<?xml version="1.0" encoding="ISO8859-1" standalone="no"?>
<!DOCTYPE book PUBLIC "-//FreeBSD//DTD DocBook XML V4.2-Based Extension//EN"
	"../../../share/sgml/freebsd42.dtd" [
<!ENTITY % entities PUBLIC "-//FreeBSD//ENTITIES DocBook FreeBSD Entity Set//EN" "../../share/sgml/entities.ent">
%entities;
<!ENTITY bibliography SYSTEM "../../../share/sgml/bibliography.sgml">
<!ENTITY rel.head "<emphasis>10-CURRENT</emphasis>">
<!ENTITY rel.head.relx "10.<replaceable>X</replaceable>">
<!ENTITY rel.head.releng "<symbol>HEAD</symbol>">
<!ENTITY rel.head.packages "packages-10-current">
<!ENTITY rel.relx "9.<replaceable>X</replaceable>">
<!ENTITY rel.stable "<emphasis>9-STABLE</emphasis>">
<!ENTITY rel.releng "<symbol>RELENG_9</symbol>">
<!ENTITY rel.relengdate "September 2011">
<!ENTITY rel.packages "packages-9-stable">
<!ENTITY rel2.relx "8.<replaceable>X</replaceable>">
<!ENTITY rel2.stable "<emphasis>8-STABLE</emphasis>">
<!ENTITY rel2.releng "<symbol>RELENG_8</symbol>">
<!ENTITY rel2.relengdate "August 2009">
<!ENTITY rel2.packages "packages-8-stable">
<!ENTITY rel3.current "7.4">
<!ENTITY rel3.relx "7.<replaceable>X</replaceable>">
<!ENTITY rel3.stable "<emphasis>7-STABLE</emphasis>">
<!ENTITY rel3.releng "<symbol>RELENG_7</symbol>">
<!ENTITY rel3.relengdate "October 2007">
<!ENTITY rel3.packages "packages-7-stable">
]>

<book lang='en'>
  <bookinfo>
    <title>Frequently Asked Questions for &os;
      &rel3.relx;, &rel2.relx;, and &rel.relx;</title>

    <corpauthor>The &os; Documentation Project</corpauthor>

    <copyright>
      <year>1995</year>
      <year>1996</year>
      <year>1997</year>
      <year>1998</year>
      <year>1999</year>
      <year>2000</year>
      <year>2001</year>
      <year>2002</year>
      <year>2003</year>
      <year>2004</year>
      <year>2005</year>
      <year>2006</year>
      <year>2007</year>
      <year>2008</year>
      <year>2009</year>
      <year>2010</year>
      <year>2011</year>
      <year>2012</year>
      <holder>The &os; Documentation Project</holder>
    </copyright>

    &legalnotice;

    <legalnotice id="trademarks" role="trademarks">
      &tm-attrib.freebsd;
      &tm-attrib.3com;
      &tm-attrib.adobe;
      &tm-attrib.creative;
      &tm-attrib.cvsup;
      &tm-attrib.ibm;
      &tm-attrib.ieee;
      &tm-attrib.intel;
      &tm-attrib.iomega;
      &tm-attrib.linux;
      &tm-attrib.microsoft;
      &tm-attrib.mips;
      &tm-attrib.netscape;
      &tm-attrib.opengroup;
      &tm-attrib.oracle;
      &tm-attrib.sgi;
      &tm-attrib.sparc;
      &tm-attrib.sun;
      &tm-attrib.usrobotics;
      &tm-attrib.general;
    </legalnotice>

    <releaseinfo>$FreeBSD$</releaseinfo>

    <abstract>
      <para>This is the FAQ for &os; versions
	&rel3.relx;, &rel2.relx; and &rel.relx;.
	All entries are assumed to be
	relevant to &os; &rel3.relx; and later,
	unless otherwise noted.  If you are interested in helping with
	this project, send email to the &a.doc;.  The latest version of
	this document is always available from the <ulink
	  url="http://www.FreeBSD.org/doc/en_US.ISO8859-1/books/faq/index.html">&os; World Wide Web server</ulink>.
	It may also be downloaded as one large <ulink
	  url="book.html">HTML</ulink> file with HTTP or as plain text,
	&postscript;, PDF, etc. from the <ulink
	  url="ftp://ftp.FreeBSD.org/pub/FreeBSD/doc/">&os; FTP
	server</ulink>.  You may also want to <ulink
	  url="&url.base;/search/index.html">Search the FAQ</ulink>.
	</para>
    </abstract>
  </bookinfo>

  <chapter id="introduction">
    <title>Introduction</title>

    <para>Welcome to the &os;
      &rel3.relx;&nbsp;-, &rel2.relx;&nbsp;- and &rel.relx;&nbsp;-
      FAQ!</para>

    <para>As is usual with Usenet FAQs, this document aims to cover the
      most frequently asked questions concerning the &os; operating
      system (and of course answer them!).  Although originally intended
      to reduce bandwidth and avoid the same old questions being asked
      over and over again, FAQs have become recognized as valuable
      information resources.</para>

    <para>Every effort has been made to make this FAQ as informative as
      possible; if you have any suggestions as to how it may be
      improved, please feel free to mail them to the &a.doc;.</para>

    <qandaset>
      <qandaentry>
	<question id="what-is-FreeBSD">
	  <para>What is &os;?</para>
	</question>

	<answer>
	  <para>Briefly, &os; is a &unix;&nbsp;like operating system for
	    AMD64 and &intel; EM64T, &i386; PC-98, IA-64, &arm;,
	    &powerpc; and &ultrasparc; platforms based on U.C.
	    Berkeley's <quote>4.4BSD-Lite</quote> release, with some
	    <quote>4.4BSD-Lite2</quote> enhancements.  It is also based
	    indirectly on William Jolitz's port of U.C. Berkeley's
	    <quote>Net/2</quote> to the &i386;, known as
	    <quote>386BSD</quote>, though very little of the 386BSD code
	    remains.  A fuller description of what &os; is and how it
	    can work for you may be found on the <ulink
	      url="&url.base;/index.html">&os; home page</ulink>.
	    </para>

	  <para>&os; is used by companies, Internet Service Providers,
	    researchers, computer professionals, students and home users
	    all over the world in their work, education and
	    recreation.</para>

	  <para>For more detailed information on &os;, please see the
	    <ulink
	      url="&url.books.handbook;/index.html">&os; Handbook</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="FreeBSD-goals">
	  <para>What is the goal of the &os; Project?</para>
	</question>

	<answer>
	  <para>The goal of the &os; Project is to provide software that
	    may be used for any purpose and without strings attached.
	    Many of us have a significant investment in the code (and
	    project) and would certainly not mind a little financial
	    compensation now and then, but we definitely do not insist
	    on it.  We believe that our first and foremost
	    <quote>mission</quote> is to provide code to any and all
	    comers, and for whatever purpose, so that the code gets the
	    widest possible use and provides the widest possible
	    benefit.  This is, we believe, one of the most fundamental
	    goals of Free Software and one that we enthusiastically
	    support.</para>

	  <para>That code in our source tree which falls under the
	    <ulink
	      url="http://www.FreeBSD.org/copyright/COPYING">GNU General Public License (GPL)</ulink>
	    or <ulink
	      url="http://www.FreeBSD.org/copyright/COPYING.LIB">GNU Library General Public License (LGPL)</ulink>
	    comes with slightly more strings attached, though at least
	    on the side of enforced access rather than the usual
	    opposite.  Due to the additional complexities that can
	    evolve in the commercial use of GPL software, we do,
	    however, endeavor to replace such software with submissions
	    under the more relaxed <ulink
	      url="http://www.FreeBSD.org/copyright/freebsd-license.html">&os; license</ulink>
	    whenever possible.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="bsd-license-restrictions">
	  <para>Does the &os; license have any restrictions?</para>
	</question>

	<answer>
	  <para>Yes.  Those restrictions do not control how you use the
	    code, merely how you treat the &os; Project itself.  If you
	    have serious license concerns, read the actual <ulink
	      url="http://www.FreeBSD.org/copyright/freebsd-license.html">license</ulink>.
	    For the simply curious, the license can be summarized like
	    this.</para>

	  <itemizedlist>
	    <listitem>
	      <para>Do not claim that you wrote this.</para>
	    </listitem>

	    <listitem>
	      <para>Do not sue us if it breaks.</para>
	    </listitem>
	  </itemizedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="replace-current-OS">
	  <para>Can &os; replace my current operating system?</para>
	</question>

	<answer>
	  <para>For most people, yes.  But this question is not quite
	    that cut-and-dried.</para>

	  <para>Most people do not actually use an operating system.
	    They use applications.  The applications are what really use
	    the operating system.  &os; is designed to provide a robust
	    and full-featured environment for applications.  It supports
	    a wide variety of web browsers, office suites, email
	    readers, graphics programs, programming environments,
	    network servers, and just about everything else you might
	    want.  Most of these applications can be managed through the
	    <ulink
	      url="http://www.FreeBSD.org/ports/">Ports Collection</ulink>.
	    </para>

	  <para>If you need to use an application that is only available
	    on one operating system, you simply cannot replace that
	    operating system.  Chances are there is a very similar
	    application on &os;, however.  If you want a solid office or
	    Internet server, a reliable workstation, or just the ability
	    to do your job without interruptions, &os; will almost
	    certainly do everything you need.  Many computer users
	    across the world, including both novices and experienced
	    &unix; administrators, use &os; as their only desktop
	    operating system.</para>

	  <para>If you are migrating to &os; from some other &unix;
	    environment, you already know most of what you need to.  If
	    your background is in graphic-driven operating systems such
	    as &windows; and older versions of &macos;, expect to invest
	    additional time learning the &unix; way of doing things.
	    This FAQ and the <ulink
	      url="&url.books.handbook;/index.html">&os; Handbook</ulink>
	    are excellent places to start.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="why-called-FreeBSD">
	  <para>Why is it called &os;?</para>
	</question>

	<answer>
	  <itemizedlist>
	    <listitem>
	      <para>It may be used free of charge, even by commercial
		users.</para>
	    </listitem>

	    <listitem>
	      <para>Full source for the operating system is freely
		available, and the minimum possible restrictions have
		been placed upon its use, distribution and incorporation
		into other work (commercial or non-commercial).</para>
	    </listitem>

	    <listitem>
	      <para>Anyone who has an improvement or bug fix is free to
		submit their code and have it added to the source tree
		(subject to one or two obvious provisions).</para>
	    </listitem>
	  </itemizedlist>

	  <para>It is worth pointing out that the word
	    <quote>free</quote> is being used in two ways here, one
	    meaning <quote>at no cost</quote>, the other meaning
	    <quote>you can do whatever you like</quote>.  Apart from one
	    or two things you <emphasis>cannot</emphasis> do with the
	    &os; code, for example pretending you wrote it, you can
	    really do whatever you like with it.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="differences-to-other-bsds">
	  <para>What are the differences between &os; and NetBSD,
	    OpenBSD, and other open source BSD operating systems?</para>
	</question>

	<answer>
	  <para>James Howard wrote a good explanation of the history and
	    differences between the various projects,
	    called <ulink
	      url="http://www.freebsdworld.gr/freebsd/bsd-family-tree.html">The BSD Family Tree</ulink>
	    which goes a fair way to answering this question.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="latest-version">
	  <para>What is the latest version of &os;?</para>
	</question>

<!--
  This answer is a hack to deal with the fact that for now there are
  multiple "latest" versions of FreeBSD.
-->

	<answer>
	  <para>At any point in the development of &os;, there can be
	    multiple parallel branches.  &rel.relx; releases are
	    made from the &rel.stable; branch, and &rel2.relx;
	    releases are made from the &rel2.stable; branch.</para>

	  <para>Up until the release of 8.0, the
	    &rel3.relx; series was the one known as
	    <emphasis>-STABLE</emphasis>.  However, as of 8.0, the
	    &rel3.relx; branch will be designated for
	    an <quote>extended support</quote> status and receive only
	    fixes for major problems, such as security-related fixes.
	    There will be no more releases made from the
	    &rel3.stable; branch, and it is considered a
	    <quote>legacy</quote> branch and most current work will only
	    become a part of &rel.stable; and &rel2.stable;.</para>

	  <para>Version <ulink
	      url="ftp://ftp.FreeBSD.org/pub/FreeBSD/releases/i386/i386/&rel.current;-RELEASE/">&rel.current;</ulink>
	    is the latest release from the &rel.stable;
	    branch; it was released in &rel.current.date;.  Version
	    <ulink
	      url="ftp://ftp.FreeBSD.org/pub/FreeBSD/releases/i386/&rel2.current;-RELEASE/">&rel2.current;</ulink>
	    is the latest release from the &rel2.stable;
	    branch; it was released in &rel2.current.date;.</para>

	  <para>Briefly, <emphasis>-STABLE</emphasis> is aimed at the
	    ISP, corporate user, or any user who wants stability and a
	    minimal number of changes compared to the new (and possibly
	    unstable) features of the latest
	    <emphasis>-CURRENT</emphasis> snapshot.  Releases can come
	    from either branch, but <emphasis>-CURRENT</emphasis> should
	    only be used if you are prepared for its increased
	    volatility (relative to <emphasis>-STABLE</emphasis>, that
	    is).</para>

	  <para>Releases are made <link
	      linkend="release-freq">every few months</link>.  While
	    many people stay more up-to-date with the &os; sources (see
	    the questions on <link
	      linkend="current">&os.current;</link> and <link
	      linkend="stable">&os.stable;</link>) than that, doing so
	    is more of a commitment, as the sources are a moving
	    target.</para>

	  <para>More information on &os; releases can be found on the
	    <ulink
	      url="http://www.FreeBSD.org/releng/index.html">Release Engineering page</ulink>
	    on the &os; Web site.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="current">
	  <para>What is <emphasis>&os;-CURRENT</emphasis>?</para>
	</question>

	<answer>
	  <para><ulink
	      url="&url.books.handbook;/current-stable.html#CURRENT">&os.current;</ulink>
	    is the development version of the operating system, which
	    will in due course become the new &os.stable; branch.  As
	    such, it is really only of interest to developers working on
	    the system and die-hard hobbyists.  See the <ulink
	      url="&url.books.handbook;/current-stable.html#CURRENT">relevant section</ulink>
	    in the <ulink
	      url="&url.books.handbook;/index.html">Handbook</ulink> for
	    details on running <emphasis>-CURRENT</emphasis>.</para>

	  <para>If you are not familiar with the operating system or are
	    not capable of identifying the difference between a real
	    problem and a temporary problem, you should not use
	    &os.current;.  This branch sometimes evolves quite quickly
	    and can be un-buildable sometimes.
	    People that use &os.current; are expected to be able to
	    analyze any problems and only report them if they are deemed
	    to be mistakes rather than <quote>glitches</quote>.
	    Questions such as <quote>make world produces some error
	    about groups</quote> on the &a.current; may be treated with
	    contempt.</para>

	  <para>Every month, <ulink
	      url="&url.base;/snapshots/">snapshot</ulink>
	    releases are made based on the current state of the
	    <emphasis>-CURRENT</emphasis> and
	    <emphasis>-STABLE</emphasis> branches.  The goals behind
	    each snapshot release are:</para>

	  <itemizedlist>
	    <listitem>
	      <para>To test the latest version of the installation
		software.</para>
	    </listitem>

	    <listitem>
	      <para>To give people who would like to run
		<emphasis>-CURRENT</emphasis> or
		<emphasis>-STABLE</emphasis> but who do not have the
		time or bandwidth to follow it on a day-to-day basis an
		easy way of bootstrapping it onto their systems.</para>
	    </listitem>

	    <listitem>
	      <para>To preserve a fixed reference point for the code in
		question, just in case we break something really badly
		later.  (Although CVS normally prevents anything
		horrible like this happening.)</para>
	    </listitem>

	    <listitem>
	      <para>To ensure that all new features and fixes in need of
		testing have the greatest possible number of potential
		testers.</para>
	    </listitem>
	  </itemizedlist>

	  <para>No claims are made that any
	    <emphasis>-CURRENT</emphasis> snapshot can be considered
	    <quote>production quality</quote> for any purpose.  If you
	    want to run a stable and fully tested system, you will have
	    to stick to full releases, or use the
	    <emphasis>-STABLE</emphasis> snapshots.</para>

	  <para>Snapshot releases are directly available from <ulink
	      url="&url.base;/snapshots/">snapshot</ulink>.
	    </para>

	  <para>Official snapshots are generated each month on a regular
	    basis for all actively developed branches.  There are also
	    daily snapshot builds of the popular &arch.i386; and
	    &arch.amd64; branches, hosted on <ulink
	      url="http://snapshots.us.freebsd.org/"></ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="stable">
	  <para>What is the <emphasis>&os;-STABLE</emphasis>
	    concept?</para>
	</question>

	<answer>
	  <para>Back when &os;&nbsp;2.0.5 was released, &os; development
	    branched in two.  One branch was named <ulink
	      url="&url.books.handbook;/current-stable.html#STABLE">-STABLE</ulink>,
	    one <ulink
	      url="&url.books.handbook;/current-stable.html#CURRENT">-CURRENT</ulink>.
	    <emphasis>&os;-STABLE</emphasis> is intended for Internet
	    Service Providers and other commercial enterprises for whom
	    sudden shifts or experimental features are quite
	    undesirable.  It receives only well-tested bug fixes and
	    other small incremental enhancements.
	    <emphasis>&os;-CURRENT</emphasis>, on the other hand, has
	    been one unbroken line since 2.0 was released, leading
	    towards &rel.current;-RELEASE and beyond.  For more detailed
	    information on branches see <quote><ulink
	      url="&url.articles.releng;/release-proc.html#REL-BRANCH">&os; Release Engineering: Creating the Release Branch</ulink></quote>,
	    the status of the branches and the upcoming release schedule
	    can be found on the <ulink
	      url="http://www.FreeBSD.org/releng">Release Engineering Information</ulink> page.
	    </para>

	  <para>The 2.2-STABLE branch was retired with the release of
	    2.2.8.  The 3-STABLE branch has ended with the release of
	    3.5.1, the final 3.<replaceable>X</replaceable> release.
	    The 4-STABLE branch has ended with the release of 4.11, the
	    final 4.<replaceable>X</replaceable> release.  The only
	    changes made to either of these branches will be, for the
	    most part, security-related bug fixes.  Support for the
	    5-STABLE branches has ended with the release of 5.5, the
	    final 5.<replaceable>X</replaceable> release.  Support for
	    the 6-STABLE branch will continue for some time but focus
	    primarily on security-related bug fixes and other serious
	    issues.</para>

	  <para>&rel.current;-STABLE is the actively developed
	    <emphasis>-STABLE</emphasis> branch.  The latest release on
	    the &rel.current;-STABLE branch is &rel.current;-RELEASE,
	    which was released in &rel.current.date;.</para>

	  <para>The &rel.head; branch is the actively developed
	    <emphasis>-CURRENT</emphasis> branch toward the next
	    generation of &os;.  See <link
	      linkend="current">What is &os;-CURRENT?</link> for more
	    information on this branch.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="release-freq">
	  <para>When are &os; releases made?</para>
	</question>

	<answer>
	  <para>The &a.re; releases a new major version of &os; about
	    every 18 months and a new minor version about every 8 months,
	    on average.  Release dates are announced well in advance, so
	    that the people working on the system know when their
	    projects need to be finished and tested.  A testing period
	    precedes each release, in order to ensure that the addition
	    of new features does not compromise the stability of the
	    release.  Many users regard this caution as one of the best
	    things about &os;, even though waiting for all the latest
	    goodies to reach <emphasis>-STABLE</emphasis> can be a
	    little frustrating.</para>

	  <para>More information on the release engineering process
	    (including a schedule of upcoming releases) can be found on
	    the <ulink
	      url="http://www.FreeBSD.org/releng/index.html">release engineering</ulink>
	    pages on the &os; Web site.</para>

	  <para>For people who need or want a little more excitement,
	    binary snapshots are made daily as discussed above.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="responsible">
	  <para>Who is responsible for &os;?</para>
	</question>

	<answer>
	  <para>The key decisions concerning the &os; project, such as
	    the overall direction of the project and who is allowed to
	    add code to the source tree, are made by a <ulink
	      url="&url.base;/administration.html#t-core">core team</ulink> of
	    9 people.  There is a much larger team of more than 350
	    <ulink
	      url="&url.articles.contributors;/article.html#STAFF-COMMITTERS">committers</ulink>
	    who are authorized to make changes directly to the &os;
	    source tree.</para>

	  <para>However, most non-trivial changes are discussed in
	    advance in the <link linkend="mailing">mailing lists</link>,
	    and there are no restrictions on who may take part in the
	    discussion.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="where-get">
	  <para>Where can I get &os;?</para>
	</question>

	<answer>
	  <para>Every significant release of &os; is available via
	    anonymous FTP from the <ulink
	      url="ftp://ftp.FreeBSD.org/pub/FreeBSD/"> &os; FTP site</ulink>:
	    </para>

	  <itemizedlist>
	    <listitem>
	      <para>The latest &rel.stable; release, &rel.current;-RELEASE
		can be found in the <ulink
		  url="ftp://ftp.FreeBSD.org/pub/FreeBSD/releases/i386/i386/&rel.current;-RELEASE/">&rel.current;-RELEASE directory</ulink>.
		</para>
	    </listitem>

	    <listitem>
	      <para><ulink url="&url.base;/snapshots/"> Snapshot</ulink>
		releases are made monthly for the <link
		  linkend="current">-CURRENT</link> and <link
		  linkend="stable">-STABLE</link> branch, these being of
		service purely to bleeding-edge testers and
		developers.</para>
	    </listitem>

	    <listitem>
	      <para>The latest &rel2.stable; release, &rel2.current;-RELEASE
		can be found in the <ulink
		  url="ftp://ftp.FreeBSD.org/pub/FreeBSD/releases/i386/&rel2.current;-RELEASE/">&rel2.current;-RELEASE directory</ulink>.
		</para>
	    </listitem>

	    <listitem>
	      <para>The latest &rel3.stable; release, &rel3.current;-RELEASE
		can be found in the <ulink
		  url="ftp://ftp.FreeBSD.org/pub/FreeBSD/releases/i386/&rel3.current;-RELEASE/">&rel3.current;-RELEASE directory</ulink>.
		</para>
	    </listitem>
	  </itemizedlist>

	  <para>Information about obtaining &os; on CD, DVD, and other
	    media can be found in <ulink
	      url="&url.books.handbook;/mirrors.html">the Handbook</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="access-pr">
	  <para>How do I access the Problem Report database?</para>
	</question>

	<answer>
	  <para>The Problem Report database of all user change requests
	    may be queried by using our web-based PR <ulink
	      url="http://www.FreeBSD.org/cgi/query-pr.cgi?query">query</ulink>
	    interface.</para>

	  <para>The &man.send-pr.1; command can be used to submit
	    problem reports and change requests via electronic mail.
	    Alternatively, the <ulink
	      url="http://www.freebsd.org/send-pr.html">web-based problem report submission interface</ulink>
	    can be used to submit problem reports through a web
	    browser.</para>

	  <para>Before submitting a problem report, please read <ulink
	      url="&url.articles.problem-reports;/article.html">Writing &os; Problem Reports</ulink>,
	    an article on how to write good problem reports.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="other-info-sources">
	  <para>What other sources of information are there?</para>
	</question>

	<answer>
	  <para>Please check the <ulink
	      url="http://www.FreeBSD.org/docs.html">Documentation</ulink>
	    list on the main <ulink
	      url="http://www.FreeBSD.org">&os;</ulink> web site.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="support">
    <title>Documentation and Support</title>

    <qandaset>
      <qandaentry>
	<question id="books">
	  <para>What good books are there about &os;?</para>
	</question>

	<answer>
	  <para>The project produces a wide range of documentation,
	    available online from this link: <ulink
	      url="http://www.FreeBSD.org/docs.html"></ulink>.  In addition, <link
	      linkend="bibliography">the Bibliography</link> at the end of this
	    FAQ, and <ulink
	      url="&url.books.handbook;/bibliography.html">the one in the Handbook</ulink>
	    reference other recommended books.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="doc-formats">
	  <para>Is the documentation available in other formats, such as
	    plain text (ASCII), or &postscript;?</para>
	</question>

	<answer>
	  <para>Yes.  The documentation is available in a number of
	    different formats and compression schemes on the &os; FTP
	    site, in the <ulink
	      url="ftp://ftp.FreeBSD.org/pub/FreeBSD/doc/">/pub/FreeBSD/doc/</ulink>
	    directory.</para>

	  <para>The documentation is categorized in a number of
	    different ways.  These include:</para>

	  <itemizedlist>
	    <listitem>
	      <para>The document's name, such as <literal>faq</literal>,
		or <literal>handbook</literal>.</para>
	    </listitem>

	    <listitem>
	      <para>The document's language and encoding.  These are
		based on the locale names you will find under
		<filename class="directory">/usr/share/locale</filename> on your &os;
		system.  The current languages and encodings that we
		have for documentation are as follows:</para>

	      <informaltable frame="none" pgwide="1">
		<tgroup cols="2">
		  <thead>
		    <row>
		      <entry>Name</entry>

		      <entry>Meaning</entry>
		    </row>
		  </thead>

		  <tbody>
		    <row>
		      <entry><literal>en_US.ISO8859-1</literal></entry>

		      <entry>English (United States)</entry>
		    </row>

		    <row>
		      <entry><literal>bn_BD.ISO10646-1</literal></entry>

		      <entry>Bengali or Bangla (Bangladesh)</entry>
		    </row>

		    <row>
		      <entry><literal>da_DK.ISO8859-1</literal></entry>

		      <entry>Danish (Denmark)</entry>
		    </row>

		    <row>
		      <entry><literal>de_DE.ISO8859-1</literal></entry>

		      <entry>German (Germany)</entry>
		    </row>

		    <row>
		      <entry><literal>el_GR.ISO8859-7</literal></entry>

		      <entry>Greek (Greece)</entry>
		    </row>

		    <row>
		      <entry><literal>es_ES.ISO8859-1</literal></entry>

		      <entry>Spanish (Spain)</entry>
		    </row>

		    <row>
		      <entry><literal>fr_FR.ISO8859-1</literal></entry>

		      <entry>French (France)</entry>
		    </row>

		    <row>
		      <entry><literal>hu_HU.ISO8859-2</literal></entry>

		      <entry>Hungarian (Hungary)</entry>
		    </row>

		    <row>
		      <entry><literal>it_IT.ISO8859-15</literal></entry>

		      <entry>Italian (Italy)</entry>
		    </row>

		    <row>
		      <entry><literal>ja_JP.eucJP</literal></entry>

		      <entry>Japanese (Japan, EUC encoding)</entry>
		    </row>

		    <row>
		      <entry><literal>mn_MN.UTF-8</literal></entry>

		      <entry>Mongolian (Mongolia, UTF-8 encoding)</entry>
		    </row>

		    <row>
		      <entry><literal>nl_NL.ISO8859-1</literal></entry>

		      <entry>Dutch (Netherlands)</entry>
		    </row>

		    <row>
		      <entry><literal>no_NO.ISO8859-1</literal></entry>

		      <entry>Norwegian (Norway)</entry>
		    </row>

		    <row>
		      <entry><literal>pl_PL.ISO8859-2</literal></entry>

		      <entry>Polish (Poland)</entry>
		    </row>

		    <row>
		      <entry><literal>pt_BR.ISO8859-1</literal></entry>

		      <entry>Portuguese (Brazil)</entry>
		    </row>

		    <row>
		      <entry><literal>ru_RU.KOI8-R</literal></entry>

		      <entry>Russian (Russia, KOI8-R encoding)</entry>
		    </row>

		    <row>
		      <entry><literal>sr_YU.ISO8859-2</literal></entry>

		      <entry>Serbian (Serbia)</entry>
		    </row>

		    <row>
		      <entry><literal>tr_TR.ISO8859-9</literal></entry>

		      <entry>Turkish (Turkey)</entry>
		    </row>

		    <row>
		      <entry><literal>zh_CN.GB2312</literal></entry>

		      <entry>Simplified Chinese (China, GB2312
			encoding)</entry>
		    </row>

		    <row>
		      <entry><literal>zh_TW.Big5</literal></entry>

		      <entry>Traditional Chinese (Taiwan, Big5 encoding)</entry>
		    </row>
		  </tbody>
		</tgroup>
	      </informaltable>

	      <note>
		<para>Some documents may not be available in all
		  languages.</para>
	      </note>
	    </listitem>

	    <listitem>
	      <para>The document's format.  We produce the documentation
		in a number of different output formats.  Each format
		has its own advantages and disadvantages.  Some formats
		are better suited for online reading, while others are
		meant to be aesthetically pleasing when printed on
		paper.  Having the documentation available in any of
		these formats ensures that our readers will be able to
		read the parts they are interested in, either on their
		monitor, or on paper after printing the documents.  The
		currently available formats are:</para>

	      <informaltable frame="none" pgwide="1">
		<tgroup cols="2">
		  <thead>
		    <row>
		      <entry>Format</entry>

		      <entry>Meaning</entry>
		    </row>
		  </thead>

		  <tbody>
		    <row>
		      <entry><literal>html-split</literal></entry>

		      <entry>A collection of small, linked, HTML
			files.</entry>
		    </row>

		    <row>
		      <entry><literal>html</literal></entry>

		      <entry>One large HTML file containing the entire
			document</entry>
		    </row>

		    <row>
		      <entry><literal>pdf</literal></entry>

		      <entry>Adobe's Portable Document Format</entry>
		    </row>

		    <row>
		      <entry><literal>ps</literal></entry>

		      <entry>&postscript;</entry>
		    </row>

		    <row>
		      <entry><literal>rtf</literal></entry>

		      <entry>Microsoft's Rich Text Format</entry>
		    </row>

		    <row>
		      <entry><literal>txt</literal></entry>

		      <entry>Plain text</entry>
		    </row>
		  </tbody>
		</tgroup>
	      </informaltable>

	      <note>
		<para>Page numbers are not automatically updated when
		  loading Rich Text Format into Word.  Press <keycombo
		    action="simul"><keycap>Ctrl</keycap><keycap>A</keycap></keycombo>,
		  <keycombo
		    action="simul"><keycap>Ctrl</keycap><keycap>End</keycap></keycombo>,
		  <keycap>F9</keycap> after loading the document, to
		  update the page numbers.</para>
	      </note>
	    </listitem>

	    <listitem>
	      <para>The compression and packaging scheme.  There are
		three of these currently in use.</para>

	      <orderedlist>
		<listitem>
		  <para>Where the format is
		    <literal>html-split</literal>, the files are bundled
		    up using &man.tar.1;.  The resulting
		    <filename>.tar</filename> file is then compressed
		    using the compression schemes detailed in the next
		    point.</para>
		</listitem>

		<listitem>
		  <para>All the other formats generate one file, called
		    <filename><replaceable>type</replaceable>.<replaceable>format</replaceable></filename>
		    (i.e., <filename>article.pdf</filename>,
		    <filename>book.html</filename>, and so on).</para>

		  <para>These files are then compressed using two
		    compression schemes.</para>

		  <informaltable frame="none" pgwide="1">
		    <tgroup cols="2">
		      <thead>
			<row>
			  <entry>Scheme</entry>

			  <entry>Description</entry>
			</row>
		      </thead>

		      <tbody>
			<row>
			  <entry><literal>zip</literal></entry>

			  <entry>The <literal>zip</literal> format.  If you want to
			    uncompress this on &os; you will need to
			    install the <filename
			      role="package">archivers/unzip</filename>
			    port first.</entry>
			</row>

			<row>
			  <entry><literal>bz2</literal></entry>

			  <entry>The <literal>bzip2</literal> format.  Less widespread than
			    <literal>zip</literal>, but generally gives smaller files.
			    Install the <filename
			      role="package">archivers/bzip2</filename>
			    port to uncompress these files.</entry>
			</row>
		      </tbody>
		    </tgroup>
		  </informaltable>

		  <para>So the &postscript; version of the Handbook,
		    compressed using <literal>bzip2</literal> will be stored in a file
		    called <filename>book.ps.bz2</filename> in the
		    <filename class="directory">handbook/</filename> directory.</para>
		</listitem>
	      </orderedlist>
	    </listitem>
	  </itemizedlist>

	  <para>After choosing the format and compression mechanism that
	    you want to download, you will have to download the compressed
	    files yourself, uncompress them, and then copy the
	    appropriate documents into place.</para>

	  <para>For example, the split HTML version of the FAQ,
	    compressed using &man.bzip2.1;, can be found in the
	    <filename>doc/en_US.ISO8859-1/books/faq/book.html-split.tar.bz2</filename>
	    file.  To download and uncompress that file you would have
	    to do this.</para>

	  <screen>&prompt.root; <userinput>fetch ftp://ftp.FreeBSD.org/pub/FreeBSD/doc/en_US.ISO8859-1/books/faq/book.html-split.tar.bz2</userinput>
&prompt.root; <userinput>bzip2 -d book.html-split.tar.bz2</userinput>
&prompt.root; <userinput>tar xvf book.html-split.tar</userinput></screen>

	  <para>You will be left with a collection of
	    <filename>.html</filename> files.  The main one is called
	    <filename>index.html</filename>, which will contain the
	    table of contents, introductory material, and links to the
	    other parts of the document.  You can then copy or move
	    these to their final location as necessary.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="mailing">
	  <para>Where do I find info on the &os; mailing lists?</para>
	</question>

	<answer>
	  <para>You can find full information in the <ulink
	      url="&url.books.handbook;/eresources.html#ERESOURCES-MAIL">Handbook entry on mailing-lists</ulink>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="newsgroups">
	  <para>What &os; news groups are available?</para>
	</question>

	<answer>
	  <para>You can find full information in the <ulink
	      url="&url.books.handbook;/eresources-news.html">Handbook entry on newsgroups</ulink>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="irc">
	  <para>Are there &os; IRC (Internet Relay Chat)
	    channels?</para>
	</question>

	<answer>
	  <para>Yes, most major IRC networks host a &os; chat
	    channel:</para>

	  <itemizedlist>
	    <listitem>
	      <para>Channel <literal>#FreeBSD</literal> on <ulink
		  url="http://www.efnet.org/index.php">EFNet</ulink> is
		a &os; forum, but do not go there for tech support or
		try to get folks there to help you avoid the pain of
		reading manual pages or doing your own research.  It is
		a chat channel, first and foremost, and topics there are
		just as likely to involve sex, sports or nuclear weapons
		as they are &os;.  You Have Been Warned!  Available at
		server <hostid>irc.efnet.org</hostid>.</para>
	    </listitem>

	    <listitem>
	      <para>Channel <literal>#FreeBSDhelp</literal> on <ulink
		  url="http://www.efnet.org/index.php">EFNet</ulink> is
		a channel dedicated to helping &os; users.  They are
		much more sympathetic to questions than
		<literal>#FreeBSD</literal> is.</para>
	    </listitem>

	    <listitem>
	      <para>Channel <literal>##FreeBSD</literal> on <ulink
		  url="http://freenode.net/">Freenode</ulink> is a
		general help channel with many users at any time.
		The conversations have been known to run off-topic for a
		while, but priority is given to users with &os;
		questions.  We are good about helping you understand the
		basics, referring to the Handbook whenever possible, and
		directing you where to learn more about the topic you
		need help with.  We are a primarily English speaking
		channel, though we have users from all over the world.
		If you would like to speak in your native language, try
		to ask the question in English and then relocate to
		another channel
		<literal>##freebsd-<replaceable>lang</replaceable></literal>
		as appropriate.</para>
	    </listitem>

	    <listitem>
	      <para>Channel <literal>#FreeBSD</literal> on <ulink
		  url="http://www.dal.net/">DALNET</ulink> is available at
		<hostid>irc.dal.net</hostid> in the US and
		<hostid>irc.eu.dal.net</hostid> in Europe.</para>
	    </listitem>

