7 100 * J
/var/log/lpd-errs 644 7 100 * JC
/var/log/maillog 640 7 * @T00 JC
/var/log/messages 644 5 100 @0101T JC
/var/log/monthly.log 640 12 * $M1D0 JN
/var/log/pflog 600 3 100 * JB /var/run/pflogd.pid
/var/log/ppp.log root:network 640 3 100 * JC
/var/log/security 600 10 100 * JC
/var/log/sendmail.st 640 10 * 168 B
/var/log/utx.log 644 3 * @01T05 B
/var/log/weekly.log 640 5 1 $W6D0 JN
/var/log/xferlog 600 7 100 * JC</programlisting>
<para>Each line starts with the name of the file to be
rotated, optionally followed by an owner and group for both
rotated and newly created files. The
<literal>mode</literal> field sets the permissions on the
log file and <literal>count</literal> denotes how many
rotated log files should be kept. The
<literal>size</literal> and <literal>when</literal> fields
tell &man.newsyslog.8; when to rotate the file. A log
file is rotated when either its size is larger than the
<literal>size</literal> field, or when the time in the
<literal>when</literal> filed has passed.
<literal>*</literal> means that this field is ignored. The
<replaceable>flags</replaceable> field gives
&man.newsyslog.8; further instructions, such as how to
compress the rotated file or to create the log file if it
is missing. The last two fields are optional, and
specify the <acronym
role="Process Identifier">PID</acronym> file of a process
and a signal number to send to that process when the file
is rotated. For more information on all fields, valid
flags, and how to specify the rotation time, refer to
&man.newsyslog.conf.5;. Since &man.newsyslog.8; is run
from &man.cron.8;, it can not rotate files more often than
it is run from &man.cron.8;.</para>
</sect3>
</sect2>
</sect1>
<sect1 id="configtuning-configfiles">
<title>Configuration Files</title>
<sect2>
<title><filename class="directory">/etc</filename>
Layout</title>
<para>There are a number of directories in which configuration
information is kept. These include:</para>
<informaltable frame="none" pgwide="1">
<tgroup cols="2">
<colspec colwidth="1*"/>
<colspec colwidth="2*"/>
<tbody>
<row>
<entry><filename
class="directory">/etc</filename></entry>
<entry>Generic system-specific configuration
information.</entry>
</row>
<row>
<entry><filename
class="directory">/etc/defaults</filename></entry>
<entry>Default versions of system configuration
files.</entry>
</row>
<row>
<entry><filename
class="directory">/etc/mail</filename></entry>
<entry>Extra &man.sendmail.8; configuration and other
<acronym>MTA</acronym> configuration files.</entry>
</row>
<row>
<entry><filename
class="directory">/etc/ppp</filename></entry>
<entry>Configuration for both user- and kernel-ppp
programs.</entry>
</row>
<row>
<entry><filename
class="directory">/etc/namedb</filename></entry>
<entry>Default location for &man.named.8; data.
Normally <filename>named.conf</filename> and zone
files are stored here.</entry>
</row>
<row>
<entry><filename
class="directory">/usr/local/etc</filename></entry>
<entry>Configuration files for installed applications.
May contain per-application subdirectories.</entry>
</row>
<row>
<entry><filename
class="directory">/usr/local/etc/rc.d</filename></entry>
<entry>&man.rc.8; scripts for installed
applications.</entry>
</row>
<row>
<entry><filename
class="directory">/var/db</filename></entry>
<entry>Automatically generated system-specific database
files, such as the package database and the
&man.locate.1; database.</entry>
</row>
</tbody>
</tgroup>
</informaltable>
</sect2>
<sect2>
<title>Hostnames</title>
<indexterm><primary>hostname</primary></indexterm>
<indexterm><primary>DNS</primary></indexterm>
<sect3>
<title><filename>/etc/resolv.conf</filename></title>
<indexterm>
<primary><filename>resolv.conf</filename></primary>
</indexterm>
<para>How a
&os; system accesses the Internet Domain Name System
(<acronym>DNS</acronym>) is controlled by
&man.resolv.conf.5;.</para>
<para>The most common entries to
<filename>/etc/resolv.conf</filename> are:</para>
<informaltable frame="none" pgwide="1">
<tgroup cols="2">
<colspec colwidth="1*"/>
<colspec colwidth="2*"/>
<tbody>
<row>
<entry><literal>nameserver</literal></entry>
<entry>The <acronym>IP</acronym> address of a name
server the resolver should query. The servers are
queried in the order listed with a maximum of
three.</entry>
</row>
<row>
<entry><literal>search</literal></entry>
<entry>Search list for hostname lookup. This is
normally determined by the domain of the local
hostname.</entry>
</row>
<row>
<entry><literal>domain</literal></entry>
<entry>The local domain name.</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>A typical <filename>/etc/resolv.conf</filename> looks
like this:</para>
<programlisting>search example.com
nameserver 147.11.1.11
nameserver 147.11.100.30</programlisting>
<note>
<para>Only one of the <literal>search</literal> and
<literal>domain</literal> options should be used.</para>
</note>
<para>When using <acronym>DHCP</acronym>, &man.dhclient.8;
usually rewrites <filename>/etc/resolv.conf</filename>
with information received from the <acronym>DHCP</acronym>
server.</para>
</sect3>
<sect3>
<title><filename>/etc/hosts</filename></title>
<indexterm><primary>hosts</primary></indexterm>
<para><filename>/etc/hosts</filename> is a simple text
database which works in conjunction with
<acronym>DNS</acronym> and
<acronym>NIS</acronym> to provide host name to
<acronym>IP</acronym> address mappings. Entries for local
computers connected via a <acronym>LAN</acronym> can be
added to this file for simplistic naming purposes instead
of setting up a &man.named.8; server. Additionally,
<filename>/etc/hosts</filename> can be used to provide a
local record of Internet names, reducing the need to query
external <acronym>DNS</acronym> servers for commonly
accessed names.</para>
<programlisting># $&os;$
#
#
# Host Database
#
# This file should contain the addresses and aliases for local hosts that
# share this file. Replace 'my.domain' below with the domainname of your
# machine.
#
# In the presence of the domain name service or NIS, this file may
# not be consulted at all; see /etc/nsswitch.conf for the resolution order.
#
#
::1 localhost localhost.my.domain
127.0.0.1 localhost localhost.my.domain
#
# Imaginary network.
#10.0.0.2 myname.my.domain myname
#10.0.0.3 myfriend.my.domain myfriend
#
# According to RFC 1918, you can use the following IP networks for
# private nets which will never be connected to the Internet:
#
# 10.0.0.0 - 10.255.255.255
# 172.16.0.0 - 172.31.255.255
# 192.168.0.0 - 192.168.255.255
#
# In case you want to be able to connect to the Internet, you need
# real official assigned numbers. Do not try to invent your own network
# numbers but instead get one from your network provider (if any) or
# from your regional registry (ARIN, APNIC, LACNIC, RIPE NCC, or AfriNIC.)
#</programlisting>
<para>The format of <filename>/etc/hosts</filename> is as
follows:</para>
<programlisting>[Internet address] [official hostname] [alias1] [alias2] ...</programlisting>
<para>For example:</para>
<programlisting>10.0.0.1 myRealHostname.example.com myRealHostname foobar1 foobar2</programlisting>
<para>Consult &man.hosts.5; for more information.</para>
</sect3>
</sect2>
</sect1>
<sect1 id="configtuning-sysctl">
<title>Tuning with &man.sysctl.8;</title>
<indexterm><primary>sysctl</primary></indexterm>
<indexterm>
<primary>tuning</primary>
<secondary>with sysctl</secondary>
</indexterm>
<para>&man.sysctl.8; is used to make changes to a running &os;
system. This includes many advanced options of the
<acronym>TCP/IP</acronym> stack and virtual memory system
that can dramatically improve performance for an experienced
system administrator. Over five hundred system variables can
be read and set using &man.sysctl.8;.</para>
<para>At its core, &man.sysctl.8; serves two functions: to read
and to modify system settings.</para>
<para>To view all readable variables:</para>
<screen>&prompt.user; <userinput>sysctl -a</userinput></screen>
<para>To read a particular variable, specify its name:</para>
<screen>&prompt.user; <userinput>sysctl kern.maxproc</userinput>
kern.maxproc: 1044</screen>
<para>To set a particular variable, use the
<replaceable>variable</replaceable>=<replaceable>value</replaceable>
syntax:</para>
<screen>&prompt.root; <userinput>sysctl kern.maxfiles=5000</userinput>
kern.maxfiles: 2088 -> 5000</screen>
<para>Settings of sysctl variables are usually either strings,
numbers, or booleans, where a a boolean is <literal>1</literal>
for yes or <literal>0</literal> for no.</para>
<para>To automatically set some variables each time the machine
boots, add them to <filename>/etc/sysctl.conf</filename>. For
more information, refer to &man.sysctl.conf.5; and <xref
linkend="configtuning-sysctlconf"/>.</para>
<sect2 id="configtuning-sysctlconf">
<title><filename>sysctl.conf</filename></title>
<indexterm><primary>sysctl.conf</primary></indexterm>
<indexterm><primary>sysctl</primary></indexterm>
<para>The configuration file for &man.sysctl.8;,
<filename>/etc/sysctl.conf</filename>, looks much like
<filename>/etc/rc.conf</filename>. Values are set in a
<literal>variable=value</literal> form. The specified values
are set after the system goes into multi-user mode. Not all
variables are settable in this mode.</para>
<para>For example, to turn off logging of fatal signal exits
and prevent users from seeing processes started by other
users, the following tunables can be set in
<filename>/etc/sysctl.conf</filename>:</para>
<programlisting># Do not log fatal signal exits (e.g., sig 11)