	    <listitem>
	      <para>Channel <literal>#FreeBSD</literal> on <ulink
		  url="http://www.undernet.org/">UNDERNET</ulink> is
		available at <hostid>us.undernet.org</hostid> in the US
		and <hostid>eu.undernet.org</hostid> in Europe.  Since
		it is a help channel, be prepared to read the documents
		you are referred to.</para>
	    </listitem>

	    <listitem>
	      <para>Channel <literal>#FreeBSD</literal> on
		<ulink url="http://www.rusnet.org.ru/">RUSNET</ulink>
		is a russian-language oriented channel dedicated
		to helping &os; users.  This is also good place
		for non-technical discussions.</para>
	    </listitem>

	    <listitem>
	      <para>Channel <literal>#bsdchat</literal> on <ulink
		  url="http://freenode.net/">Freenode</ulink> is a
		Traditional-Chinese (UTF-8 encoding) language oriented
		channel dedicated to helping &os; users.  This is also
		good place for non-technical discussions.</para>
	    </listitem>
	  </itemizedlist>

	  <para>Each of these channels are distinct and are not
	    connected to each other.  Their chat styles also differ, so
	    you may need to try each to find one suited to your chat
	    style.  As with <emphasis>all</emphasis> types of IRC
	    traffic, if you are easily offended or cannot deal with lots
	    of young people (and more than a few older ones) doing the
	    verbal equivalent of jello wrestling, do not even bother
	    with it.</para>
	</answer>
      </qandaentry>

      <qandaentry>
    	<question id="forums">
    	  <para>Are there any web based forums to discuss &os;?</para>
    	</question>
    	<answer>
    	  <para>The official &os; forums are located at <ulink
	        url="http://forums.FreeBSD.org/">http://forums.FreeBSD.org/</ulink>.</para>
    	</answer>
      </qandaentry>

      <qandaentry>
	<question id="training">
	  <para>Where can I get commercial &os; training and
	    support?</para>
	</question>

	<answer>
	  <para><ulink
	      url="http://www.ixsystems.com">iXsystems, Inc.</ulink>,
	    parent company of the <ulink
	      url="http://www.freebsdmall.com/">&os; Mall</ulink>,
	    provides commercial &os; and PC-BSD software <ulink
	      url="http://www.ixsystems.com/bsdsupport">support</ulink>,
	    in addition to &os; development and tuning solutions.</para>

	  <para>BSD Certification Group, Inc. provides system
	    administration certifications for DragonFly&nbsp;BSD, &os;, NetBSD,
	    OpenBSD.  If you are interested in them, visit <ulink
	      url="http://www.BSDCertification.org">their site</ulink>.
	    </para>

	  <para>Any other organizations providing training and support
	    should contact the Project in order to be listed here.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="install">
    <chapterinfo>
      <author>
	<firstname>Nik</firstname>
	<surname>Clayton</surname>
	<affiliation>
	  <address><email>nik@FreeBSD.org</email></address>
	</affiliation>
      </author>
    </chapterinfo>

    <title>Installation</title>

    <qandaset>
      <qandaentry>
	<question id="floppy-download">
	  <para>Which file do I download to get &os;?</para>
	</question>

	<answer>
	  <para>You need three floppy images:
	    <filename>floppies/boot.flp</filename>,
	    <filename>floppies/kern1.flp</filename>, and
	    <filename>floppies/kern2.flp</filename>.  These images need
	    to be copied onto floppies by tools like
	    <command>fdimage</command> or &man.dd.1;.</para>

	  <para>If you need to download the distributions yourself (for
	    a DOS file system install, for instance), below are some
	    recommendations for distributions to grab:</para>

	  <itemizedlist>
	    <listitem>
	      <para>base/</para>
	    </listitem>

	    <listitem>
	      <para>manpages/</para>
	    </listitem>

	    <listitem>
	      <para>compat*/</para>
	    </listitem>

	    <listitem>
	      <para>doc/</para>
	    </listitem>

	    <listitem>
	      <para>src/ssys.*</para>
	    </listitem>
	  </itemizedlist>

	  <para>Full instructions on this procedure and a little bit
	    more about installation issues in general can be found in
	    the <ulink
	      url="&url.books.handbook;/install.html">Handbook entry on installing &os;</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="floppy-image-too-large">
	  <para>What do I do if the floppy images does not fit on a
	    single floppy?</para>
	</question>

	<answer>
	  <para>A 3.5&nbsp;inch (1.44&nbsp;MB) floppy can accommodate
	    1,474,560&nbsp;bytes of data.  The boot image is exactly
	    1,474,560&nbsp;bytes in size.</para>

	  <para>Common mistakes when preparing the boot floppy
	    are:</para>

	  <itemizedlist>
	    <listitem>
	      <para>Not downloading the floppy image in
		<emphasis>binary</emphasis> mode when using
		<acronym>FTP</acronym>.</para>

	      <para>Some FTP clients default their transfer mode to
		<emphasis>ascii</emphasis> and attempt to change any
		end-of-line characters received to match the conventions
		used by the client's system.  This will almost
		invariably corrupt the boot image.  Check the size of
		the downloaded boot image: if it is not
		<emphasis>exactly</emphasis> that on the server, then
		the download process is suspect.</para>

	      <para>To workaround: type <emphasis>binary</emphasis> at
		the FTP command prompt after getting connected to the
		server and before starting the download of the
		image.</para>
	    </listitem>

	    <listitem>
	      <para>Using the DOS <command>copy</command> command (or
		equivalent GUI tool) to transfer the boot image to
		floppy.</para>

	      <para>Programs like <command>copy</command> will not work
		as the boot image has been created to be booted into
		directly.  The image has the complete content of the
		floppy, track for track, and is not meant to be placed
		on the floppy as a regular file.  You have to transfer
		it to the floppy <quote>raw</quote>, using the low-level
		tools (e.g. <command>fdimage</command> or
		<command>rawrite</command>) described in the <ulink
		  url="&url.books.handbook;/install.html">installation guide to &os;</ulink>.
		</para>
	    </listitem>
	  </itemizedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="install-instructions-location">
	  <para>Where are the instructions for installing &os;?</para>
	</question>

	<answer>
	  <para>Installation instructions can be found in the <ulink
	      url="&url.books.handbook;/install.html">Handbook entry on installing &os;</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="need-to-run">
	  <para>What do I need in order to run &os;?</para>
	</question>

	<answer>
	  <para>For &os; you will need a 486 or better PC, with
	    64&nbsp;MB or more of RAM and at least 1&nbsp;GB of hard
	    disk space.</para>

	  <para>See also <xref linkend="hardware"/>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="custom-boot-floppy">
	  <para>How can I make my own custom install disk?</para>
	</question>

	<answer>
	  <para>Customized &os; installation media can be created by
	    building a custom release.  Follow the instructions in the
	    <ulink
	      url="&url.articles.releng;/article.html">Release Engineering</ulink>
	    article.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="multiboot">
	  <para>Can I have more than one operating system on my
	    PC?</para>
	</question>

	<answer>
	  <para>Have a look at <ulink
	      url="&url.articles.multi-os;/index.html"> the multi-OS page</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="windows-coexist">
	  <para>Can &windows; co-exist with &os;?</para>
	</question>

	<answer>
	  <para>Install &windows; first, then &os;.  &os;'s boot manager
	    will then manage to boot &windows; and &os;.  If you install
	    &windows; second, it will boorishly overwrite your boot
	    manager without even asking.  If that happens, see the next
	    section.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="win95-damaged-boot-manager">
	  <para>&windows; killed my boot manager!  How do I get it
	    back?</para>
	</question>

	<answer>
	  <para>You can reinstall the boot manager &os; comes with in
	    one of three ways:</para>

	  <itemizedlist>
	    <listitem>
	      <para>Running DOS, go into the <filename
		  class="directory">tools</filename> directory of your
		&os; distribution and look for
		<filename>bootinst.exe</filename>.  You run it like
		so:</para>

	      <screen><prompt>...\TOOLS&gt;</prompt> <userinput>bootinst.exe boot.bin</userinput></screen>

	      <para>and the boot manager will be reinstalled.</para>
	    </listitem>

	    <listitem>
	      <para>Boot the &os; boot floppy again and go to the
		<guimenuitem>Custom</guimenuitem> menu item for custom
		installation.  Choose
		<guimenuitem>Partition</guimenuitem>.  Select the drive
		which used to contain your boot manager (likely the
		first one) and when you come to the partition editor for
		it, as the very first thing (e.g.  do not make any
		changes) press <keycap>W</keycap>.  This will ask for
		confirmation, select &gui.yes;, and when you get the
		Boot Manager selection prompt, be sure to select the
		<application>&os; Boot Manager</application>.  This will
		re-write the boot manager to disk.  Now quit out of the
		installation menu and reboot off the hard disk as
		normal.</para>
	    </listitem>

	    <listitem>
	      <para>Boot the &os; boot floppy (or CD-ROM) and choose the
		<guimenuitem>Fixit</guimenuitem> menu item.  Select
		either the Fixit floppy or CD-ROM #2 (the
		<quote>live</quote> file system option) as appropriate
		and enter the fixit shell.  Then execute the following
		command:</para>

	      <screen><prompt>Fixit#</prompt> <userinput>fdisk -B -b /boot/boot0 <replaceable>bootdevice</replaceable></userinput></screen>

	      <para>substituting <replaceable>bootdevice</replaceable>
		for your real boot device such as
		<devicename>ad0</devicename> (first IDE disk),
		<devicename>ad4</devicename> (first IDE disk on
		auxiliary controller), <devicename>da0</devicename>
		(first SCSI disk), etc.</para>
	    </listitem>
	  </itemizedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="boot-on-thinkpad">
	  <para>My A, T, or X series IBM Thinkpad locks up when I first
	    booted up my &os; installation.  How can I solve
	    this?</para>
	</question>

	<answer>
	  <para>A bug in early revisions of IBM's BIOS on these machines
	    mistakenly identifies the &os; partition as a potential FAT
	    suspend-to-disk partition.  When the BIOS tries to parse the
	    &os; partition it hangs.</para>

	  <para>According to IBM<footnote>
	      <para>In an email from Keith Frechette
		<email>kfrechet@us.ibm.com</email>.</para></footnote>,
	    the following model/BIOS release numbers incorporate the
	    fix.</para>

	  <informaltable frame="none" pgwide="1">
	    <tgroup cols="2">
	      <thead>
		<row>
		  <entry>Model</entry>

		  <entry>BIOS revision</entry>
		</row>
	      </thead>

	      <tbody>
		<row>
		  <entry>T20</entry>

		  <entry>IYET49WW or later</entry>
		</row>

		<row>
		  <entry>T21</entry>

		  <entry>KZET22WW or later</entry>
		</row>

		<row>
		  <entry>A20p</entry>

		  <entry>IVET62WW or later</entry>
		</row>

		<row>
		  <entry>A20m</entry>

		  <entry>IWET54WW or later</entry>
		</row>

		<row>
		  <entry>A21p</entry>

		  <entry>KYET27WW or later</entry>
		</row>

		<row>
		  <entry>A21m</entry>

		  <entry>KXET24WW or later</entry>
		</row>

		<row>
		  <entry>A21e</entry>

		  <entry>KUET30WW</entry>
		</row>
	      </tbody>
	    </tgroup>
	  </informaltable>

	  <para>It has been reported that later IBM BIOS revisions may
	    have reintroduced the bug.  <ulink
	      url="http://docs.FreeBSD.org/cgi/mid.cgi?20010427133759.A71732">This message</ulink>
	    from &a.nectar; to the &a.mobile; describes a procedure
	    which may work if your newer IBM laptop does not boot &os;
	    properly, and you can upgrade or downgrade the BIOS.</para>

	  <para>If you have an earlier BIOS, and upgrading is not an
	    option, a workaround is to install &os;, change the partition
	    ID &os; uses, and install new boot blocks that can handle
	    the different partition ID.</para>

	  <para>First, you will need to restore the machine to a state
	    where it can get through its self-test screen.  Doing this
	    requires powering up the machine without letting it find a
	    &os; partition on its primary disk.  One way is to remove
	    the hard disk and temporarily move it to an older ThinkPad
	    (such as a ThinkPad 600) or a desktop PC with an appropriate
	    conversion cable.  Once it is there, you can delete the &os;
	    partition and move the hard disk back.  The ThinkPad should
	    now be in a bootable state again.</para>

	  <para>With the machine functional again, you can use the
	    workaround procedure described here to get a working &os;
	    installation.</para>

	  <procedure>
	    <step>
	      <para>Download <filename>boot1</filename> and
		<filename>boot2</filename> from <ulink
		  url="http://people.FreeBSD.org/~bmah/ThinkPad/"></ulink>.
		Put these files somewhere you will be able to retrieve
		them later.</para>
	    </step>

	    <step>
	      <para>Install &os; as normal on to the ThinkPad.
		<emphasis>Do not</emphasis> use <literal>Dangerously
		Dedicated</literal> mode.  <emphasis>Do not</emphasis>
		reboot when the install has finished.</para>
	    </step>

	    <step>
	      <para>Either switch to the <quote>Emergency Holographic
		Shell</quote> (<keycombo
		  action="simul"><keycap>Alt</keycap><keycap>F4</keycap></keycombo>)
		or start a <quote>fixit</quote> shell.</para>
	    </step>

	    <step>
	      <para>Use &man.fdisk.8; to change the &os; partition ID
		from <literal>165</literal> to <literal>166</literal>
		(this is the type used by OpenBSD).</para>
	    </step>

	    <step>
	      <para>Bring the <filename>boot1</filename> and
		<filename>boot2</filename> files to the local file
		system.</para>
	    </step>

	    <step>
	      <para>Use &man.disklabel.8; to write
		<filename>boot1</filename> and <filename>boot2</filename>
		to your &os; slice.</para>

	      <screen>&prompt.root; <userinput>disklabel -B -b boot1 -s boot2 ad0s<replaceable>n</replaceable></userinput></screen>

	      <para><replaceable>n</replaceable> is the number of the
		slice where you installed &os;.</para>
	    </step>

	    <step>
	      <para>Reboot.  At the boot prompt you will be given the
		option of booting <literal>OpenBSD</literal>.  This will
		actually boot &os;.</para>
	    </step>
	  </procedure>

	  <para>Getting this to work in the case where you want to dual
	    boot OpenBSD and &os; on the same laptop is left as an
	    exercise for the reader.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="install-bad-blocks">
	  <para>Can I install on a disk with bad blocks?</para>
	</question>

	<answer>
	  <para>You can, but it is a bad idea.</para>

	  <para>If you are seeing bad block errors with a modern IDE
	    drive, chances are the drive is going to die very soon (the
	    drive's internal remapping functions are no longer
	    sufficient to fix the bad blocks, which means the disk is
	    heavily corrupted); we suggest you buy a new hard
	    drive.</para>

	  <para>If you have a SCSI drive with bad blocks, see <link
	      linkend="awre">this answer</link>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="boot-floppy-strangeness">
	  <para>Strange things happen when I boot the install floppy!
	    What is happening?</para>
	</question>

	<answer>
	  <para>If you are seeing things like the machine grinding to a
	    halt or spontaneously rebooting when you try to boot the
	    install floppy, here are three questions to ask
	    yourself:</para>

	  <orderedlist>
	    <listitem>
	      <para>Did you use a new, freshly-formatted, error-free
		floppy (preferably a brand-new one straight out of the
		box, as opposed to the magazine cover disk that has been
		lying under the bed for the last three years)?</para>
	    </listitem>

	    <listitem>
	      <para>Did you download the floppy image in binary (or
		image) mode?  (do not be embarrassed, even the best of us
		have accidentally downloaded a binary file in ASCII mode
		at least once!)</para>
	    </listitem>

	    <listitem>
	      <para>If you are using &windows;&nbsp;95 or
		&windows;&nbsp;98 did you run <command>fdimage</command>
		or <command>rawrite</command> in pure DOS mode?  These
		operating systems can interfere with programs that write
		directly to hardware, which the disk creation program
		does; even running it inside a DOS shell in the GUI can
		cause this problem.</para>
	    </listitem>
	  </orderedlist>

	  <para>There have also been reports of &netscape; causing
	    problems when downloading the boot floppy, so it is probably
	    best to use a different FTP client if you can.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="no-install-cdrom">
	  <para>I booted from my ATAPI CD-ROM, but the install program
	    says no CD-ROM is found.  Where did it go?</para>
	</question>

	<answer>
	  <para>The usual cause of this problem is a mis-configured
	    CD-ROM drive.  Many PCs now ship with the CD-ROM as the slave
	    device on the secondary IDE controller, with no master
	    device on that controller.  This is illegal according to the
	    ATAPI specification, but &windows; plays fast and loose with
	    the specification, and the BIOS ignores it when booting.
	    This is why the BIOS was able to see the CD-ROM to boot from
	    it, but why &os; cannot see it to complete the
	    install.</para>

	  <para>Reconfigure your system so that the CD-ROM is either the
	    master device on the IDE controller it is attached to, or
	    make sure that it is the slave on an IDE controller that
	    also has a master device.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="install-PLIP">
	  <para>Can I install on my laptop over PLIP (Parallel Line
	    IP)?</para>
	</question>

	<answer>
	  <para>Yes.  Use a standard Laplink cable.  If necessary, you
	    can check out the <ulink
	      url="&url.books.handbook;/network-plip.html">PLIP section of the Handbook</ulink>
	    for details on parallel port networking.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="geometry">
	  <para>Which geometry should I use for a disk drive?</para>
	</question>

	<answer>
	    <note>
	      <para>By the <quote>geometry</quote> of a disk, we mean
		the number of cylinders, heads and sectors/track on a
		disk.  We will refer to this as C/H/S for convenience.
		This is how the PC's BIOS works out which area on a disk
		to read/write from.</para>
	    </note>

	  <para>This causes a lot of confusion among new system
	    administrators.  First of all, the
	    <emphasis>physical</emphasis> geometry of a SCSI drive is
	    totally irrelevant, as &os; works in term of disk blocks.
	    In fact, there is no such thing as <quote>the</quote>
	    physical geometry, as the sector density varies across the
	    disk.  What manufacturers claim is the <quote>physical
	    geometry</quote> is usually the geometry that they have
	    determined wastes the least space.  For IDE disks, &os; does
	    work in terms of C/H/S, but all modern drives internally
	    convert this into block references.</para>

	  <para>All that matters is the <emphasis>logical</emphasis>
	    geometry.  This is the answer that the BIOS gets when it
	    asks the drive <quote>what is your geometry?</quote> It then
	    uses this geometry to access the disk.  As &os; uses the
	    BIOS when booting, it is very important to get this right.
	    In particular, if you have more than one operating system on
	    a disk, they must all agree on the geometry.  Otherwise you
	    will have serious problems booting!</para>

	  <para>For SCSI disks, the geometry to use depends on whether
	    extended translation support is turned on in your controller
	    (this is often referred to as <quote>support for DOS disks
	    &gt;1GB</quote> or something similar).  If it is turned off,
	    then use <replaceable>N</replaceable> cylinders, 64 heads
	    and 32 sectors/track, where <replaceable>N</replaceable> is
	    the capacity of the disk in MB.  For example, a 2&nbsp;GB disk
	    should pretend to have 2048 cylinders, 64 heads and 32
	    sectors/track.</para>

	  <para>If it <emphasis>is</emphasis> turned on (it is often
	    supplied this way to get around certain limitations in
	    &ms-dos;) and the disk capacity is more than 1&nbsp;GB, use
	    <replaceable>M</replaceable> cylinders, 63 sectors per track
	    (<emphasis>not</emphasis> 64), and 255 heads, where
	    <replaceable>M</replaceable> is the disk capacity in MB
	    divided by 7.844238 (!).  So our example 2&nbsp;GB drive
	    would have 261 cylinders, 63 sectors per track and 255
	    heads.</para>

	  <para>If you are not sure about this, or &os; fails to detect
	    the geometry correctly during installation, the simplest way
	    around this is usually to create a small DOS partition on
	    the disk.  The BIOS should then detect the correct geometry,
	    and you can always remove the DOS partition in the partition
	    editor if you do not want to keep it.  You might want to
	    leave it around for programming network cards and the like,
	    however.</para>

	  <para>Alternatively, there is a freely available utility
	    distributed with &os; called
	    <filename>pfdisk.exe</filename>.  You can find it in the
	    <filename class="directory">tools</filename> subdirectory on
	    the &os; CD-ROM or on the various &os; FTP sites.  This
	    program can be used to work out what geometry the other
	    operating systems on the disk are using.  You can then enter
	    this geometry in the partition editor.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="disk-divide-restrictions">
	  <para>Are there any restrictions on how I divide the disk
	    up?</para>
	</question>

	<answer>
	  <para>Yes.  You must make sure that your root partition is
	    below 1024 cylinders so the BIOS can boot the kernel from it.
	    (Note that this is a limitation in the PC's BIOS, not
	    &os;).</para>

	  <para>For a SCSI drive, this will normally imply that the root
	    partition will be in the first 1024&nbsp;MB (or in the first
	    4096&nbsp;MB if extended translation is turned on &mdash; see
	    previous question).  For IDE, the corresponding figure is
	    504&nbsp;MB.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="disk-manager">
	  <para>Is &os; compatible with any disk managers?</para>
	</question>

	<answer>
	  <para>&os; recognizes the <application>Ontrack Disk
	    Manager</application> and makes allowances for it.  Other disk
	    managers are not supported.</para>

	  <para>If you just want to use the disk with &os; you do not
	    need a disk manager.  Just configure the disk for as much
	    space as the BIOS can deal with (usually
	    504&nbsp;megabytes), and &os; should figure out how much
	    space you really have.  If you are using an old disk with an
	    MFM controller, you may need to explicitly tell &os; how
	    many cylinders to use.</para>

	  <para>If you want to use the disk with &os; and another
	    operating system, you may be able to do without a disk
	    manager: just make sure the &os; boot partition and the
	    slice for the other operating system are in the first 1024
	    cylinders.  If you are reasonably careful, a
	    20&nbsp;megabyte boot partition should be plenty.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="missing-os">
	  <para>When I boot &os; for the first time after install I get
	    <errorname>Missing Operating System</errorname>.  What is
	    happening?</para>
	</question>

	<answer>
	  <para>This is classically a case of &os; and DOS or some other
	    OS conflicting over their ideas of disk <link
	      linkend="geometry">geometry</link>.  You will have to
	    reinstall &os;, but obeying the instructions given above
	    will almost always get you going.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="stop-at-boot-manager">
	  <para>Why can I not get past the boot manager's
	    <prompt>F?</prompt> prompt?</para>
	</question>

	<answer>
	  <para>This is another symptom of the problem described in the
	    preceding question.  Your BIOS geometry and &os; geometry
	    settings do not agree!  If your controller or BIOS supports
	    cylinder translation (often marked as <quote>&gt;1GB drive
	    support</quote>), try toggling its setting and reinstalling
	    &os;.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="need-complete-sources">
	  <para>Do I need to install the source?</para>
	</question>

	<answer>
	  <para>In general, no.  There is nothing in the base
	    system which requires the presence of the source to
	    operate.  Some ports, like <filename
	      role="package">sysutils/lsof</filename>, will not build
	    unless the source is installed.  In particular, if the
	    port builds a kernel module or directly operates on kernel
	    structures, the source must be installed.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="need-kernel">
	  <para>Do I need to build a kernel?</para>
	</question>

	<answer>
	  <para>Usually not.  The supplied <literal>GENERIC</literal>
	    kernel contains the drivers an ordinary computer will
	    need.  &man.freebsd-update.8;, the &os; binary upgrade
	    tool, cannot upgrade custom kernels, another reason
	    to stick with the GENERIC kernel when possible.
	    For computers with very limited RAM, such as
	    embedded systems, it may be worthwhile to build a
	    smaller custom kernel containing just the required
	    drivers.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="password-encryption">
	  <para>Should I use DES, Blowfish, or MD5 passwords and how do
	    I specify which form my users receive?</para>
	</question>

	<answer>
	  <para>&os;&nbsp;7 and 8 use MD5 password hashing by
	    default.  Recent versions
	    of &os; use <emphasis>SHA512</emphasis> by default.
	    These are
	    believed to be more secure than the traditional &unix;
	    password format, which used a scheme based on the
	    <emphasis>DES</emphasis> algorithm.  DES passwords are still
	    available if you need to share your password file with
	    legacy operating systems which still use the less secure
	    password format.  &os; also allows you to use the Blowfish
	    password format, which is more secure.  Which password
	    format to use for new passwords is controlled by the
	    <literal>passwd_format</literal> login capability in
	    <filename>/etc/login.conf</filename>, which takes values of
	    <literal>des</literal>, <literal>blf</literal> (if these are
	    available) or <literal>md5</literal>.  See the
	    &man.login.conf.5; manual page for more information about
	    login capabilities.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="boot-floppy-hangs">
	  <para>Why does the boot floppy start, but hang at the
	    <literal>Probing Devices...</literal> screen?</para>
	</question>

	<answer>
	  <para>If you have a IDE &iomegazip; or &jaz; drive installed,
	    remove it and try again.  The boot floppy can get confused by
	    the drives.  After the system is installed you can reconnect
	    the drive.  Hopefully this will be fixed in a later
	    release.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="panic-on-install-reboot">
	  <para>Why do I get a <errorname>panic: can't mount
	    root</errorname> error when rebooting the system after
	    installation?</para>
	</question>

	<answer>
	  <para>This error comes from confusion between the boot block's
	    and the kernel's understanding of the disk devices.  The
	    error usually manifests on two-disk IDE systems, with the
	    hard disks arranged as the master or single device on
	    separate IDE controllers, with &os; installed on the
	    secondary IDE controller.  The boot blocks think the system
	    is installed on <devicename>ad0</devicename> (the second
	    BIOS disk) while the kernel assigns the first disk on the
	    secondary controller device, <devicename>ad2</devicename>.
	    After the device probing, the kernel tries to mount what the
	    boot blocks think is the boot disk,
	    <devicename>ad0</devicename>, while it is really
	    <devicename>ad2</devicename>, and fails.</para>

	  <para>To fix the problem, do one of the following:</para>

	  <orderedlist>
	    <listitem>
	      <para>Reboot the system and hit <keycap>Enter</keycap> at
		the <literal>Booting kernel in 10 seconds; hit [Enter]
		to interrupt</literal> prompt.  This will drop you into
		the boot loader.</para>

	      <para>Then type
		<userinput>set root_disk_unit="<replaceable>disk_number</replaceable>"</userinput>.
		<replaceable>disk_number</replaceable> will be
		<literal>0</literal> if &os; is installed on the master
		drive on the first IDE controller, <literal>1</literal>
		if it is installed on the slave on the first IDE
		controller, <literal>2</literal> if it is installed on
		the master of the second IDE controller, and
		<literal>3</literal> if it is installed on the slave of
		the second IDE controller.</para>

	      <para>Then type <userinput>boot</userinput>, and your
		system should boot correctly.</para>

	      <para>To make this change permanent (i.e, so you do not
		have to do this every time you reboot or turn on your
		&os; machine), put the line
		<userinput>root_disk_unit="<replaceable>disk_number</replaceable>"</userinput>
		in <filename>/boot/loader.conf.local</filename>.</para>
	    </listitem>

	    <listitem>
		<para>Move the &os; disk onto the primary IDE
		  controller, so the hard disks are consecutive.</para>
	    </listitem>
	  </orderedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="memory-limits">
	  <para>What are the limits for memory?</para>
	</question>

	<answer>
	  <para>Memory limits depend on the platform used.  On a
	    standard &i386; install, the limit is 4&nbsp;GB but more
	    memory can be supported through &man.pae.4;.  See <link
	      linkend="memory-i386-over-4gb">instructions for using 4&nbsp;GB or more memory on &i386;</link>.
	    </para>

	  <para>&os;/pc98 has a limit of 4&nbsp;GB memory, and PAE can
	    not be used with it.  Other architectures supported by &os;
	    have much higher theoretical limits on maximum memory (many
	    terabytes).</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ffs-limits">
	  <para>What are the limits for FFS file systems?</para>
	</question>

	<answer>
	  <para>For FFS file systems, the maximum theoretical limit is
	    8&nbsp;TB (2&nbsp;G blocks), or 16&nbsp;TB for the default
	    block size of 8&nbsp;KB.  In practice, there is a soft limit
	    of 1&nbsp;TB, but with modifications file systems with
	    4&nbsp;TB are possible (and exist).</para>

	  <para>The maximum size of a single FFS file is approximately
	    1&nbsp;G blocks, or 4&nbsp;TB with a block size of
	    4&nbsp;KB.</para>

	  <table>
	    <title>Maximum File Sizes</title>

	    <tgroup cols="3">
	      <thead>
		<row>
		  <entry>FS Block Size</entry>

		  <entry>Works</entry>

		  <entry>Should Work</entry>
		</row>
	      </thead>

	      <tbody>
		<row>
		  <entry>4&nbsp;KB</entry>

		  <entry>&gt;&nbsp;4&nbsp;GB</entry>

		  <entry>4&nbsp;TB&nbsp;-&nbsp;1</entry>
		</row>

		<row>
		  <entry>8&nbsp;KB</entry>

		  <entry>&gt;&nbsp;32&nbsp;GB</entry>

		  <entry>32&nbsp;TB&nbsp;-&nbsp;1</entry>
		</row>

		<row>
		  <entry>16&nbsp;KB</entry>

		  <entry>&gt;&nbsp;128&nbsp;GB</entry>

		  <entry>32&nbsp;TB&nbsp;-&nbsp;1</entry>
		</row>

		<row>
		  <entry>32&nbsp;KB</entry>

		  <entry>&gt;&nbsp;512&nbsp;GB</entry>

		  <entry>64&nbsp;TB&nbsp;-&nbsp;1</entry>
		</row>

		<row>
		  <entry>64&nbsp;KB</entry>

		  <entry>&gt;&nbsp;2048&nbsp;GB</entry>

		  <entry>128&nbsp;TB&nbsp;-&nbsp;1</entry>
		</row>
	      </tbody>
	    </tgroup>
	  </table>

	  <para>When the FS block size is 4&nbsp;KB, triple indirect
	    blocks work and everything should be limited by the maximum FS
	    block number that can be represented using triple indirect
	    blocks (approx.
	    1024<superscript>3</superscript>&nbsp;+&nbsp;1024<superscript>2</superscript>&nbsp;+&nbsp;1024),
	    but everything is limited by a (wrong) limit of
	    1&nbsp;G&nbsp;-&nbsp;1 on FS block numbers.  The limit on FS
	    block numbers should be 2&nbsp;G&nbsp;-&nbsp;1.  There are
	    some bugs for FS block numbers near 2&nbsp;G&nbsp;-&nbsp;1,
	    but such block numbers are unreachable when the FS block
	    size is 4&nbsp;KB.</para>

	  <para>For block sizes of 8&nbsp;KB and larger, everything
	    should be limited by the 2&nbsp;G&nbsp;-&nbsp;1 limit on FS
	    block numbers, but is actually limited by the
	    1&nbsp;G&nbsp;-&nbsp;1 limit on FS block numbers.  Using the
	    correct limit of 2&nbsp;G&nbsp;-&nbsp;1 blocks does cause
	    problems.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="archsw-readin-failed-error">
	  <para>Why do I get an error message,
	    <errorname>archsw.readin.failed</errorname> after compiling
	    and booting a new kernel?</para>
	</question>

	<answer>
	  <para>Because your world and kernel are out of sync.  This is
	    not supported.  Be sure you use <command>make <maketarget>buildworld</maketarget></command>
	    and <command>make <maketarget>buildkernel</maketarget></command>
	    to update your kernel.</para>

	  <para>You can boot by specifying the kernel directly at the
	    second stage, pressing any key when the <literal>|</literal>
	    shows up before loader is started.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="boot-acpi">
	  <para>Installation crashes while booting, what can I do?</para>
	</question>

	<answer>
	  <para>Try disabling ACPI support.  When the bootloader loads,
	    press the <keycap>Space</keycap> key.  The system will display
	    the following:</para>

	  <screen>OK</screen>

	  <para>Type:</para>

	  <screen><userinput>unset acpi_load</userinput></screen>

	  <para>And then type:</para>

	  <screen><userinput>boot</userinput></screen>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="hardware">
    <title>Hardware Compatibility</title>

    <sect1 id="compatibility-general">
      <title>General</title>

      <qandaset>
	<qandaentry>
	  <question id="which-hardware-to-get">
	    <para>I want to get a piece of hardware for my &os; system.
	      Which model/brand/type is best?</para>
	  </question>

	  <answer>
	    <para>This is discussed continually on the &os; mailing
	      lists.  Since hardware changes so quickly, however, we
	      expect this.  We <emphasis>still</emphasis> strongly
	      recommend that you read through the Hardware&nbsp;Notes
	      for &os; <ulink
		url="&rel.current.hardware;">&rel.current;</ulink> or
	      <ulink
		url="&rel2.current.hardware;">&rel2.current;</ulink> and
	      search the mailing list <ulink
		url="http://www.FreeBSD.org/search/#mailinglists">archives</ulink>
	      before asking about the latest and greatest hardware.
	      Chances are a discussion about the type of hardware you
	      are looking for took place just last week.</para>

	    <para>If you are looking for a laptop, check the &a.mobile;
	      archives.  Otherwise, you probably want the archives for
	      the &a.questions;, or possibly a specific mailing list for
	      a particular hardware type.</para>
	  </answer>
	</qandaentry>
      </qandaset>
    </sect1>

    <sect1 id="compatibility-memory">
      <title>Memory</title>

      <qandaset>
	<qandaentry>
	  <question id="memory-upper-limitation">
	    <para>Does &os; support more than 4&nbsp;GB of memory (RAM)?
	      More than 16&nbsp;GB?  More than 48&nbsp;GB?</para>
	  </question>

	  <answer>
	    <para>Yes.  &os; as an operating system generally supports
	      as much physical memory (RAM) as the platform it is running
	      on does.  Keep in mind that different platforms have
	      different limits for memory; for example &i386; without
	      <acronym>PAE</acronym> supports at most 4&nbsp;GB of
	      memory (and usually less than that because of PCI address
	      space) and &i386; with PAE supports at most 64&nbsp;GB
	      memory.  AMD64 platforms currently deployed support up to
	      1&nbsp;TB of physical memory.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="memory-i386-over-4gb">
	    <para>Why does &os; report less than 4&nbsp;GB memory when
	      installed on an &i386; machine?</para>
	  </question>

	  <answer>
	    <para>The total address space on &i386; machines is 32-bit,
	      meaning that at most 4&nbsp;GB of memory is addressable (can
	      be accessed).  Furthermore, some addresses in this range
	      are reserved by hardware for different purposes, for
	      example for using and controlling PCI devices, for
	      accessing video memory, and so on.  Therefore, the total
	      amount of memory usable by the operating system for its
	      kernel and applications is limited to significantly less
	      than 4&nbsp;GB.  Usually, 3.2&nbsp;GB to 3.7&nbsp;GB is
	      the maximum usable physical memory in this
	      configuration.</para>

	    <para>To access more than 3.2&nbsp;GB to 3.7&nbsp;GB of
	      installed memory (meaning up to 4&nbsp;GB but also more than
	      4&nbsp;GB), a special tweak called <acronym>PAE</acronym>
	      must be used.  PAE stands for Physical Address Extension
	      and is a way for 32-bit x86 CPUs to address more than
	      4&nbsp;GB of memory.  It remaps the memory that would
	      otherwise be overlaid by address reservations for
	      hardware devices above the 4&nbsp;GB range and uses it as
	      additional physical memory (see &man.pae.4;).  Using PAE
	      has some drawbacks; this mode of memory access is a little
	      bit slower than the normal (without PAE) mode and loadable
	      modules (see &man.kld.4;) are not supported.  This means
	      all drivers must be compiled into the kernel.</para>