kern.logsigexit=0
# Prevent users from seeing information about processes that
# are being run under another UID.
security.bsd.see_other_uids=0</programlisting>
</sect2>
<sect2 id="sysctl-readonly">
<sect2info>
<authorgroup>
<author>
<firstname>Tom</firstname>
<surname>Rhodes</surname>
<contrib>Contributed by </contrib>
<!-- 31 January 2003 -->
</author>
</authorgroup>
</sect2info>
<title>&man.sysctl.8; Read-only</title>
<para>In some cases it may be desirable to modify read-only
&man.sysctl.8; values, which will require a reboot of the
system.</para>
<para>For instance, on some laptop models the &man.cardbus.4;
device will not probe memory ranges and will fail with errors
similar to:</para>
<screen>cbb0: Could not map register memory
device_probe_and_attach: cbb0 attach returned 12</screen>
<para>The fix requires the modification of a read-only
&man.sysctl.8; setting. Add
<option>hw.pci.allow_unsupported_io_range=1</option> to
<filename>/boot/loader.conf</filename> and reboot. Now
&man.cardbus.4; should work properly.</para>
</sect2>
</sect1>
<sect1 id="configtuning-disk">
<title>Tuning Disks</title>
<para>The following section will discuss various tuning
mechanisms and options which may be applied to disk
devices. In many cases, disks with mechanical parts,
such as <acronym>SCSI</acronym> drives, will be the
bottleneck driving down the overall system performance. While
a solution is to install a drive without mechanical parts,
such as a solid state drive, mechanical drives are not
going away anytime in the near future. When tuning disks,
it is advisable to utilize the features of the &man.iostat.8;
command to test various changes to the system. This
command will allow the user to obtain valuable information
on system <acronym>IO</acronym>.</para>
<sect2>
<title>Sysctl Variables</title>
<sect3>
<title><varname>vfs.vmiodirenable</varname></title>
<indexterm>
<primary><varname>vfs.vmiodirenable</varname></primary>
</indexterm>
<para>The <varname>vfs.vmiodirenable</varname> &man.sysctl.8;
variable
may be set to either <literal>0</literal> (off) or
<literal>1</literal> (on). It is set to
<literal>1</literal> by default. This variable controls
how directories are cached by the system. Most directories
are small, using just a single fragment (typically 1 K)
in the file system and typically 512 bytes in the
buffer cache. With this variable turned off, the buffer
cache will only cache a fixed number of directories, even
if the system has a huge amount of memory. When turned on,
this &man.sysctl.8; allows the buffer cache to use the
<acronym>VM</acronym> page cache to cache the directories,
making all the memory available for caching directories.
However, the minimum in-core memory used to cache a
directory is the physical page size (typically 4 K)
rather than 512 bytes. Keeping this option enabled
is recommended if the system is running any services which
manipulate large numbers of files. Such services can
include web caches, large mail systems, and news systems.
Keeping this option on will generally not reduce
performance, even with the wasted memory, but one should
experiment to find out.</para>
</sect3>
<sect3>
<title><varname>vfs.write_behind</varname></title>
<indexterm>
<primary><varname>vfs.write_behind</varname></primary>
</indexterm>
<para>The <varname>vfs.write_behind</varname> &man.sysctl.8;
variable
defaults to <literal>1</literal> (on). This tells the file
system to issue media writes as full clusters are collected,
which typically occurs when writing large sequential files.
This avoids saturating the buffer cache with dirty buffers
when it would not benefit I/O performance. However, this
may stall processes and under certain circumstances should
be turned off.</para>
</sect3>
<sect3>
<title><varname>vfs.hirunningspace</varname></title>
<indexterm>
<primary><varname>vfs.hirunningspace</varname></primary>
</indexterm>
<para>The <varname>vfs.hirunningspace</varname> &man.sysctl.8;
variable determines how much outstanding write I/O may be
queued to disk controllers system-wide at any given
instance. The default is usually sufficient, but on
machines with many disks, try bumping it up to four or five
<emphasis>megabytes</emphasis>. Setting too high a value
which exceeds the buffer cache's write threshold can lead
to bad clustering performance. Do not set this value
arbitrarily high as higher write values may add latency to
reads occurring at the same time.</para>
<para>There are various other buffer cache and
<acronym>VM</acronym> page cache related &man.sysctl.8;
values. Modifying these values is not recommended as the
<acronym>VM</acronym> system does a good job of
automatically tuning itself.</para>
</sect3>
<sect3>
<title><varname>vm.swap_idle_enabled</varname></title>
<indexterm>
<primary><varname>vm.swap_idle_enabled</varname></primary>
</indexterm>
<para>The <varname>vm.swap_idle_enabled</varname>
&man.sysctl.8; variable is useful in large multi-user
systems with many active login users and lots of idle
processes. Such systems tend to generate continuous
pressure on free memory reserves. Turning this feature on
and tweaking the swapout hysteresis (in idle seconds) via
<varname>vm.swap_idle_threshold1</varname> and
<varname>vm.swap_idle_threshold2</varname> depresses the
priority of memory pages associated with idle processes more
quickly then the normal pageout algorithm. This gives a
helping hand to the pageout daemon. Only turn this option
on if needed, because the tradeoff is essentially pre-page
memory sooner rather than later which eats more swap and
disk bandwidth. In a small system this option will have a
determinable effect, but in a large system that is already
doing moderate paging, this option allows the
<acronym>VM</acronym> system to stage whole processes into
and out of memory easily.</para>
</sect3>
<sect3>
<title><varname>hw.ata.wc</varname></title>
<indexterm>
<primary><varname>hw.ata.wc</varname></primary>
</indexterm>
<para>Turning off <acronym>IDE</acronym> write caching reduces
write bandwidth to <acronym>IDE</acronym> disks, but may
sometimes be necessary due to data consistency issues
introduced by hard drive vendors. The problem is that
some <acronym>IDE</acronym> drives lie about when a write
completes. With <acronym>IDE</acronym> write caching
turned on, <acronym>IDE</acronym> hard drives write data
to disk out of order and will sometimes delay writing some
blocks indefinitely when under heavy disk load. A crash or
power failure may cause serious file system corruption.
Check the default on the system by observing the
<varname>hw.ata.wc</varname> &man.sysctl.8; variable. If
<acronym>IDE</acronym> write caching is turned off, one can
set this read-only variable to
<literal>1</literal> in
<filename>/boot/loader.conf</filename> in order to enable
it at boot time.</para>
<para>For more information, refer to &man.ata.4;.</para>
</sect3>
<sect3>
<title><literal>SCSI_DELAY</literal>
(<varname>kern.cam.scsi_delay</varname>)</title>
<indexterm>
<primary><varname>kern.cam.scsi_delay</varname></primary>
</indexterm>
<indexterm>
<primary>kernel options</primary>
<secondary><literal>SCSI DELAY</literal></secondary>
</indexterm>
<para>The <literal>SCSI_DELAY</literal> kernel configuration
option may be used to reduce system boot times. The
defaults are fairly high and can be responsible for
<literal>15</literal> seconds of delay in the boot process.