	    <para>The most common way to enable PAE is to build a new
	      kernel with the special ready-provided kernel configuration
	      file called <filename>PAE</filename>, which is already
	      configured to build a safe kernel.  Note that some entries
	      in this kernel configuration file are too conservative and
	      some drivers marked as unready to be used with PAE are
	      actually usable.  A rule of thumb is that if the driver is
	      usable on 64-bit architectures (like AMD64), it is also
	      usable with PAE.  If you wish to create your own kernel
	      configuration file, you can enable PAE by adding the
	      following line to your configuration:</para>

	    <programlisting>options       PAE</programlisting>

	    <para>PAE is not much used nowadays because most new x86
	      hardware also supports running in 64-bit mode, known as
	      AMD64 or &intel;&nbsp;64.  It has a much larger address
	      space and does not need such tweaks.  &os; supports AMD64
	      and it is recommended that this version of &os; be used
	      instead of the &i386; version if 4&nbsp;GB or more memory
	      is required.</para>
	  </answer>
	</qandaentry>
      </qandaset>
    </sect1>

    <sect1 id="compatibility-processors">
      <title>Architectures and Processors</title>

      <qandaset>
	<qandaentry>
	  <question id="architectures">
	    <para>Does &os; support architectures other than the
	      x86?</para>
	  </question>

	  <answer>
	    <para>Yes.  &os; currently runs on the Intel x86 and the
	      AMD64 architectures.  The Intel EM64T, IA-64, &arm;,
	      &powerpc;, and &sparc64; architectures are also
	      supported.  Upcoming platforms are &mips; and &s390;, join
	      the &a.mips; for more information about ongoing work on
	      the &mips; platform.  For general discussion on new
	      architectures, join the &a.platforms;.</para>

	    <para>If your machine has a different architecture and you
	      need something right now, we suggest you look at <ulink
		url="http://www.netbsd.org/">NetBSD</ulink> or <ulink
		url="http://www.openbsd.org/">OpenBSD</ulink>.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="smp-support">
	    <para>Does &os; support Symmetric Multiprocessing
	      (SMP)?</para>
	  </question>

	  <answer>
	    <para>Symmetric multi-processor (SMP) systems are generally
	      supported by &os;, although in some cases, BIOS or
	      motherboard bugs may generate some problems.</para>

	    <para>&os; will take advantage of HyperThreading (HTT)
	      support on Intel CPUs that support this feature.  A kernel
	      with the <literal>options SMP</literal> feature enabled
	      will automatically detect the additional logical
	      processors.</para>

	    <para>&man.smp.4; has more details.</para>
	  </answer>
	</qandaentry>
      </qandaset>
    </sect1>

    <sect1 id="compatibility-drives">
      <title>Hard Drives, Tape Drives, and CD and DVD Drives</title>

      <qandaset>
	<qandaentry>
	  <question id="supported-hard-drives">
	    <para>What kind of hard drives does &os; support?</para>
	  </question>

	  <answer>
	    <para>&os; supports EIDE, SATA, SCSI, and SAS drives (with a
	      compatible controller; see the next section), and all
	      drives using the original <quote>Western Digital</quote>
	      interface (MFM, RLL, ESDI, and of course IDE).  A few ESDI
	      controllers that use proprietary interfaces may not work:
	      stick to WD1002/3/6/7 interfaces and clones.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="supported-scsi-controllers">
	    <para>Which SCSI or SAS controllers are supported?</para>
	  </question>

	  <answer>
	    <para>See the complete list in the Hardware Notes for &os;
	      <ulink url="&rel.current.hardware;">&rel.current;</ulink>
	      or <ulink
		url="&rel2.current.hardware;">&rel2.current;</ulink>.
	      </para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="tape-support">
	    <para>What types of tape drives are supported?</para>
	  </question>

	  <answer>
	    <para>&os; supports SCSI and QIC-36 (with a QIC-02
	      interface).  This includes 8-mm (aka Exabyte) and DAT
	      drives.</para>

	    <para>Some of the early 8-mm drives are not quite compatible
	      with SCSI-2, and may not work well with &os;.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="tape-changer-support">
	    <para>Does &os; support tape changers?</para>
	  </question>

	  <answer>
	    <para>&os; supports SCSI changers using the &man.ch.4; device
	      and the &man.chio.1; command.  The details of how you
	      actually control the changer can be found in the
	      &man.chio.1; manual page.</para>

	    <para>If you are not using <application>AMANDA</application>
	      or some other product that already understands changers,
	      remember that they only know how to move a tape from one
	      point to another, so you need to keep track of which slot a
	      tape is in, and which slot the tape currently in the drive
	      needs to go back to.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="supported-cdrom-drives">
	    <para>Which CD-ROM drives are supported by &os;?</para>
	  </question>

	  <answer>
	    <para>Any SCSI drive connected to a supported controller is
	      supported.  Most ATAPI compatible IDE CD-ROMs are
	      supported.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="supported-cdrw-drives">
	    <para>Which CD-RW drives are supported by &os;?</para>
	  </question>

	  <answer>
	    <para>&os; supports any ATAPI-compatible IDE CD-R or CD-RW
	      drive.  See &man.burncd.8; for details.</para>

	    <para>&os; also supports any SCSI CD-R or CD-RW drives.
	      Install and use the <command>cdrecord</command> command
	      from the ports or packages system, and make sure that you
	      have the <devicename>pass</devicename> device compiled in
	      your kernel.</para>
	  </answer>
	</qandaentry>
      </qandaset>
    </sect1>

    <sect1 id="compatibility-kbd-mice">
      <title>Keyboards and Mice</title>

      <qandaset>
	<qandaentry>
	  <question id="usbkbd">
	    <para>Does &os; support my USB keyboard?</para>
	  </question>

	  <answer>
	    <para>&os; supports USB keyboards out-of-the-box.  Once you
	      have USB keyboard support enabled on your system, the AT
	      keyboard becomes <devicename>/dev/kbd0</devicename> and
	      the USB keyboard becomes
	      <devicename>/dev/kbd1</devicename>, if both are connected
	      to the system.  If there is the USB keyboard only, it will
	      be <devicename>/dev/ukbd0</devicename>.</para>

	    <para>If you want to use the USB keyboard in the console,
	      you have to explicitly tell the console driver to use the
	      existing USB keyboard.  This can be done by running the
	      following command as a part of system
	      initialization.</para>

	    <screen>&prompt.root; <userinput>kbdcontrol -k /dev/kbd1 &lt; /dev/console &gt; /dev/null</userinput></screen>

	    <para>Note that if the USB keyboard is the only keyboard, it
	      is accessed as <devicename>/dev/ukbd0</devicename>, thus,
	      the command should look like:</para>

	    <screen>&prompt.root; <userinput>kbdcontrol -k /dev/ukbd0 &lt; /dev/console &gt; /dev/null</userinput></screen>

	    <note>
	      <para>To make this change permanent across reboots, add
		<literal>keyboard="/dev/ukbd0"</literal> to
		<filename>/etc/rc.conf</filename>.</para>
	    </note>

	    <para>Once this is done, the USB keyboard should work in the
	      X environment as well without any special settings.</para>

	    <para>If you want to switch back to the default keyboard,
	      use this command:</para>

	    <screen>&prompt.root; <userinput>kbdcontrol -k /dev/kbd0 &gt; /dev/null</userinput></screen>

	    <para>To allow using both the second USB keyboard and the
	      first AT keyboard at the same time on a console via
	      &man.kbdmux.4; driver type the following commands:</para>

	    <screen>&prompt.root; <userinput>kbdcontrol -K &lt; /dev/console &gt; /dev/null</userinput>
&prompt.root; <userinput>kbdcontrol -a atkbd0 &lt; /dev/kbdmux0 &gt; /dev/null</userinput>
&prompt.root; <userinput>kbdcontrol -a ukbd1 &lt; /dev/kbdmux0 &gt; /dev/null</userinput>
&prompt.root; <userinput>kbdcontrol -k /dev/kbdmux0 &lt; /dev/console &gt; /dev/null</userinput></screen>

	    <para>See the &man.ukbd.4;, &man.kbdcontrol.1; and
	      &man.kbdmux.4; manual pages for more information.</para>

	    <note>
	      <para>Hot-plugging and unplugging of the USB keyboard may
		not work quite right yet.  We recommend connecting the
		keyboard before starting the system and leaving it
		connected until the system is shutdown to avoid
		issues.</para>
	    </note>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="busmouse">
	    <para>I have an unusual bus mouse.  How do I set it
	      up?</para>
	  </question>

	  <answer>
	    <para>&os; supports the bus mouse and the InPort bus mouse
	      from such manufacturers as Microsoft, Logitech and ATI.  The
	      <filename>GENERIC</filename> kernel does not include the
	      device driver.  To build a custom kernel with the bus mouse
	      driver, add the following line to the kernel config
	      file:</para>

	    <programlisting>device mse0 at isa? port 0x23c irq5</programlisting>

	    <para>Bus mice usually come with dedicated interface cards.
	      These cards may allow you to set the port address and the
	      IRQ number other than shown above.  Refer to the manual of
	      your mouse and the &man.mse.4; manual page for more
	      information.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="ps2mouse">
	    <para>How do I use my PS/2 (<quote>mouse port</quote> or
	      <quote>keyboard</quote>) mouse?</para>
	  </question>

	  <answer>
	    <para>The PS/2 mouse is supported out-of-the-box.  The
	      necessary device driver, <devicename>psm</devicename>, is
	      included in the kernel.</para>

	    <para>If your custom kernel does not have this, add the
	      following line to your kernel configuration and compile a
	      new kernel.</para>

	    <programlisting>device psm0 at atkbdc? irq 12</programlisting>

	    <para>Once the kernel detects <devicename>psm0</devicename>
	      correctly at boot time, a device node
	      <devicename>psm0</devicename> will be created
	      automatically.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="moused">
	    <para>Is it possible to use a mouse in any way outside the X
	      Window system?</para>
	  </question>

	  <answer>
	    <para>If you are using the default console driver,
	      &man.syscons.4;, you can use a mouse pointer in text
	      consoles to cut &amp; paste text.  Run the mouse daemon,
	      &man.moused.8;, and turn on the mouse pointer in the
	      virtual console:</para>

	    <screen>&prompt.root; <userinput>moused -p /dev/<replaceable>xxxx</replaceable> -t <replaceable>yyyy</replaceable></userinput>
&prompt.root; <userinput>vidcontrol -m on</userinput></screen>

	    <para>Where <replaceable>xxxx</replaceable> is the mouse
	      device name and <replaceable>yyyy</replaceable> is a
	      protocol type for the mouse.  The mouse daemon can
	      automatically determine the protocol type of most mice,
	      except old serial mice.  Specify the
	      <literal>auto</literal> protocol to invoke automatic
	      detection.  If automatic detection does not work, see the
	      &man.moused.8; manual page for a list of supported
	      protocol types.</para>

	    <para>If you have a PS/2 mouse, just add
	      <literal>moused_enable="YES"</literal> to
	      <filename>/etc/rc.conf</filename> to start the mouse
	      daemon at boot-time.  Additionally, if you would like to
	      use the mouse daemon on all virtual terminals instead of
	      just the console, add
	      <literal>allscreens_flags="-m on"</literal> to
	      <filename>/etc/rc.conf</filename>.</para>

	    <para>When the mouse daemon is running, access to the mouse
	      must be coordinated between the mouse daemon and other
	      programs such as X Windows.  Refer to the FAQ <link
		linkend="x-and-moused">Why does my mouse not work with X?</link>
	      for more details on this issue.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="text-mode-cut-paste">
	    <para>How do I cut and paste text with a mouse in the text
	      console?</para>
	  </question>

	  <answer>
	    <para>Once you get the mouse daemon running (see the <link
		linkend="moused">previous section</link>), hold down the
	      button 1 (left button) and move the mouse to select a region
	      of text.  Then, press the button 2 (middle button) to paste
	      it at the text cursor.  Pressing button 3 (right button)
	      will <quote>extend</quote> the selected region of
	      text.</para>

	    <para>If your mouse does not have a middle button, you may
	      wish to emulate one or remap buttons using mouse daemon
	      options.  See the &man.moused.8; manual page for
	      details.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="mouse-wheel-buttons">
	    <para>My mouse has a fancy wheel and buttons.  Can I use
	      them in &os;?</para>
	  </question>

	  <answer>
	    <para>The answer is, unfortunately, <quote>It
	      depends</quote>.  These mice with additional features
	      require specialized driver in most cases.  Unless the
	      mouse device driver or the user program has specific
	      support for the mouse, it will act just like a standard
	      two, or three button mouse.</para>

	    <para>For the possible usage of wheels in the X Window
	      environment, refer to <link
		linkend="x-and-wheel">that section</link>.
	      </para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="laptop-mouse-trackball">
	    <para>How do I use the mouse/trackball/touchpad on my
	      laptop?</para>
	  </question>

	  <answer>
	    <para>Please refer to <link linkend="ps2mouse">the answer to
	      the previous question</link>.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="keyboard-delete-key">
	    <para>How do I use my delete key in <command>sh</command>
	      and <command>csh</command>?</para>
	  </question>

	  <answer>
	    <para>For the <application>Bourne Shell</application>, add
	      the following lines to your <filename>.shrc</filename>.  See
	      &man.sh.1; and &man.editrc.5;.</para>

	    <programlisting>bind ^? ed-delete-next-char # for console
bind ^[[3~ ed-delete-next-char # for xterm</programlisting>

	    <para>For the <application>C Shell</application>, add the
	      following lines to your <filename>.cshrc</filename>.  See
	      &man.csh.1;.</para>

	    <programlisting>bindkey ^? delete-char # for console
bindkey ^[[3~ delete-char # for xterm</programlisting>

	    <para>For more information, see <ulink
		url="http://www.ibb.net/~anne/keyboard.html">this page</ulink>.
	      </para>
	  </answer>
	</qandaentry>
      </qandaset>
    </sect1>

    <sect1 id="compatibility-networking">
      <title>Networking and Serial Devices</title>

      <qandaset>
	<qandaentry>
	  <question id="network-cards">
	    <para>Which network cards does &os; support?</para>
	  </question>

	  <answer>
	    <para>See the Hardware Notes supplied with each release of
	      &os; for a more complete list.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="support-broadcom">
	    <para>Is there a native driver for the Broadcom 43xx
	      cards?</para>
	  </question>

	  <answer>
	    <para>Yes, many Broadcom 43xx cards are supported by the
	      &man.bwn.4; and &man.bwi.4; drivers.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="multiport-serial-support">
	    <para>Which multi-port serial cards are supported by
	      &os;?</para>
	  </question>

	  <answer>
	    <para>There is a list of these in the <ulink
		url="&url.books.handbook;/serial.html">Serial Communications</ulink>
	      chapter of the handbook.</para>

	    <para>Some unnamed clone cards have also been known to work,
	      especially those that claim to be AST compatible.</para>

	    <para>Check the &man.sio.4; manual page to get more
	      information on configuring such cards.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="serial-console-prompt">
	    <para>How do I get the boot: prompt to show on the serial
	      console?</para>
	  </question>

	  <answer>
	    <para>See <ulink
		url="&url.books.handbook;/serialconsole-setup.html">this section of the handbook</ulink>.
	      </para>
	  </answer>
	</qandaentry>
      </qandaset>
    </sect1>

    <sect1 id="compatibility-sound">
      <title>Sound Devices</title>

      <qandaset>
	<qandaentry>
	  <question id="sound-card-support">
	    <para>Which sound cards are supported by &os;?</para>
	  </question>

	  <answer>
	    <para>&os; supports various sound cards (for more details,
	      see <ulink
		url="&url.base;/releases/">&os; Release Information</ulink>
	      and the &man.snd.4; manual page).  There is also limited
	      support for MPU-401 and compatible MIDI cards.  Cards
	      conforming to the &microsoft; Sound System specification
	      are also supported.</para>

	    <note>
	      <para>This is only for sound!  This driver does not
		support CD-ROMs, SCSI or joysticks on these cards, except
		for the &soundblaster;.  The &soundblaster; SCSI
		interface and some non-SCSI CD-ROMs are supported, but
		you cannot boot off this device.</para>
	    </note>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="es1370-silent-pcm">
	    <para>Workarounds for no sound from my &man.pcm.4; sound
	      card?</para>
	  </question>

	  <answer>
	    <para>Some sound cards set their output volume to 0 at every
	      boot.  Run the following command every time the machine
	      boots:</para>

	    <screen>&prompt.root; <userinput>mixer pcm 100 vol 100 cd 100</userinput></screen>
	  </answer>
	</qandaentry>
      </qandaset>
    </sect1>

    <sect1 id="compatibility-other">
      <title>Other Hardware</title>

      <qandaset>
	<qandaentry>
	  <question id="power-management-support">
	    <para>Does &os; support power management on my
	      laptop?</para>
	  </question>

	  <answer>
	    <para>&os; supports <acronym>APM</acronym> on certain
	      machines.  Further information can be found in
	      &man.apm.4;.</para>

	    <para>&os; also supports the <acronym>ACPI</acronym>
	      features found in most modern hardware.  Further
	      information can be found in &man.acpi.4;.  If a system
	      supports both <acronym>APM</acronym> and
	      <acronym>ACPI</acronym>, either can be used.  We suggest
	      you try both and choose the one that best fits your
	      needs.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="disable-acpi">
	    <para>How do I disable ACPI?</para>
	  </question>

	  <answer>
	    <para>Add following line
	      <screen>hint.acpi.0.disabled="1"</screen> into your
	      <filename>/boot/device.hints</filename> file.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="micron-hang-boot">
	    <para>Why does my Micron system hang at boot time?</para>
	  </question>

	  <answer>
	    <para>Certain Micron motherboards have a non-conforming PCI
	      BIOS implementation that causes grief when &os; boots
	      because PCI devices do not get configured at their
	      reported addresses.</para>

	    <para>Disable the <quote>Plug and Play Operating
	      System</quote> flag in the BIOS to work around this
	      problem.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="asusk7v-boot-failure">
	    <para>The boot floppy hangs on a system with an ASUS K7V
	      motherboard.  How do I fix this?</para>
	  </question>

	  <answer>
	    <para>Go into the BIOS setup and disable the <quote>boot
	      virus protection</quote>.</para>
	  </answer>
	</qandaentry>

	<qandaentry>
	  <question id="micron-3comnic-failure">
	    <para>Why does my &tm.3com; PCI network card not work with my
	      Micron computer?</para>
	  </question>

	  <answer>
	    <para>See <link
		linkend="micron-hang-boot">the previous answer</link>.
	      </para>
	  </answer>
	</qandaentry>
      </qandaset>
    </sect1>
  </chapter>

  <chapter id="troubleshoot">
    <title>Troubleshooting</title>

    <qandaset>
      <qandaentry>
	<question id="pae">
	  <para>Why is &os; finding the wrong amount of memory on &i386;
	    hardware?</para>
	</question>

	<answer>
	  <para>The most likely reason is the difference between
	    physical memory addresses and virtual addresses.</para>

	  <para>The convention for most PC hardware is to use the memory
	    area between 3.5&nbsp;GB and 4&nbsp;GB for a special purpose
	    (usually for PCI).  This address space is used to access PCI
	    hardware.  As a result real, physical memory can not be
	    accessed by that address space.</para>

	  <para>What happens to the memory that should appear in that
	    location is dependent on your hardware.  Unfortunately, some
	    hardware does nothing and the ability to use that last
	    500&nbsp;MB of RAM is entirely lost.</para>

	  <para>Luckily, most hardware remaps the memory to a higher
	    location so that it can still be used.  However, this can
	    cause some confusion if you watch the boot messages.</para>

	  <para>On a 32-bit version of &os;, the memory appears
	    lost, since it will be remapped above 4&nbsp;GB, which a
	    32-bit kernel is unable to access.  In this case, the
	    solution is to build a PAE enabled kernel.  See <link
	      linkend="memory-limits">the entry on memory limits</link>
	    and <link linkend="memory-upper-limitation">about different
	    memory limits on different platforms</link> for more
	    information.</para>

	  <para>On a 64-bit version of &os;, or when running a
	    PAE-enabled kernel, &os; will correctly detect and remap the
	    memory so it is usable.  During boot, however, it may seem
	    as if &os; is detecting more memory than the system really
	    has, due to the described remapping.  This is normal and the
	    available memory will be corrected as the boot process
	    completes.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="awre">
	  <para>What do I do when I have bad blocks on my hard
	    drive?</para>
	</question>

	<answer>
	  <para>With SCSI drives, the drive should be capable of
	    re-mapping these automatically.  However, many drives ship
	    with this feature disabled.</para>

	  <para>To enable bad block remapping edit the first device page
	    mode, which can be done by giving the command (as
	    <username>root</username>)</para>

	  <screen>&prompt.root; <userinput>camcontrol modepage sd0 -m 1 -e -P 3</userinput></screen>

	  <para>and changing the values of AWRE and ARRE from 0 to
	    1:</para>

	  <programlisting>AWRE (Auto Write Reallocation Enbld):  1
ARRE (Auto Read Reallocation Enbld):  1</programlisting>

	  <para>Modern IDE drives also have bad block remapping features
	    in the controller, and they ship with this feature turned
	    on.</para>

	  <para>If you see warnings about bad blocks (on either type of
	    drive), it is time to consider replacing the drive.  You
	    might be able to use the drive manufacturer's diagnostic
	    program to lock out those bad blocks, but at best this will
	    buy you some time.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="hpnetserver-scsi-failure">
	  <para>Why does &os; not detect my HP Netserver's SCSI
	    controller?</para>
	</question>

	<answer>
	  <para>This is basically a known problem.  The EISA on-board
	    SCSI controller in the HP Netserver machines occupies EISA
	    slot number 11, so all the <quote>true</quote> EISA slots
	    are in front of it.  Alas, the address space for EISA slots
	    &gt;= 10 collides with the address space assigned to PCI,
	    and &os;'s auto-configuration currently cannot handle this
	    situation very well.</para>

	  <para>So now, the best you can do is to pretend there is no
	    address range clash :), by bumping the kernel option
	    <literal>EISA_SLOTS</literal> to a value of 12.  Configure
	    and compile a kernel, as described in the <ulink
	      url="&url.books.handbook;/kernelconfig.html">Handbook entry on configuring the kernel</ulink>.
	    </para>

	  <para>Of course, this does present you with a chicken-and-egg
	    problem when installing on such a machine.  In order to work
	    around this problem, a special hack is available inside
	    <emphasis>UserConfig</emphasis>.  Do not use the
	    <quote>visual</quote> interface, but the plain command-line
	    interface there.  Simply type the following command at the
	    prompt and install your system as usual:</para>

	  <programlisting>eisa 12
quit</programlisting>

	  <para>While it is recommended you compile and install a custom
	    kernel anyway.</para>

	  <para>Hopefully, future versions will have a proper fix for
	    this problem.</para>

	  <note>
	    <para>You cannot use a <literal>dangerously
	      dedicated</literal> disk with an HP Netserver.  See <link
		linkend="dedicate">this note</link> for more info.
	      </para>
	  </note>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ed1-timeout">
	  <para>I keep seeing messages like <errorname>ed1:
	    timeout</errorname>.  What do these messages mean?</para>
	</question>

	<answer>
	  <para>This is usually caused by an interrupt conflict (e.g.,
	    two boards using the same IRQ).  Boot with the
	    <option>-c</option> option and change the
	    <devicename>ed0</devicename>/<devicename>de0</devicename>/...
	    entry to match your board.</para>

	  <para>If you are using the BNC connector on your network card,
	    you may also see device timeouts because of bad termination.
	    To check this, attach a terminator directly to the NIC (with
	    no cable) and see if the error messages go away.</para>

	  <para>Some NE2000 compatible cards will give this error if
	    there is no link on the UTP port or if the cable is
	    disconnected.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="bad-3c509">
	  <para>Why did my &tm.3com; 3C509 card stop working for no
	    apparent reason?</para>
	</question>

	<answer>
	  <para>This card has a bad habit of losing its configuration
	    information.  Refresh your card's settings with the DOS
	    utility <command>3c5x9.exe</command>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="printer-slow">
	  <para>My parallel printer is ridiculously slow.  What can I
	    do?</para>
	</question>

	<answer>
	  <para>If the only problem is that the printer is terribly
	    slow, try changing your <ulink
	      url="&url.books.handbook;/printing-intro-setup.html#PRINTING-PARALLEL-PORT-MODE">printer port mode</ulink>
	    as discussed in the <ulink
	      url="&url.books.handbook;/printing-intro-setup.html">Printer Setup</ulink>
	    section of the Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="signal11">
	  <para>Why do my programs occasionally die with
	    <errorname>Signal 11</errorname> errors?</para>
	</question>

	<answer>
	  <para>Signal 11 errors are caused when your process has
	    attempted to access memory which the operating system has not
	    granted it access to.  If something like this is happening
	    at seemingly random intervals then you need to start
	    investigating things very carefully.</para>

	  <para>These problems can usually be attributed to
	    either:</para>

	  <orderedlist>
	    <listitem>
	      <para>If the problem is occurring only in a specific
		application that you are developing yourself it is
		probably a bug in your code.</para>
	    </listitem>

	    <listitem>
	      <para>If it is a problem with part of the base &os;
		system, it may also be buggy code, but more often than not
		these problems are found and fixed long before us
		general FAQ readers get to use these bits of code (that
		is what -current is for).</para>
	    </listitem>
	  </orderedlist>

	  <para>In particular, a dead giveaway that this is
	    <emphasis>not</emphasis> a &os; bug is if you see the
	    problem when you are compiling a program, but the activity
	    that the compiler is carrying out changes each time.</para>

	  <para>For example, suppose you are running
	    <command>make <maketarget>buildworld</maketarget></command>,
	    and the compile fails while trying to compile
	    <filename>ls.c</filename> into <filename>ls.o</filename>.
	    If you then run
	    <command>make <maketarget>buildworld</maketarget></command>
	    again, and the compile fails in the same place then this is
	    a broken build &mdash; try updating your sources and try
	    again.  If the compile fails elsewhere then this is almost
	    certainly hardware.</para>

	  <para>What you should do:</para>

	  <para>In the first case you can use a debugger e.g.
	    &man.gdb.1; to find the point in the program which is
	    attempting to access a bogus address and then fix it.</para>

	  <para>In the second case you need to verify that it is not
	    your hardware at fault.</para>

	  <para>Common causes of this include:</para>

	  <orderedlist>
	    <listitem>
	      <para>Your hard disks might be overheating: Check the fans
		in your case are still working, as your disk (and perhaps
		other hardware might be overheating).</para>
	    </listitem>

	    <listitem>
	      <para>The processor running is overheating: This might be
		because the processor has been overclocked, or the fan
		on the processor might have died.  In either case you
		need to ensure that you have hardware running at what it
		is specified to run at, at least while trying to solve
		this problem.  i.e.  Clock it back to the default
		settings.</para>

	      <para>If you are overclocking then note that it is far
		cheaper to have a slow system than a fried system that
		needs replacing!  Also the wider community is not often
		sympathetic to problems on overclocked systems, whether
		you believe it is safe or not.</para>
	    </listitem>

	    <listitem>
	      <para>Dodgy memory: If you have multiple memory
		SIMMS/DIMMS installed then pull them all out and try
		running the machine with each SIMM or DIMM individually
		and narrow the problem down to either the problematic
		DIMM/SIMM or perhaps even a combination.</para>
	    </listitem>

	    <listitem>
	      <para>Over-optimistic Motherboard settings: In your BIOS
		settings, and some motherboard jumpers you have options
		to set various timings, mostly the defaults will be
		sufficient, but sometimes, setting the wait states on
		RAM too low, or setting the <quote>RAM Speed:
		Turbo</quote> option, or similar in the BIOS will cause
		strange behavior.  A possible idea is to set to BIOS
		defaults, but it might be worth noting down your
		settings first!</para>
	    </listitem>

	    <listitem>
	      <para>Unclean or insufficient power to the motherboard.
		If you have any unused I/O boards, hard disks, or CD-ROMs
		in your system, try temporarily removing them or
		disconnecting the power cable from them, to see if your
		power supply can manage a smaller load.  Or try another
		power supply, preferably one with a little more power
		(for instance, if your current power supply is rated at
		250&nbsp;Watts try one rated at 300&nbsp;Watts).</para>
	    </listitem>
	  </orderedlist>

	  <para>You should also read the SIG11 FAQ (listed below) which
	    has excellent explanations of all these problems, albeit from
	    a &linux; viewpoint.  It also discusses how memory testing
	    software or hardware can still pass faulty memory.</para>

	  <para>Finally, if none of this has helped it is possible that
	    you have just found a bug in &os;, and you should follow the
	    instructions to send a problem report.</para>

	  <para>There is an extensive FAQ on this at <ulink
	      url="http://www.bitwizard.nl/sig11/">the SIG11 problem FAQ</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="trap-12-panic">
	  <para>My system crashes with either <errorname>Fatal trap 12:
	    page fault in kernel mode</errorname>, or
	    <errorname>panic:</errorname>, and spits out a bunch of
	    information.  What should I do?</para>
	</question>

	<answer>
	  <para>The &os; developers are very interested in these
	    errors, but need some more information than just the error
	    you see.  Copy your full crash message.  Then consult the
	    FAQ section on <link
	      linkend="kernel-panic-troubleshooting">kernel panics</link>,
	    build a debugging kernel, and get a backtrace.  This might
	    sound difficult, but you do not need any programming skills;
	    you just have to follow the instructions.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="screen-loses-sync">
	  <para>Why does the screen go black and lose sync when I
	    boot?</para>
	</question>

	<answer>
	  <para>This is a known problem with the ATI&nbsp;Mach64 video
	    card.  The problem is that this card uses address
	    <literal>2e8</literal>, and the fourth serial port does too.
	    Due to a bug (feature?) in the &man.sio.4; driver it will
	    touch this port even if you do not have the fourth serial
	    port, and <emphasis>even</emphasis> if you disable
	    <devicename>sio3</devicename> (the fourth port) which
	    normally uses this address.</para>

	  <para>Until the bug has been fixed, you can use this
	    workaround:</para>

	  <orderedlist>
	    <listitem>
	      <para>Enter <option>-c</option> at the boot prompt.
		(This will put the kernel into configuration
		mode).</para>
	    </listitem>

	    <listitem>
	      <para>Disable <devicename>sio0</devicename>,
		<devicename>sio1</devicename>,
		<devicename>sio2</devicename> and
		<devicename>sio3</devicename> (all of them).  This way
		the &man.sio.4; driver does not get activated &mdash; no
		problems.</para>
	    </listitem>

	    <listitem>
	      <para>Type exit to continue booting.</para>
	    </listitem>
	  </orderedlist>

	  <para>If you want to be able to use your serial ports, you
	    will have to build a new kernel with the following
	    modification: in
	    <filename>/usr/src/sys/dev/sio/sio.c</filename> (or in
	    <filename>/usr/src/sys/pc98/cbus/sio.c</filename> for pc98)
	    find the one occurrence of the string
	    <literal>0x2e8</literal> and remove that string and the
	    preceding comma (keep the trailing comma).  Now follow the
	    normal procedure of building a new kernel.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="reallybigram">
	  <para>Why does &os; only use 64&nbsp;MB of RAM when my system
	    has 128&nbsp;MB of RAM installed?</para>
	</question>

	<answer>
	  <para>Due to the manner in which &os; gets the memory size
	    from the BIOS, it can only detect 16&nbsp;bits worth of
	    Kbytes in size (65535&nbsp;Kbytes = 64&nbsp;MB) (or less...
	    some BIOSes peg the memory size to 16&nbsp;MB).  If you have
	    more than 64&nbsp;MB, &os; will attempt to detect it;
	    however, the attempt may fail.</para>

	  <para>To work around this problem, you need to use the kernel
	    option specified below.  There is a way to get complete
	    memory information from the BIOS, but we do not have room in
	    the bootblocks to do it.  Someday when lack of room in the
	    bootblocks is fixed, we will use the extended BIOS functions
	    to get the full memory information... but for now we are
	    stuck with the kernel option.</para>

	  <programlisting>options MAXMEM=<replaceable>n</replaceable></programlisting>

	  <para>Where <replaceable>n</replaceable> is your memory in
	    Kilobytes.  For a 128&nbsp;MB machine, you would want to use
	    <literal>131072</literal>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="kmem-map-too-small">
	  <para>My system has more than 1&nbsp;GB of RAM, and I'm
	    getting panics with <errorname>kmem_map too small</errorname>
	    messages.  What is wrong?</para>
	</question>

	<answer>
	  <para>Normally, &os; determines a number of kernel parameters,
	    such as as the maximum number of files that can be open
	    concurrently, from the amount of memory installed in the
	    system.  On systems with one gigabyte of RAM or more, this
	    <quote>auto sizing</quote> mechanism may choose values that
	    are too high: while starting up, the kernel allocates
	    various tables and other structures that fill up most of the
	    available kernel memory.  Later on, while the system is
	    running, the kernel has no more space left for dynamic
	    memory allocations, and panics.</para>

	  <para>Compile your own kernel, and add the
	    <option>VM_KMEM_SIZE_MAX</option> to your kernel
	    configuration file, increasing the maximum size to
	    400&nbsp;MB (<option>options
	    VM_KMEM_SIZE_MAX=419430400</option>).  400&nbsp;MB appears
	    to be sufficient for machines with up to 6&nbsp;GB of
	    memory.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="panic-kmemmap-too-small">
	  <para>My system does not have 1&nbsp;GB of RAM, and &os; still
	    panics with <errorname>kmem_map too
	    small</errorname>!</para>
	</question>

	<answer>
	  <para>The panic indicates that the system ran out of virtual
	    memory for network buffers (specifically, mbuf clusters).
	    You can increase the amount of VM available for mbuf
	    clusters by following the instructions in the <ulink
	      url="&url.books.handbook;/configtuning-kernel-limits.html#NMBCLUSTERS">Network Limits</ulink>
	    section of the Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="proc-table-full">
	  <para>Why do I get the error <errorname>kernel: proc: table is
	    full</errorname>?</para>
	</question>

	<answer>
	  <para>The &os; kernel will only allow a certain number of
	    processes to exist at one time.  The number is based on the
	    <varname>kern.maxusers</varname> &man.sysctl.8; variable.
	    <varname>kern.maxusers</varname> also affects various other
	    in-kernel limits, such as network buffers (see <link
	      linkend="panic-kmemmap-too-small">this</link> earlier
	    question).  If your machine is heavily loaded, you probably
	    want to increase <varname>kern.maxusers</varname>.  This
	    will increase these other system limits in addition to the
	    maximum number of processes.</para>

	  <para>To adjust your <varname>kern.maxusers</varname> value,
	    see the <ulink
	      url="&url.books.handbook;/configtuning-kernel-limits.html#KERN-MAXFILES">File/Process Limits</ulink>
	    section of the Handbook.  (While that section refers to open
	    files, the same limits apply to processes.)</para>

	  <para>If your machine is lightly loaded, and you are simply
	    running a very large number of processes, you can adjust
	    this with the <varname>kern.maxproc</varname> tunable.  If
	    this tunable needs adjustment it needs to be defined in
	    <filename>/boot/loader.conf</filename>.  The tunable will
	    not get adjusted until the system is rebooted.  For more
	    information about tuning tunables, you should see the
	    &man.loader.conf.5; and &man.sysctl.conf.5; manual pages.
	    If these processes are being run by a single user, you will
	    also need to adjust <varname>kern.maxprocperuid</varname> to
	    be one less than your new <varname>kern.maxproc</varname>
	    value.  (It must be at least one less because one system
	    program, &man.init.8;, must always be running.)</para>