Reducing it to <literal>5</literal> seconds usually works
with modern drives. The
<varname>kern.cam.scsi_delay</varname> boot time tunable
should be used. The tunable and kernel configuration
option accept values in terms of
<emphasis>milliseconds</emphasis> and
<emphasis>not</emphasis>
<emphasis>seconds</emphasis>.</para>
</sect3>
</sect2>
<sect2 id="soft-updates">
<title>Soft Updates</title>
<indexterm><primary>Soft Updates</primary></indexterm>
<indexterm><primary>&man.tunefs.8;</primary></indexterm>
<para>To fine-tune a file system, use &man.tunefs.8;. This
program has many different options. To toggle Soft Updates
on and off, use:</para>
<screen>&prompt.root; <userinput>tunefs -n enable /filesystem</userinput>
&prompt.root; <userinput>tunefs -n disable /filesystem</userinput></screen>
<para>A file system cannot be modified with &man.tunefs.8; while
it is mounted. A good time to enable Soft Updates is before
any partitions have been mounted, in single-user mode.</para>
<para>Soft Updates is recommended for <acronym>UFS</acronym>
file systems as it drastically improves meta-data performance,
mainly file creation and deletion, through the use of a memory
cache. There are two downsides to Soft Updates to be aware
of. First, Soft Updates guarantee file system consistency
in the case of a crash, but could easily be several seconds
or even a minute behind updating the physical disk. If the
system crashes, unwritten data may be lost. Secondly, Soft
Updates delay the freeing of file system blocks. If the
root file system is almost full, performing a major update,
such as <command>make installworld</command>, can cause the
file system to run out of space and the update to fail.</para>
<sect3>
<title>More Details About Soft Updates</title>
<indexterm>
<primary>Soft Updates</primary>
<secondary>details</secondary>
</indexterm>
<para>Meta-data updates are updates to non-content data like
inodes or directories. There are two traditional approaches
to writing a file system's meta-data back to disk.</para>
<para>Historically, the default behavior was to write out
meta-data updates synchronously. If a directory changed,
the system waited until the change was actually written to
disk. The file data buffers (file contents) were passed
through the buffer cache and backed up to disk later on
asynchronously. The advantage of this implementation is
that it operates safely. If there is a failure during an
update, meta-data is always in a consistent state. A
file is either created completely or not at all. If the
data blocks of a file did not find their way out of the
buffer cache onto the disk by the time of the crash,
&man.fsck.8; recognizes this and repairs the file system
by setting the file length to
<literal>0</literal>. Additionally, the implementation is
clear and simple. The disadvantage is that meta-data
changes are slow. For example, <command>rm -r</command>
touches all the files in a directory sequentially, but each
directory change will be written synchronously to the
disk. This includes updates to the directory itself, to
the inode table, and possibly to indirect blocks allocated
by the file. Similar considerations apply for unrolling
large hierarchies using <command>tar -x</command>.</para>
<para>The second approach is to use asynchronous meta-data
updates. This is the default for a <acronym>UFS</acronym>
file system mounted with <command>mount -o async</command>.
Since all meta-data updates are also passed through the
buffer cache, they will be intermixed with the updates of
the file content data. The advantage of this
implementation is there is no need to wait until each
meta-data update has been written to disk, so all operations
which cause huge amounts of meta-data updates work much
faster than in the synchronous case. This implementation
is still clear and simple, so there is a low risk for bugs
creeping into the code. The disadvantage is that there is
no guarantee for a consistent state of the file system.
If there is a failure during an operation that updated
large amounts of meta-data, like a power failure or someone
pressing the reset button, the file system will be left
in an unpredictable state. There is no opportunity to
examine the state of the file system when the system comes
up again as the data blocks of a file could already have
been written to the disk while the updates of the inode
table or the associated directory were not. It is
impossible to implement a &man.fsck.8; which is able to
clean up the resulting chaos because the necessary
information is not available on the disk. If the file
system has been damaged beyond repair, the only choice
is to reformat it and restore from backup.</para>
<para>The usual solution for this problem is to implement
<emphasis>dirty region logging</emphasis>, which is also
referred to as <emphasis>journaling</emphasis>.
Meta-data updates are still written synchronously, but only
into a small region of the disk. Later on, they are moved
to their proper location. Because the logging area is a
small, contiguous region on the disk, there are no long
distances for the disk heads to move, even during heavy
operations, so these operations are quicker than synchronous
updates. Additionally, the complexity of the implementation
is limited, so the risk of bugs being present is low. A
disadvantage is that all meta-data is written twice, once
into the logging region and once to the proper location, so
performance <quote>pessimization</quote> might result. On
the other hand, in case of a crash, all pending meta-data
operations can be either quickly rolled back or completed
from the logging area after the system comes up again,
resulting in a fast file system startup.</para>
<para>Kirk McKusick, the developer of Berkeley
<acronym>FFS</acronym>, solved this problem with Soft
Updates. All pending meta-data updates are kept in memory
and written out to disk in a sorted sequence
(<quote>ordered meta-data updates</quote>). This has the
effect that, in case of heavy meta-data operations, later
updates to an item <quote>catch</quote> the earlier ones
which are still in memory and have not already been written
to disk. All operations are generally performed in memory
before the update is written to disk and the data blocks are
sorted according to their position so that they will not be
on the disk ahead of their meta-data. If the system
crashes, an implicit <quote>log rewind</quote> causes all
operations which were not written to the disk appear as if
they never happened. A consistent file system state is
maintained that appears to be the one of 30 to 60 seconds
earlier. The algorithm used guarantees that all resources
in use are marked as such in their blocks and inodes.
After a crash, the only resource allocation error that
occurs is that resources are marked as <quote>used</quote>
which are actually <quote>free</quote>. &man.fsck.8;
recognizes this situation, and frees the resources that
are no longer used. It is safe to ignore the dirty state
of the file system after a crash by forcibly mounting it
with <command>mount -f</command>. In order to free
resources that may be unused, &man.fsck.8; needs to be run
at a later time. This is the idea behind the
<emphasis>background &man.fsck.8;</emphasis>: at system
startup time, only a <emphasis>snapshot</emphasis> of the
file system is recorded and &man.fsck.8; is run afterwards.
All file systems can then be mounted
<quote>dirty</quote>, so the system startup proceeds in
multi-user mode. Then, background &man.fsck.8; is
scheduled for all file systems where this is required, to
free resources that may be unused. File systems that do
not use Soft Updates still need the usual foreground
&man.fsck.8;.</para>
<para>The advantage is that meta-data operations are nearly
as fast as asynchronous updates and are faster than
<emphasis>logging</emphasis>, which has to write the
meta-data twice. The disadvantages are the complexity of
the code, a higher memory consumption, and some
idiosyncrasies. After a crash, the state of the file
system appears to be somewhat <quote>older</quote>. In
situations where the standard synchronous approach would
have caused some zero-length files to remain after the
&man.fsck.8;, these files do not exist at all with Soft
Updates because neither the meta-data nor the file contents
have been written to disk. Disk space is not released until
the updates have been written to disk, which may take place
some time after running &man.rm.1;. This may cause problems
when installing large amounts of data on a file system
that does not have enough free space to hold all the files
twice.</para>
</sect3>
</sect2>
</sect1>
<sect1 id="configtuning-kernel-limits">
<title>Tuning Kernel Limits</title>
<indexterm>
<primary>tuning</primary>
<secondary>kernel limits</secondary>
</indexterm>
<sect2 id="file-process-limits">
<title>File/Process Limits</title>
<sect3 id="kern-maxfiles">
<title><varname>kern.maxfiles</varname></title>
<indexterm>
<primary><varname>kern.maxfiles</varname></primary>
</indexterm>
<para>The <varname>kern.maxfiles</varname> &man.sysctl.8;
variable can be raised or lowered based upon system
requirements. This variable indicates the maximum number
of file descriptors on the system. When the file descriptor
table is full, <errorname>file: table is full</errorname>
will show up repeatedly in the system message buffer, which
can be viewed using &man.dmesg.8;.</para>
<para>Each open file, socket, or fifo uses one file
descriptor. A large-scale production server may easily
require many thousands of file descriptors, depending on the
kind and number of services running concurrently.</para>
<para>In older &os; releases, the default value of
<varname>kern.maxfiles</varname> is derived from
<option>maxusers</option> in the kernel configuration file.