	  <para>To make a sysctl change permanent place the proper value
	    in <filename>/etc/sysctl.conf</filename>.  More information
	    about system tuning with &man.sysctl.8; can be found at the
	    <ulink
	      url="&url.books.handbook;/configtuning-sysctl.html">Tuning with sysctl</ulink>
	    section of the Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="cmap-busy-panic">
	  <para>Why do I get an error reading <errorname>CMAP
	    busy</errorname> when rebooting with a new kernel?</para>
	</question>

	<answer>
	  <para>The logic that attempts to detect an out of date
	    <filename>/var/db/kvm_*.db</filename> files sometimes fails
	    and using a mismatched file can sometimes lead to
	    panics.</para>

	  <para>If this happens, reboot single-user and do:</para>

	  <screen>&prompt.root; <userinput>rm /var/db/kvm_*.db</userinput></screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="brkadrint-illegal-host-access">
	  <para>What does the message <errorname>ahc0: brkadrint,
	    Illegal Host Access at seqaddr 0x0</errorname> mean?</para>
	</question>

	<answer>
	  <para>This is a conflict with an Ultrastor SCSI Host
	    Adapter.</para>

	  <para>During the boot process enter the kernel configuration
	    menu and disable <devicename>uha0</devicename>, which is
	    causing the problem.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="aci0-illegal-cable">
	  <para>When I boot my system, I get the error <errorname>ahc0:
	    illegal cable configuration</errorname>.  My cabling is
	    correct.  What is going on?</para>
	</question>

	<answer>
	  <para>Your motherboard lacks the external logic to support
	    automatic termination.  Switch your SCSI BIOS to specify the
	    correct termination for your configuration rather than
	    automatic termination.  The &man.ahc.4; driver cannot
	    determine if the external logic for cable detection (and
	    thus auto-termination) is available.  The driver simply
	    assumes that this support must exist if the configuration
	    contained in the serial EEPROM is set to <quote>automatic
	    termination</quote>.  Without the external cable detection
	    logic the driver will often configure termination
	    incorrectly, which can compromise the reliability of the
	    SCSI bus.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="mail-loopback">
	  <para>Why does <application>sendmail</application> give me an
	    error reading <errorname>mail loops back to
	    myself</errorname>?</para>
	</question>

	<answer>
	  <para>You can find a detailed answer for this question in the
	    <ulink
	      url="&url.books.handbook;/mail-trouble.html#Q26.5.2.">Handbook</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="remote-fullscreen">
	  <para>Why do full screen applications on remote machines
	    misbehave?</para>
	</question>

	<answer>
	  <para>The remote machine may be setting your terminal type to
	    something other than the <literal>cons25</literal> terminal
	    type required by the &os; console.</para>

	  <para>There are a number of possible work-arounds for this
	    problem:</para>

	  <itemizedlist>
	    <listitem>
	      <para>After logging on to the remote machine, set your
		<envar>TERM</envar> shell variable to
		<literal>ansi</literal> or <literal>sco</literal> if
		the remote machine knows about these terminal
		types.</para>
	    </listitem>

	    <listitem>
	      <para>Use a VT100 emulator like
		<application>screen</application> at the &os; console.
		<application>screen</application> offers you the
		ability to run multiple concurrent sessions from one
		terminal, and is a neat program in its own right.
		Each <application>screen</application> window behaves
		like a VT100 terminal, so the <envar>TERM</envar>
		variable at the remote end should be set to
		<literal>vt100</literal>.</para>
	    </listitem>

	    <listitem>
	      <para>Install the <literal>cons25</literal> terminal
		database entry on the remote machine.  The way to do
		this depends on the operating system on the remote
		machine.  The system administration manuals for the
		remote system should be able to help you here.</para>
	    </listitem>

	    <listitem>
	      <para>Fire up an X server at the &os; end and login to the
		remote machine using an X based terminal emulator such
		as <command>xterm</command> or <command>rxvt</command>.
		The <envar>TERM</envar> variable at the remote host
		should be set to <literal>xterm</literal> or
		<literal>vt100</literal>.</para>
	    </listitem>
	  </itemizedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="pnp-not-found">
	  <para>Why is my PnP card not found (or found as
	    <literal>unknown</literal>)?</para>
	</question>

	<answer>
	  <para>The reasons for this behavior are explained by the
	    following email, posted to the &a.questions; by &a.peter;, in
	    answer to a question about an internal modem that was no
	    longer found after an upgrade to
	    &os;&nbsp;4.<replaceable>X</replaceable> (the comments in
	    <literal>[]</literal> have been added to clarify the
	    context).</para>

	  <note>
	    <para>The contents of this quotation has been updated from
	      its original text.</para>
	  </note>

	  <blockquote>
	    <para>The PNP bios preconfigured it [the modem] and left it
	      laying around in port space, so [in
	      3.<replaceable>X</replaceable>] the old-style ISA probes
	      <quote>found</quote> it there.</para>

	    <para>Under 4.0, the ISA code is much more PnP-centric.  It
	      was possible [in 3.<replaceable>X</replaceable>] for an ISA
	      probe to find a <quote>stray</quote> device and then for
	      the PNP device ID to match and then fail due to resource
	      conflicts.  So, it disables the programmable cards first
	      so this double probing cannot happen.  It also means that
	      it needs to know the PnP IDs for supported PnP hardware.
	      Making this more user tweakable is on the TODO
	      list.</para>
	  </blockquote>

	  <para>To get the device working again requires finding its PnP
	    ID and adding it to the list that the ISA probes use to
	    identify PnP devices.  This is obtained using
	    &man.pnpinfo.8; to probe the device, for example this is the
	    output from &man.pnpinfo.8; for an internal modem:</para>

	  <screen>&prompt.root; <userinput>pnpinfo</userinput>
Checking for Plug-n-Play devices...

Card assigned CSN #1
Vendor ID PMC2430 (0x3024a341), Serial Number 0xffffffff
PnP Version 1.0, Vendor Version 0
Device Description: Pace 56 Voice Internal Plug &amp; Play Modem

Logical Device ID: PMC2430 0x3024a341 #0
        Device supports I/O Range Check
TAG Start DF
    I/O Range 0x3f8 .. 0x3f8, alignment 0x8, len 0x8
        [16-bit addr]
    IRQ: 4  - only one type (true/edge)</screen>

	  <para>[more TAG lines elided]</para>

	  <screen>TAG End DF
End Tag

Successfully got 31 resources, 1 logical fdevs
-- card select # 0x0001

CSN PMC2430 (0x3024a341), Serial Number 0xffffffff

Logical device #0
IO:  0x03e8 0x03e8 0x03e8 0x03e8 0x03e8 0x03e8 0x03e8 0x03e8
IRQ 5 0
DMA 4 0
IO range check 0x00 activate 0x01</screen>

	  <para>The information you require is in the <literal>Vendor
	    ID</literal> line at the start of the output.  The
	    hexadecimal number in parentheses
	    (<literal>0x3024a341</literal> in this example) is the PnP
	    ID and the string immediately before this
	    (<literal>PMC2430</literal>) is a unique ASCII ID.</para>

	  <para>Alternatively, if &man.pnpinfo.8; does not list the card
	    in question, &man.pciconf.8; can be used instead.  This is
	    part of the output from <command>pciconf -vl</command> for
	    an onboard sound chip:</para>

	  <screen>&prompt.root; <userinput>pciconf -vl</userinput>
chip1@pci0:31:5:        class=0x040100 card=0x00931028 chip=0x24158086 rev=0x02 hdr=0x00
    vendor   = 'Intel Corporation'
    device   = '82801AA 8xx Chipset AC'97 Audio Controller'
    class    = multimedia
    subclass = audio</screen>

	  <para>Here, you would use the <varname>chip</varname> value,
	    <literal>0x24158086</literal>.</para>

	  <para>This information (<literal>Vendor ID</literal> or
	    <varname>chip</varname> value) needs adding to the file
	    <filename>/usr/src/sys/dev/sio/sio_isa.c</filename>.</para>

	  <para>You should first make a backup of
	    <filename>sio_isa.c</filename> just in case things go wrong.
	    You will also need it to make the patch to submit with your
	    PR (you are going to submit a PR, are you not?) then edit
	    <filename>sio_isa.c</filename> and search for the
	    line:</para>

	  <programlisting>static struct isa_pnp_id sio_ids[] = {</programlisting>

	  <para>Then scroll down to find the correct place to add the
	    entry for your device.  The entries look like this, and are
	    sorted on the ASCII Vendor ID string which should be
	    included in the comment to the right of the line of code
	    along with all (if it will fit) or part of the
	    <emphasis>Device Description</emphasis> from the output of
	    &man.pnpinfo.8;:</para>

	  <programlisting>{0x0f804f3f, NULL},     /* OZO800f - Zoom 2812 (56k Modem) */
{0x39804f3f, NULL},     /* OZO8039 - Zoom 56k flex */
{0x3024a341, NULL},     /* PMC2430 - Pace 56 Voice Internal Modem */
{0x1000eb49, NULL},     /* ROK0010 - Rockwell ? */
{0x5002734a, NULL},     /* RSS0250 - 5614Jx3(G) Internal Modem */</programlisting>

	  <para>Add the hexadecimal Vendor ID for your device in the
	    correct place, save the file, rebuild your kernel, and
	    reboot.  Your device should now be found as an
	    <devicename>sio</devicename> device.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="nlist-failed">
	  <para>Why do I get the error <errorname>nlist
	    failed</errorname> when running, for example,
	    <command>top</command> or <command>systat</command>?</para>
	</question>

	<answer>
	  <para>The problem is that the application you are trying to
	    run is looking for a specific kernel symbol, but, for whatever
	    reason, cannot find it; this error stems from one of two
	    problems:</para>

	  <itemizedlist>
	    <listitem>
	      <para>Your kernel and userland are not synchronized (i.e.,
		you built a new kernel but did not do an
		<maketarget>installworld</maketarget>, or vice versa),
		and thus the symbol table is different from what the
		user application thinks it is.  If this is the case,
		simply complete the upgrade process (see
		<filename>/usr/src/UPDATING</filename> for the correct
		sequence).</para>
	    </listitem>

	    <listitem>
	      <para>You are not using <command>/boot/loader</command> to
		load your kernel, but doing it directly from
		<filename>boot2</filename> (see &man.boot.8;).  While
		there is nothing wrong with bypassing
		<command>/boot/loader</command>, it generally does a
		better job of making the kernel symbols available to
		user applications.</para>
	    </listitem>
	  </itemizedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="connection-delay">
	  <para>Why does it take so long to connect to my computer via
	    <command>ssh</command> or <command>telnet</command>?</para>
	</question>

	<answer>
	  <para>The symptom: there is a long delay between the time the
	    TCP connection is established and the time when the client
	    software asks for a password (or, in &man.telnet.1;'s case,
	    when a login prompt appears).</para>

	  <para>The problem: more likely than not, the delay is caused
	    by the server software trying to resolve the client's IP
	    address into a hostname.  Many servers, including the
	    <application>Telnet</application> and
	    <application>SSH</application> servers that come with &os;,
	    do this in order to, among other things, store the hostname
	    in a log file for future reference by the
	    administrator.</para>

	  <para>The remedy: if the problem occurs whenever you connect
	    from your computer (the client) to any server, the problem is
	    with the client; likewise, if the problem only occurs when
	    someone connects to your computer (the server) the problem
	    is with the server.</para>

	  <para>If the problem is with the client, the only remedy is to
	    fix the DNS so the server can resolve it.  If this is on a
	    local network, consider it a server problem and keep
	    reading; conversely, if this is on the global Internet, you
	    will most likely need to contact your ISP and ask them to
	    fix it for you.</para>

	  <para>If the problem is with the server, and this is on a
	    local network, you need to configure the server to be able to
	    resolve address-to-hostname queries for your local address
	    range.  See the &man.hosts.5; and &man.named.8; manual pages
	    for more information.  If this is on the global Internet,
	    the problem may be that your server's resolver is not
	    functioning correctly.  To check, try to look up another
	    host &mdash; say, <hostid>www.yahoo.com</hostid>.  If it
	    does not work, that is your problem.</para>

	  <para>Following a fresh install of &os;, it is also possible
	    that domain and name server information is missing from
	    <filename>/etc/resolv.conf</filename>.  This will often
	    cause a delay in <application>SSH</application>, as the
	    option <literal>UseDNS</literal> is set to
	    <literal>yes</literal> by default in the
	    <filename>sshd_config</filename> file in
	    <filename class="directory">/etc/ssh</filename>.  If this is causing the
	    problem, you will either need to fill in the missing
	    information in <filename>/etc/resolv.conf</filename> or set
	    <literal>UseDNS</literal> to <literal>no</literal> in
	    <filename>sshd_config</filename> as a temporary
	    workaround.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="stray-irq">
	  <para>What does <errorname>stray IRQ</errorname> mean?</para>
	</question>

	<answer>
	  <para>Stray IRQs are indications of hardware IRQ glitches,
	    mostly from hardware that removes its interrupt request in
	    the middle of the interrupt request acknowledge
	    cycle.</para>

	  <para>One has three options for dealing with this:</para>

	  <itemizedlist>
	    <listitem>
	      <para>Live with the warnings.  All except the first 5 per
		irq are suppressed anyway.</para>
	    </listitem>

	    <listitem>
	      <para>Break the warnings by changing the value of
		<varname>MAX_STRAY_LOG</varname> from
		<literal>5</literal> to <literal>0</literal> in your
		platform's (e.g.  &i386;)
		<filename>intr_machdep.c</filename> file and rebuild the
		new kernel and all the warnings will be
		suppressed.</para>
	    </listitem>

	    <listitem>
	      <para>Break the warnings by installing parallel port
		hardware that uses IRQ&nbsp;7 and the PPP driver for it
		(this happens on most systems), and install an ide drive
		or other hardware that uses IRQ&nbsp;15 and a suitable
		driver for it.</para>
	    </listitem>
	  </itemizedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="file-table-full">
	  <para>Why does <errorname>file: table is full</errorname> show
	    up repeatedly in &man.dmesg.8;?</para>
	</question>

	<answer>
	  <para>This error message indicates you have exhausted the
	    number of available file descriptors on your system.  Please
	    see the <ulink
	      url="&url.books.handbook;/configtuning-kernel-limits.html#KERN-MAXFILES">kern.maxfiles</ulink>
	    section of the <ulink
	      url="&url.books.handbook;/configtuning-kernel-limits.html">Tuning Kernel Limits</ulink>
	    section of the Handbook for a discussion and
	    solution.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="calcru-negative-runtime">
	  <para>Why are <errorname>calcru: negative runtime</errorname>
	    or <errorname>calcru: runtime went backwards</errorname>
	    messages pounding the console?</para>
	</question>

	<answer>
	  <para>There is a known problem when enabling &intel; Enhanced
	    SpeedStep from the BIOS: it causes the kernel to start printing
	    <errorname>calcru</errorname> messages like this:</para>

	  <screen>calcru: runtime went backwards from 6 usec to 3 usec for pid 37 (pagezero)
calcru: runtime went backwards from 6 usec to 3 usec for pid 36 (vmdaemon)
calcru: runtime went backwards from 170 usec to 138 usec for pid 35 (pagedaemon)
calcru: runtime went backwards from 553 usec to 291 usec for pid 15 (swi6: task queue)
calcru: runtime went backwards from 15521 usec to 10366 usec for pid 2 (g_event)
calcru: runtime went backwards from 25 usec to 12 usec for pid 11 (swi1: net)
calcru: runtime went backwards from 4417 usec to 3960 usec for pid 1 (init)
calcru: runtime went backwards from 2084385 usec to 1793542 usec for pid 1 (init)
calcru: runtime went backwards from 408 usec to 204 usec for pid 0 (swapper)</screen>

	  <para>It is because &intel; SpeedStep (EIST) is incompatible
	    with some motherboards.</para>

	  <para>Workaround: Disable the EIST feature in the BIOS.  You
	    can still achieve ACPI-based processor frequency throttling
	    by using &man.powerd.8;.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="computer-clock-skew">
	  <para>Why does the clock on my computer keep incorrect time?</para>
	</question>

	<answer>
	  <para>Your computer has two or more clocks, and &os; has
	    chosen to use the wrong one.</para>

	  <para>Run &man.dmesg.8;, and check for lines that contain
	    <literal>Timecounter</literal>.  The one with the highest
	    quality value that &os; chose.</para>

	  <screen>&prompt.root; <userinput>dmesg | grep Timecounter</userinput>
Timecounter "i8254" frequency 1193182 Hz quality 0
Timecounter "ACPI-fast" frequency 3579545 Hz quality 1000
Timecounter "TSC" frequency 2998570050 Hz quality 800
Timecounters tick every 1.000 msec</screen>

	  <para>You can confirm this by checking the
	    <varname>kern.timecounter.hardware</varname>
	    &man.sysctl.3;.</para>

	  <screen>&prompt.root; <userinput>sysctl kern.timecounter.hardware</userinput>
kern.timecounter.hardware: ACPI-fast</screen>

	  <para>It may be a broken ACPI timer.  The simplest solution is
	    to disable the ACPI timer in
	    <filename>/boot/loader.conf</filename>:</para>

	  <programlisting>debug.acpi.disabled="timer"</programlisting>

	  <para>Or the BIOS may modify the TSC clock&mdash;perhaps to
	    change the speed of the processor when running from batteries,
	    or going into a power saving mode, but &os; is unaware of
	    these adjustments, and appears to gain or lose time.</para>

	  <para>In this example, the <literal>i8254</literal> clock is
	    also available, and can be selected by writing its name to the
	    <varname>kern.timecounter.hardware</varname>
	    &man.sysctl.3;.</para>

	  <screen>&prompt.root; <userinput>sysctl -w kern.timecounter.hardware=i8254</userinput>
kern.timecounter.hardware: TSC -&gt; i8254</screen>

	  <para>Your computer should now start keeping more accurate
	    time.</para>

	  <para>To have this change automatically run at boot time, add
	    the following line to
	    <filename>/etc/sysctl.conf</filename>:</para>

	  <programlisting>kern.timecounter.hardware=i8254</programlisting>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="null-null">
	  <para>Why did my laptop fail to correctly probe PC
	    cards?</para>
	</question>

	<answer>
	  <para>This problem is common on laptops that boot more than
	    one operating system.  Some non-BSD operating systems leave
	    PC card hardware in an inconsistent state.  &man.pccardd.8;
	    will detect the card as
	    <errorname>"(null)""(null)"</errorname> instead of its
	    actual model.</para>

	  <para>You must remove all power from the PC card slot to fully
	    reset the hardware.  Completely power off the laptop.  (Do
	    not suspend it, do not let it go into standby; the power
	    needs to be completely off.) Wait a few moments, and reboot.
	    Your PC card should work now.</para>

	  <para>Some laptop hardware lies when it claims to be off.  If
	    the above does not work shut down, remove the battery, wait
	    a moment, replace the battery, and reboot.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="boot-read-error">
	  <para>Why does &os;'s boot loader display <errorname>Read
	    error</errorname> and stop after the BIOS screen?</para>
	</question>

	<answer>
	  <para>&os;'s boot loader is incorrectly recognizing the hard
	    drive's geometry.  This must be manually set within
	    &man.fdisk.8; when creating or modifying &os;'s
	    slice.</para>

	  <para>The correct drive geometry values can be found within
	    the machine's BIOS.  Look for the number of cylinders, heads
	    and sectors for the particular drive.</para>

	  <para>Within &man.sysinstall.8;'s fdisk, hit
	    <keycap>G</keycap> to set the drive geometry.</para>

	  <para>A dialog will pop up requesting the number of cylinders,
	    heads and sectors.  Type the numbers found from the BIOS
	    separated by forward slashes.  For example, values of 5000
	    cylinders, 250 heads, and 60 sectors would be entered as
	    <userinput>5000/250/60</userinput>.</para>

	  <para>Press <keycap>Enter</keycap> to set the values, and hit
	    <keycap>W</keycap> to write the new partition table to the
	    drive.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="bootmanager-restore">
	  <para>Another operating system destroyed my Boot Manager.  How
	    do I get it back?</para>
	</question>

	<answer>
	  <para>Enter &man.sysinstall.8; and choose
	    <guimenuitem>Configure</guimenuitem>, then
	    <guimenuitem>Fdisk</guimenuitem>.  Select the disk the Boot
	    Manager resided on with the <keycap>Space</keycap> key.
	    Press <keycap>W</keycap> to write changes to the drive.  A
	    prompt will appear asking which boot loader to install.
	    Select this, and it will be restored.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="indefinite-wait-buffer">
	  <para>What does the error <errorname>swap_pager: indefinite
	    wait buffer:</errorname> mean?</para>
	</question>

	<answer>
	  <para>This means that a process is trying to page memory to
	    disk, and the page attempt has hung trying to access the
	    disk for more than 20 seconds.  It might be caused by bad
	    blocks on the disk drive, disk wiring, cables, or any other
	    disk I/O-related hardware.  If the drive itself is actually
	    bad, you will also see disk errors in
	    <filename>/var/log/messages</filename> and in the output of
	    <command>dmesg</command>.  Otherwise, check your cables and
	    connections.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="udma-icrc">
	  <para>What are <errorname>UDMA ICRC</errorname> errors, and
	    how do I fix them?</para>
	</question>

	<answer>
	  <para>The &man.ata.4; driver reports <errorname>UDMA
	    ICRC</errorname> errors when a DMA transfer to or from a drive
	    is corrupted.  The driver will retry the operation a few
	    times.  Should the retries fail, it will switch from DMA to
	    the slower PIO mode of communication with the device.</para>

	  <para>The problem can be caused by many factors, although
	    perhaps the most common cause is faulty or incorrect
	    cabling.  Check that the ATA cables are undamaged and rated
	    for the Ultra&nbsp;DMA mode in use.  If you are using
	    removable drive trays, they must also be compatible.  Be
	    sure that all connections are making good contact.  Problems
	    have also been noticed when an old drive is installed on the
	    same ATA channel as an Ultra&nbsp;DMA&nbsp;66 (or faster)
	    drive.  Lastly, these errors can indicate that the drive is
	    failing.  Most drive vendors provide testing software for
	    their drives, so test your drive, and, if necessary, back up
	    your data and replace it.</para>

	  <para>The &man.atacontrol.8; utility can be used to show and
	    select the DMA or PIO modes used for each ATA device.  In
	    particular, <command>atacontrol mode
	    <replaceable>channel</replaceable></command> will show the
	    modes in use on a particular ATA channel, where the primary
	    channel is numbered 0, and so on.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="lock-order-reversal">
	  <para>What is a <errorname>lock order
	    reversal</errorname>?</para>
	</question>

	<answer>
	  <para>An answer for this question can be found in the &os;
	    Glossary, see <ulink
	      url="&url.books.handbook;/freebsd-glossary.html#LOR-GLOSSARY">LOR</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="called-with-non-sleepable-locks-held">
	  <para>What does <errorname>Called ...  with the following
	    non-sleepable locks held</errorname> mean?</para>
	</question>

	<answer>
	  <para>This means that a function that may sleep was called
	    while a mutex (or other unsleepable) lock was held.</para>

	  <para>The reason this is an error is because mutexes are not
	    intended to be held for long periods of time; they are
	    supposed to only be held to maintain short periods of
	    synchronization.  This programming contract allows device
	    drivers to use mutexes to synchronize with the rest of the
	    kernel during interrupts.  Interrupts (under &os;) may not
	    sleep.  Hence it is imperative that no subsystem in the
	    kernel block for an extended period while holding a
	    mutex.</para>

	  <para>To catch such errors, assertions may be added to the
	    kernel that interact with the &man.witness.4; subsystem to
	    emit a warning or fatal error (depending on the system
	    configuration) when a potentially blocking call is made
	    while holding a mutex.</para>

	  <para>In summary, such warnings are non-fatal, however with
	    unfortunate timing they could cause undesirable effects
	    ranging from a minor blip in the system's responsiveness to
	    a complete system lockup.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="touch-not-found">
	  <para>Why does
	    <maketarget>buildworld</maketarget>/<maketarget>installworld</maketarget>
	    die with the message <errorname>touch: not
	    found</errorname>?</para>
	</question>

	<answer>
	  <para>This error does not mean that the &man.touch.1; utility
	    is missing.  The error is instead probably due to the dates
	    of the files being set sometime in the future.  If your
	    CMOS-clock is set to local time you need to run the command
	    <command>adjkerntz&nbsp;-i</command> to adjust the kernel
	    clock when booting into single user mode.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="commercial">
    <title>Commercial Applications</title>

    <note>
      <para>This section is still very sparse, though we are hoping, of
	course, that companies will add to it! :)  The &os; group has no
	financial interest in any of the companies listed here but
	simply lists them as a public service (and feels that commercial
	interest in &os; can have very positive effects on &os;'s
	long-term viability).  We encourage commercial software vendors
	to send their entries here for inclusion.  See <ulink
	  url="&url.base;/commercial/index.html">the Vendors page</ulink>
	for a longer list.</para>
    </note>

    <qandaset>
      <qandaentry>
	<question id="officesuite">
	  <para>Where can I get an Office Suite for &os;?</para>
	</question>

	<answer>
	  <para>The open-source <application><ulink
	      url="http://www.openoffice.org">OpenOffice.org</ulink></application>
	    and <application><ulink
              url="http://www.libreoffice.org">LibreOffice</ulink></application>
            office suites work natively on &os;.  The &linux; version of
	    <application><ulink
	      url="http://www.oracle.com/us/products/applications/open-office/index.html">Oracle Open Office</ulink></application>,
	    the value-added closed-source version of OpenOffice.org,
	    also works on &os;.</para>

	  <para>&os; also includes a variety of text editors,
	    spreadsheets, and drawing programs in the Ports
	    Collection.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="motif">
	  <para>Where can I get <application>&motif;</application> for
	    &os;?</para>
	</question>

	<answer>
	  <para>The Open Group has released the source code to
	    <application>&motif;</application>.  You can
	    install the <filename
	      role="package">x11-toolkits/open-motif</filename> package,
	    or compile it from ports.  Refer to <ulink
	      url="&url.books.handbook;/ports.html">the ports section of the Handbook</ulink>
	    for more information on how to do this.</para>

	  <note>
	    <para>The <application>Open&nbsp;&motif;</application>
	      distribution only allows redistribution if it is running
	      on an <ulink
		url="http://www.opensource.org/">open source</ulink>
	      operating system.</para>
	  </note>

	  <para>In addition, there are commercial distributions of the
	    <application>&motif;</application> software available.  These,
	    however, are not for free, but their license allows them to
	    be used in closed-source software.  Contact <link
	    linkend="apps2go">Apps2go</link> for the least expensive ELF
	    <application>&motif;&nbsp;2.1.20</application> distribution
	    for &os; (&i386;).<anchor id="apps2go"/></para>

	  <para>There are two distributions, the <quote>development
	    edition</quote> and the <quote>runtime edition</quote> (for
	    much less).  These distributions includes:</para>

	  <itemizedlist>
	    <listitem>
	      <para><application>OSF/&motif; manager</application>,
		<application>xmbind</application>,
		<application>panner</application>,
		<application>wsm</application>.</para>
	    </listitem>

	    <listitem>
	      <para>Development kit with uil, mrm, xm, xmcxx, include
		and <application>Imake</application> files.</para>
	    </listitem>

	    <listitem>
	      <para>Static and dynamic ELF libraries.</para>
	    </listitem>

	    <listitem>
	      <para>Demonstration applets.</para>
	    </listitem>
	  </itemizedlist>

	  <para>Be sure to specify that you want the &os; version of
	    <application>&motif;</application> when ordering (do not
	    forget to mention the architecture you want too)!  Versions
	    for NetBSD and OpenBSD are also sold by
	    <emphasis>Apps2go</emphasis>.  This is currently a FTP only
	    download.</para>

	  <variablelist>
	    <varlistentry>
	      <term>More info</term>

	      <listitem>
		<para><ulink url="http://www.apps2go.com/"> Apps2go
		  WWW page</ulink></para>
	      </listitem>
	    </varlistentry>

	    <varlistentry>
	      <term>or</term>

	      <listitem>
		<para><email>sales@apps2go.com</email> or
		  <email>support@apps2go.com</email></para>
	      </listitem>
	    </varlistentry>

	    <varlistentry>
	      <term>or</term>

	      <listitem>
		<para>phone (817)&nbsp;431&nbsp;8775 or
		  +1&nbsp;817&nbsp;431-8775</para>
	      </listitem>
	    </varlistentry>
	  </variablelist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="cde">
	  <para>Where can I get <application>CDE</application> for
	    &os;?</para>
	</question>

	<answer>
	  <para><emphasis>Xi Graphics</emphasis> used to sell
	    <application>CDE</application> for &os;, but no longer
	    do.</para>

	  <para><ulink
	      url="http://www.kde.org/"><application>KDE</application></ulink>
	    is an open source X11 desktop which is similar to
	    <application>CDE</application> in many respects.  You might
	    also like the look and feel of <ulink
	      url="http://www.xfce.org/"><application>xfce</application></ulink>.
	    <application>KDE</application> and
	    <application>xfce</application> are both in the <ulink
	      url="&url.base;/ports/index.html">ports system</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="database-systems">
	  <para>Are there any Database systems for &os;?</para>
	</question>

	<answer>
	  <para>Yes!  See the <ulink
	      url="&url.base;/commercial/software_bycat.html#CATEGORY_DATABASE">Commercial Vendors</ulink>
	    section of &os;'s Web site.</para>

	  <para>Also see the <ulink
	      url="&url.base;/ports/databases.html">Databases</ulink>
	    section of the Ports Collection.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="oracle-support">
	  <para>Can I run <application>&oracle;</application> on
	    &os;?</para>
	</question>

	<answer>
	  <para>Yes.  Instructions on how to set up &linux;
	    <application>&oracle;</application> on &os; can be found
	    under <ulink
	    url="http://www.shadowcom.net/freebsd-oracle9i/">http://www.shadowcom.net/freebsd-oracle9i/</ulink>.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="applications">
    <title>User Applications</title>

    <qandaset>
      <qandaentry>
	<question id="user-apps">
	  <para>So, where are all the user applications?</para>
	</question>

	<answer>
	  <para>Please take a look at <ulink
	      url="&url.base;/ports/index.html">the ports page</ulink>
	    for info on software packages ported to &os;.  The list
	    currently tops &os.numports; and is growing daily, so come
	    back to check often or subscribe to the &a.announce; for
	    periodic updates on new entries.</para>

	  <para>Most ports should work on the
	    &rel3.relx;, &rel2.relx;, and &rel.relx; branches.
	    Each time a &os;
	    release is made, a snapshot of the ports tree at the time of
	    release in also included in the <filename class="directory">ports/</filename>
	    directory.</para>

	  <para>We also support the concept of a <quote>package</quote>,
	    essentially no more than a compressed binary distribution
	    with a little extra intelligence embedded in it for doing
	    whatever custom installation work is required.  A package
	    can be installed and uninstalled again easily without having
	    to know the gory details of which files it includes.</para>

	  <para>Use the <guimenuitem>Packages</guimenuitem> package
	    installation menu in &man.sysinstall.8; (under the
	    <guimenuitem>Configure</guimenuitem> menu item) or invoke
	    the &man.pkg.add.1; command on the specific package files
	    you are interested in installing.  Package files can usually
	    be identified by their <filename>.tbz</filename> suffix and
	    CD-ROM distribution people will have a
	    <filename class="directory">packages/All</filename> directory on their CD
	    which contains such files.  They can also be downloaded over
	    the net for various versions of &os; at the following
	    locations:</para>

	  <variablelist>
	    <varlistentry>
	      <term>for &rel3.relx;&nbsp;-RELEASE/&rel3.stable;</term>

	      <listitem>
		<para><ulink
		    url="ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/&rel3.packages;/">ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/&rel3.packages;</ulink>
		  </para>
	      </listitem>
	    </varlistentry>

	    <varlistentry>
	      <term>for &rel2.relx;&nbsp;-RELEASE/&rel2.stable;</term>

	      <listitem>
		<para><ulink
		    url="ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/&rel2.packages;/">ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/&rel2.packages;</ulink>
		  </para>
	      </listitem>
	    </varlistentry>

	    <varlistentry>
	      <term>for &rel.relx;&nbsp;-RELEASE/&rel.stable;</term>

	      <listitem>
		<para><ulink
		    url="ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/&rel.packages;/">ftp://ftp.FreeBSD.org/pub/FreeBSD/ports/i386/&rel.packages;</ulink>
		  </para>
	      </listitem>
	    </varlistentry>
	  </variablelist>

	  <para>or your nearest local mirror site.</para>

	  <para>Note that all ports may not be available as packages
	    since new ones are constantly being added.  It is always a
	    good idea to check back periodically to see which packages
	    are available at the <ulink
	      url="ftp://ftp.FreeBSD.org/pub/FreeBSD/">ftp.FreeBSD.org</ulink>
	    master site.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="configure-inn">
	  <para>How do I configure INN (Internet News) for my
	    machine?</para>
	</question>

	<answer>
	  <para>After installing the <filename
	      role="package">news/inn</filename> package or port, an
	    excellent place to start are the <ulink
	      url="http://www.eyrie.org/~eagle/faqs/inn.html">INN
	    FAQ</ulink>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="java">
	  <para>Does &os; support &java;?</para>
	</question>

	<answer>
	  <para>Yes.  Please see <ulink
	      url="&url.base;/java/index.html">http://www.FreeBSD.org/java/</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ports-4x">
	  <para>Why can I not build this port on my
	    &rel3.relx;&nbsp;-, &rel2.relx;&nbsp;-, or
	    &rel.relx;&nbsp;-STABLE machine?</para>
	</question>

	<answer>
	  <para>If you are running a &os; version that lags
	    significantly behind <emphasis>-CURRENT</emphasis> or
	    <emphasis>-STABLE</emphasis>, you may need to update your
	    Ports Collection; see the <ulink
	      url="&url.books.porters-handbook;/keeping-up.html">Keeping Up</ulink>
	    section of the Porter's Handbook for further information on
	    how to do this.  If you are up to date, then someone might
	    have committed a change to the port which works for
	    <emphasis>-CURRENT</emphasis> but which broke the port for
	    <emphasis>-STABLE</emphasis>.  Please submit a bug report on
	    this with the &man.send-pr.1; command, since the Ports
	    Collection is supposed to work for both the
	    <emphasis>-CURRENT</emphasis> and
	    <emphasis>-STABLE</emphasis> branches.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="make-index">
	  <para>I just tried to build <filename>INDEX</filename> using
	    <command>make <maketarget>index</maketarget></command>, and
	    it failed.  Why?</para>
	</question>

	<answer>
	  <para>First, always make sure that you have a completely
	    up-to-date Ports Collection.  Errors that affect building
	    <filename>INDEX</filename> from an up-to-date copy of the
	    Ports Collection are high-visibility and are thus almost
	    always fixed immediately.</para>

	  <para>However, if you are up-to-date, perhaps you are seeing
	    another problem.  <command>make <maketarget>index</maketarget></command>
	    has a known bug in dealing with incomplete copies of the
	    Ports Collection.  It assumes that you have a local copy of
	    every single port that every other port that you have a
	    local copy of depends on.  To explain, if you have a copy of
	    <filename>foo/bar</filename> on your disk, and
	    <filename>foo/bar</filename> depends on
	    <filename>baz/quux</filename>, then you must also have a
	    copy of <filename>baz/quux</filename> on your disk, and the
	    ports <filename>baz/quux</filename> depends on, and so on.
	    Otherwise, <command>make <maketarget>index</maketarget></command>
	    has insufficient information to create its dependency
	    tree.</para>

	  <para>This is particularly a problem for &os; users who
	    utilize &man.csup.1; to track the Ports
	    Collection but choose not to install certain categories by
	    specifying them in <filename>refuse</filename>.  In theory,
	    one should be able to refuse categories, but in practice
	    there are too many ports that depend on ports in other
	    categories.  Until someone comes up with a solution for this
	    problem, the general rule is that if you want to build
	    <filename>INDEX</filename>, you must have a complete copy of
	    the Ports Collection.</para>

	  <para>There are rare cases where <filename>INDEX</filename>
	    will not build due to odd cases involving
	    <makevar>WITH_<replaceable>*</replaceable></makevar> or
	    <makevar>WITHOUT_<replaceable>*</replaceable></makevar>
	    variables being set in <filename>make.conf</filename>.  If
	    you suspect that this is the case, please try to make
	    <filename>INDEX</filename> with those make variables turned
	    off before reporting it to &a.ports;.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="cvsup-in-base">
	  <para>Why is <application>CVSup</application> not integrated
	    in the main &os; tree?</para>
	</question>

	<answer>
	  <para>The &os; base system is designed as self-hosting &mdash;
	    it should be possible to build the whole operating system
	    starting with a very limited set of tools.  Thus, the actual
	    build tools needed to compile the &os; sources are bundled
	    with the sources themselves.  This includes a C compiler
	    (&man.gcc.1;), &man.make.1;, &man.awk.1;, and similar
	    tools.</para>

	  <para>Since <application>CVSup</application> is written in
	    Modula-3, adding it to the &os; base system would also require
	    adding and maintaining a Modula-3 compiler.  This would lead
	    to both an increase in the disk space consumed by the &os;
	    sources and additional maintenance work.  Thus, it is much
	    easier for both the developers and users to keep
	    <application>CVSup</application> as a separate port, which
	    can be easily installed as a package bundled on the &os;
	    installation CDs.</para>

	  <para>However, &os; users are not without an integrated
	    <application>CVSup</application> compatible client anymore
	    since &os;&nbsp;6.2-RELEASE.  Thanks to &a.mux;,
	    <application>CVSup</application> was rewritten in C as
	    &man.csup.1; and it is the part of the base system by now.
	    Although it does not implement all the features of
	    <application>CVSup</application> at the moment, it is good
	    enough (and really fast!) to keep your sources synchronized.
	    For systems earlier than 6.2, it can be installed as a port
	    or package (see <filename
	      role="package">net/csup</filename>).</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ports-update">
	  <para>I updated the sources, now how do I update my installed
	    ports?</para>
	</question>

	<answer>
	  <para>&os; does not include a port upgrading tool, but it does
	    have some tools to make the upgrade process somewhat easier.
	    You can also install additional tools to simplify port
	    handling, see the <ulink
	      url="&url.books.handbook;/ports-using.html">Upgrading Ports</ulink>
	    section in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ports-major-upgrade">
	  <para>Do I need to recompile every port each time I perform a
	    major version update?</para>
	</question>

	<answer>
	  <para>By all means!  While a recent system will run with
	    software compiled under an older release, you will end up with
	    things randomly crashing and failing to work once you start
	    installing other ports or updating a portion of what you
	    already have.</para>

	  <para>When the system is upgraded, various shared libraries,
	    loadable modules, and other parts of the system will be
	    replaced with newer versions.  Applications linked against
	    the older versions may fail to start or, in other cases,
	    fail to function properly.</para>

	  <para>For more information, see <ulink
	      url="&url.books.handbook;/updating-upgrading-freebsdupdate.html#FREEBSDUPDATE-UPGRADE">the section on upgrades</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ports-minor-upgrade">
	  <para>Do I need to recompile every port each time I perform a
	    minor version update?</para>
	</question>

	<answer>
	  <para>In general, no.  &os; developers do their utmost to
	    guarantee binary compatibility across all releases with the
	    same major version number.  Any exceptions will be
	    documented in the Release Notes, and advice given there
	    should be followed.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="minimal-sh">
	  <para>Why is <command>/bin/sh</command> so minimal?  Why does
	    &os; not use <command>bash</command> or another
	    shell?</para>
	</question>

	<answer>
	  <para>Because &posix; says that there shall be such a
	    shell.</para>

	  <para>The more complicated answer: many people need to write
	    shell scripts which will be portable across many systems.
	    That is why &posix; specifies the shell and utility commands
	    in great detail.  Most scripts are written in Bourne shell,
	    and because several important programming interfaces
	    (&man.make.1;, &man.system.3;, &man.popen.3;, and analogues
	    in higher-level scripting languages like Perl and Tcl) are
	    specified to use the Bourne shell to interpret commands.
	    Because the Bourne shell is so often and widely used, it is
	    important for it to be quick to start, be deterministic in
	    its behavior, and have a small memory footprint.</para>

	  <para>The existing implementation is our best effort at
	    meeting as many of these requirements simultaneously as we
	    can.  In order to keep <command>/bin/sh</command> small, we
	    have not provided many of the convenience features that
	    other shells have.  That is why the Ports Collection
	    includes more featureful shells like
	    <command>bash</command>, <command>scsh</command>,
	    <command>tcsh</command>, and <command>zsh</command>.  (You
	    can compare for yourself the memory utilization of all these
	    shells by looking at the <quote>VSZ</quote> and
	    <quote>RSS</quote> columns in a <command>ps
	    <option>-u</option></command> listing.)</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="netscape-slow-startup">
	  <para>Why do <application>&netscape;</application> and
	    <application>Opera</application> take so long to start?</para>
	</question>

	<answer>
	  <para>The usual answer is that DNS on your system is
	    misconfigured.  Both <application>&netscape;</application>
	    and <application>Opera</application> perform DNS checks when
	    starting up.  The browser will not appear on your desktop
	    until the program either gets a response or determines that
	    the system has no network connection.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ports-base-update">
	  <para>I updated parts of the Ports Collection using
	    <application>CVSup</application>, and now many ports fail to
	    build with mysterious error messages!  What happened?  Is
	    the Ports Collection broken in some major way?</para>
	</question>

	<answer>
	  <para>If you only update parts of the Ports Collection, using
	    one of its <application>CVSup</application> subcollections
	    and not the <literal>ports-all</literal>
	    <application>CVSup</application> collection, you should
	    <emphasis>always</emphasis> update the
	    <literal>ports-base</literal> subcollection too!  The
	    reasons are described <ulink
	      url="&url.books.handbook;/cvsup.html#CVSUP-COLLEC-PBASE-WARN">in the Handbook</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="midi-sound-files">
	  <para>How do I create audio CDs from my MIDI files?</para>
	</question>

	<answer>
	  <para>To create audio CDs from MIDI files, first install
	    <filename role="package">audio/timidity++</filename> from
	    ports then install manually the GUS patches set by Eric A.
	    Welsh, available at <ulink
	      url="http://alleg.sourceforge.net/digmid.html"></ulink>.
	    After <application>TiMidity++</application> has been installed
	    properly, MIDI files may be converted to WAV files with the
	    following command line:</para>

	  <screen>&prompt.user; <userinput>timidity -Ow -s 44100 -o <replaceable>/tmp/juke/01.wav</replaceable> <replaceable>01.mid</replaceable></userinput></screen>

	  <para>The WAV files can then be converted to other formats or
	    burned onto audio CDs, as described in the <ulink
	      url="&url.books.handbook;/creating-cds.html">&os; Handbook</ulink>.
	    </para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="kernelconfig">
    <title>Kernel Configuration</title>

    <qandaset>
      <qandaentry>
	<question id="make-kernel">
	  <para>I would like to customize my kernel.  Is it
	    difficult?</para>
	</question>

	<answer>
	  <para>Not at all!  Check out the <ulink
	      url="&url.books.handbook;/kernelconfig.html">kernel config section of the Handbook</ulink>.
	    </para>

	  <note>
	    <para>The new <filename>kernel</filename> will be installed
	      to the <filename class="directory">/boot/kernel</filename> directory along
	      with its modules, while the old kernel and its modules
	      will be moved to the <filename class="directory">/boot/kernel.old</filename>
	      directory, so if you make a mistake the next time you play
	      with your configuration you can boot the previous version
	      of your kernel.</para>
	  </note>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="missing-hw-float">
	  <para>My kernel compiles fail because
	    <literal>_hw_float</literal> is missing.  How do I solve
	    this problem?</para>
	</question>

	<answer>
	  <para>You probably removed <devicename>npx0</devicename> (see
	    &man.npx.4;) from your kernel configuration file because you
	    do not have a math co-processor.  The
	    <devicename>npx0</devicename> device is
	    <emphasis>MANDATORY</emphasis>.  Somewhere inside your
	    hardware lies a device that provides hardware floating-point
	    support, even if it is no longer a separate device as used
	    in the good old 386 days.  You <emphasis>must</emphasis>
	    include the <devicename>npx0</devicename> device.  Even if
	    you manage to build a kernel without
	    <devicename>npx0</devicename> support, it will not boot
	    anyway.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="why-kernel-big">
	  <para>Why is my kernel so big (over 10&nbsp;MB)?</para>
	</question>

	<answer>
	  <para>Chances are, you compiled your kernel in <emphasis>debug
	    mode</emphasis>.  Kernels built in debug mode contain many
	    symbols that are used for debugging, thus greatly increasing
	    the size of the kernel.  Note that there will be little or
	    no performance decrease from running a debug kernel, and it
	    is useful to keep one around in case of a system
	    panic.</para>

	  <para>However, if you are running low on disk space, or you
	    simply do not want to run a debug kernel, make sure that
	    both of the following are true:</para>

	  <itemizedlist>
	    <listitem>
	      <para>You do not have a line in your kernel configuration
		file that reads:</para>

	      <programlisting>makeoptions DEBUG=-g</programlisting>
	    </listitem>

	    <listitem>
	      <para>You are not running &man.config.8; with the
		<option>-g</option> option.</para>
	    </listitem>
	  </itemizedlist>

	  <para>Either of the above settings will cause your kernel to
	    be built in debug mode.  As long as you make sure you follow
	    the steps above, you can build your kernel normally, and you
	    should notice a fairly large size decrease; most kernels
	    tend to be around 1.5&nbsp;MB to 2&nbsp;MB.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="multiport-serial-interrupts">
	  <para>Why do I get interrupt conflicts with multi-port serial
	    code?</para>
	</question>

	<answer>
	  <para>When I compile a kernel with multi-port serial code, it
	    tells me that only the first port is probed and the rest
	    skipped due to interrupt conflicts.  How do I fix
	    this?</para>

	  <para>The problem here is that &os; has code built-in to keep
	    the kernel from getting trashed due to hardware or software
	    conflicts.  The way to fix this is to leave out the IRQ
	    settings on all but one port.  Here is an example:</para>

	  <programlisting>#
# Multiport high-speed serial line - 16550 UARTS
#
device sio2 at isa? port 0x2a0 tty irq 5 flags 0x501 vector siointr
device sio3 at isa? port 0x2a8 tty flags 0x501 vector siointr
device sio4 at isa? port 0x2b0 tty flags 0x501 vector siointr
device sio5 at isa? port 0x2b8 tty flags 0x501 vector siointr</programlisting>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="generic-kernel-build-failure">
	  <para>Why does every kernel I try to build fail to compile,
	    even <filename>GENERIC</filename>?</para>
	</question>

	<answer>
	  <para>There are a number of possible causes for this problem.
	    They are, in no particular order:</para>

	  <itemizedlist>
	    <listitem>
	      <para>You are not using the
		<command>make <maketarget>buildkernel</maketarget></command> and
		<command>make <maketarget>installkernel</maketarget></command>
		targets, and your source tree is different from the one
		used to build the currently running system (e.g., you
		are compiling &rel.current;-RELEASE on a
		&rel2.current;-RELEASE system).  If you are attempting
		an upgrade, please read the
		<filename>/usr/src/UPDATING</filename> file, paying
		particular attention to the <quote>COMMON ITEMS</quote>
		section at the end.</para>
	    </listitem>

	    <listitem>
	      <para>You are using the
		<command>make <maketarget>buildkernel</maketarget></command>
		and
		<command>make <maketarget>installkernel</maketarget></command>
		targets, but you failed to assert the completion of the
		<command>make <maketarget>buildworld</maketarget></command>
		target.  The
		<command>make <maketarget>buildkernel</maketarget></command>
		target relies on files generated by the
		<command>make <maketarget>buildworld</maketarget></command>
		target to complete its job correctly.</para>
	    </listitem>

	    <listitem>
	      <para>Even if you are trying to build <link
		  linkend="stable">&os;-STABLE</link>, it is possible that
		you fetched the source tree at a time when it was either
		being modified, or broken for other reasons; only
		releases are absolutely guaranteed to be buildable,
		although <link linkend="stable">&os;-STABLE</link>
		builds fine the majority of the time.  If you have not
		already done so, try re-fetching the source tree and see
		if the problem goes away.  Try using a different server
		in case the one you are using is having problems.</para>
	    </listitem>
	  </itemizedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="scheduler-in-use">
	  <para>How can I verify which scheduler is in use on a running
	    system?</para>
	</question>

	<answer>
	  <para>Check for the existence of the
	    <varname>kern.sched.quantum</varname> sysctl.  If you have
	    it, you should see something like this:</para>

	  <screen>&prompt.user; sysctl <replaceable>kern.sched.quantum</replaceable>
kern.sched.quantum: 99960</screen>

	  <para>If the <varname>kern.sched.quantum</varname> sysctl
	    exists, you are using the 4BSD scheduler (&man.sched.4bsd.4;).
	    If not, you will get an error printed by &man.sysctl.8;
	    (which you can safely ignore):</para>

	  <screen>&prompt.user; sysctl <replaceable>kern.sched.quantum</replaceable>
sysctl: unknown oid 'kern.sched.quantum'</screen>

	  <para>The name of the scheduler currently being used is
	    directly available as the value of the
	    <varname>kern.sched.name</varname> sysctl:</para>

	  <screen>&prompt.user; sysctl <replaceable>kern.sched.name</replaceable>
kern.sched.name: 4BSD</screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="scheduler-kern-quantum">
	  <para>What is <varname>kern.sched.quantum</varname>?</para>
	</question>

	<answer>
	  <para><varname>kern.sched.quantum</varname> is the maximum
	    number of ticks a process can run without being preempted.  It
	    is specific to the 4BSD scheduler, so you can use its
	    presence or absence to determine which scheduler is in
	    use.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="disks">
    <title>Disks, File Systems, and Boot Loaders</title>

    <qandaset>
      <qandaentry>
	<question id="adding-disks">
	  <para>How can I add my new hard disk to my &os; system?</para>
	</question>

	<answer>
	  <para>See the <ulink
	      url="&url.books.handbook;/disks-adding.html">Adding Disks</ulink>
	    section in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="new-huge-disk">
	  <para>How do I move my system over to my huge new disk?</para>
	</question>

	<answer>
	  <para>The best way is to reinstall the OS on the new disk,
	    then move the user data over.  This is highly recommended if
	    you have been tracking <emphasis>-STABLE</emphasis> for more
	    than one release, or have updated a release instead of
	    installing a new one.  You can install booteasy on both
	    disks with &man.boot0cfg.8;, and dual boot them until you
	    are happy with the new configuration.  Skip the next
	    paragraph to find out how to move the data after doing
	    this.</para>

	  <para>Should you decide not to do a fresh install, you need to
	    partition and label the new disk with either
	    &man.sysinstall.8;, or &man.fdisk.8; and &man.disklabel.8;.
	    You should also install booteasy on both disks with
	    &man.boot0cfg.8;, so that you can dual boot to the old or
	    new system after the copying is done.  See the <ulink
	      url="&url.articles.formatting-media;/index.html">formatting-media article</ulink>
	    for details on this process.</para>

	  <para>Now you have the new disk set up, and are ready to move
	    the data.  Unfortunately, you cannot just blindly copy the
	    data.  Things like device files (in
	    <filename class="directory">/dev</filename>), flags, and links tend to screw
	    that up.  You need to use tools that understand these
	    things, which means &man.dump.8;.  Although it is suggested
	    that you move the data in single user mode, it is not
	    required.</para>

	  <para>You should never use anything but &man.dump.8; and
	    &man.restore.8; to move the root file system.  The
	    &man.tar.1; command may work &mdash; then again, it may not.
	    You should also use &man.dump.8; and &man.restore.8; if you
	    are moving a single partition to another empty partition.
	    The sequence of steps to use <command>dump</command> to move
	    a partitions data to a new partition is:</para>

	  <procedure>
	    <step>
	      <para><command>newfs</command> the new partition.</para>
	    </step>

	    <step>
	      <para><command>mount</command> it on a temporary mount
		point.</para>
	    </step>

	    <step>
	      <para><command>cd</command> to that directory.</para>
	    </step>

	    <step>
	      <para><command>dump</command> the old partition, piping
		output to the new one.</para>
	    </step>
	  </procedure>

	  <para>For example, if you are going to move root to
	    <devicename>/dev/<replaceable>ad1s1a</replaceable></devicename>,
	    with <filename class="directory"><replaceable>/mnt</replaceable></filename> as
	    the temporary mount point, it is:</para>

	  <screen>&prompt.root; <userinput>newfs /dev/<replaceable>ad1s1a</replaceable></userinput>
&prompt.root; <userinput>mount /dev/<replaceable>ad1s1a</replaceable> <replaceable>/mnt</replaceable></userinput>
&prompt.root; <userinput>cd <replaceable>/mnt</replaceable></userinput>
&prompt.root; <userinput>dump 0af - / | restore rf -</userinput></screen>

	  <para>Rearranging your partitions with <command>dump</command>
	    takes a bit more work.  To merge a partition like
	    <filename class="directory">/var</filename> into its parent, create the new
	    partition large enough for both, move the parent partition
	    as described above, then move the child partition into the
	    empty directory that the first move created:</para>

	  <screen>&prompt.root; <userinput>newfs /dev/<replaceable>ad1s1a</replaceable></userinput>
&prompt.root; <userinput>mount /dev/<replaceable>ad1s1a</replaceable> <replaceable>/mnt</replaceable></userinput>
&prompt.root; <userinput>cd <replaceable>/mnt</replaceable></userinput>
&prompt.root; <userinput>dump 0af - / | restore rf -</userinput>
&prompt.root; <userinput>cd var</userinput>
&prompt.root; <userinput>dump 0af - /var | restore rf -</userinput></screen>

	  <para>To split a directory from its parent, say putting
	    <filename class="directory">/var</filename> on its own partition when it was
	    not before, create both partitions, then mount the child
	    partition on the appropriate directory in the temporary
	    mount point, then move the old single partition:</para>

	  <screen>&prompt.root; <userinput>newfs /dev/<replaceable>ad1s1a</replaceable></userinput>
&prompt.root; <userinput>newfs /dev/<replaceable>ad1s1d</replaceable></userinput>
&prompt.root; <userinput>mount /dev/<replaceable>ad1s1a</replaceable> <replaceable>/mnt</replaceable></userinput>
&prompt.root; <userinput>mkdir <replaceable>/mnt</replaceable>/var</userinput>
&prompt.root; <userinput>mount /dev/<replaceable>ad1s1d</replaceable> <replaceable>/mnt</replaceable>/var</userinput>
&prompt.root; <userinput>cd <replaceable>/mnt</replaceable></userinput>
&prompt.root; <userinput>dump 0af - / | restore rf -</userinput></screen>

	  <para>You might prefer &man.cpio.1;, &man.pax.1;, &man.tar.1;
	    to &man.dump.8; for user data.  At the time of this writing,
	    these are known to lose file flag information, so use them
	    with caution.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="dangerously-dedicated">
	  <para>Will a <quote>dangerously dedicated</quote> disk
	    endanger my health?</para>
	</question>

	<answer>
	  <para><anchor id="dedicate"/>The installation procedure allows
	    you to chose two different methods in partitioning your hard
	    disk(s).  The default way makes it compatible with other
	    operating systems on the same machine, by using
	    &man.fdisk.8; table entries (called <quote>slices</quote> in
	    &os;), with a &os; slice that employs partitions of its own.
	    Optionally, one can chose to install a boot-selector to
	    switch between the possible operating systems on the
	    disk(s).  The alternative uses the entire disk for &os;, and
	    makes no attempt to be compatible with other operating
	    systems.</para>

	  <para>So why it is called <quote>dangerous</quote>?  A disk in
	    this mode does not contain what normal PC utilities would
	    consider a valid &man.fdisk.8; table.  Depending on how well
	    they have been designed, they might complain at you once
	    they are getting in contact with such a disk, or even worse,
	    they might damage the BSD bootstrap without even asking or
	    notifying you.  In addition, the <quote>dangerously
	    dedicated</quote> disk's layout is known to confuse many
	    BIOSes, including those from AWARD (e.g. as found in HP
	    Netserver and Micronics systems as well as many others) and
	    Symbios/NCR (for the popular 53C8xx range of SCSI
	    controllers).  This is not a complete list, there are more.
	    Symptoms of this confusion include the <errorname>read
	    error</errorname> message printed by the &os; bootstrap when
	    it cannot find itself, as well as system lockups when
	    booting.</para>

	  <para>Why have this mode at all then?  It only saves a few
	    kbytes of disk space, and it can cause real problems for a new
	    installation.  <quote>Dangerously dedicated</quote> mode's
	    origins lie in a desire to avoid one of the most common
	    problems plaguing new &os; installers &mdash; matching the
	    BIOS <quote>geometry</quote> numbers for a disk to the disk
	    itself.</para>

	  <para><quote>Geometry</quote> is an outdated concept, but one
	    still at the heart of the PC's BIOS and its interaction with
	    disks.  When the &os; installer creates slices, it has to
	    record the location of these slices on the disk in a fashion
	    that corresponds with the way the BIOS expects to find them.
	    If it gets it wrong, you will not be able to boot.</para>

	  <para><quote>Dangerously dedicated</quote> mode tries to work
	    around this by making the problem simpler.  In some cases,
	    it gets it right.  But it is meant to be used as a
	    last-ditch alternative &mdash; there are better ways to
	    solve the problem 99 times out of 100.</para>

	  <para>So, how do you avoid the need for <quote>DD</quote> mode
	    when you are installing?  Start by making a note of the
	    geometry that your BIOS claims to be using for your disks.
	    You can arrange to have the kernel print this as it boots by
	    specifying <option>-v</option> at the
	    <literal>boot:</literal> prompt, or using
	    <command>boot -v</command> in the loader.  Just before the
	    installer starts, the kernel will print a list of BIOS
	    geometries.  Do not panic &mdash; wait for the installer to
	    start and then use scrollback to read the numbers.
	    Typically the BIOS disk units will be in the same order that
	    &os; lists your disks, first IDE, then SCSI.</para>

	  <para>When you are slicing up your disk, check that the disk
	    geometry displayed in the FDISK screen is correct (ie.  it
	    matches the BIOS numbers); if it is wrong, use the
	    <keycap>G</keycap> key to fix it.  You may have to do this
	    if there is absolutely nothing on the disk, or if the disk
	    has been moved from another system.  Note that this is only
	    an issue with the disk that you are going to boot from; &os;
	    will sort itself out just fine with any other disks you may
	    have.</para>

	  <para>Once you have got the BIOS and &os; agreeing about the
	    geometry of the disk, your problems are almost guaranteed to
	    be over, and with no need for <quote>DD</quote> mode at all.
	    If, however, you are still greeted with the dreaded
	    <errorname>read error</errorname> message when you try to
	    boot, it is time to cross your fingers and go for it &mdash; there
	    is nothing left to lose.</para>

	  <para>To return a <quote>dangerously dedicated</quote> disk
	    for normal PC use, there are basically two options.  The
	    first is, you write enough NULL bytes over the MBR to make
	    any subsequent installation believe this to be a blank disk.
	    You can do this for example with the following
	    command:</para>

	  <screen>&prompt.root; <userinput>dd if=/dev/zero of=/dev/<replaceable>rda0</replaceable> count=15</userinput></screen>

	  <para>Alternatively, the undocumented DOS
	    <quote>feature</quote></para>

	  <screen><prompt>C:\&gt;</prompt> <userinput>fdisk /mbr</userinput></screen>

	  <para>will to install a new master boot record as well, thus
	    clobbering the BSD bootstrap.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="safe-softupdates">
	  <para>Which partitions can safely use Soft Updates?  I have
	    heard that Soft Updates on <filename class="directory">/</filename> can cause
	    problems.</para>
	</question>

	<answer>
	  <para>Short answer: you can usually use Soft Updates safely on
	    all partitions.</para>

	  <para>Long answer: There used to be some concern over using
	    Soft Updates on the root partition.  Soft Updates has two
	    characteristics that caused this.  First, a Soft Updates
	    partition has a small chance of losing data during a system
	    crash.  (The partition will not be corrupted; the data will
	    simply be lost.) Also, Soft Updates can cause temporary
	    space shortages.</para>

	  <para>When using Soft Updates, the kernel can take up to
	    thirty seconds to actually write changes to the physical
	    disk.  If you delete a large file, the file still resides on
	    disk until the kernel actually performs the deletion.  This
	    can cause a very simple race condition.  Suppose you delete
	    one large file and immediately create another large file.
	    The first large file is not yet actually removed from the
	    physical disk, so the disk might not have enough room for
	    the second large file.  You get an error that the partition
	    does not have enough space, although you know perfectly well
	    that you just released a large chunk of space!  When you try
	    again mere seconds later, the file creation works as you
	    expect.  This has left more than one user scratching his
	    head and doubting his sanity, the &os; file system, or
	    both.</para>

	  <para>If a system should crash after the kernel accepts a
	    chunk of data for writing to disk, but before that data is
	    actually written out, data could be lost or corrupted.  This
	    risk is extremely small, but generally manageable.  Use of
	    IDE write caching greatly increases this risk; it is
	    strongly recommended that you disable IDE write caching when
	    using Soft Updates.</para>

	  <para>These issues affect all partitions using Soft Updates.
	    So, what does this mean for the root partition?</para>

	  <para>Vital information on the root partition changes very
	    rarely.  Files such as
	    <filename>/boot/kernel/kernel</filename> and the contents of
	    <filename class="directory">/etc</filename> only change during system
	    maintenance, or when users change their passwords.  If the
	    system crashed during the thirty-second window after such a
	    change is made, it is possible that data could be lost.
	    This risk is negligible for most applications, but you
	    should be aware that it exists.  If your system cannot
	    tolerate this much risk, do not use Soft Updates on the root
	    file system!</para>

	  <para><filename class="directory">/</filename> is traditionally one of the
	    smallest partitions.  If you put the
	    <filename class="directory">/tmp</filename> directory on
	    <filename class="directory">/</filename> and you have a busy
	    <filename class="directory">/tmp</filename>, you might see intermittent space
	    problems.  Symlinking <filename class="directory">/tmp</filename> to
	    <filename class="directory">/var/tmp</filename> will solve this
	    problem.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="inappropriate-ccd">
	  <para>What is inappropriate about my &man.ccd.4;?</para>
	</question>

	<answer>
	  <para>The symptom of this is:</para>

	  <screen>&prompt.root; <userinput>ccdconfig -C</userinput>
ccdconfig: ioctl (CCDIOCSET): /dev/<replaceable>ccd0c</replaceable>: Inappropriate file type or format</screen>

	  <para>This usually happens when you are trying to concatenate
	    the <literal>c</literal> partitions, which default to type
	    <literal>unused</literal>.  The &man.ccd.4; driver requires
	    the underlying partition type to be
	    <literal>FS_BSDFFS</literal>.  Edit the disk label of the
	    disks you are trying to concatenate and change the types of
	    partitions to <literal>4.2BSD</literal>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ccd-disk-label">
	  <para>Why can I not edit the disk label on my
	    &man.ccd.4;?</para>
	</question>

	<answer>
	  <para>The symptom of this is:</para>

	  <screen>&prompt.root; <userinput>disklabel <replaceable>ccd0</replaceable></userinput>
(it prints something sensible here, so let us try to edit it)
&prompt.root; <userinput>disklabel -e <replaceable>ccd0</replaceable></userinput>
(edit, save, quit)
disklabel: ioctl DIOCWDINFO: No disk label on disk;
use "disklabel -r" to install initial label</screen>

	  <para>This is because the disk label returned by &man.ccd.4;
	    is actually a <quote>fake</quote> one that is not really on
	    the disk.  You can solve this problem by writing it back
	    explicitly, as in:</para>

	  <screen>&prompt.root; <userinput>disklabel <replaceable>ccd0</replaceable> &gt; <replaceable>/tmp/disklabel.tmp</replaceable></userinput>
&prompt.root; <userinput>disklabel -Rr <replaceable>ccd0</replaceable> <replaceable>/tmp/disklabel.tmp</replaceable></userinput>
&prompt.root; <userinput>disklabel -e <replaceable>ccd0</replaceable></userinput>
(this will work now)</screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="mount-foreign-fs">
	  <para>Can I mount other foreign file systems under
	    &os;?</para>
	</question>

	<answer>
	  <para>&os; supports a variety of other file systems.</para>

	  <variablelist>
	    <varlistentry>
	      <term>UFS</term>

	      <listitem>
		<para>UFS CD-ROMs can be mounted directly on &os;.
		  Mounting disk partitions from Digital UNIX and other
		  systems that support UFS may be more complex,
		  depending on the details of the disk partitioning for
		  the operating system in question.</para>
	      </listitem>
	    </varlistentry>

	    <varlistentry>
	      <term>ext2/ext3</term>

	      <listitem>
		<para>&os; supports <literal>ext2fs</literal> and
		  <literal>ext3fs</literal> partitions.  See
		  &man.mount.ext2fs.8; for more information.</para>
	      </listitem>
	    </varlistentry>

	    <varlistentry>
	      <term>NTFS</term>

	      <listitem>
		<para>&os; includes a read-only NTFS driver.  For more
		  information, see &man.mount.ntfs.8;.  A port of <ulink
		    url="http://www.tuxera.com/community/"><application>ntfs-3g</application></ulink>
		  supports write operations on NTFS (see <filename
		    role="package">sysutils/fusefs-ntfs</filename>).</para>
	      </listitem>
	    </varlistentry>

	    <varlistentry>
	      <term>FAT</term>

	      <listitem>
		<para>&os; includes a read-write FAT driver.  For more
		  information, see &man.mount.msdosfs.8;.</para>
	      </listitem>
	    </varlistentry>

	    <varlistentry>
	      <term>ZFS</term>

	      <listitem>
		<para>&os; includes a port of
		  &sun;'s ZFS driver.  The current recommendation is to
		  use it only on &arch.amd64; platforms with sufficient
		  memory.  For more information, see &man.zfs.8;.</para>
	      </listitem>
	    </varlistentry>
	  </variablelist>

	  <para>&os; also supports network file systems such as NFS (see
	    &man.mount.nfs.8;), NetWare (see &man.mount.nwfs.8;), and
	    Microsoft-style SMB file systems (see &man.mount.smbfs.8;).
	    You can find ports based on FUSE (<filename
	      role="package">sysutils/fusefs-kmod</filename>) for many
	    other file systems.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="mount-dos">
	  <para>How do I mount a secondary DOS partition?</para>
	</question>

	<answer>
	  <para>The secondary DOS partitions are found after
	    <emphasis>all</emphasis> the primary partitions.  For
	    example, if you have an <quote>E</quote> partition as the
	    second DOS partition on the second SCSI drive, there will be
	    a device file for <quote>slice 5</quote> in
	    <filename class="directory">/dev</filename>, so simply mount it:</para>

	  <screen>&prompt.root; <userinput>mount -t msdosfs /dev/da1s5 /dos/e</userinput></screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="crypto-file-system">
	  <para>Is there a cryptographic file system for &os;?</para>
	</question>

	<answer>
	  <para>Yes.  You can use either &man.gbde.8; or &man.geli.8;,
	    see the <ulink
	      url="&url.books.handbook;/disks-encrypting.html">Encrypting Disk Partitions</ulink>
	    section of the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="nt-bootloader">
	  <para>How can I use the &windowsnt; loader to boot
	    &os;?</para>
	</question>

	<answer>
	  <para>The general idea is that you copy the first sector of
	    your native root &os; partition into a file in the
	    DOS/&windowsnt; partition.  Assuming you name that file
	    something like <filename>c:\bootsect.bsd</filename>
	    (inspired by <filename>c:\bootsect.dos</filename>), you can
	    then edit the <filename>c:\boot.ini</filename> file to come
	    up with something like this:</para>

	  <programlisting>[boot loader]
timeout=30
default=multi(0)disk(0)rdisk(0)partition(1)\WINDOWS
[operating systems]
multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Windows NT"
C:\BOOTSECT.BSD="&os;"
C:\="DOS"</programlisting>

	  <para>If &os; is installed on the same disk as the &windowsnt;
	    boot partition simply copy <filename>/boot/boot1</filename> to
	    <filename>C:\BOOTSECT.BSD</filename>.  However, if &os; is
	    installed on a different disk
	    <filename>/boot/boot1</filename> will not work,
	    <filename>/boot/boot0</filename> is needed.</para>

	  <para><filename>/boot/boot0</filename> needs to be installed
	    using &man.sysinstall.8; by selecting the &os; boot manager
	    on the screen which asks if you wish to use a boot manager.
	    This is because <filename>/boot/boot0</filename> has the
	    partition table area filled with NULL characters but
	    &man.sysinstall.8; copies the partition table before copying
	    <filename>/boot/boot0</filename> to the MBR.</para>

	    <warning>
	      <para><emphasis>Do not simply copy
		<filename>/boot/boot0</filename> instead of
		<filename>/boot/boot1</filename>; you will overwrite
		your partition table and render your computer
		un-bootable!</emphasis></para>
	    </warning>

	  <para>When the &os; boot manager runs it records the last OS
	    booted by setting the active flag on the partition table
	    entry for that OS and then writes the whole 512-bytes of
	    itself back to the MBR so if you just copy
	    <filename>/boot/boot0</filename> to
	    <filename>C:\BOOTSECT.BSD</filename> then it writes an empty
	    partition table, with the active flag set on one entry, to
	    the MBR.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="lilo-bootloader">
	  <para>How do I boot &os; and &linux; from LILO?</para>
	</question>

	<answer>
	  <para>If you have &os; and &linux; on the same disk, just
	    follow LILO's installation instructions for booting a
	    non-&linux; operating system.  Very briefly, these
	    are:</para>

	  <para>Boot &linux;, and add the following lines to
	    <filename>/etc/lilo.conf</filename>:</para>

	  <programlisting>other=/dev/hda2
        table=/dev/hda
        label=&os;</programlisting>

	  <para>(the above assumes that your &os; slice is known to
	    &linux; as <devicename>/dev/hda2</devicename>; tailor to
	    suit your setup).  Then, run <command>lilo</command> as
	    <username>root</username> and you should be done.</para>

	  <para>If &os; resides on another disk, you need to add
	    <literal>loader=/boot/chain.b</literal> to the LILO entry.
	    For example:</para>

	  <programlisting>other=/dev/dab4
        table=/dev/dab
        loader=/boot/chain.b
        label=&os;</programlisting>

	  <para>In some cases you may need to specify the BIOS drive
	    number to the &os; boot loader to successfully boot off the
	    second disk.  For example, if your &os; SCSI disk is probed
	    by BIOS as BIOS disk 1, at the &os; boot loader prompt you
	    need to specify:</para>

	  <screen>Boot: <userinput>1:da(0,a)/boot/kernel/kernel</userinput></screen>

	  <para>You can configure &man.boot.8; to automatically do this
	    for you at boot time.</para>

	  <para>The <ulink
	      url="http://tldp.org/HOWTO/Linux+FreeBSD.html">&linux;+&os; mini-HOWTO</ulink>
	    is a good reference for &os; and &linux; interoperability
	    issues.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="grub-loader">
	  <para>How do I boot &os; and &linux; using GRUB?</para>
	</question>