<varname>kern.maxfiles</varname> grows proportionally to the
value of <option>maxusers</option>. When compiling a custom
kernel, consider setting this kernel configuration option
according to the use of the system. From this number, the
kernel is given most of its pre-defined limits. Even though
a production machine may not have 256 concurrent users, the
resources needed may be similar to a high-scale web
server.</para>
<para>The read-only &man.sysctl.8; variable
<varname>kern.maxusers</varname> is automatically sized at
boot based on the amount of memory available in the system,
and may be determined at run-time by inspecting the value
of <varname>kern.maxusers</varname>. Some systems require
larger or smaller values of
<varname>kern.maxusers</varname> and values of
<literal>64</literal>, <literal>128</literal>, and
<literal>256</literal> are not uncommon. Going above
<literal>256</literal> is not recommended unless a huge
number of file descriptors is needed. Many of the tunable
values set to their defaults by
<varname>kern.maxusers</varname> may be individually
overridden at boot-time or run-time in
<filename>/boot/loader.conf</filename>. Refer to
&man.loader.conf.5; and
<filename>/boot/defaults/loader.conf</filename> for more
details and some hints.</para>
<para>In older releases, the system will auto-tune
<literal>maxusers</literal> if it is set to
<literal>0</literal>.
<footnote><para>The auto-tuning algorithm sets
<literal>maxusers</literal> equal to the amount of
memory in the system, with a minimum of
<literal>32</literal>, and a maximum of
<literal>384</literal>.</para></footnote>. When
setting this option, set <literal>maxusers</literal> to
at least <literal>4</literal>, especially if the system
runs <application>&xorg;</application> or is used to
compile software. The most important table set by
<literal>maxusers</literal> is the maximum number of
processes, which is set to
<literal>20 + 16 * maxusers</literal>. If
<literal>maxusers</literal> is set to <literal>1</literal>,
there can only be
<literal>36</literal> simultaneous processes, including
the <literal>18</literal> or so that the system starts up
at boot time and the <literal>15</literal> or so used by
<application>&xorg;</application>. Even a simple task like
reading a manual page will start up nine processes to
filter, decompress, and view it. Setting
<literal>maxusers</literal> to <literal>64</literal> allows
up to <literal>1044</literal> simultaneous processes, which
should be enough for nearly all uses. If, however, the
<errortype>proc table full</errortype> error is displayed
when trying to start another program, or a server is
running with a large number of simultaneous users, increase
the number and rebuild.</para>
<note>
<para><literal>maxusers</literal> does
<emphasis>not</emphasis> limit the number of users which
can log into the machine. It instead sets various table
sizes to reasonable values considering the maximum number
of users on the system and how many processes each user
will be running.</para>
</note>
</sect3>
<sect3>
<title><varname>kern.ipc.somaxconn</varname></title>
<indexterm>
<primary><varname>kern.ipc.somaxconn</varname></primary>
</indexterm>
<para>The <varname>kern.ipc.somaxconn</varname> &man.sysctl.8;
variable limits the size of the listen queue for accepting
new <literal>TCP</literal> connections. The default value
of <literal>128</literal> is typically too low for robust
handling of new connections on a heavily loaded web server.
For such environments, it is recommended to increase this
value to <literal>1024</literal> or higher. A service
such as &man.sendmail.8;, or
<application>Apache</application> may itself limit the
listen queue size, but will often have a directive in its
configuration file to adjust the queue size. Large listen
queues do a better job of avoiding Denial of Service
(<acronym>DoS</acronym>) attacks.</para>
</sect3>
</sect2>
<sect2 id="nmbclusters">
<title>Network Limits</title>
<para>The <literal>NMBCLUSTERS</literal> kernel configuration
option dictates the amount of network Mbufs available to the
system. A heavily-trafficked server with a low number of
Mbufs will hinder performance. Each cluster represents
approximately 2 K of memory, so a value of
<literal>1024</literal> represents <literal>2</literal>
megabytes of kernel memory reserved for network buffers. A
simple calculation can be done to figure out how many are
needed. A web server which maxes out at
<literal>1000</literal> simultaneous connections where each
connection uses a 6 K receive and 16 K send buffer,
requires approximately 32 MB worth of network buffers
to cover the web server. A good rule of thumb is to multiply
by <literal>2</literal>, so
2x32 MB / 2 KB =
64 MB / 2 kB =
<literal>32768</literal>. Values between
<literal>4096</literal> and <literal>32768</literal> are
recommended for machines with greater amounts of memory.
Never specify an arbitrarily high value for this parameter
as it could lead to a boot time crash. To observe network
cluster usage, use <option>-m</option> with
&man.netstat.1;.</para>
<para>The <varname>kern.ipc.nmbclusters</varname> loader tunable
should be used to tune this at boot time. Only older versions
of &os; will require the use of the
<literal>NMBCLUSTERS</literal> kernel &man.config.8;
option.</para>
<para>For busy servers that make extensive use of the
&man.sendfile.2; system call, it may be necessary to increase
the number of &man.sendfile.2; buffers via the
<literal>NSFBUFS</literal> kernel configuration option or by
setting its value in <filename>/boot/loader.conf</filename>
(see &man.loader.8; for details). A common indicator that
this parameter needs to be adjusted is when processes are seen
in the <literal>sfbufa</literal> state. The &man.sysctl.8;
variable <varname>kern.ipc.nsfbufs</varname> is read-only.
This parameter nominally scales with
<varname>kern.maxusers</varname>, however it may be necessary
to tune accordingly.</para>
<important>
<para>Even though a socket has been marked as non-blocking,
calling &man.sendfile.2; on the non-blocking socket may
result in the &man.sendfile.2; call blocking until enough
<literal>struct sf_buf</literal>'s are made
available.</para>
</important>
<sect3>
<title><varname>net.inet.ip.portrange.*</varname></title>
<indexterm>
<primary>net.inet.ip.portrange.*</primary>
</indexterm>
<para>The <varname>net.inet.ip.portrange.*</varname>
&man.sysctl.8;
variables control the port number ranges automatically bound
to <literal>TCP</literal> and <literal>UDP</literal>
sockets. There are three ranges: a low range, a default
range, and a high range. Most network programs use the
default range which is controlled by
<varname>net.inet.ip.portrange.first</varname> and
<varname>net.inet.ip.portrange.last</varname>, which default
to <literal>1024</literal> and <literal>5000</literal>,
respectively. Bound port ranges are used for outgoing
connections and it is possible to run the system out of
ports under certain circumstances. This most commonly
occurs when running a heavily loaded web proxy. The port
range is not an issue when running a server which handles
mainly incoming connections, such as a web server, or has
a limited number of outgoing connections, such as a mail
relay. For situations where there is a shortage of ports,
it is recommended to increase
<varname>net.inet.ip.portrange.last</varname> modestly. A
value of <literal>10000</literal>, <literal>20000</literal>
or <literal>30000</literal> may be reasonable. Consider
firewall effects when changing the port range. Some
firewalls may block large ranges of ports, usually
low-numbered ports, and expect systems to use higher ranges
of ports for outgoing connections. For this reason, it
is not recommended that the value of
<varname>net.inet.ip.portrange.first</varname> be
lowered.</para>
</sect3>
<sect3>
<title><literal>TCP</literal> Bandwidth Delay Product</title>
<indexterm>
<primary><literal>TCP</literal> Bandwidth Delay Product
Limiting</primary>
<secondary><varname>net.inet.tcp.inflight.enable</varname></secondary>
</indexterm>
<para><literal>TCP</literal> bandwidth delay product limiting
can be enabled by setting the
<varname>net.inet.tcp.inflight.enable</varname>
&man.sysctl.8; variable to <literal>1</literal>. This
instructs the system to attempt to calculate the bandwidth
delay product for each connection and limit the amount of
data queued to the network to just the amount required to
maintain optimum throughput.</para>
<para>This feature is useful when serving data over modems,
Gigabit Ethernet, high speed <literal>WAN</literal> links,
or any other link with a high bandwidth delay product,
especially when also using window scaling or when a large
send window has been configured. When enabling this option,
also set <varname>net.inet.tcp.inflight.debug</varname> to
<literal>0</literal> to disable debugging. For production
use, setting <varname>net.inet.tcp.inflight.min</varname>
to at least <literal>6144</literal> may be beneficial.
Setting high minimums may effectively disable bandwidth
limiting, depending on the link. The limiting feature
reduces the amount of data built up in intermediate route
and switch packet queues and reduces the amount of data
built up in the local host's interface queue. With fewer
queued packets, interactive connections, especially over
slow modems, will operate with lower
<emphasis>Round Trip Times</emphasis>. This feature only
effects server side data transmission such as uploading.
It has no effect on data reception or downloading.</para>
<para>Adjusting <varname>net.inet.tcp.inflight.stab</varname>
is <emphasis>not</emphasis> recommended. This parameter
defaults to <literal>20</literal>, representing 2 maximal
packets added to the bandwidth delay product window
calculation. The additional window is required to stabilize
the algorithm and improve responsiveness to changing
conditions, but it can also result in higher &man.ping.8;
times over slow links, though still much lower than without
the inflight algorithm. In such cases, try reducing this
parameter to <literal>15</literal>,
<literal>10</literal>, or <literal>5</literal> and
reducing <varname>net.inet.tcp.inflight.min</varname>
to a value such as <literal>3500</literal> to get the
desired effect. Reducing these parameters should be done
as a last resort only.</para>
</sect3>
</sect2>
<sect2>
<title>Virtual Memory</title>
<sect3>
<title><varname>kern.maxvnodes</varname></title>
<para>A vnode is the internal representation of a file or
directory. Increasing the number of vnodes available to
the operating system reduces disk I/O. Normally, this is
handled by the operating system and does not need to be
changed. In some cases where disk I/O is a bottleneck and
the system is running out of vnodes, this setting needs
to be increased. The amount of inactive and free
<acronym>RAM</acronym> will need to be taken into
account.</para>
<para>To see the current number of vnodes in use:</para>
<screen>&prompt.root; <userinput>sysctl vfs.numvnodes</userinput>
vfs.numvnodes: 91349</screen>
<para>To see the maximum vnodes:</para>
<screen>&prompt.root; <userinput>sysctl kern.maxvnodes</userinput>
kern.maxvnodes: 100000</screen>
<para>If the current vnode usage is near the maximum, try
increasing <varname>kern.maxvnodes</varname> by a value of
<literal>1000</literal>. Keep an eye on the number of
<varname>vfs.numvnodes</varname>. If it climbs up to the
maximum again, <varname>kern.maxvnodes</varname> will need
to be increased further. Otherwise, a shift in memory
usage as reported by &man.top.1; should be visible and
more memory should be active.</para>
</sect3>
</sect2>
</sect1>
<sect1 id="adding-swap-space">
<title>Adding Swap Space</title>
<para>Sometimes a system requires more swap space. There are
three ways to increase swap space: add a new hard drive,
enable swap over <literal>NFS</literal>, or create a swap file
on an existing partition.</para>
<para>For information on how to encrypt swap space, which options
exist, and why it should be done, refer to <xref
linkend="swap-encrypting"/>.</para>
<sect2 id="new-drive-swap">
<title>Swap on a New or Existing Hard Drive</title>
<para>Adding a new hard drive for swap gives better performance
than adding a partition on an existing drive. Setting up
partitions and hard drives is explained in <xref
linkend="disks-adding"/> while <xref
linkend="configtuning-initial"/> discusses partition
layouts and swap partition size considerations.</para>
<para>Use &man.swapon.8; to add a swap partition to the system.
For example:</para>
<screen>&prompt.root; <userinput>swapon<replaceable> /dev/ada1s1b</replaceable></userinput></screen>
<warning>
<para>It is possible to use any partition not currently
mounted, even if it already contains data. Using
&man.swapon.8; on a partition that contains data will
overwrite and destroy that data. Make sure that the
partition to be added as swap is really the intended
partition before running &man.swapon.8;.</para>
</warning>
<para>To automatically add this swap partition on boot, add an
entry to <filename>/etc/fstab</filename>:</para>
<programlisting><replaceable>/dev/ada1s1b</replaceable> none swap sw 0 0</programlisting>
<para>See &man.fstab.5; for an explanation of the entries in
<filename>/etc/fstab</filename>.</para>
</sect2>
<sect2 id="nfs-swap">
<title>Swapping over <literal>NFS</literal></title>
<para>Swapping over <literal>NFS</literal> is only recommended
when there is no local hard disk to swap to.
<literal>NFS</literal> swapping will be limited by the
available network bandwidth and puts an additional burden
on &man.nfsd.8;.</para>
</sect2>
<sect2 id="create-swapfile">
<title>Swapfiles</title>
<para>To create a swap file, specify its size. The following
example creates a 64MB file named
<filename>/usr/swap0</filename>.</para>
<example>
<title>Creating a Swapfile on &os;</title>
<orderedlist>
<listitem>
<para>The <filename>GENERIC</filename> kernel already
includes the memory disk driver (&man.md.4;) required
for this operation. When building a custom kernel,
make sure to include the following line in the custom
configuration file:</para>
<programlisting>device md</programlisting>
<para>For information on building a custom kernel, refer
to <xref linkend="kernelconfig"/>.</para>
</listitem>
<listitem>
<para>First, create the swapfile
<filename>/usr/swap0</filename>:</para>
<screen>&prompt.root; <userinput>dd if=/dev/zero of=/usr/swap0 bs=1024k count=64</userinput></screen>
</listitem>
<listitem>
<para>Then, set proper permissions on
<filename>/usr/swap0</filename>:</para>
<screen>&prompt.root; <userinput>chmod 0600 /usr/swap0</userinput></screen>
</listitem>
<listitem>
<para>Enable the swap file in
<filename>/etc/rc.conf</filename>:</para>
<programlisting>swapfile="/usr/swap0" # Set to name of swapfile if aux swapfile desired.</programlisting>
</listitem>
<listitem>
<para>Reboot the machine or, to enable the swap file
immediately, type:</para>
<screen>&prompt.root; <userinput>mdconfig -a -t vnode -f /usr/swap0 -u 0 && swapon /dev/md0</userinput></screen>
</listitem>
</orderedlist>
</example>
</sect2>
</sect1>
<sect1 id="acpi-overview">
<sect1info>
<authorgroup>
<author>
<firstname>Hiten</firstname>
<surname>Pandya</surname>
<contrib>Written by </contrib>
</author>
<author>
<firstname>Tom</firstname>
<surname>Rhodes</surname>
</author>
</authorgroup>
</sect1info>
<title>Power and Resource Management</title>
<para>It is important to utilize hardware resources in an
efficient manner. Before the Advanced Configuration and Power
Interface (<acronym>ACPI</acronym>) was introduced, it was
difficult and inflexible for operating systems to manage the
power usage and thermal properties of a system. The hardware
was managed by the <acronym>BIOS</acronym> and the user had less
control and visibility into the power management settings. Some
limited configurability was available via <emphasis>Advanced
Power Management (<acronym>APM</acronym>)</emphasis>. Power
and resource management allows the operating system to monitor
system limits and to possibly provide an alert if the system
temperature increases unexpectedly.</para>
<para>This section provides comprehensive information about
<acronym>ACPI</acronym>. References will be provided for further
reading.</para>
<sect2 id="acpi-intro">
<title>What Is ACPI?</title>
<indexterm>
<primary>ACPI</primary>
</indexterm>
<indexterm>
<primary>APM</primary>
</indexterm>
<para><acronym>ACPI</acronym> is a standard written by an
alliance of vendors to provide a standard interface for
hardware resources and power management. It is a key
element in <emphasis>Operating System-directed configuration
and Power Management</emphasis> as it provides more control
and flexibility to the operating system. Modern systems
<quote>stretched</quote> the limits of the current Plug and
Play interfaces prior to the introduction of
<acronym>ACPI</acronym>. <acronym>ACPI</acronym> is the
direct successor to <acronym>APM</acronym>.</para>
</sect2>
<sect2 id="acpi-old-spec">
<title>Shortcomings of Advanced Power Management</title>
<para>The <acronym>APM</acronym> facility controls the power
usage of a system based on its activity. The
<acronym>APM</acronym> <acronym>BIOS</acronym> is supplied
by the vendor and is specific to the hardware platform. An
<acronym>APM</acronym> driver in the operating system
mediates access to the <emphasis><acronym>APM</acronym>
Software Interface</emphasis>, which allows management of
power levels. <acronym>APM</acronym> should still be used
for systems manufactured at or before the year 2000.</para>
<para>There are four major problems in <acronym>APM</acronym>.
First, power management is done by the vendor-specific
<acronym>BIOS</acronym>, separate from the operating system.