	<answer>
	  <para>Booting &os; using GRUB is very simple.  Just add the
	    following to your configuration file
	    <filename>/boot/grub/menu.lst</filename> (or
	    <filename>/boot/grub/grub.conf</filename> in some systems,
	    e.g. Red Hat Linux and its derivatives).</para>

	  <programlisting>title &os; 6.1
	root <replaceable>(hd0,a)</replaceable>
	kernel /boot/loader
	  </programlisting>

	  <para>Where <replaceable>hd0,a</replaceable> points to your
	    root partition on the first disk.  If you need to specify
	    which slice number should be used, use something like this
	    <replaceable>(hd0,2,a)</replaceable>.  By default, if the
	    slice number is omitted, GRUB searches the first slice which
	    has <literal>a</literal> partition.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="booteasy-loader">
	  <para>How do I boot &os; and &linux; using
	    <application>BootEasy?</application></para>
	</question>

	<answer>
	  <para>Install LILO at the start of your &linux; boot partition
	    instead of in the Master Boot Record.  You can then boot
	    LILO from <application>BootEasy</application>.</para>

	  <para>If you are running &windows; and &linux; this is
	    recommended anyway, to make it simpler to get &linux; booting
	    again if you should need to reinstall &windows; (which is a
	    Jealous Operating System, and will bear no other Operating
	    Systems in the Master Boot Record).</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="changing-bootprompt">
	  <para>How do I change the boot prompt from
	    <literal>???</literal> to something more meaningful?</para>
	</question>

	<answer>
	  <para>You can not do that with the standard boot manager
	    without rewriting it.  There are a number of other boot
	    managers in the <filename class="directory">sysutils</filename> ports category
	    that provide this functionality.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="removable-drives">
	  <para>I have a new removable drive, how do I use it?</para>
	</question>

	<answer>
	  <para>Whether it is a removable drive like a &iomegazip; or an
	    EZ drive (or even a floppy, if you want to use it that way),
	    or a new hard disk, once it is installed and recognized by
	    the system, and you have your cartridge/floppy/whatever
	    slotted in, things are pretty much the same for all
	    devices.</para>

	  <para>(this section is based on <ulink
	      url="http://www.vmunix.com/mark/FreeBSD/ZIP-FAQ.html">Mark Mayo's ZIP FAQ</ulink>)
	    </para>

	  <para>If it is a ZIP drive or a floppy, you have already got a
	    DOS file system on it, you can use a command like this:</para>

	  <screen>&prompt.root; <userinput>mount -t msdosfs /dev/fd0c /floppy</userinput></screen>

	  <para>if it is a floppy, or this:</para>

	  <screen>&prompt.root; <userinput>mount -t msdosfs /dev/da2s4 /zip</userinput></screen>

	  <para>for a ZIP disk with the factory configuration.</para>

	  <para>For other disks, see how they are laid out using
	    &man.fdisk.8; or &man.sysinstall.8;.</para>

	  <para>The rest of the examples will be for a ZIP drive on
	    <devicename>da2</devicename>, the third SCSI disk.</para>

	  <para>Unless it is a floppy, or a removable you plan on
	    sharing with other people, it is probably a better idea to
	    stick a BSD file system on it.  You will get long filename
	    support, at least a 2X improvement in performance, and a lot
	    more stability.  First, you need to redo the DOS-level
	    partitions/file systems.  You can either use &man.fdisk.8;
	    or &man.sysinstall.8;, or for a small drive that you do not
	    want to bother with multiple operating system support on,
	    just blow away the whole FAT partition table (slices) and
	    just use the BSD partitioning:</para>

	  <screen>&prompt.root; <userinput>dd if=/dev/zero of=/dev/rda2 count=2</userinput>
&prompt.root; <userinput>disklabel -Brw da2 auto</userinput></screen>

	  <para>You can use &man.disklabel.8; or &man.sysinstall.8; to
	    create multiple BSD partitions.  You will certainly want to
	    do this if you are adding swap space on a fixed disk, but it
	    is probably irrelevant on a removable drive like a
	    ZIP.</para>

	  <para>Finally, create a new file system, this one is on our
	    ZIP drive using the whole disk:</para>

	  <screen>&prompt.root; <userinput>newfs /dev/rda2c</userinput></screen>

	  <para>and mount it:</para>

	  <screen>&prompt.root; <userinput>mount /dev/da2c /zip</userinput></screen>

	  <para>and it is probably a good idea to add a line like this
	    to <filename>/etc/fstab</filename> (see &man.fstab.5;) so
	    you can just type <command>mount /zip</command> in the
	    future:</para>

	  <programlisting>/dev/da2c /zip ffs rw,noauto 0 0</programlisting>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="mount-cd-superblock">
	  <para>Why do I get <errorname>Incorrect super
	    block</errorname> when mounting a CD-ROM?</para>
	</question>

	<answer>
	  <para>You have to tell &man.mount.8; the type of the device
	    that you want to mount.  This is described in the <ulink
	      url="&url.books.handbook;/creating-cds.html"> Handbook section on optical media</ulink>,
	    specifically the section <ulink
	      url="&url.books.handbook;/creating-cds.html#MOUNTING-CD">Using Data CDs</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="cdrom-not-configured">
	  <para>Why do I get <errorname>Device not
	    configured</errorname> when mounting a CD-ROM?</para>
	</question>

	<answer>
	  <para>This generally means that there is no CD-ROM in the
	    CD-ROM drive, or the drive is not visible on the bus.
	    Please see the <ulink
	      url="&url.books.handbook;/creating-cds.html#MOUNTING-CD">Using Data CDs</ulink>
	    section of the Handbook for a detailed discussion of this
	    issue.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="cdrom-unicode-filenames">
	  <para>Why do all non-English characters in filenames show up
	    as <quote>?</quote> on my CDs when mounted in &os;?</para>
	</question>

	<answer>
	  <para>Your CD-ROM probably uses the <quote>Joliet</quote>
	    extension for storing information about files and
	    directories.  This is discussed in the Handbook chapter on
	    <ulink
	      url="&url.books.handbook;/creating-cds.html">creating and using CD-ROMs</ulink>,
	    specifically the section on <ulink
	      url="&url.books.handbook;/creating-cds.html#MOUNTING-CD">Using Data CD-ROMs</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="burncd-isofs">
	  <para>I burned a CD under &os; and now I can not read it under
	    any other operating system.  Why?</para>
	</question>

	<answer>
	  <para>You most likely burned a raw file to your CD, rather
	    than creating an ISO&nbsp;9660 file system.  Take a look at
	    the <ulink
	      url="&url.books.handbook;/creating-cds.html">Handbook chapter on creating CD-ROMs</ulink>,
	    particularly the section on <ulink
	      url="&url.books.handbook;/creating-cds.html#RAWDATA-CD">burning raw data CDs</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="copy-cd">
	  <para>How can I create an image of a data CD?</para>
	</question>

	<answer>
	  <para>This is discussed in the Handbook section on <ulink
	      url="&url.books.handbook;/creating-cds.html#IMAGING-CD">duplicating data CDs</ulink>.
	    For more on working with CD-ROMs, see the <ulink
	      url="&url.books.handbook;/creating-cds.html">Creating CDs Section</ulink>
	    in the Storage chapter in the Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="mount-audio-CD">
	  <para>Why can I not <command>mount</command> an audio
	  CD?</para>
	</question>

	<answer>
	  <para>If you try to mount an audio CD, you will get an error
	    like <errorname>cd9660: /dev/acd0c: Invalid
	    argument</errorname>.  This is because
	    <command>mount</command> only works on file systems.  Audio
	    CDs do not have file systems; they just have data.  You need
	    a program that reads audio CDs, such as the <filename
	      role="package">audio/xmcd</filename> port.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="multi-session-CD">
	  <para>How do I <command>mount</command> a multi-session
	    CD?</para>
	</question>

	<answer>
	  <para>By default, &man.mount.8; will attempt to mount the last
	    data track (session) of a CD.  If you would like to load an
	    earlier session, you must use the <option>-s</option>
	    command line argument.  Please see &man.mount.cd9660.8; for
	    specific examples.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="user-floppymount">
	  <para>How do I let ordinary users mount floppies, CD-ROMs and
	    other removable media?</para>
	</question>

	<answer>
	  <para>Ordinary users can be permitted to mount devices.  Here
	    is how:</para>

	  <procedure>
	    <step>
	      <para>As <username>root</username> set the sysctl variable
		<varname>vfs.usermount</varname> to
		<literal>1</literal>.</para>

	      <screen>&prompt.root; <userinput>sysctl -w vfs.usermount=1</userinput></screen>
	    </step>

	    <step>
	      <para>As <username>root</username> assign the appropriate
		permissions to the block device associated with the
		removable media.</para>

	      <para>For example, to allow users to mount the first
		floppy drive, use:</para>

	      <screen>&prompt.root; <userinput>chmod 666 /dev/fd0</userinput></screen>

	      <para>To allow users in the group
		<groupname>operator</groupname> to mount the CD-ROM
		drive, use:</para>

	      <screen>&prompt.root; <userinput>chgrp operator /dev/acd0c</userinput>
&prompt.root; <userinput>chmod 640 /dev/acd0c</userinput></screen>
	    </step>

	    <step>
	      <para>You will need to alter
		<filename>/etc/devfs.conf</filename> to make these
		changes permanent across reboots.</para>

	      <para>As <username>root</username>, add the necessary
		lines to <filename>/etc/devfs.conf</filename>.  For
		example, to allow users to mount the first floppy drive
		add:</para>

	      <programlisting># Allow all users to mount the floppy disk.
own       /dev/fd0	  root:operator
perm      /dev/fd0	  0666</programlisting>

	      <para>To allow users in the group
		<groupname>operator</groupname> to mount the CD-ROM drive
		add:</para>

	      <programlisting># Allow members of the group operator to mount CD-ROMs.
own       /dev/acd0	  root:operator
perm      /dev/acd0	  0660</programlisting>
	    </step>

	    <step>
	      <para>Finally, add the line
		<literal><varname>vfs.usermount</varname>=1</literal> to
		the file <filename>/etc/sysctl.conf</filename> so that
		it is reset at system boot time.</para>
	    </step>
	  </procedure>

	  <para>All users can now mount the floppy
	    <devicename>/dev/fd0</devicename> onto a directory that they
	    own:</para>

	  <screen>&prompt.user; <userinput>mkdir <replaceable>~/my-mount-point</replaceable></userinput>
&prompt.user; <userinput>mount -t msdosfs /dev/fd0 <replaceable>~/my-mount-point</replaceable></userinput></screen>

	  <para>Users in group <groupname>operator</groupname> can now
	    mount the CD-ROM <devicename>/dev/acd0c</devicename> onto a
	    directory that they own:</para>

	  <screen>&prompt.user; <userinput>mkdir <replaceable>~/my-mount-point</replaceable></userinput>
&prompt.user; <userinput>mount -t cd9660 /dev/acd0c <replaceable>~/my-mount-point</replaceable></userinput></screen>

	  <para>Unmounting the device is simple:</para>

	  <screen>&prompt.user; <userinput>umount <replaceable>~/my-mount-point</replaceable></userinput></screen>

	  <para>Enabling <varname>vfs.usermount</varname>, however, has
	    negative security implications.  A better way to access
	    &ms-dos; formatted media is to use the <filename
	      role="package">emulators/mtools</filename> package in the
	    Ports Collection.</para>

	  <note>
	    <para>The device name used in the previous examples must be
	      changed according to your configuration.</para>
	  </note>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="du-vs-df">
	  <para>The <command>du</command> and <command>df</command>
	    commands show different amounts of disk space available.
	    What is going on?</para>
	</question>

	<answer>
	  <para>You need to understand what <command>du</command> and
	    <command>df</command> really do.  <command>du</command> goes
	    through the directory tree, measures how large each file is,
	    and presents the totals.  <command>df</command> just asks
	    the file system how much space it has left.  They seem to be
	    the same thing, but a file without a directory entry will
	    affect <command>df</command> but not
	    <command>du</command>.</para>

	  <para>When a program is using a file, and you delete the file,
	    the file is not really removed from the file system until
	    the program stops using it.  The file is immediately deleted
	    from the directory listing, however.  You can see this
	    easily enough with a program such as
	    <command>more</command>.  Assume you have a file large
	    enough that its presence affects the output of
	    <command>du</command> and <command>df</command>.  (Since
	    disks can be so large today, this might be a
	    <emphasis>very</emphasis> large file!) If you delete this
	    file while using <command>more</command> on it,
	    <command>more</command> does not immediately choke and
	    complain that it cannot view the file.  The entry is simply
	    removed from the directory so no other program or user can
	    access it.  <command>du</command> shows that it is gone
	    &mdash; it has walked the directory tree and the file is not
	    listed.  <command>df</command> shows that it is still there,
	    as the file system knows that <command>more</command> is
	    still using that space.  Once you end the
	    <command>more</command> session, <command>du</command> and
	    <command>df</command> will agree.</para>

	  <para>Note that Soft Updates can delay the freeing of disk
	    space; you might need to wait up to 30 seconds for the
	    change to be visible!</para>

	  <para>This situation is common on web servers.  Many people
	    set up a &os; web server and forget to rotate the log files.
	    The access log fills up <filename class="directory">/var</filename>.  The new
	    administrator deletes the file, but the system still
	    complains that the partition is full.  Stopping and
	    restarting the web server program would free the file,
	    allowing the system to release the disk space.  To prevent
	    this from happening, set up &man.newsyslog.8;.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="add-swap-space">
	  <para>How can I add more swap space?</para>
	</question>

	<answer>
	  <para>In the <ulink
	      url="&url.books.handbook;/config-tuning.html">Configuration and Tuning</ulink>
	    section of the Handbook, you will find a <ulink
	      url="&url.books.handbook;/adding-swap-space.html">section</ulink>
	    describing how to do this.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="manufacturer-disk-size">
	  <para>Why does &os; see my disk as smaller than the
	    manufacturer says it is?</para>
	</question>

	<answer>
	  <para>Disk manufacturers calculate gigabytes as a billion
	    bytes each, whereas &os; calculates them as
	    1,073,741,824&nbsp;bytes each.  This explains why, for
	    example, &os;'s boot messages will report a disk that
	    supposedly has 80&nbsp;GB as holding 76,319&nbsp;MB.</para>

	  <para>Also note that &os; will (by default) <link
	      linkend="disk-more-than-full">reserve</link> 8% of the disk
	    space.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="disk-more-than-full">
	  <para>How is it possible for a partition to be more than 100%
	    full?</para>
	</question>

	<answer>
	  <para>A portion of each UFS partition (8%, by default) is
	    reserved for use by the operating system and the
	    <username>root</username> user.  &man.df.1; does not count
	    that space when calculating the <literal>Capacity</literal>
	    column, so it can exceed 100%.  Also, you will notice that
	    the <literal>Blocks</literal> column is always greater than
	    the sum of the <literal>Used</literal> and
	    <literal>Avail</literal> columns, usually by a factor of
	    8%.</para>

	  <para>For more details, look up the <option>-m</option> option
	    in &man.tunefs.8;.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="admin">
    <title>System Administration</title>

    <qandaset>
      <qandaentry>
	<question id="startup-config-files">
	  <para>Where are the system start-up configuration
	    files?</para>
	</question>

	<answer>
	  <para>The primary configuration file is
	    <filename>/etc/defaults/rc.conf</filename> (see
	    &man.rc.conf.5;).  System startup scripts such as
	    <filename class="directory">/etc/rc</filename> and
	    <filename class="directory">/etc/rc.d</filename> (see &man.rc.8;) just include
	    this file.  <emphasis>Do not edit this file!</emphasis>
	    Instead, if there is any entry in
	    <filename>/etc/defaults/rc.conf</filename> that you want to
	    change, you should copy the line into
	    <filename>/etc/rc.conf</filename> and change it
	    there.</para>

	  <para>For example, if you wish to start &man.named.8;, the
	    included DNS server, all you need to do is:</para>

	  <screen>&prompt.root; <userinput>echo 'named_enable="YES"' &gt;&gt; /etc/rc.conf</userinput></screen>

	  <para>To start up local services, place shell scripts in the
	    <filename class="directory">/usr/local/etc/rc.d</filename> directory.  These
	    shell scripts should be set executable, the default file
	    mode is <literal>555</literal>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="adding-users">
	  <para>How do I add a user easily?</para>
	</question>

	<answer>
	  <para>Use the &man.adduser.8; command, or the &man.pw.8;
	    command for more complicated situations.</para>

	  <para>To remove the user, use the &man.rmuser.8; command or,
	    if necessary, &man.pw.8;.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="root-not-found-cron-errors">
	  <para>Why do I keep getting messages like <errorname>root: not
	    found</errorname> after editing my
	    <filename>crontab</filename> file?</para>
	</question>

	<answer>
	  <para>This is normally caused by editing the system crontab
	    (<filename>/etc/crontab</filename>) and then using
	    &man.crontab.1; to install it:</para>

	  <screen>&prompt.root; <userinput>crontab /etc/crontab</userinput></screen>

	  <para>This is not the correct way to do things.  The system
	    crontab has a different format to the per-user crontabs
	    which &man.crontab.1; updates (the &man.crontab.5; manual
	    page explains the differences in more detail).</para>

	  <para>If this is what you did, the extra crontab is simply a
	    copy of <filename>/etc/crontab</filename> in the wrong
	    format it.  Delete it with the command:</para>

	  <screen>&prompt.root; <userinput>crontab -r</userinput></screen>

	  <para>Next time, when you edit
	    <filename>/etc/crontab</filename>, you should not do
	    anything to inform &man.cron.8; of the changes, since it
	    will notice them automatically.</para>

	  <para>If you want something to be run once per day, week, or
	    month, it is probably better to add shell scripts
	    <filename class="directory">/usr/local/etc/periodic</filename>, and let the
	    &man.periodic.8; command run from the system
	    <command>cron</command> schedule it with the other periodic
	    system tasks.</para>

	  <para>The actual reason for the error is that the system
	    crontab has an extra field, specifying which user to run the
	    command as.  In the default system crontab provided with
	    &os;, this is <username>root</username> for all entries.
	    When this crontab is used as the <username>root</username>
	    user's crontab (which is <emphasis>not</emphasis> the same
	    as the system crontab), &man.cron.8; assumes the string
	    <literal>root</literal> is the first word of the command to
	    execute, but no such command exists.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="su-wheel-group">
	  <para>Why do I get the error, <errorname>you are not in the
	    correct group to su root</errorname> when I try to
	    <command>su</command> to <username>root</username>?</para>
	</question>

	<answer>
	  <para>This is a security feature.  In order to
	    <command>su</command> to <username>root</username> (or any
	    other account with superuser privileges), you must be in the
	    <groupname>wheel</groupname> group.  If this feature were
	    not there, anybody with an account on a system who also
	    found out <username>root</username>'s password would be able
	    to gain superuser level access to the system.  With this
	    feature, this is not strictly true; &man.su.1; will prevent
	    them from even trying to enter the password if they are not
	    in <groupname>wheel</groupname>.</para>

	  <para>To allow someone to <command>su</command> to
	    <username>root</username>, simply put them in the
	    <groupname>wheel</groupname> group.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="rcconf-readonly">
	  <para>I made a mistake in <filename>rc.conf</filename>, or
	    another startup file, and now I cannot edit it because the
	    file system is read-only.  What should I do?</para>
	</question>

	<answer>
	  <para>Restart the system using <userinput>boot -s</userinput>
	    at the loader prompt to enter Single User mode.  When
	    prompted for a shell pathname, simply press
	    <keycap>Enter</keycap>, and run
	    <command>mount -urw /</command> to re-mount the root file
	    system in read/write mode.  You may also need to run
	    <command>mount -a -t ufs</command> to mount the file system
	    where your favorite editor is defined.  If your favorite
	    editor is on a network file system, you will need to either
	    configure the network manually before you can mount network
	    file systems, or use an editor which resides on a local file
	    system, such as &man.ed.1;.</para>

	  <para>If you intend to use a full screen editor such as
	    &man.vi.1; or &man.emacs.1;, you may also need to run
	    <command>export TERM=cons25</command> so that these editors
	    can load the correct data from the &man.termcap.5;
	    database.</para>

	  <para>Once you have performed these steps, you can edit
	    <filename>/etc/rc.conf</filename> as you usually would to
	    fix the syntax error.  The error message displayed
	    immediately after the kernel boot messages should tell you
	    the number of the line in the file which is at fault.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="printer-setup">
	  <para>Why am I having trouble setting up my printer?</para>
	</question>

	<answer>
	  <para>See the <ulink
	      url="&url.books.handbook;/printing.html">Handbook entry on printing</ulink>.
	    It should cover most of your problem.</para>

	  <para>Some printers require a host-based driver to do any kind
	    of printing.  These so-called <quote>WinPrinters</quote> are
	    not natively supported by &os;.  If your printer does not
	    work in DOS or &windows;, it is probably a WinPrinter.  Your
	    only hope of getting one of these to work is to check if the
	    <filename role="package">print/pnm2ppa</filename> port
	    supports it.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="keyboard-mappings">
	  <para>How can I correct the keyboard mappings for my
	    system?</para>
	</question>

	<answer>
	  <para>Please see the Handbook section on <ulink
	      url="&url.books.handbook;/using-localization.html">using localization</ulink>,
	    specifically the section on <ulink
	      url="&url.books.handbook;/using-localization.html#SETTING-CONSOLE">console setup</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="pnp-resources">
	  <para>Why do I get messages like: <errorname>unknown:
	    &lt;PNP0303&gt; can't assign resources</errorname> on
	    boot?</para>
	</question>

	<answer>
	  <para>The following is an excerpt from a post to the
	    &a.current;.</para>

	  <blockquote>
	    <attribution>&a.wollman;, 24 April 2001</attribution>

	    <para>The <quote>can't assign resources</quote> messages
	      indicate that the devices are legacy ISA devices for which
	      a non-PnP-aware driver is compiled into the kernel.  These
	      include devices such as keyboard controllers, the
	      programmable interrupt controller chip, and several other
	      bits of standard infrastructure.  The resources cannot be
	      assigned because there is already a driver using those
	      addresses.</para>
	  </blockquote>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="user-quotas">
	  <para>Why can I not get user quotas to work properly?</para>
	</question>

<!-- XXX This may be the worst answer in the entire
     document.
-->

	<answer>
	  <orderedlist>
	    <listitem>
	      <para>It is possible that your kernel is not configured
		to use quotas.  If this is the case, you will need to
		add the following line to your kernel configuration
		file and recompile:</para>

	      <programlisting>options QUOTA</programlisting>

	      <para>Please read the <ulink
		  url="&url.books.handbook;/quotas.html">Handbook entry on quotas</ulink>
		for full details.</para>
	    </listitem>

	    <listitem>
	      <para>Do not turn on quotas on
		<filename class="directory">/</filename>.</para>
	    </listitem>

	    <listitem>
	      <para>Put the quota file on the file system that the
		quotas are to be enforced on, i.e.:</para>

	      <informaltable frame="none" pgwide="1">
		<tgroup cols="2">
		  <thead>
		    <row>
		      <entry>File System</entry>

		      <entry>Quota file</entry>
		    </row>
		  </thead>

		  <tbody>
		    <row>
		      <entry><filename class="directory">/usr</filename></entry>

		      <entry><filename>/usr/admin/quotas</filename></entry>
		    </row>

		    <row>
		      <entry><filename class="directory">/home</filename></entry>

		      <entry><filename>/home/admin/quotas</filename></entry>
		    </row>

		    <row>
		      <entry>&hellip;</entry>

		      <entry>&hellip;</entry>
		    </row>
		  </tbody>
		</tgroup>
	      </informaltable>
	    </listitem>
	  </orderedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="sysv-ipc">
	  <para>Does &os; support System V IPC primitives?</para>
	</question>

	<answer>
	  <para>Yes, &os; supports System V-style IPC, including shared
	    memory, messages and semaphores, in the
	    <filename>GENERIC</filename> kernel.  In a custom kernel,
	    enable this support by adding the following lines to your
	    kernel config.</para>

	  <programlisting>options    SYSVSHM          # enable shared memory
options    SYSVSEM          # enable for semaphores
options    SYSVMSG          # enable for messaging</programlisting>

	  <para>Recompile and install your kernel.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="sendmail-alternative">
	  <para>What other mail-server software can I use instead of
	    <application>sendmail</application>?</para>
	</question>

	<answer>
	  <para>The <ulink
	      url="http://www.sendmail.org/"><application>sendmail</application></ulink>
	    server is the default mail-server software for &os;, but you
	    can easily replace it with one of the other MTA (for
	    instance, an MTA installed from the ports).</para>

	  <para>There are various alternative MTAs in the ports tree
	    already, with <filename role="package">mail/exim</filename>,
	    <filename role="package">mail/postfix</filename>, <filename
	      role="package">mail/qmail</filename>, and <filename
	      role="package">mail/zmailer</filename> being some of the
	    most popular choices.</para>

	  <para>Diversity is nice, and the fact that you have many
	    different mail-servers to chose from is considered a good
	    thing; therefore try to avoid asking questions like
	    <quote>Is <application>sendmail</application> better than
	    <application>qmail</application>?</quote> in the mailing
	    lists.  If you do feel like asking, first check the mailing
	    list archives.  The advantages and disadvantages of each and
	    every one of the available MTAs have already been discussed
	    a few times.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="forgot-root-pw">
	  <para>I have forgotten the <username>root</username> password!
	    What do I do?</para>
	</question>

	<answer>
	  <para>Do not panic!  Restart the system, type
	    <userinput>boot -s</userinput> at the
	    <literal>Boot:</literal> prompt to enter Single User mode.
	    At the question about the shell to use, hit
	    <keycap>Enter</keycap>.  You will be dropped to a
	    &prompt.root; prompt.  Enter <command>mount -urw /</command>
	    to remount your root file system read/write, then run
	    <command>mount -a</command> to remount all the file systems.
	    Run <command>passwd root</command> to change the
	    <username>root</username> password then run &man.exit.1; to
	    continue booting.</para>

	  <note>
	    <para>If you are still prompted to give the
	      <username>root</username> password when entering the
	      Single User mode, it means that the console has been
	      marked as <literal>insecure</literal> in
	      <filename>/etc/ttys</filename>.  In this case it will be
	      required to boot from a &os; installation disk, choose
	      the <guimenuitem>Fixit</guimenuitem> shell from
	      &man.sysinstall.8; and issue the commands mentioned
	      above.</para>
	  </note>

	  <note>
	    <para>If you cannot mount your root partition from Single
	      User mode, it is possible that the partitions are
	      encrypted and it is impossible to mount them without the
	      access keys.  Your chances depend on the chosen
	      implementation.  For more information see the section
	      about encrypted disks in the &os; <ulink
		url="&url.books.handbook;/disks-encrypting.html">Handbook</ulink>.</para>
	  </note>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="CAD-reboot">
	  <para>How do I keep <keycombo
	      action="simul"><keycap>Control</keycap><keycap>Alt</keycap><keycap>Delete</keycap></keycombo>
	    from rebooting the system?</para>
	</question>

	<answer>
	  <para>If you are using &man.syscons.4; (the default console
	    driver) build and install a new kernel with the line in the
	    configuration file:</para>

	  <programlisting>options SC_DISABLE_REBOOT</programlisting>

	  <para>This can also be done by setting the following
	    &man.sysctl.8; which does not require a reboot or kernel
	    recompile:</para>

	  <screen>&prompt.root; <userinput>sysctl hw.syscons.kbd_reboot=0</userinput></screen>

	  <note>
	    <para>The above two methods are exclusive:  The &man.sysctl.8;
	      does not exist if you compile your kernel with the
	      <literal>SC_DISABLE_REBOOT</literal> option.</para>
	  </note>

	  <para>If you use the &man.pcvt.4; console driver, use the
	    following kernel configuration line instead and rebuild the
	    kernel:</para>

	  <programlisting>options PCVT_CTRL_ALT_DEL</programlisting>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="dos-to-unix-txt">
	  <para>How do I reformat DOS text files to &unix; ones?</para>
	</question>

	<answer>
	  <para>Use this &man.perl.1; command:</para>

	  <screen>&prompt.user; <userinput>perl -i.bak -npe 's/\r\n/\n/g' <replaceable>file(s)</replaceable></userinput></screen>

	  <para>where <replaceable>file(s)</replaceable> is one or more
	    files to process.  The modification is done in-place, with the
	    original file stored with a <filename>.bak</filename>
	    extension.</para>

	  <para>Alternatively you can use the &man.tr.1; command:</para>

	  <screen>&prompt.user; <userinput>tr -d '\r' &lt; <replaceable>dos-text-file</replaceable> &gt; <replaceable>unix-file</replaceable></userinput></screen>

	  <para><replaceable>dos-text-file</replaceable> is the file
	    containing DOS text while
	    <replaceable>unix-file</replaceable> will contain the
	    converted output.  This can be quite a bit faster than using
	    <command>perl</command>.</para>

	  <para>Yet another way to reformat DOS text files is to use the
	    <filename role="package">converters/dosunix</filename> port
	    from the Ports Collection.  Consult its documentation about
	    the details.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="kill-by-name">
	  <para>How do I kill processes by name?</para>
	</question>

	<answer>
	  <para>Use &man.pkill.1;.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="root-acl">
	  <para>Why is &man.su.1; bugging me about not being in
	    <username>root</username>'s ACL?</para>
	</question>

	<answer>
	  <para>The error comes from the
	    <application>Kerberos</application> distributed authentication
	    system.  The problem is not fatal but annoying.  You can
	    either run su with the <option>-K</option> option, or
	    uninstall <application>Kerberos</application> as described
	    in the next question.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="uninstall-kerberos">
	  <para>How do I uninstall
	    <application>Kerberos</application>?</para>
	</question>

	<answer>
	  <para>To remove <application>Kerberos</application> from the
	    system, reinstall the <literal>base</literal> distribution
	    for the release you are running.  If you have the CD-ROM,
	    you can mount it (we will assume on <filename
	      class="directory">/cdrom</filename>) and run the commands
	    below:</para>

	  <screen>&prompt.root; <userinput>cd /cdrom/base</userinput>
&prompt.root; <userinput>./install.sh</userinput></screen>

	  <para>Alternately, you can include the
	    <makevar>NO_KERBEROS</makevar> option in your
	    <filename>/etc/make.conf</filename> and rebuild
	    world.</para>
	</answer>
      </qandaentry>
      <qandaentry>
	<question id="add-pty">
	  <para>How do I add pseudoterminals to the system?</para>
	</question>

	<answer>
	  <para>If you have a lot of <command>telnet</command>,
	    <command>ssh</command>, X, or <command>screen</command>
	    users, you might run out of pseudoterminals.  By default,
	    &os;&nbsp;6.2 and earlier support 256 pseudoterminals, while
	    &os;&nbsp;6.3 and later support 512 pseudoterminals.</para>

	  <tip>
	    <para>If needed, more pseudoterminals can be added.
	      However, this requires patching the standard C library,
	      the kernel, and <filename>/etc/ttys</filename>.  For
	      example, <ulink
		url="http://www.freebsd.org/~jhb/patches/pty_1152.patch"></ulink>
	      expands the number of pseudoterminals to 1152.  Note that
	      the patch will only apply cleanly to &os;&nbsp;6.3 or
	      later.</para>
	  </tip>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="reread-rc">
	  <para>How do I re-read <filename>/etc/rc.conf</filename> and
	    re-start <filename>/etc/rc</filename> without a
	    reboot?</para>
	</question>

	<answer>
	  <para>Go into single user mode and then back to multi user
	    mode.</para>

	  <para>On the console do:</para>

	  <screen>&prompt.root; <userinput>shutdown now</userinput>
(Note: without -r or -h)

&prompt.root; <userinput>return</userinput>
&prompt.root; <userinput>exit</userinput></screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="release-candidate">
	  <para>I tried to update my system to the latest
	    <emphasis>-STABLE</emphasis>, but got
	    <emphasis>-BETA<replaceable>x</replaceable></emphasis>,
	    <emphasis>-RC</emphasis> or
	    <emphasis>-PRERELEASE</emphasis>!  What is going on?</para>
	</question>

	<answer>
	  <para>Short answer: it is just a name.
	    <emphasis>RC</emphasis> stands for <quote>Release
	    Candidate</quote>.  It signifies that a release is imminent.
	    In &os;, <emphasis>-PRERELEASE</emphasis> is typically
	    synonymous with the code freeze before a release.  (For some
	    releases, the <emphasis>-BETA</emphasis> label was used in
	    the same way as <emphasis>-PRERELEASE</emphasis>.)</para>

	  <para>Long answer: &os; derives its releases from one of two
	    places.  Major, dot-zero, releases, such as 7.0-RELEASE and
	    8.0-RELEASE, are branched from the head of the development
	    stream, commonly referred to as <link
	    linkend="current">-CURRENT</link>.  Minor releases, such as
	    6.3-RELEASE or 5.2-RELEASE, have been snapshots of the
	    active <link linkend="stable">-STABLE</link> branch.
	    Starting with 4.3-RELEASE, each release also now has its own
	    branch which can be tracked by people requiring an extremely
	    conservative rate of development (typically only security
	    advisories).</para>

	  <para>When a release is about to be made, the branch from
	    which it will be derived from has to undergo a certain
	    process.  Part of this process is a code freeze.  When a
	    code freeze is initiated, the name of the branch is changed
	    to reflect that it is about to become a release.  For
	    example, if the branch used to be called 6.2-STABLE, its
	    name will be changed to 6.3-PRERELEASE to signify the code
	    freeze and signify that extra pre-release testing should be
	    happening.  Bug fixes can still be committed to be part of
	    the release.  When the source code is in shape for the
	    release the name will be changed to 6.3-RC to signify that a
	    release is about to be made from it.  Once in the RC stage,
	    only the most critical bugs found can be fixed.  Once the
	    release (6.3-RELEASE in this example) and release branch
	    have been made, the branch will be renamed to
	    6.3-STABLE.</para>

	  <para>For more information on version numbers and the various
	    CVS branches, refer to the <ulink
	      url="&url.articles.releng;/article.html">Release Engineering</ulink>
	    article.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="kernel-chflag-failure">
	  <para>I tried to install a new kernel, and the &man.chflags.1;
	    failed.  How do I get around this?</para>
	</question>

	<answer>
	  <para>Short answer: You are probably at security level greater
	    than 0.  Reboot directly to Single User mode to install the
	    kernel.</para>

	  <para>Long answer: &os; disallows changing system flags at
	    security levels greater than 0.  You can check your security
	    level with the command:</para>

	  <screen>&prompt.root; <userinput>sysctl kern.securelevel</userinput></screen>

	  <para>You cannot lower the security level; you have to boot to
	    Single Mode to install the kernel, or change the security
	    level in <filename>/etc/rc.conf</filename> then reboot.  See
	    the &man.init.8; manual page for details on
	    <literal>securelevel</literal>, and see
	    <filename>/etc/defaults/rc.conf</filename> and the
	    &man.rc.conf.5; manual page for more information on
	    <filename>rc.conf</filename>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="kernel-securelevel-time">
	  <para>I cannot change the time on my system by more than one
	    second!  How do I get around this?</para>
	</question>

	<answer>
	  <para>Short answer: You are probably at security level greater
	    than 1.  Reboot directly to Single User mode to change the
	    date.</para>

	  <para>Long answer: &os; disallows changing the time by more
	    that one second at security levels greater than 1.  You can
	    check your security level with the command:</para>

	  <screen>&prompt.root; <userinput>sysctl kern.securelevel</userinput></screen>

	  <para>You cannot lower the security level; you have to boot to
	    Single User mode to change the date, or change the security
	    level in <filename>/etc/rc.conf</filename> then reboot.  See
	    the &man.init.8; manual page for details on
	    <literal>securelevel</literal>, and see
	    <filename>/etc/defaults/rc.conf</filename> and the
	    &man.rc.conf.5; manual page for more information on
	    <filename>rc.conf</filename>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="statd-mem-leak">
	  <para>Why is <command>rpc.statd</command> using 256&nbsp;MB of
	    memory?</para>
	</question>