For example, the user can set idle-time values for a hard
drive in the <acronym>APM</acronym> <acronym>BIOS</acronym>
so that, when exceeded, the <acronym>BIOS</acronym> spins
down the hard drive without the consent of the operating
system. Second, the <acronym>APM</acronym> logic is embedded
in the <acronym>BIOS</acronym>, and it operates outside the
scope of the operating system. This means that users can
only fix problems in the <acronym>APM</acronym>
<acronym>BIOS</acronym> by flashing a new one into the
<acronym>ROM</acronym>, which is a dangerous procedure with
the potential to leave the system in an unrecoverable state
if it fails. Third, <acronym>APM</acronym> is a
vendor-specific technology, meaning that there is a lot of
duplication of efforts and bugs found in one vendor's
<acronym>BIOS</acronym> may not be solved in others. Lastly,
the <acronym>APM</acronym> <acronym>BIOS</acronym> did not
have enough room to implement a sophisticated power policy
or one that can adapt well to the purpose of the
machine.</para>
<para>The <emphasis>Plug and Play <acronym>BIOS</acronym>
(<acronym>PNPBIOS</acronym>)</emphasis> was unreliable in
many situations. <acronym>PNPBIOS</acronym> is 16-bit
technology, so the operating system has to use 16-bit
emulation in order to interface with
<acronym>PNPBIOS</acronym> methods.</para>
<para>The &os; <acronym>APM</acronym> driver is documented in
&man.apm.4;.</para>
</sect2>
<sect2 id="acpi-config">
<title>Configuring <acronym>ACPI</acronym></title>
<para>The &man.acpi.4; driver is loaded by default at start
up by &man.loader.8; and should
<emphasis>not</emphasis> be compiled into the kernel. The
reasoning is that modules are easier to work with and do not
require a kernel rebuild. This has the advantage of making
testing easier. Another reason is that starting
<acronym>ACPI</acronym> after a system has been brought up
often does not work well. If experiencing problems,
<acronym>ACPI</acronym> can be disabled altogether. This
driver should not and can not be unloaded because the system
bus uses it for various hardware interactions.
<acronym>ACPI</acronym> can be disabled by rebooting after
setting <literal>hint.acpi.0.disabled="1"</literal> in
<filename>/boot/loader.conf</filename> or by setting this
variable at the &man.loader.8; prompt.</para>
<note>
<para><acronym>ACPI</acronym> and <acronym>APM</acronym>
cannot coexist and should be used separately. The last one
to load will terminate if the driver notices the other is
running.</para>
</note>
<para><acronym>ACPI</acronym> can be used to put the system into
a sleep mode with &man.acpiconf.8;, the <option>-s</option>
flag, and a <literal>1-5</literal> option. Most users
only need <literal>1</literal> (quick suspend to
<acronym>RAM</acronym>) or <literal>3</literal> (suspend to
<acronym>RAM</acronym>). Option <literal>5</literal> performs
a soft-off which is the same action as:</para>
<screen>&prompt.root; <userinput>halt -p</userinput></screen>
<para>Other options are available via &man.sysctl.8;. Refer to
&man.acpi.4; and &man.acpiconf.8; for more information.</para>
</sect2>
</sect1>
<sect1 id="ACPI-debug">
<sect1info>
<authorgroup>
<author>
<firstname>Nate</firstname>
<surname>Lawson</surname>
<contrib>Written by </contrib>
</author>
</authorgroup>
<authorgroup>
<author>
<firstname>Peter</firstname>
<surname>Schultz</surname>
<contrib>With contributions from </contrib>
</author>
<author>
<firstname>Tom</firstname>
<surname>Rhodes</surname>
</author>
</authorgroup>
</sect1info>
<title>Using and Debugging &os; <acronym>ACPI</acronym></title>
<indexterm>
<primary>ACPI</primary>
<secondary>problems</secondary>
</indexterm>
<para><acronym>ACPI</acronym> is a fundamentally new way of
discovering devices, managing power usage, and providing
standardized access to various hardware previously managed by
the <acronym>BIOS</acronym>. Progress is being made toward
<acronym>ACPI</acronym> working on all systems, but bugs in some
motherboards' <firstterm><acronym>ACPI</acronym> Machine
Language</firstterm> (<acronym>AML</acronym>) bytecode,
incompleteness in &os;'s kernel subsystems, and bugs in the
&intel; <acronym>ACPI-CA</acronym> interpreter continue to
appear.</para>
<para>This section is intended to help users assist the &os;
<acronym>ACPI</acronym> maintainers in identifying the root
cause of problems and in debugging and developing a
solution.</para>
<sect2 id="ACPI-submitdebug">
<title>Submitting Debugging Information</title>
<note>
<para>Before submitting a problem, ensure the latest
<acronym>BIOS</acronym> version is installed and, if
available, the embedded controller firmware version.</para>
</note>
<para>When submitting a problem, send the following information
to <ulink url="mailto:freebsd-acpi@FreeBSD.org">
freebsd-acpi@FreeBSD.org</ulink>:</para>
<itemizedlist>
<listitem>
<para>Description of the buggy behavior, including system
type and model and anything that causes the bug to appear.
Note as accurately as possible when the bug began
occurring if it is new.</para>
</listitem>
<listitem>
<para>The output of &man.dmesg.8; after running
<command>boot -v</command>, including any error messages
generated by the bug.</para>
</listitem>
<listitem>
<para>The &man.dmesg.8; output from <command>boot
-v</command> with <acronym>ACPI</acronym> disabled,
if disabling it helps to fix the problem.</para>
</listitem>
<listitem>
<para>Output from <command>sysctl hw.acpi</command>. This
lists which features the system offers.</para>
</listitem>
<listitem>
<para>The <acronym>URL</acronym> to a pasted version of the
<firstterm><acronym>ACPI</acronym> Source
Language</firstterm> (<acronym>ASL</acronym>). Do
<emphasis>not</emphasis> send the
<acronym>ASL</acronym> directly to the list as it can be
very large. Generate a copy of the
<acronym>ASL</acronym> by running this command:</para>
<screen>&prompt.root; <userinput>acpidump -dt > <replaceable>name</replaceable>-<replaceable>system</replaceable>.asl</userinput></screen>
<para>Substitute the login name for
<replaceable>name</replaceable> and manufacturer/model for
<replaceable>system</replaceable>. For example, use
<filename>njl-FooCo6000.asl</filename>.</para>
</listitem>
</itemizedlist>
<para>Most &os; developers watch &a.current;, but one should
submit problems to &a.acpi.name; to be sure it is seen. Be
patient when waiting for a response. If the bug is not
immediately apparent, submit a
<acronym>PR</acronym> using &man.send-pr.1;. When entering a
<acronym>PR</acronym>, include the same information as
requested above. This helps developers to track the problem
and resolve it. Do not send a <acronym>PR</acronym> without
emailing &a.acpi.name; first as it is likely that the problem
has been reported before.</para>
</sect2>
<sect2 id="ACPI-background">
<title>Background</title>
<indexterm>
<primary><acronym>ACPI</acronym></primary>
</indexterm>
<para><acronym>ACPI</acronym> is present in all modern computers
that conform to the ia32 (x86), ia64 (Itanium), and amd64
(AMD) architectures. The full standard has many features
including <acronym>CPU</acronym> performance management, power
planes control, thermal zones, various battery systems,
embedded controllers, and bus enumeration. Most systems
implement less than the full standard. For instance, a
desktop system usually only implements bus enumeration
while a laptop might have cooling and battery management
support as well. Laptops also have suspend and resume, with
their own associated complexity.</para>
<para>An <acronym>ACPI</acronym>-compliant system has various
components. The <acronym>BIOS</acronym> and chipset vendors
provide various fixed tables, such as <acronym>FADT</acronym>,
in memory that specify things like the <acronym>APIC</acronym>
map (used for <acronym>SMP</acronym>), config registers, and
simple configuration values. Additionally, a bytecode table,
the <firstterm>Differentiated System Description
Table</firstterm> <acronym>DSDT</acronym>, specifies a
tree-like name space of devices and methods.</para>
<para>The <acronym>ACPI</acronym> driver must parse the fixed
tables, implement an interpreter for the bytecode, and modify
device drivers and the kernel to accept information from the
<acronym>ACPI</acronym> subsystem. For &os;, &intel; has
provided an interpreter (<acronym>ACPI-CA</acronym>) that is
shared with &linux; and NetBSD. The path to the
<acronym>ACPI-CA</acronym> source code is <filename
class="directory">src/sys/contrib/dev/acpica</filename>.
The glue code that allows <acronym>ACPI-CA</acronym> to work
on &os; is in <filename
class="directory">src/sys/dev/acpica/Osd</filename>.
Finally, drivers that implement various
<acronym>ACPI</acronym> devices are found in <filename
class="directory">src/sys/dev/acpica</filename>.</para>
</sect2>
<sect2 id="ACPI-comprob">
<title>Common Problems</title>
<indexterm>
<primary>ACPI</primary>
<secondary>problems</secondary>
</indexterm>
<para>For <acronym>ACPI</acronym> to work correctly, all the
parts have to work correctly. Here are some common problems,
in order of frequency of appearance, and some possible
workarounds or fixes.</para>
<sect3>
<title>Mouse Issues</title>
<para>In some cases, resuming from a suspend operation will
cause the mouse to fail. A known work around is to add
<literal>hint.psm.0.flags="0x3000"</literal> to
<filename>/boot/loader.conf</filename>. If this does not
work, consider sending a bug report using
&man.send-pr.1;.</para>
</sect3>
<sect3>
<title>Suspend/Resume</title>
<para><acronym>ACPI</acronym> has three suspend to
<acronym>RAM</acronym> (<acronym>STR</acronym>) states,
<literal>S1</literal>-<literal>S3</literal>, and one suspend
to disk state (<literal>STD</literal>), called
<literal>S4</literal>. <literal>S5</literal> is
<quote>soft off</quote> and is the normal state the system
is in when plugged in but not powered up.
<literal>S4</literal> can be implemented in two separate
ways. <literal>S4</literal><acronym>BIOS</acronym> is a
<acronym>BIOS</acronym>-assisted suspend to disk.
<literal>S4</literal><acronym>OS</acronym> is implemented
entirely by the operating system.</para>
<para>Start by checking <command>sysctl hw.acpi</command>
for the suspend-related items. Here are the results for a
Thinkpad:</para>
<screen>hw.acpi.supported_sleep_state: S3 S4 S5
hw.acpi.s4bios: 0</screen>
<para>Use <command>acpiconf -s</command> to test
<literal>S3</literal>,
<literal>S4</literal><acronym>OS</acronym>, and
<literal>S5</literal>. An <option>s4bios</option> of one
(<literal>1</literal>), indicates
<literal>S4</literal><acronym>BIOS</acronym> support instead
of <literal>S4</literal> <acronym>OS</acronym>.</para>
<para>When testing suspend/resume, start with
<literal>S1</literal>, if supported. This state is most
likely to work since it does not require much driver
support. No one has implemented <literal>S2</literal>,
which is similar to <literal>S1</literal>. Next, try
<literal>S3</literal>. This is the deepest
<acronym>STR</acronym> state and requires a lot of driver
support to properly reinitialize the hardware. If there are
problems resuming, email &a.acpi.name;. However, the
problem may not be resolved quickly since due to the amount
of drivers and hardware that need more testing and
work.</para>
<para>A common problem with suspend/resume is that many device
drivers do not save, restore, or reinitialize their
firmware, registers, or device memory properly. As a first
attempt at debugging the problem, try:</para>
<screen>&prompt.root; <userinput>sysctl debug.bootverbose=1</userinput>
&prompt.root; <userinput>sysctl debug.acpi.suspend_bounce=1</userinput>
&prompt.root; <userinput>acpiconf -s 3</userinput></screen>
<para>This test emulates the suspend/resume cycle of all
device drivers without actually going into
<literal>S3</literal> state. In some cases, problems such
as losing firmware state, device watchdog time out, and
retrying forever, can be captured with this method. Note
that the system will not really enter <literal>S3</literal>
state, which means devices may not lose power, and many
will work fine even if suspend/resume methods are totally
missing, unlike real <literal>S3</literal> state.</para>
<para>Harder cases require additional hardware, such as a
serial port and cable for debugging through a serial
console, a Firewire port and cable for using &man.dcons.4;,
and kernel debugging skills.</para>
<para>To help isolate the problem, remove as many drivers
from the kernel as possible. If it works, narrow down which
driver is the problem by loading drivers until it fails
again. Typically, binary drivers like
<filename>nvidia.ko</filename>, display drivers, and
<acronym>USB</acronym> will have the most problems while
Ethernet interfaces usually work fine. If drivers can be
properly loaded and unloaded, automate this by putting the
appropriate commands in
<filename>/etc/rc.suspend</filename> and
<filename>/etc/rc.resume</filename>.
Try setting <option>hw.acpi.reset_video</option> to
<literal>0</literal> if the display is messed up after
resume. Try setting longer or shorter values for
<option>hw.acpi.sleep_delay</option> to see if that
helps.</para>
<para>Try loading a recent &linux; distribution to see if
suspend/resume works on the same hardware. If it works on
&linux;, it is likely a &os; driver problem. Narrowing down
which driver causes the problem will assist developers in
fixing the problem. Since the <acronym>ACPI</acronym>
maintainers rarely maintain other drivers, such as sound
or <acronym>ATA</acronym>, any driver problems should also
be posted to the &a.current.name; list and mailed to the
driver maintainer. Advanced users can include debugging
&man.printf.3;s in a problematic driver to track down where
in its resume function it hangs.</para>
<para>Finally, try disabling <acronym>ACPI</acronym> and
enabling <acronym>APM</acronym> instead. If suspend/resume
works with <acronym>APM</acronym>, stick with
<acronym>APM</acronym>, especially on older hardware
(pre-2000). It took vendors a while to get
<acronym>ACPI</acronym> support correct and older hardware
is more likely to have <acronym>BIOS</acronym> problems with
<acronym>ACPI</acronym>.</para>
</sect3>
<sect3>
<title>System Hangs</title>
<para>Most system hangs are a result of lost interrupts or an
interrupt storm. Chipsets may have problems based on boot,
how the <acronym>BIOS</acronym> configures interrupts before
correctness of the <acronym>APIC</acronym>
(<acronym>MADT</acronym>) table, and routing of the
<firstterm>System Control Interrupt</firstterm>
(<acronym>SCI</acronym>).</para>
<indexterm>
<primary>interrupt storms</primary>
</indexterm>
<para>Interrupt storms can be distinguished from lost
interrupts by checking the output of
<command>vmstat -i</command> and looking at the line that
has <literal>acpi0</literal>. If the counter is increasing
at more than a couple per second, there is an interrupt
storm. If the system appears hung, try breaking to
<acronym>DDB</acronym> (<keycombo action="simul">
<keycap>CTRL</keycap>
<keycap>ALT</keycap>
<keycap>ESC</keycap>
</keycombo> on console) and type
<literal>show interrupts</literal>.</para>
<indexterm>
<primary>APIC</primary>
<secondary>disabling</secondary>
</indexterm>
<para>When dealing with interrupt problems, try disabling
<acronym>APIC</acronym> support with
<literal>hint.apic.0.disabled="1"</literal> in
<filename>/boot/loader.conf</filename>.</para>
</sect3>
<sect3>
<title>Panics</title>
<para>Panics are relatively rare for <acronym>ACPI</acronym>
and are the top priority to be fixed. The first step is to
isolate the steps to reproduce the panic, if possible, and
get a backtrace. Follow the advice for enabling
<literal>options DDB</literal> and setting up a serial
console in <xref linkend="serialconsole-ddb"/> or setting
up a &man.dump.8; partition. To get a backtrace in
<acronym>DDB</acronym>, use <literal>tr</literal>. When
handwriting the backtrace, get at least the last five
and the top five lines in the trace.</para>
<para>Then, try to isolate the problem by booting with
<acronym>ACPI</acronym> disabled. If that works, isolate
the <acronym>ACPI</acronym> subsystem by using various
values of <option>debug.acpi.disable</option>. See
&man.acpi.4; for some examples.</para>
</sect3>
<sect3>
<title>System Powers Up After Suspend or Shutdown</title>
<para>First, try setting
<literal>hw.acpi.disable_on_poweroff="0"</literal> in
&man.loader.conf.5;. This keeps <acronym>ACPI</acronym>
from disabling various events during the shutdown process.