	<answer>
	  <para>No, there is no memory leak, and it is not using
	    256&nbsp;MB of memory.  For convenience,
	    <command>rpc.statd</command> maps an obscene amount of
	    memory into its address space.  There is nothing terribly
	    wrong with this from a technical standpoint; it just throws
	    off things like &man.top.1; and &man.ps.1;.</para>

	  <para>&man.rpc.statd.8; maps its status file (resident on
	    <filename class="directory">/var</filename>) into its address space; to save
	    worrying about remapping it later when it needs to grow, it
	    maps it with a generous size.  This is very evident from the
	    source code, where one can see that the length argument to
	    &man.mmap.2; is <literal>0x10000000</literal>, or one
	    sixteenth of the address space on an IA32, or exactly
	    256&nbsp;MB.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="unsetting-schg">
	  <para>Why can I not unset the <literal>schg</literal> file
	    flag?</para>
	</question>

	<answer>
	  <para>You are running at an elevated (i.e., greater than 0)
	    securelevel.  Lower the securelevel and try again.  For more
	    information, see <link linkend="securelevel">the FAQ entry
	    on securelevel</link> and the &man.init.8; manual
	    page.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ssh-shosts">
	  <para>Why does <application>SSH</application> authentication
	    through <filename>.shosts</filename> not work by default in
	    recent versions of &os;?</para>
	</question>

	<answer>
	  <para>The reason why <filename>.shosts</filename>
	    authentication does not work by default in more recent
	    versions of &os; is because &man.ssh.1; is not installed
	    suid <username>root</username> by default.  To
	    <quote>fix</quote> this, you can do one of the
	    following:</para>

	  <itemizedlist>
	    <listitem>
	      <para>As a permanent fix, set
		<makevar>ENABLE_SUID_SSH</makevar> to
		<literal>true</literal> in
		<filename>/etc/make.conf</filename> then rebuild and
		install &man.ssh.1; (or run
		<command>make <maketarget>world</maketarget></command>).
		</para>
	    </listitem>

	    <listitem>
	      <para>As a temporary fix, change the mode on
		<filename>/usr/bin/ssh</filename> to
		<literal>4555</literal> by running
		<command>chmod 4555 /usr/bin/ssh</command> as
		<username>root</username>.  Then add
		<literal><makevar>ENABLE_SUID_SSH</makevar>= true</literal>
		to <filename>/etc/make.conf</filename> so
		the change takes effect the next time
		<command>make <maketarget>world</maketarget></command>
		is run.</para>
	    </listitem>
	  </itemizedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="vnlru">
	  <para>What is <literal>vnlru</literal>?</para>
	</question>

	<answer>
	  <para><literal>vnlru</literal> flushes and frees vnodes when
	    the system hits the <varname>kern.maxvnodes</varname> limit.
	    This kernel thread sits mostly idle, and only activates if
	    you have a huge amount of RAM and are accessing tens of
	    thousands of tiny files.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="top-memory-states">
	  <para>What do the various memory states displayed by
	    <command>top</command> mean?</para>
	</question>

<!-- Provided by John Dyson via Usenet -->
	<answer>
	  <itemizedlist>
	    <listitem>
	      <para><literal>Active</literal>: pages recently
		statistically used.</para>
	    </listitem>

	    <listitem>
	      <para><literal>Inactive</literal>: pages recently
		statistically unused.</para>
	    </listitem>

	    <listitem>
	      <para><literal>Cache</literal>: (most often) pages that
		have percolated from inactive to a status where they
		maintain their data, but can often be immediately reused
		(either with their old association, or reused with a new
		association).  There can be certain immediate transitions
		from <literal>active</literal> to
		<literal>cache</literal> state if the page is known to
		be clean (unmodified), but that transition is a matter
		of policy, depending upon the algorithm choice of the VM
		system maintainer.</para></listitem>

	    <listitem>
	      <para><literal>Free</literal>: pages without data content,
		and can be immediately used in certain circumstances
		where cache pages might be ineligible.  Free pages can
		be reused at interrupt or process
		state.</para>
	    </listitem>

	    <listitem>
	      <para><literal>Wired</literal>: pages that are fixed into
		memory, usually for kernel purposes, but also sometimes
		for special use in processes.</para>
	    </listitem>
	  </itemizedlist>

	  <para>Pages are most often written to disk (sort of a VM sync)
	    when they are in the inactive state, but active pages can
	    also be synced.  This depends upon the CPU tracking of the
	    modified bit being available, and in certain situations
	    there can be an advantage for a block of VM pages to be
	    synced, whether they are active or inactive.  In most common
	    cases, it is best to think of the inactive queue to be a
	    queue of relatively unused pages that might or might not be
	    in the process of being written to disk.  Cached pages are
	    already synced, not mapped, but available for immediate
	    process use with their old association or with a new
	    association.  Free pages are available at interrupt level,
	    but cached or free pages can be used at process state for
	    reuse.  Cache pages are not adequately locked to be
	    available at interrupt level.</para>

	  <para>There are some other flags (e.g., busy flag or busy
	    count) that might modify some of the described rules.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="free-memory-amount">
	  <para>How much free memory is available?</para>
	</question>

<!-- Provided by John Dyson via Usenet -->
	<answer>
	  <para>There are a couple of kinds of <quote>free
	    memory</quote>.  One kind is the amount of memory
	    immediately available without paging anything else out.
	    That is approximately the size of cache queue + size of free
	    queue (with a derating factor, depending upon system
	    tuning).  Another kind of <quote>free memory</quote> is the
	    total amount of <acronym>VM</acronym> space.  That can be
	    complex, but is dependent upon the amount of swap space and
	    memory.  Other kinds of <quote>free memory</quote>
	    descriptions are also possible, but it is relatively useless
	    to define these, but rather it is important to make sure
	    that the paging rate is kept low, and to avoid running out
	    of swap space.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="var-empty">
	  <para>What is <filename class="directory">/var/empty</filename>?  I can not
	    delete it!</para>
	</question>

	<answer>
	  <para><filename class="directory">/var/empty</filename> is a directory that the
	    &man.sshd.8; program uses when performing privilege separation.
	    The <filename class="directory">/var/empty</filename> directory is empty, owned by
	    <username>root</username> and has the <literal>schg</literal>
	    flag set.</para>

	  <para>Although it is not recommended to delete this directory, to
	    do so you will need to unset the <literal>schg</literal> flag
	    first.  See the &man.chflags.1; manual page for more information
	    (and bear in mind the answer to <link linkend="unsetting-schg">
	    the question on unsetting the schg flag</link>).
	  </para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="x">
    <title>The X Window System and Virtual Consoles</title>

    <qandaset>
      <qandaentry>
	<question id="whatis-X">
	  <para>What is the X Window System?</para>
	</question>

	<answer>
	  <para>The X Window System (commonly <literal>X11</literal>) is
	    the most widely available windowing system capable of running
	    on &unix; or &unix;&nbsp;like systems, including &os;.
	    <ulink url= "http://www.x.org/wiki/">The X.Org Foundation</ulink>
	    administers the <ulink
	      url="http://en.wikipedia.org/wiki/X_Window_System_core_protocol">X protocol standards</ulink>,
	    with the current reference implementation, version 11
	    release &xorg.version;, so you will often see references
	    shortened to <literal>X11</literal>.</para>

	  <para>Many implementations are available for different
	    architectures and operating systems.  An implementation of
	    the server-side code is properly known as an <literal>X
	    server</literal>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="running-X">
	  <para>I want to run X, how do I go about it?</para>
	</question>

	<answer>
	  <para>If you would like to add X to an existing installation,
	    you should use either the <filename
	      role="package">x11/xorg</filename> meta-port, which will
	    build and install all the necessary components, or install
	    &xorg; from &os; packages:</para>

	  <screen><userinput>&prompt.root; pkg_add -r xorg</userinput></screen>

	  <para>It is also possible to install &xorg; from
	    &man.sysinstall.8; by choosing
	    <guimenuitem>Configure</guimenuitem>, then
	    <guimenuitem>Distributions</guimenuitem>, then
	    <guimenuitem>The X.Org Distribution</guimenuitem>.</para>

	  <para>After the installation of &xorg; was successful, follow
	    the instructions from the <ulink
	      url="&url.books.handbook;/x-config.html">X11 Configuration</ulink> section of
	    the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="running-X-securelevels">
	  <para>I <emphasis>tried</emphasis> to run X, but I get an
	    <errorname>KDENABIO failed (Operation not
	    permitted)</errorname> error when I type
	    <command>startx</command>.  What do I do now?</para>
	</question>

	<answer>
	  <para>Your system is probably running at a raised
	    <literal>securelevel</literal>.  It is not possible to start X
	    at a raised <literal>securelevel</literal> because X
	    requires write access to &man.io.4;.  For more information,
	    see at the &man.init.8; manual page.</para>

	  <para>So the question is what else you should do instead, and
	    you basically have two choices: set your
	    <literal>securelevel</literal> back down to zero (usually
	    from <filename>/etc/rc.conf</filename>), or run &man.xdm.1;
	    at boot time (before the <literal>securelevel</literal> is
	    raised).</para>

	  <para>See <xref linkend="xdm-boot"/> for more information about
	    running &man.xdm.1; at boot time.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="x-and-moused">
	  <para>Why does my mouse not work with X?</para>
	</question>

	<answer>
	  <para>If you are using &man.syscons.4; (the default console
	    driver), you can configure &os; to support a mouse pointer on
	    each virtual screen.  In order to avoid conflicting with X,
	    &man.syscons.4; supports a virtual device called
	    <devicename>/dev/sysmouse</devicename>.  All mouse events
	    received from the real mouse device are written to the
	    &man.sysmouse.4; device via &man.moused.8;.  If you wish to
	    use your mouse on one or more virtual consoles,
	    <emphasis>and</emphasis> use X, see <xref
	      linkend="moused" remap="another section"/> and set up
	    &man.moused.8;.</para>

	  <para>Then edit <filename>/etc/X11/xorg.conf</filename> and
	    make sure you have the following lines:</para>

	  <programlisting>Section "InputDevice"
   Option          "Protocol" "SysMouse"
   Option          "Device" "/dev/sysmouse"
.....</programlisting>

	  <para>Starting with &xorg; version 7.4, the
	    <literal>InputDevice</literal> sections in
	    <filename>xorg.conf</filename> are ignored in favor of
	    autodetected devices.  To restore the old behavior, add the
	    following line to the <literal>ServerLayout</literal> or
	    <literal>ServerFlags</literal> section:</para>

	  <programlisting>Option "AutoAddDevices" "false"</programlisting>

	  <para>Some people prefer to use
	    <devicename>/dev/mouse</devicename> under X.  To make this
	    work, <devicename>/dev/mouse</devicename> should be linked
	    to <devicename>/dev/sysmouse</devicename> (see
	    &man.sysmouse.4;) by adding the following line to
	    <filename>/etc/devfs.conf</filename> (see
	    &man.devfs.conf.5;):</para>

	  <programlisting>link    sysmouse    mouse</programlisting>

	  <para>This link can be created by restarting &man.devfs.5;
	    with the following command (as
	    <username>root</username>):</para>

	  <screen>&prompt.root; <userinput>/etc/rc.d/devfs restart</userinput></screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="x-and-wheel">
	  <para>My mouse has a fancy wheel.  Can I use it in X?</para>
	</question>

	<answer>
	  <para>Yes.</para>

	  <para>You need to tell X that you have a 5 button mouse.  To
	    do this, simply add the lines <literal>Buttons 5</literal>
	    and <literal>ZAxisMapping 4 5</literal> to the
	    <quote>InputDevice</quote> section of
	    <filename>/etc/X11/xorg.conf</filename>.  For example, you
	    might have the following <quote>InputDevice</quote> section
	    in <filename>/etc/X11/xorg.conf</filename>.</para>

	  <example>
	    <title><quote>InputDevice</quote> Section for Wheeled Mouse
	      in &xorg; Configuration File</title>

	    <programlisting>Section "InputDevice"
   Identifier      "Mouse1"
   Driver          "mouse"
   Option          "Protocol" "auto"
   Option          "Device" "/dev/sysmouse"
   Option          "Buttons" "5"
   Option          "ZAxisMapping" "4 5"
EndSection</programlisting>
	  </example>

	  <example>
	    <title><quote>.emacs</quote> Example for Naive Page
	      Scrolling with Wheeled Mouse (optional)</title>

	    <programlisting>;; wheel mouse
(global-set-key [mouse-4] 'scroll-down)
(global-set-key [mouse-5] 'scroll-up)</programlisting>
	  </example>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="no-remote-x11">
	  <para>How do I use remote X displays?</para>
	</question>

	<answer>
	  <para>For security reasons, the default setting is to not
	    allow a machine to remotely open a window.</para>

	  <para>To enable this feature, simply start
	    <application>X</application> with the optional
	    <option>-listen_tcp</option> argument:</para>

	  <screen>&prompt.user; <userinput>startx -listen_tcp</userinput></screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="virtual-console">
	  <para>What is a virtual console and how do I make more?</para>
	</question>

	<answer>
	  <para>Virtual consoles, put simply, enable you to have several
	    simultaneous sessions on the same machine without doing
	    anything complicated like setting up a network or running
	    X.</para>

	  <para>When the system starts, it will display a login prompt
	    on the monitor after displaying all the boot messages.  You
	    can then type in your login name and password and start
	    working (or playing!) on the first virtual console.</para>

	  <para>At some point, you will probably wish to start another
	    session, perhaps to look at documentation for a program you
	    are running or to read your mail while waiting for an FTP
	    transfer to finish.  Just do <keycombo
	      action="simul"><keycap>Alt</keycap><keycap>F2</keycap></keycombo>
	    (hold down the <keycap>Alt</keycap> key and press the
	    <keycap>F2</keycap> key), and you will find a login prompt
	    waiting for you on the second <quote>virtual
	    console</quote>!  When you want to go back to the original
	    session, do <keycombo
	      action="simul"><keycap>Alt</keycap><keycap>F1</keycap></keycombo>.
	    </para>

	  <para>The default &os; installation has eight virtual consoles
	    enabled.  <keycombo
	      action="simul"><keycap>Alt</keycap><keycap>F1</keycap></keycombo>,
	    <keycombo
	      action="simul"><keycap>Alt</keycap><keycap>F2</keycap></keycombo>,
	    <keycombo
	      action="simul"><keycap>Alt</keycap><keycap>F3</keycap></keycombo>,
	    and so on will switch between these virtual consoles.</para>

	  <para>To enable more of them, edit
	    <filename>/etc/ttys</filename> (see &man.ttys.5;) and add
	    entries for <devicename>ttyv8</devicename> to
	    <devicename>ttyvc</devicename> after the comment on
	    <quote>Virtual terminals</quote>:</para>

	  <programlisting># Edit the existing entry for ttyv8 in /etc/ttys and change
# "off" to "on".
ttyv8   "/usr/libexec/getty Pc"         cons25  on secure
ttyv9   "/usr/libexec/getty Pc"         cons25  on secure
ttyva   "/usr/libexec/getty Pc"         cons25  on secure
ttyvb   "/usr/libexec/getty Pc"         cons25  on secure</programlisting>

	  <para>Use as many or as few as you want.  The more virtual
	    terminals you have, the more resources that are used; this
	    can be important if you have 8&nbsp;MB RAM or less.  You may
	    also want to change the <literal>secure</literal> to
	    <literal>insecure</literal>.</para>

	  <important>
	    <para>If you want to run an X server you
	      <emphasis>must</emphasis> leave at least one virtual
	      terminal unused (or turned off) for it to use.  That is to
	      say that if you want to have a login prompt pop up for all
	      twelve of your Alt-function keys, you are out of luck
	      &mdash; you can only do this for eleven of them if you
	      also want to run an X server on the same machine.</para>
	  </important>

	  <para>The easiest way to disable a console is by turning it
	    off.  For example, if you had the full 12 terminal
	    allocation mentioned above and you wanted to run X, you
	    would change settings for virtual terminal 12 from:</para>

	  <programlisting>ttyvb   "/usr/libexec/getty Pc"         cons25  on  secure</programlisting>

	  <para>to:</para>

	  <programlisting>ttyvb   "/usr/libexec/getty Pc"         cons25  off secure</programlisting>

	  <para>If your keyboard has only ten function keys, you would
	    end up with:</para>

	  <programlisting>ttyv9   "/usr/libexec/getty Pc"         cons25  off secure
ttyva   "/usr/libexec/getty Pc"         cons25  off secure
ttyvb   "/usr/libexec/getty Pc"         cons25  off secure</programlisting>

	  <para>(You could also just delete these lines.)</para>

	  <para>Next, the easiest (and cleanest) way to activate the
	    virtual consoles is to reboot.  However, if you really do
	    not want to reboot, you can just shut down the X Window
	    system and execute (as <username>root</username>):</para>

	  <screen>&prompt.root; <userinput>kill -HUP 1</userinput></screen>

	  <para>It is imperative that you completely shut down X Window
	    if it is running, before running this command.  If you do not,
	    your system will probably appear to hang or lock up after
	    executing the <command>kill</command> command.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="vty-from-x">
	  <para>How do I access the virtual consoles from X?</para>
	</question>

	<answer>
	  <para>Use <keycombo
	      action="simul"><keycap>Ctrl</keycap><keycap>Alt</keycap><keycap>F<replaceable>n</replaceable></keycap></keycombo>
	    to switch back to a virtual console.  <keycombo
	      action="simul"><keycap>Ctrl</keycap><keycap>Alt</keycap><keycap>F1</keycap></keycombo>
	    would return you to the first virtual console.</para>

	  <para>Once you are back to a text console, you can then use
	    <keycombo
	      action="simul"><keycap>Alt</keycap><keycap>F<replaceable>n</replaceable></keycap></keycombo>
	    as normal to move between them.</para>

	  <para>To return to the X session, you must switch to the
	    virtual console running X.  If you invoked X from the
	    command line, (e.g., using <command>startx</command>) then
	    the X session will attach to the next unused virtual
	    console, not the text console from which it was invoked.  If
	    you have eight active virtual terminals then X will be
	    running on the ninth, and you would use <keycombo
	      action="simul"><keycap>Alt</keycap><keycap>F9</keycap></keycombo>
	    to return.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="xdm-boot">
	  <para>How do I start <application>XDM</application> on
	    boot?</para>
	</question>

	<answer>
	  <para>There are two schools of thought on how to start
	    &man.xdm.1;.  One school starts <command>xdm</command> from
	    <filename>/etc/ttys</filename> (see &man.ttys.5;) using the
	    supplied example, while the other simply runs
	    <command>xdm</command> from from
	    <filename>rc.local</filename> (see &man.rc.8;) or from an
	    <filename>X</filename> script in
	    <filename class="directory">/usr/local/etc/rc.d</filename>.  Both are equally
	    valid, and one may work in situations where the other does
	    not.  In both cases the result is the same: X will pop up a
	    graphical login prompt.</para>

	  <para>The &man.ttys.5; method has the advantage of documenting
	    which vty X will start on and passing the responsibility of
	    restarting the X server on logout to &man.init.8;.  The
	    &man.rc.8; method makes it easy to <command>kill</command>
	    <command>xdm</command> if there is a problem starting the X
	    server.</para>

	  <para>If loaded from &man.rc.8;, <command>xdm</command> should
	    be started without any arguments (i.e., as a daemon).  The
	    <command>xdm</command> command must start
	    <emphasis>after</emphasis> &man.getty.8; runs, or else
	    <command>getty</command> and <command>xdm</command> will
	    conflict, locking out the console.  The best way around this
	    is to have the script sleep 10 seconds or so then launch
	    <command>xdm</command>.</para>

	  <para>If you are to start <command>xdm</command> from
	    <filename>/etc/ttys</filename>, there still is a chance of
	    conflict between <command>xdm</command> and &man.getty.8;.
	    One way to avoid this is to add the <literal>vt</literal>
	    number in the
	    <filename>/usr/local/lib/X11/xdm/Xservers</filename>
	    file:</para>

	  <programlisting>:0 local /usr/local/bin/X vt4</programlisting>

	  <para>The above example will direct the X server to run in
	    <devicename>/dev/ttyv3</devicename>.  Note the number is
	    offset by one.  The X server counts the vty from one,
	    whereas the &os; kernel numbers the vty from zero.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="xconsole-failure">
	  <para>Why do I get <errorname>Couldn't open
	    console</errorname> when I run
	    <command>xconsole</command>?</para>
	</question>

	<answer>
	  <para>If you start <application>X</application> with
	    <command>startx</command>, the permissions on
	    <devicename>/dev/console</devicename> will
	    <emphasis>not</emphasis> get changed, resulting in things
	    like <command>xterm -C</command> and
	    <command>xconsole</command> not working.</para>

	  <para>This is because of the way console permissions are set
	    by default.  On a multi-user system, one does not
	    necessarily want just any user to be able to write on the
	    system console.  For users who are logging directly onto a
	    machine with a VTY, the &man.fbtab.5; file exists to solve
	    such problems.</para>

	  <para>In a nutshell, make sure an uncommented line of the form
	    is in <filename>/etc/fbtab</filename> (see
	    &man.fbtab.5;):</para>

	  <programlisting>/dev/ttyv0 0600 /dev/console</programlisting>

	  <para>It will ensure that whomever logs in on
	    <devicename>/dev/ttyv0</devicename> will own the
	    console.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="xfree86-root">
	  <para>Before, I was able to run &xorg; as a regular user.
	    Why does it now say that I must be
	    <username>root</username>?</para>
	</question>

	<answer>
	  <para>All X servers need to be run as
	    <username>root</username> in order to get direct access to
	    your video hardware.</para>

	  <para>There are two ways to be able to use &xorg;
	    as a regular user.  The first is to use
	    <command>xdm</command> or another display manager (e.g.,
	    <command>kdm</command>); the second is to use the
	    <command>Xwrapper</command>.</para>

	  <para><command>xdm</command> is a daemon that handles
	    graphical logins.  It is usually started at boot time, and is
	    responsible for authenticating users and starting their
	    sessions; it is essentially the graphical counterpart of
	    &man.getty.8; and &man.login.1;.  For more information on
	    <command>xdm</command> see <ulink
	      url="http://www.x.org/wiki/UserDocumentation">the &xorg; documentation</ulink>,
	    and the <link
	      linkend="xdm-boot">the FAQ entry</link> on it.</para>

	  <para><command>Xwrapper</command> is the X server wrapper; it
	    is a small utility to enable one to manually run an X server
	    while maintaining reasonable safety.  It performs some
	    sanity checks on the command line arguments given, and if
	    they pass, runs the appropriate X server.  If you do not
	    want to run a display manager for whatever reason, this is
	    for you.  If you have installed the complete Ports
	    Collection, you can find the port in <filename
	      role="package">x11/wrapper</filename>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ps2-x">
	  <para>Why does my PS/2 mouse misbehave under X?</para>
	</question>

	<answer>
	  <para>Your mouse and the mouse driver may have somewhat become
	    out of synchronization.</para>

	  <para> In rare cases the driver may erroneously report
	    synchronization problem and you may see the kernel
	    message:</para>

	  <programlisting>psmintr: out of sync (xxxx != yyyy)</programlisting>

	  <para>and notice that your mouse does not work
	    properly.</para>

	  <para>If this happens, disable the synchronization check code
	    by setting the driver flags for the PS/2 mouse driver to
	    <literal>0x100</literal>.  Enter
	    <emphasis>UserConfig</emphasis> by giving the
	    <option>-c</option> option at the boot prompt:</para>

	  <screen>boot: <userinput>-c</userinput></screen>

	  <para>Then, in the <emphasis>UserConfig</emphasis> command
	    line, type:</para>

	  <screen>UserConfig&gt; <userinput>flags psm0 0x100</userinput>
UserConfig&gt; <userinput>quit</userinput></screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ps2-mousesystems">
	  <para>Why does my PS/2 mouse from MouseSystems not
	    work?</para>
	</question>

	<answer>
	  <para>There have been some reports that certain model of PS/2
	    mouse from MouseSystems works only if it is put into the
	    <quote>high resolution</quote> mode.  Otherwise, the mouse
	    cursor may jump to the upper-left corner of the screen every
	    so often.</para>

	  <para>Specify the flags <literal>0x04</literal> to the PS/2
	    mouse driver to put the mouse into the high resolution mode.
	    Enter <emphasis>UserConfig</emphasis> by giving the
	    <option>-c</option> option at the boot prompt:</para>

	  <screen>boot: <userinput>-c</userinput></screen>

	  <para>Then, in the <emphasis>UserConfig</emphasis> command
	    line, type:</para>

	  <screen>UserConfig&gt; <userinput>flags psm0 0x04</userinput>
UserConfig&gt; <userinput>quit</userinput></screen>

	  <para>See the previous section for another possible cause of
	    mouse problems.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="mouse-button-reverse">
	  <para>How do I reverse the mouse buttons?</para>
	</question>

	<answer>
	  <para>Run the command <command>xmodmap -e "pointer = 3 2 1"</command>
	    from your <filename>.xinitrc</filename> or
	    <filename>.xsession</filename>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="install-splash">
	  <para>How do I install a splash screen and where do I find
	    them?</para>
	</question>

	<answer>
	  <para>The detailed answer for this question can be found in
	    the <ulink
	      url="&url.books.handbook;/boot-blocks.html#BOOT-SPLASH">Boot Time Splash Screens</ulink>
	    section of the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="windows-keys">
	  <para>Can I use the <keycap>Windows</keycap> keys on my
	    keyboard in X?</para>
	</question>

	<answer>
	  <para>Yes.  All you need to do is use &man.xmodmap.1; to
	    define what function you wish them to perform.</para>

	  <para>Assuming all <quote>Windows</quote> keyboards are
	    standard then the keycodes for these three keys are the
	    following:</para>

	  <itemizedlist>
	    <listitem>
	      <para><keycode>115</keycode> &mdash;
		<keycap>Windows</keycap> key, between the left-hand
		<keycap>Ctrl</keycap> and <keycap>Alt</keycap>
		keys</para>
	    </listitem>

	    <listitem>
	      <para><keycode>116</keycode> &mdash;
		<keycap>Windows</keycap> key, to the right of the
		<keycap>AltGr</keycap> key</para>
	    </listitem>

	    <listitem>
	      <para><keycode>117</keycode> &mdash; <keycap>Menu</keycap>
		key, to the left of the right-hand <keycap>Ctrl</keycap>
		key</para>
	    </listitem>
	  </itemizedlist>

	  <para>To have the left <keycap>Windows</keycap> key print a
	    comma, try this.</para>

	  <screen>&prompt.root; <userinput>xmodmap -e "keycode 115 = comma"</userinput></screen>

	  <para>You will probably have to re-start your window manager
	    to see the result.</para>

	  <para>To have the <keycap>Windows</keycap> key-mappings
	    enabled automatically every time you start X either put the
	    <command>xmodmap</command> commands in your
	    <filename>~/.xinitrc</filename> file or, preferably, create
	    a file <filename>~/.xmodmaprc</filename> and include the
	    <command>xmodmap</command> options, one per line, then add
	    the following line to your
	    <filename>~/.xinitrc</filename>:</para>

	  <programlisting>xmodmap $HOME/.xmodmaprc</programlisting>

	  <para>For example, you could map the 3 keys to be
	    <keycap>F13</keycap>, <keycap>F14</keycap>, and
	    <keycap>F15</keycap>, respectively.  This would make it easy
	    to map them to useful functions within applications or your
	    window manager, as demonstrated further down.</para>

	  <para>To do this put the following in
	    <filename>~/.xmodmaprc</filename>.</para>

	  <programlisting>keycode 115 = F13
keycode 116 = F14
keycode 117 = F15</programlisting>

	  <para>If you use the <filename
	      role="package">x11-wm/fvwm2</filename> port, for example,
	    you could map the keys so that <keycap>F13</keycap>
	    iconifies (or de-iconifies) the window the cursor is in,
	    <keycap>F14</keycap> brings the window the cursor is in to
	    the front or, if it is already at the front, pushes it to
	    the back, and <keycap>F15</keycap> pops up the main
	    Workplace (application) menu even if the cursor is not on
	    the desktop, which is useful if you do not have any part of
	    the desktop visible (and the logo on the key matches its
	    functionality).</para>

	  <para>The following entries in <filename>~/.fvwmrc</filename>
	    implement the aforementioned setup:</para>

	  <programlisting>Key F13        FTIWS    A        Iconify
Key F14        FTIWS    A        RaiseLower
Key F15        A        A        Menu Workplace Nop</programlisting>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="x-3d-acceleration">
	  <para>How can I get 3D hardware acceleration for
	    &opengl;?</para>
	</question>

	<answer>
	  <para>The availability of 3D acceleration depends on the
	    version of &xorg; that you are using and the type of video
	    chip you have.  If you have an nVidia chip, you can use the
	    binary drivers provided for &os; by installing one of the
	    following ports:</para>

	  <itemizedlist>
	    <listitem>
	      <para>The latest versions of nVidia cards are supported by
		the <filename role="package">x11/nvidia-driver</filename>
		port.</para>
	    </listitem>

	    <listitem>
	      <para>nVidia cards like the GeForce2&nbsp;MX/3/4 series
		are supported by the 96XX series of drivers, available
		in the <filename
		  role="package">x11/nvidia-driver-96xx</filename>
		port.</para>
	    </listitem>

	    <listitem>
	      <para>Even older cards, like GeForce and RIVA&nbsp;TNT are
		supported by the 71XX series of drivers, available in
		the <filename
		  role="package">x11/nvidia-driver-71xx</filename>
		port.</para>
	    </listitem>
	  </itemizedlist>

	  <para>In fact, nVidia provides detailed information on which
	    card is supported by which driver.  This information is
	    available directly on their web site: <ulink
	      url="http://www.nvidia.com/object/IO_32667.html"></ulink>.
	    </para>

	  <para>For Matrox&nbsp;G200/G400, you should check the
	    <filename role="package">x11-servers/mga_hal</filename>
	    port.</para>

	  <para>For ATI&nbsp;Rage&nbsp;128 and Radeon, see the
	    &man.ati.4x;, &man.r128.4x; and &man.radeon.4x; manual
	    pages.</para>

	  <para>For 3dfx Voodoo&nbsp;3, 4, 5, and Banshee cards, there
	    is a <filename
	      role="package">x11-servers/driglide</filename>
	    port.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="networking">
    <title>Networking</title>

    <qandaset>
      <qandaentry>
	<question id="diskless-booting">
	  <para>Where can I get information on <quote>diskless
	    booting</quote>?</para>
	  </question>

	  <answer>
	    <para><quote>Diskless booting</quote> means that the &os;
	      box is booted over a network, and reads the necessary
	      files from a server instead of its hard disk.  For full
	      details, please read <ulink
		url="&url.books.handbook;/network-diskless.html">the Handbook entry on diskless booting</ulink>
	      </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="router">
	  <para>Can a &os; box be used as a dedicated network
	    router?</para>
	</question>

	<answer>
	  <para>Yes.  Please see the Handbook entry on <ulink
	      url="&url.books.handbook;/advanced-networking.html">advanced networking</ulink>,
	    specifically the section on <ulink
	      url="&url.books.handbook;/network-routing.html">routing and gateways</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="win95-connection">
	  <para>Can I connect my &windows; box to the Internet via
	    &os;?</para>
	</question>

	<answer>
	  <para>Typically, people who ask this question have two PCs at
	    home, one with &os; and one with some version of &windows;
	    the idea is to use the &os; box to connect to the Internet
	    and then be able to access the Internet from the &windows;
	    box through the &os; box.  This is really just a special
	    case of the previous question and works perfectly
	    well.</para>

	  <para>If you are using dialup to connect to the Internet
	    user-mode &man.ppp.8; contains a <option>-nat</option>
	    option.  If you run &man.ppp.8; with the
	    <option>-nat</option> option, set
	    <literal>gateway_enable</literal> to
	    <emphasis>YES</emphasis> in
	    <filename>/etc/rc.conf</filename>, and configure your
	    &windows; machine correctly, this should work fine.  For
	    more information, please see the &man.ppp.8; manual page or
	    the <ulink
	      url="&url.books.handbook;/userppp.html">Handbook entry on user PPP</ulink>.
	    </para>

	  <para>If you are using kernel-mode PPP or have an Ethernet
	    connection to the Internet, you need to use &man.natd.8;.
	    Please look at the <ulink
	      url="&url.books.handbook;/network-natd.html">natd</ulink>
	    section of the Handbook for a tutorial.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="slip-ppp-support">
	  <para>Does &os; support SLIP and PPP?</para>
	</question>

	<answer>
	  <para>Yes.  See the manual pages for &man.slattach.8;,
	    &man.sliplogin.8;, &man.ppp.8;, and &man.pppd.8;.
	    &man.ppp.8; and &man.pppd.8; provide support for both
	    incoming and outgoing connections, while &man.sliplogin.8;
	    deals exclusively with incoming connections, and
	    &man.slattach.8; deals exclusively with outgoing
	    connections.</para>

	  <para>For more information on how to use these, please see the
	    <ulink
	      url="&url.books.handbook;/ppp-and-slip.html">Handbook chapter on PPP and SLIP</ulink>.
	    </para>

	  <para>If you only have access to the Internet through a
	    <quote>shell account</quote>, you may want to have a look at
	    the <filename role="package">net/slirp</filename> package.
	    It can provide you with (limited) access to services such as
	    ftp and http direct from your local machine.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="natd">
	  <para>Does &os; support NAT or Masquerading?</para>
	</question>

	<answer>
	  <para>Yes.  If you want to use NAT over a user PPP connection,
	    please see the <ulink
	      url="&url.books.handbook;/userppp.html">Handbook entry on user PPP</ulink>.
	    If you want to use NAT over some other sort of network
	    connection, please look at the <ulink
	      url="&url.books.handbook;/network-natd.html">natd</ulink>
	    section of the Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="parallel-connect">
	  <para>How do I connect two &os; systems over a parallel line
	    using PLIP?</para>
	</question>

	<answer>
	  <para>Please see the <ulink
	      url="&url.books.handbook;/network-plip.html">PLIP section</ulink>
	    of the Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ethernet-aliases">
	  <para>How can I set up Ethernet aliases?</para>
	</question>

	<answer>
	  <para>If the alias is on the same subnet as an address already
	    configured on the interface, then add <literal>netmask
	    0xffffffff</literal> to your &man.ifconfig.8; command-line,
	    as in the following:</para>

	  <screen>&prompt.root; <userinput>ifconfig <replaceable>ed0</replaceable> alias <replaceable>192.0.2.2</replaceable> netmask 0xffffffff</userinput></screen>

	  <para>Otherwise, just specify the network address and netmask
	    as usual:</para>

	  <screen>&prompt.root; <userinput>ifconfig <replaceable>ed0</replaceable> alias <replaceable>172.16.141.5</replaceable> netmask 0xffffff00</userinput></screen>

	  <para>You can read more about this in the &os; <ulink
	      url="&url.books.handbook;/configtuning-virtual-hosts.html">Handbook</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="port-3c503">
	  <para>How do I get my 3C503 to use the other network
	    port?</para>
	</question>

	<answer>
	  <para>If you want to use the other ports, you will have to
	    specify an additional parameter on the &man.ifconfig.8;
	    command line.  The default port is <literal>link0</literal>.
	    To use the AUI port instead of the BNC one, use
	    <literal>link2</literal>.  These flags should be specified
	    using the ifconfig_* variables in
	    <filename>/etc/rc.conf</filename> (see
	    &man.rc.conf.5;).</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="nfs">
	  <para>Why am I having trouble with NFS and &os;?</para>
	</question>