Some systems need this value set to <literal>1</literal>
(the default) for the same reason. This usually fixes the
problem of a system powering up spontaneously after a
suspend or poweroff.</para>
</sect3>
<sect3>
<title>Other Problems</title>
<para>For other problems with <acronym>ACPI</acronym>, such as
it not working with a docking station or devices not being
detected, email a description to &a.acpi.name;. Some
issues may be related to unfinished parts of the
<acronym>ACPI</acronym> subsystem which might take a while
to be implemented. Be patient and prepared to test
patches.</para>
</sect3>
</sect2>
<sect2 id="ACPI-aslanddump">
<title><acronym>ASL</acronym>, &man.acpidump.8;, and
<acronym>IASL</acronym></title>
<indexterm>
<primary><acronym>ACPI</acronym></primary>
<secondary><acronym>ASL</acronym></secondary>
</indexterm>
<para>Some <acronym>BIOS</acronym> vendors provide incorrect
or buggy bytecode. This is usually manifested by kernel
console messages like this:</para>
<screen>ACPI-1287: *** Error: Method execution failed [\\_SB_.PCI0.LPC0.FIGD._STA] \\
(Node 0xc3f6d160), AE_NOT_FOUND</screen>
<para>Often, these problems may be resolved by updating the
<acronym>BIOS</acronym> to the latest revision. Most console
messages are harmless, but if there are other problems like
the battery status is not working, these messages are a
good place to start looking for problems. The bytecode,
known as <acronym>AML</acronym>, is compiled from a source
language called <acronym>ASL</acronym>. The
<acronym>AML</acronym> is found in the table known as the
<acronym>DSDT</acronym>. To get a copy of the system's
<acronym>ASL</acronym>, use &man.acpidump.8;. Include both
<option>-t</option>, to show the contents of the fixed tables,
and <option>-d</option>, to disassemble the
<acronym>AML</acronym>. Refer to <xref
linkend="ACPI-submitdebug"/> for an example syntax.</para>
<para>The simplest first check is to recompile the
<acronym>ASL</acronym> to check for errors. Warnings can
usually be ignored, but errors are bugs that will usually
prevent <acronym>ACPI</acronym> from working correctly. To
recompile the <acronym>ASL</acronym>, issue the following
command:</para>
<screen>&prompt.root; <userinput>iasl your.asl</userinput></screen>
</sect2>
<sect2 id="ACPI-fixasl">
<title>Fixing the <acronym>ASL</acronym></title>
<indexterm>
<primary><acronym>ACPI</acronym></primary>
<secondary><acronym>ASL</acronym></secondary>
</indexterm>
<para>The goal of &os; is for everyone to have working
<acronym>ACPI</acronym> without any user intervention. At
this point, workarounds are still being developed for common
mistakes made by <acronym>BIOS</acronym> vendors. The
µsoft; interpreter (<filename>acpi.sys</filename> and
<filename>acpiec.sys</filename>) does not strictly check for
adherence to the standard, and thus many
<acronym>BIOS</acronym> vendors who only test
<acronym>ACPI</acronym> under &windows; never fix their
<acronym>ASL</acronym>. &os; developers continue to identify
and document which non-standard behavior is allowed by
µsoft;'s interpreter and replicate it so that &os; can
work without forcing users to fix the <acronym>ASL</acronym>.
As a workaround, and to help identify behavior, fix the
<acronym>ASL</acronym> manually. If this works, send a
&man.diff.1; of the old and new <acronym>ASL</acronym> so
developers can possibly work around the buggy behavior in
<acronym>ACPI-CA</acronym>.</para>
<indexterm>
<primary><acronym>ACPI</acronym></primary>
<secondary>error messages</secondary>
</indexterm>
<para>Here is a list of common error messages, their cause, and
how to fix them:</para>
<sect3>
<title>Operating System Dependencies</title>
<para>Some <acronym>AML</acronym> versions assume the user is
running &windows;. To override this, set
<literal>hw.acpi.osname=<replaceable>"Windows
2001"</replaceable></literal> in
<filename>/boot/loader.conf</filename>, using the strings
in the <acronym>ASL</acronym>.</para>
</sect3>
<sect3>
<title>Missing Return Statements</title>
<para>Some methods do not explicitly return a value as the
standard requires. While <acronym>ACPI-CA</acronym>
does not handle this, &os; has a workaround that allows it
to return the value implicitly. Explicit return statements
can be added where required if the value which should be
returned is known. To force &man.iasl.8; to compile the
<acronym>ASL</acronym>, use the <option>-f</option>
flag.</para>
</sect3>
<sect3>
<title>Overriding the Default <acronym>AML</acronym></title>
<para>After customizing <filename>your.asl</filename>, compile
it with this command:</para>
<screen>&prompt.root; <userinput>iasl your.asl</userinput></screen>
<para>Adding the <option>-f</option> flag forces creation
of the <acronym>AML</acronym>, even if there are errors
during compilation. Some errors, such as missing return
statements, are automatically worked around by the
interpreter.</para>
<para>The default output filename for &man.iasl.8; is
<filename>DSDT.aml</filename>. Load this file instead of
the <acronym>BIOS</acronym>'s buggy copy, which is still
present in flash memory, by editing
<filename>/boot/loader.conf</filename> as follows:</para>
<programlisting>acpi_dsdt_load="YES"
acpi_dsdt_name="/boot/DSDT.aml"</programlisting>
<para>Be sure to copy <filename>DSDT.aml</filename> to
<filename class="directory">/boot</filename>.</para>
</sect3>
</sect2>
<sect2 id="ACPI-debugoutput">
<title>Getting Debugging Output from
<acronym>ACPI</acronym></title>
<indexterm>
<primary>ACPI</primary>
<secondary>problems</secondary>
</indexterm>
<indexterm>
<primary>ACPI</primary>
<secondary>debugging</secondary>
</indexterm>
<para>The <acronym>ACPI</acronym> driver has a flexible
debugging facility. A set of subsystems and the level of
verbosity can be specified. The subsystems to debug are
specified as <quote>layers</quote> and are broken down into
<acronym>ACPI-CA</acronym> components (ACPI_ALL_COMPONENTS)
and <acronym>ACPI</acronym> hardware support
(ACPI_ALL_DRIVERS). The verbosity of debugging output is
specified as the <quote>level</quote> and ranges from
ACPI_LV_ERROR (just report errors) to ACPI_LV_VERBOSE
(everything). The <quote>level</quote> is a bitmask so
multiple options can be set at once, separated by spaces. In
practice, a serial console should be used to log the output
so it is not lost as the console message buffer flushes.
A full list of the individual layers and levels is found in
&man.acpi.4;.</para>
<para>Debugging output is not enabled by default. To enable it,
add <literal>options ACPI_DEBUG</literal> to the kernel
configuration file if <acronym>ACPI</acronym> is compiled into
the kernel. Add <literal>ACPI_DEBUG=1</literal> to
<filename>/etc/make.conf</filename> to enable it globally.
If it is a module, recompile just the
<filename>acpi.ko</filename> module as follows:</para>
<screen>&prompt.root; <userinput>cd /sys/modules/acpi/acpi
&& make clean &&
make ACPI_DEBUG=1</userinput></screen>
<para>Install <filename>acpi.ko</filename> in <filename
class="directory">/boot/kernel</filename> and add the
desired level and layer to
<filename>/boot/loader.conf</filename>. This example enables
debug messages for all <acronym>ACPI-CA</acronym> components
and all <acronym>ACPI</acronym> hardware drivers such as
(<acronym>CPU</acronym> and <acronym>LID</acronym>. It only
outputs error messages at the least verbose level.</para>
<programlisting>debug.acpi.layer="ACPI_ALL_COMPONENTS ACPI_ALL_DRIVERS"
debug.acpi.level="ACPI_LV_ERROR"</programlisting>
<para>If the required information is triggered by a specific
event, such as a suspend and then resume, leave out changes to
<filename>/boot/loader.conf</filename> and instead use
&man.sysctl.8; to specify the layer and level after booting
and preparing the system for the specific event. The
variables which can be set using &man.sysctl.8; are named
the same as the tunables in
<filename>/boot/loader.conf</filename>.</para>
</sect2>
<sect2 id="ACPI-References">
<title>References</title>
<para>More information about <acronym>ACPI</acronym> may be
found in the following locations:</para>
<itemizedlist>
<listitem>
<para>The &a.acpi;</para>
</listitem>
<listitem>
<para>The <acronym>ACPI</acronym> Mailing List Archives
<ulink
url="http://lists.freebsd.org/pipermail/freebsd-acpi/"></ulink></para>
</listitem>
<listitem>
<para>The old <acronym>ACPI</acronym> Mailing List Archives
<ulink
url="http://home.jp.FreeBSD.org/mail-list/acpi-jp/"></ulink></para>
</listitem>
<listitem>
<para>The <acronym>ACPI</acronym> 2.0 Specification
<ulink url="http://acpi.info/spec.htm"></ulink></para>
</listitem>
<listitem>
<para>&man.acpi.4;,
&man.acpi.thermal.4;, &man.acpidump.8;, &man.iasl.8;,
and &man.acpidb.8;</para>
</listitem>
<listitem>
<para><ulink
url="http://www.cpqlinux.com/acpi-howto.html#fix_broken_dsdt">
<acronym>DSDT</acronym> debugging
resource</ulink>.</para>
</listitem>
</itemizedlist>
</sect2>
</sect1>
</chapter>