	<answer>
	  <para>Certain PC network cards are better than others (to put
	    it mildly) and can sometimes cause problems with network
	    intensive applications like NFS.</para>

	  <para>See <ulink
	      url="&url.books.handbook;/network-nfs.html">the Handbook entry on NFS</ulink>
	    for more information on this topic.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="nfs-linux">
	  <para>Why can I not NFS-mount from a &linux; box?</para>
	</question>

	<answer>
	  <para>Some versions of the &linux; NFS code only accept mount
	    requests from a privileged port; try to issue the following
	    command:</para>

	  <screen>&prompt.root; <userinput>mount -o -P <replaceable>linuxbox:/blah</replaceable> <replaceable>/mnt</replaceable></userinput></screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="nfs-sun">
	  <para>Why can I not NFS-mount from a &sun; box?</para>
	</question>

	<answer>
	  <para>&sun; workstations running
	    &sunos;&nbsp;4.<replaceable>X</replaceable> only accept
	    mount requests from a privileged port; try the following
	    command:</para>

	  <screen>&prompt.root; <userinput>mount -o -P <replaceable>sunbox:/blah</replaceable> <replaceable>/mnt</replaceable></userinput></screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="exports-errors">
	  <para>Why does <command>mountd</command> keep telling me it
	    <errorname>can't change attributes</errorname> and that I
	    have a <errorname>bad exports list</errorname> on my &os;
	    NFS server?</para>
	</question>

	<answer>
	  <para>The most frequent problem is not understanding the
	    correct format of <filename>/etc/exports</filename>.  Please
	    review &man.exports.5; and the <ulink
	      url="&url.books.handbook;/network-nfs.html">NFS</ulink>
	    entry in the Handbook, especially the section on <ulink
	      url="&url.books.handbook;/network-nfs.html#CONFIGURING-NFS">configuring NFS</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-nextstep">
	  <para>Why am I having problems talking PPP to NeXTStep
	    machines?</para>
	</question>

	<answer>
	  <para>Try disabling the TCP extensions in
	    <filename>/etc/rc.conf</filename> (see &man.rc.conf.5;) by
	    changing the following variable to
	    <literal>NO</literal>:</para>

	  <programlisting>tcp_extensions=NO</programlisting>

	  <para>Xylogic's Annex boxes are also broken in this regard and
	    you must use the above change to connect through
	    them.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ip-multicast">
	  <para>How do I enable IP multicast support?</para>
	</question>

	<answer>
	  <para>&os; supports multicast host operations by default.  If
	    you want your box to run as a multicast router, you need to
	    recompile your kernel with the <literal>MROUTING</literal>
	    option and run &man.mrouted.8;.  &os; will start
	    &man.mrouted.8; at boot time if the flag
	    <literal>mrouted_enable</literal> is set to
	    <literal>YES</literal> in
	    <filename>/etc/rc.conf</filename>.</para>

	  <note>
	    <para>In recent &os; releases, the &man.mrouted.8; multicast
	      routing daemon, the &man.map-mbone.8; and &man.mrinfo.8;
	      utilities have been removed from the base system.  These
	      programs are now available in the &os; Ports Collection as
	      <filename role="package">net/mrouted</filename>.</para>
	  </note>

	  <para>MBONE tools are available in their own ports category,
	    <ulink
	      url="http://www.FreeBSD.org/ports/mbone.html">mbone</ulink>.
	    If you are looking for the conference tools
	    <command>vic</command> and <command>vat</command>, look
	    there!</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="fqdn-hosts">
	  <para>Why do I have to use the FQDN for hosts on my
	    site?</para>
	</question>

	<answer>
	  <para>See the answer in the &os; <ulink
	      url="&url.books.handbook;/mail-trouble.html">Handbook</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="network-permission-denied">
	  <para>Why do I get an error, <errorname>Permission
	    denied</errorname>, for all networking operations?</para>
	</question>

	<answer>
	  <para>If you have compiled your kernel with the
	    <literal>IPFIREWALL</literal> option, you need to be aware
	    that the default policy is to deny all packets that are not
	    explicitly allowed.</para>

	  <para>If you had unintentionally misconfigured your system for
	    firewalling, you can restore network operability by typing
	    the following while logged in as
	    <username>root</username>:</para>

	  <screen>&prompt.root; <userinput>ipfw add 65534 allow all from any to any</userinput></screen>

	  <para>You can also set <literal>firewall_type="open"</literal>
	    in <filename>/etc/rc.conf</filename>.</para>

	  <para>For further information on configuring a &os; firewall,
	    see the <ulink
	      url="&url.books.handbook;/firewalls.html">Handbook chapter</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ipfw-fwd">
	  <para>Why is my <command>ipfw</command> <quote>fwd</quote>
	    rule to redirect a service to another machine not
	    working?</para>
	</question>

	<answer>
	  <para>Possibly because you want to do network address
	    translation (NAT) and not just forward packets.  A
	    <quote>fwd</quote> rule does exactly what it says; it
	    forwards packets.  It does not actually change the data
	    inside the packet.  Say we have a rule like:</para>

	  <screen>01000 fwd <replaceable>10.0.0.1</replaceable> from any to <replaceable>foo 21</replaceable></screen>

	  <para>When a packet with a destination address of
	    <replaceable>foo</replaceable> arrives at the machine with
	    this rule, the packet is forwarded to
	    <replaceable>10.0.0.1</replaceable>, but it still has the
	    destination address of <replaceable>foo</replaceable>!  The
	    destination address of the packet is
	    <emphasis>not</emphasis> changed to
	    <replaceable>10.0.0.1</replaceable>.  Most machines would
	    probably drop a packet that they receive with a destination
	    address that is not their own.  Therefore, using a
	    <quote>fwd</quote> rule does not often work the way the user
	    expects.  This behavior is a feature and not a bug.</para>

	  <para>See the <link
	      linkend="service-redirect">FAQ about redirecting services</link>,
	    the &man.natd.8; manual, or one of the several port
	    redirecting utilities in the <ulink
	      url="&url.base;/ports/index.html">Ports Collection</ulink>
	    for a correct way to do this.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="service-redirect">
	  <para>How can I redirect service requests from one machine to
	    another?</para>
	</question>

	<answer>
	  <para>You can redirect FTP (and other service) request with
	    the <filename role="package">sysutils/socket</filename>
	    port.  Simply replace the service's command line to call
	    <command>socket</command> instead, like so:</para>

	  <programlisting>ftp stream tcp nowait nobody /usr/local/bin/socket socket <replaceable>ftp.example.com</replaceable> <replaceable>ftp</replaceable></programlisting>

	  <para>where <replaceable>ftp.example.com</replaceable> and
	    <replaceable>ftp</replaceable> are the host and port to
	    redirect to, respectively.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="bandwidth-mgr-tool">
	  <para>Where can I get a bandwidth management tool?</para>
	</question>

	<answer>
	  <para>There are three bandwidth management tools available for
	    &os;.  &man.dummynet.4; is integrated into &os; as part of
	    &man.ipfw.4;.  <ulink
	      url="http://www.sonycsl.co.jp/person/kjc/programs.html">ALTQ</ulink>
	    has been integrated into &os; as part of &man.pf.4;.
	    Bandwidth Manager from <ulink
	      url="http://www.etinc.com/">Emerging Technologies</ulink>
	    is a commercial product.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="bpf-not-configured">
	  <para>Why do I get <errorname>/dev/bpf0: device not
	    configured</errorname>?</para>
	</question>

	<answer>
	  <para>You are running a program that requires the Berkeley
	    Packet Filter (&man.bpf.4;), but it is not in your kernel.
	    Add this to your kernel config file and build a new
	    kernel:</para>

	  <programlisting>device bpf        # Berkeley Packet Filter</programlisting>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="mount-smb-share">
	  <para>How do I mount a disk from a &windows; machine that is
	    on my network, like smbmount in &linux;?</para>
	</question>

	<answer>
	  <para>Use the <application>SMBFS</application> toolset.  It
	    includes a set of kernel modifications and a set of userland
	    programs.  The programs and information are available as
	    &man.mount.smbfs.8; in the base system.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="icmp-response-bw-limit">
	  <para>What are these messages about: <errorname>Limiting
	    icmp/open port/closed port response</errorname> in my log
	    files?</para>
	</question>

	<answer>
	  <para>This is the kernel telling you that some activity is
	    provoking it to send more ICMP or TCP reset (RST) responses
	    than it thinks it should.  ICMP responses are often
	    generated as a result of attempted connections to unused UDP
	    ports.  TCP resets are generated as a result of attempted
	    connections to unopened TCP ports.  Among others, these are
	    the kinds of activities which may cause these
	    messages:</para>

	  <itemizedlist>
	    <listitem>
	      <para>Brute-force denial of service (DoS) attacks (as
		opposed to single-packet attacks which exploit a
		specific vulnerability).</para>
	    </listitem>

	    <listitem>
	      <para>Port scans which attempt to connect to a large
		number of ports (as opposed to only trying a few
		well-known ports).</para>
	    </listitem>
	  </itemizedlist>

	  <para>The first number in the message tells you how many
	    packets the kernel would have sent if the limit was not in
	    place, and the second number tells you the limit.  You can
	    control the limit using the
	    <varname>net.inet.icmp.icmplim</varname> sysctl variable
	    like this, where <literal>300</literal> is the limit in
	    packets per second:</para>

	  <screen>&prompt.root; <userinput>sysctl -w net.inet.icmp.icmplim=300</userinput></screen>

	  <para>If you do not want to see messages about this in your
	    log files, but you still want the kernel to do response
	    limiting, you can use the
	    <varname>net.inet.icmp.icmplim_output</varname> sysctl
	    variable to disable the output like this:</para>

	  <screen>&prompt.root; <userinput>sysctl -w net.inet.icmp.icmplim_output=0</userinput></screen>

	  <para>Finally, if you want to disable response limiting, you
	    can set the <varname>net.inet.icmp.icmplim</varname> sysctl
	    variable (see above for an example) to <literal>0</literal>.
	    Disabling response limiting is discouraged for the reasons
	    listed above.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="unknown-hw-addr-format">
	  <para>What are these <errorname>arp: unknown hardware address
	    format</errorname> error messages?</para>
	</question>

	<answer>
	  <para>This means that some device on your local Ethernet is
	    using a MAC address in a format that &os; does not
	    recognize.  This is probably caused by someone experimenting
	    with an Ethernet card somewhere else on the network.  You
	    will see this most commonly on cable modem networks.  It is
	    harmless, and should not affect the performance of your &os;
	    machine.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="arp-wrong-iface">
	  <para>Why do I keep seeing messages like: <errorname>192.168.0.10 is on
	    fxp1 but got reply from 00:15:17:67:cf:82 on rl0</errorname>, and how do I
	    disable it?</para>
	</question>

	<answer>
	  <para>Because a packet is coming from outside the network
	    unexpectedly.  To disable them, set
	    <varname>net.link.ether.inet.log_arp_wrong_iface</varname>
	    to <literal>0</literal>.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="security">
    <title>Security</title>

    <qandaset>
      <qandaentry>
	<question id="sandbox">
	  <para>What is a sandbox?</para>
	</question><answer>

	  <para><quote>Sandbox</quote> is a security term.  It can mean
	    two things:</para>

	  <itemizedlist>
	    <listitem>
	      <para>A process which is placed inside a set of virtual
		walls that are designed to prevent someone who breaks
		into the process from being able to break into the wider
		system.</para>

	      <para>The process is said to be able to
		<quote>play</quote> inside the walls.  That is, nothing
		the process does in regards to executing code is
		supposed to be able to breech the walls so you do not
		have to do a detailed audit of its code to be able to
		say certain things about its security.</para>

	      <para>The walls might be a user&nbsp;ID, for example.
		This is the definition used in the &man.security.7; and
		&man.named.8; man pages.</para>

	      <para>Take the <literal>ntalk</literal> service, for
		example (see &man.inetd.8;).  This service used to run
		as user&nbsp;ID <username>root</username>.  Now it runs
		as user&nbsp;ID <username>tty</username>.  The
		<username>tty</username> user is a sandbox designed to
		make it more difficult for someone who has successfully
		hacked into the system via <literal>ntalk</literal> from
		being able to hack beyond that user&nbsp;ID.</para>
	    </listitem>

	    <listitem>
	      <para>A process which is placed inside a simulation of the
		machine.  This is more hard-core.  Basically it means
		that someone who is able to break into the process may
		believe that he can break into the wider machine but is,
		in fact, only breaking into a simulation of that machine
		and not modifying any real data.</para>

	      <para>The most common way to accomplish this is to build a
		simulated environment in a subdirectory and then run the
		processes in that directory chroot'd (i.e. <filename
		  class="directory">/</filename> for that process is this
		directory, not the real <filename
		  class="directory">/</filename> of the system).</para>

	      <para>Another common use is to mount an underlying file
		system read-only and then create a file system layer on
		top of it that gives a process a seemingly writeable
		view into that file system.  The process may believe it
		is able to write to those files, but only the process
		sees the effects &mdash; other processes in the system
		do not, necessarily.</para>

	      <para>An attempt is made to make this sort of sandbox so
		transparent that the user (or hacker) does not realize
		that he is sitting in it.</para>
	    </listitem>
	  </itemizedlist>

	  <para>&unix; implements two core sandboxes.  One is at the
	    process level, and one is at the userid level.</para>

	  <para>Every &unix; process is completely firewalled off from
	    every other &unix; process.  One process cannot modify the
	    address space of another.  This is unlike &windows; where a
	    process can easily overwrite the address space of any other,
	    leading to a crash.</para>

	  <para>A &unix; process is owned by a particular userid.  If
	    the user&nbsp;ID is not the <username>root</username> user,
	    it serves to firewall the process off from processes owned
	    by other users.  The user&nbsp;ID is also used to firewall
	    off on-disk data.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="securelevel">
	  <para>What is securelevel?</para>
	</question>

	<answer>
	  <para>The securelevel is a security mechanism implemented in
	    the kernel.  Basically, when the securelevel is positive, the
	    kernel restricts certain tasks; not even the superuser
	    (i.e., <username>root</username>) is allowed to do them.  At
	    the time of this writing, the securelevel mechanism is
	    capable of, among other things, limiting the ability
	    to:</para>

	  <itemizedlist>
	    <listitem>
	      <para>Unset certain file flags, such as
		<literal>schg</literal> (the system immutable
		flag).</para>
	    </listitem>

	    <listitem>
	      <para>Write to kernel memory via
		<devicename>/dev/mem</devicename> and
		<devicename>/dev/kmem</devicename>.</para>
	    </listitem>

	    <listitem>
	      <para>Load kernel modules.</para>
	    </listitem>

	    <listitem>
	      <para>Alter firewall rules.</para>
	    </listitem>
	  </itemizedlist>

	  <para>To check the status of the securelevel on a running
	    system, simply execute the following command:</para>

	  <screen>&prompt.root; <userinput>sysctl kern.securelevel</userinput></screen>

	  <para>The output will contain the name of the &man.sysctl.8;
	    variable (in this case, <varname>kern.securelevel</varname>)
	    and a number.  The latter is the current value of the
	    securelevel.  If it is positive (i.e., greater than 0), at
	    least some of the securelevel's protections are
	    enabled.</para>

	  <para>You cannot lower the securelevel of a running system;
	    being able to do that would defeat its purpose.  If you need
	    to do a task that requires that the securelevel be
	    non-positive (e.g., an <maketarget>installworld</maketarget>
	    or changing the date), you will have to change the
	    securelevel setting in <filename>/etc/rc.conf</filename>
	    (you want to look for the
	    <varname>kern_securelevel</varname> and
	    <varname>kern_securelevel_enable</varname> variables) and
	    reboot.</para>

	  <para>For more information on securelevel and the specific
	    things all the levels do, please consult the &man.init.8;
	    manual page.</para>

	  <warning>
	    <para>Securelevel is not a silver bullet; it has many known
	      deficiencies.  More often than not, it provides a false
	      sense of security.</para>

	    <para>One of its biggest problems is that in order for it to
	      be at all effective, all files used in the boot process up
	      until the securelevel is set must be protected.  If an
	      attacker can get the system to execute their code prior to
	      the securelevel being set (which happens quite late in the
	      boot process since some things the system must do at
	      start-up cannot be done at an elevated securelevel), its
	      protections are invalidated.  While this task of
	      protecting all files used in the boot process is not
	      technically impossible, if it is achieved, system
	      maintenance will become a nightmare since one would have
	      to take the system down, at least to single-user mode, to
	      modify a configuration file.</para>

	    <para>This point and others are often discussed on the
	      mailing lists, particularly the &a.security;.  Please
	      search the archives <ulink
		url="&url.base;/search/index.html">here</ulink> for an
	      extensive discussion.  Some people are hopeful that
	      securelevel will soon go away in favor of a more
	      fine-grained mechanism, but things are still hazy in this
	      respect.</para>

	    <para>Consider yourself warned.</para>
	  </warning>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="extra-named-port">
	  <para>BIND (<command>named</command>) is listening on
	    some high-numbered ports.  What is going on?</para>
	</question>

	<answer>
	  <para>BIND uses a random high-numbered port for outgoing
	    queries.  Recent versions of it choose a new, random UDP
	    port for each query.  This may cause problems for some
	    network configurations, especially if a firewall blocks
	    incoming UDP packets on particular ports.  If you want to
	    get past that firewall, you can try the
	    <literal>avoid-v4-udp-ports</literal> and
	    <literal>avoid-v6-udp-ports</literal> options to avoid
	    selecting random port numbers within a blocked range.</para>

	  <warning>
	    <para>If a port number (like 53) is specified via the
	      <literal>query-source</literal> or
	      <literal>query-source-v6</literal> options in
	      <filename>/etc/namedb/named.conf</filename>, randomized
	      port selection will not be used.  It is strongly
	      recommended that these options not be used to specify
	      fixed port numbers.</para>
	  </warning>

	  <para>Congratulations, by the way.  It is good practice to
	    read your &man.sockstat.1; output and notice odd
	    things!</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="sendmail-port-587">
	  <para>The <application>sendmail</application> daemon is
	    listening on port 587 as well as the standard port 25!  What
	    is going on?</para>
	</question>

	<answer>
	  <para>Recent versions of <application>sendmail</application>
	    support a mail submission feature that runs over port 587.
	    This is not yet widely supported, but is growing in
	    popularity.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="toor-account">
	  <para>What is this UID 0 <username>toor</username> account?
	    Have I been compromised?</para>
	</question>

	<answer>
	  <para>Do not worry.  <username>toor</username> is an
	    <quote>alternative</quote> superuser account (toor is root
	    spelt backwards).  Previously it was created when the
	    &man.bash.1; shell was installed but now it is created by
	    default.  It is intended to be used with a non-standard
	    shell so you do not have to change
	    <username>root</username>'s default shell.  This is
	    important as shells which are not part of the base
	    distribution (for example a shell installed from ports or
	    packages) are likely to be installed in
	    <filename class="directory">/usr/local/bin</filename> which, by default,
	    resides on a different file system.  If
	    <username>root</username>'s shell is located in
	    <filename class="directory">/usr/local/bin</filename> and
	    <filename class="directory">/usr</filename> (or whatever file system contains
	    <filename class="directory">/usr/local/bin</filename>) is not mounted for some
	    reason, <username>root</username> will not be able to log in
	    to fix a problem (although if you reboot into single user
	    mode you will be prompted for the path to a shell).</para>

	  <para>Some people use <username>toor</username> for day-to-day
	    <username>root</username> tasks with a non-standard shell,
	    leaving <username>root</username>, with a standard shell,
	    for single user mode or emergencies.  By default you cannot
	    log in using <username>toor</username> as it does not have a
	    password, so log in as <username>root</username> and set a
	    password for <username>toor</username> if you want to use
	    it.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="suidperl">
	  <para>Why is <command>suidperl</command> not working
	    properly?</para>
	</question>

	<answer>
	  <para>For security reasons, <command>suidperl</command> is not
	    installed by default.  If you want
	    <command>suidperl</command> to be built during upgrades from
	    source, edit <filename>/etc/make.conf</filename> and add
	    <literal><varname>ENABLE_SUIDPERL</varname>=true</literal>
	    before you build <command>perl</command>.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="ppp">
    <title>PPP</title>

    <qandaset>
      <qandaentry>
	<question id="userppp">
	  <para>I cannot make &man.ppp.8; work.  What am I doing
	    wrong?</para>
	</question>

	<answer>
	  <para>You should first read the &man.ppp.8; manual page and
	    the <ulink
	      url="&url.books.handbook;/ppp-and-slip.html#USERPPP">PPP section of the handbook</ulink>.
	    Enable logging with the following command:</para>

	  <programlisting>set log Phase Chat Connect Carrier lcp ipcp ccp command</programlisting>

	  <para>This command may be typed at the &man.ppp.8; command
	    prompt or it may be entered in the
	    <filename>/etc/ppp/ppp.conf</filename> configuration file
	    (the start of the <literal>default</literal> section is the
	    best place to put it).  Make sure that
	    <filename>/etc/syslog.conf</filename> (see
	    &man.syslog.conf.5;) contains the lines below and the file
	    <filename>/var/log/ppp.log</filename> exists:</para>

	  <programlisting>!ppp
*.*        /var/log/ppp.log</programlisting>

	  <para>You can now find out a lot about what is going on from
	    the log file.  Do not worry if it does not all make sense.
	    If you need to get help from someone, it may make sense to
	    them.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-hangs">
	  <para>Why does &man.ppp.8; hang when I run it?</para>
	</question>

	<answer>
	  <para>This is usually because your hostname will not resolve.
	    The best way to fix this is to make sure that
	    <filename>/etc/hosts</filename> is consulted by your
	    resolver first by editing
	    <filename>/etc/host.conf</filename> and putting the
	    <literal>hosts</literal> line first.  Then, simply put an
	    entry in <filename>/etc/hosts</filename> for your local
	    machine.  If you have no local network, change your
	    <hostid>localhost</hostid> line:</para>

	  <programlisting>127.0.0.1        foo.example.com foo localhost</programlisting>

	  <para>Otherwise, simply add another entry for your host.
	    Consult the relevant manual pages for more details.</para>

	  <para>You should be able to successfully
	    <command>ping -c1 `hostname`</command> when you are
	    done.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-nodial-auto">
	  <para>Why will &man.ppp.8; not dial in
	    <literal>-auto</literal> mode?</para>
	</question>

	<answer>
	  <para>First, check that you have got a default route.  By
	    running <command>netstat -rn</command> (see
	    &man.netstat.1;), you should see two entries like
	    this:</para>

	  <programlisting>Destination        Gateway            Flags     Refs     Use     Netif Expire
default            10.0.0.2           UGSc        0        0      tun0
10.0.0.2           10.0.0.1           UH          0        0      tun0</programlisting>

	  <para>This is assuming that you have used the addresses from
	    the handbook, the manual page, or from the
	    <filename>ppp.conf.sample</filename> file.  If you do not
	    have a default route, it may be because you forgot to add
	    the <literal>HISADDR</literal> line to the
	    <filename>ppp.conf</filename> file.</para>

	  <para>Another reason for the default route line being missing
	    is that you have mistakenly set up a default router in your
	    <filename>/etc/rc.conf</filename> (see &man.rc.conf.5;) file
	    and you have omitted the line below from
	    <filename>ppp.conf</filename>:</para>

	  <programlisting>delete ALL</programlisting>

	  <para>If this is the case, go back to the <ulink
	      url="&url.books.handbook;/userppp.html#USERPPP-FINAL">Final System Configuration</ulink>
	    section of the handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="no-route-to-host">
	  <para>What does <errorname>No route to host</errorname>
	    mean?</para>
	</question>

	<answer>
	  <para>This error is usually due that the following section is
	    missing in your <filename>/etc/ppp/ppp.linkup</filename>
	    file:</para>

	  <programlisting>MYADDR:
  delete ALL
  add 0 0 HISADDR</programlisting>

	  <para>This is only necessary if you have a dynamic IP address
	    or do not know the address of your gateway.  If you are
	    using interactive mode, you can type the following after
	    entering <literal>packet mode</literal> (packet mode is
	    indicated by the capitalized <acronym>PPP</acronym> in the
	    prompt):</para>

	  <programlisting>delete ALL
add 0 0 HISADDR</programlisting>

	  <para>Refer to the <ulink
	      url="&url.books.handbook;/userppp.html#USERPPP-DYNAMICIP">PPP and Dynamic IP addresses</ulink>
	    section of the handbook for further details.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="connection-threeminutedrop">
	  <para>Why does my connection drop after about 3
	    minutes?</para>
	</question>

	<answer>
	  <para>The default PPP timeout is 3 minutes.  This can be
	    adjusted with the following line:</para>

	  <programlisting>set timeout <replaceable>NNN</replaceable></programlisting>

	  <para>where <replaceable>NNN</replaceable> is the number of
	    seconds of inactivity before the connection is closed.  If
	    <replaceable>NNN</replaceable> is zero, the connection is
	    never closed due to a timeout.  It is possible to put this
	    command in the <filename>ppp.conf</filename> file, or to
	    type it at the prompt in interactive mode.  It is also
	    possible to adjust it on the fly while the line is active by
	    connecting to <application>ppp</application>'s server socket
	    using &man.telnet.1; or &man.pppctl.8;.  Refer to the
	    &man.ppp.8; man page for further details.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-drop-heavy-load">
	  <para>Why does my connection drop under heavy load?</para>
	</question>

	<answer>
	  <para>If you have Link Quality Reporting (LQR) configured, it
	    is possible that too many LQR packets are lost between your
	    machine and the peer.  The &man.ppp.8; program deduces that
	    the line must therefore be bad, and disconnects.  Prior to
	    &os; version 2.2.5, LQR was enabled by default.  It is now
	    disabled by default.  LQR can be disabled with the following
	    line:</para>

	  <programlisting>disable lqr</programlisting>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-drop-random">
	  <para>Why does my connection drop after a random amount of
	    time?</para>
	</question>

	<answer>
	  <para>Sometimes, on a noisy phone line or even on a line with
	    call waiting enabled, your modem may hang up because it
	    thinks (incorrectly) that it lost carrier.</para>

	  <para>There is a setting on most modems for determining how
	    tolerant it should be to temporary losses of carrier.  On a
	    &usrobotics;&nbsp;&sportster; for example, this is measured
	    by the <literal>S10</literal> register in tenths of a
	    second.  To make your modem more forgiving, you could add
	    the following send-expect sequence to your dial
	    string:</para>

	  <programlisting>set dial "...... ATS10=10 OK ......"</programlisting>

	  <para>Refer to your modem manual for details.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-hangs-random">
	  <para>Why does my connection hang after a random amount of
	    time?</para>
	</question>

	<answer>
	  <para>Many people experience hung connections with no apparent
	    explanation.  The first thing to establish is which side of
	    the link is hung.</para>

	  <para>If you are using an external modem, you can simply try
	    using &man.ping.8; to see if the <acronym>TD</acronym> light
	    is flashing when you transmit data.  If it flashes (and the
	    <acronym>RD</acronym> light does not), the problem is with
	    the remote end.  If <acronym>TD</acronym> does not flash,
	    the problem is local.  With an internal modem, you will need
	    to use the <literal>set server</literal> command in your
	    <filename>ppp.conf</filename> file.  When the hang occurs,
	    connect to &man.ppp.8; using &man.pppctl.8;.  If your
	    network connection suddenly revives (PPP was revived due to
	    the activity on the diagnostic socket) or if you cannot
	    connect (assuming the <literal>set socket</literal> command
	    succeeded at startup time), the problem is local.  If you
	    can connect and things are still hung, enable local async
	    logging with <literal>set log local async</literal> and use
	    &man.ping.8; from another window or terminal to make use of
	    the link.  The async logging will show you the data being
	    transmitted and received on the link.  If data is going out
	    and not coming back, the problem is remote.</para>

	  <para>Having established whether the problem is local or
	    remote, you now have two possibilities:</para>

	  <itemizedlist>
	    <listitem>
	      <para>If the problem is remote, read on entry <xref
		  linkend="ppp-remote-not-responding"/>.</para>
	    </listitem>

	    <listitem>
	      <para>If the problem is local, read on entry <xref
		  linkend="ppp-hung"/>.</para>
	    </listitem>
	  </itemizedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-remote-not-responding">
	  <para>The remote end is not responding.  What can I do?</para>
	</question>

	<answer>
	  <para>There is very little you can do about this.  Most ISPs
	    will refuse to help if you are not running a &microsoft; OS.
	    You can <literal>enable lqr</literal> in your
	    <filename>ppp.conf</filename> file, allowing &man.ppp.8; to
	    detect the remote failure and hang up, but this detection is
	    relatively slow and therefore not that useful.  You may want
	    to avoid telling your ISP that you are running
	    user-PPP.</para>

	  <para>First, try disabling all local compression by adding the
	    following to your configuration:</para>

	  <programlisting>disable pred1 deflate deflate24 protocomp acfcomp shortseq vj
deny pred1 deflate deflate24 protocomp acfcomp shortseq vj</programlisting>

	  <para>Then reconnect to ensure that this makes no difference.
	    If things improve or if the problem is solved completely,
	    determine which setting makes the difference through trial
	    and error.  This will provide good ammunition when you
	    contact your ISP (although it may make it apparent that you
	    are not running a &microsoft; product).</para>

	  <para>Before contacting your ISP, enable async logging locally
	    and wait until the connection hangs again.  This may use up
	    quite a bit of disk space.  The last data read from the port
	    may be of interest.  It is usually ASCII data, and may even
	    describe the problem (<errorname>Memory fault</errorname>,
	    <errorname>Core dumped</errorname>).</para>

	  <para>If your ISP is helpful, they should be able to enable
	    logging on their end, then when the next link drop occurs,
	    they may be able to tell you why their side is having a
	    problem.  Feel free to send the details to &a.brian;, or
	    even to ask your ISP to contact him directly.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-hung">
	  <para>&man.ppp.8; has hung.  What can I do?</para>
	</question>

	<answer>
	  <para>Your best bet here is to rebuild &man.ppp.8; with
	    debugging information, and then use &man.gdb.1; to grab a
	    stack trace from the <application>ppp</application> process
	    that is stuck.  To rebuild the
	    <application>ppp</application> utility with debugging
	    information, you can type:</para>

	  <screen>&prompt.root; <userinput>cd /usr/src/usr.sbin/ppp</userinput>
&prompt.root; <userinput>env DEBUG_FLAGS='-g' make clean</userinput>
&prompt.root; <userinput>env DEBUG_FLAGS='-g' make install</userinput></screen>

	  <para>Then you should restart <application>ppp</application>
	    and wait until it hangs again.  When the debug build of
	    <application>ppp</application> hangs, start
	    <application>gdb</application> on the stuck process by
	    typing:</para>

	  <screen>&prompt.root; <userinput>gdb ppp `pgrep ppp`</userinput></screen>

	  <para>At the <application>gdb</application> prompt, you can
	    use the <command>bt</command> or <command>where</command>
	    commands to get a stack trace.  Save the output of your
	    <application>gdb</application> session, and
	    <quote>detach</quote> from the running process by the
	    <command>quit</command> command of
	    <application>gdb</application>.</para>

	  <para>Finally, send the log of your
	    <application>gdb</application> session to &a.brian;.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-loginok-thennothing">
	  <para>Why does nothing happen after the <quote>Login
	    OK!</quote> message?</para>
	</question>

	<answer>
	  <para>Prior to &os; version 2.2.5, once the link was
	    established, &man.ppp.8; would wait for the peer to initiate
	    the Line Control Protocol (LCP).  Many ISPs will not
	    initiate negotiations and expect the client to do so.  To
	    force &man.ppp.8; to initiate the LCP, use the following
	    line:</para>

	  <programlisting>set openmode active</programlisting>

	  <note>
	    <para>It usually does no harm if both sides initiate
	      negotiation, so openmode is now active by default.
	      However, the next section explains when it
	      <emphasis>does</emphasis> do some harm.</para>
	  </note>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-same-magic">
	  <para>I keep seeing errors about magic being the same.  What
	    does it mean?</para>
	</question>

	<answer>
	  <para>Occasionally, just after connecting, you may see
	    messages in the log that say <errorname>Magic is
	    same</errorname>.  Sometimes, these messages are harmless,
	    and sometimes one side or the other exits.  Most PPP
	    implementations cannot survive this problem, and even if the
	    link seems to come up, you will see repeated configure
	    requests and configure acknowledgments in the log file until
	    &man.ppp.8; eventually gives up and closes the
	    connection.</para>

	  <para>This normally happens on server machines with slow disks
	    that are spawning a &man.getty.8; on the port, and executing
	    &man.ppp.8; from a login script or program after login.
	    There were reports of it happening consistently when using
	    slirp.  The reason is that in the time taken between
	    &man.getty.8; exiting and &man.ppp.8; starting, the
	    client-side &man.ppp.8; starts sending Line Control Protocol
	    (LCP) packets.  Because ECHO is still switched on for the
	    port on the server, the client &man.ppp.8; sees these
	    packets <quote>reflect</quote> back.</para>

	  <para>One part of the LCP negotiation is to establish a magic
	    number for each side of the link so that
	    <quote>reflections</quote> can be detected.  The protocol
	    says that when the peer tries to negotiate the same magic
	    number, a NAK should be sent and a new magic number should
	    be chosen.  During the period that the server port has ECHO
	    turned on, the client &man.ppp.8; sends LCP packets, sees
	    the same magic in the reflected packet and NAKs it.  It also
	    sees the NAK reflect (which also means &man.ppp.8; must
	    change its magic).  This produces a potentially enormous
	    number of magic number changes, all of which are happily
	    piling into the server's tty buffer.  As soon as &man.ppp.8;
	    starts on the server, it is flooded with magic number
	    changes and almost immediately decides it has tried enough
	    to negotiate LCP and gives up.  Meanwhile, the client, who
	    no longer sees the reflections, becomes happy just in time
	    to see a hangup from the server.</para>

	  <para>This can be avoided by allowing the peer to start
	    negotiating with the following line in your
	    <filename>ppp.conf</filename> file:</para>

	  <programlisting>set openmode passive</programlisting>

	  <para>This tells &man.ppp.8; to wait for the server to
	    initiate LCP negotiations.  Some servers however may never
	    initiate negotiations.  If this is the case, you can do
	    something like:</para>

	  <programlisting>set openmode active 3</programlisting>

	  <para>This tells &man.ppp.8; to be passive for 3 seconds, and
	    then to start sending LCP requests.  If the peer starts
	    sending requests during this period, &man.ppp.8; will
	    immediately respond rather than waiting for the full 3
	    second period.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-lcp-constant">
	  <para>LCP negotiations continue until the connection is
	    closed.  What is wrong?</para>
	</question>

	<answer>
	  <para>There is currently an implementation mis-feature in
	    &man.ppp.8; where it does not associate LCP, CCP &amp; IPCP
	    responses with their original requests.  As a result, if one
	    PPP implementation is more than 6 seconds slower than the
	    other side, the other side will send two additional LCP
	    configuration requests.  This is fatal.</para>

	  <para>Consider two implementations, <hostid>A</hostid> and
	    <hostid>B</hostid>.  <hostid>A</hostid> starts sending LCP
	    requests immediately after connecting and <hostid>B</hostid>
	    takes 7 seconds to start.  When <hostid>B</hostid> starts,
	    <hostid>A</hostid> has sent 3 LCP REQs.  We are assuming the
	    line has ECHO switched off, otherwise we would see magic
	    number problems as described in the previous section.
	    <hostid>B</hostid> sends a REQ, then an ACK to the first of
	    <hostid>A</hostid>'s REQs.  This results in
	    <hostid>A</hostid> entering the <acronym>OPENED</acronym>
	    state and sending and ACK (the first) back to
	    <hostid>B</hostid>.  In the meantime, <hostid>B</hostid>
	    sends back two more ACKs in response to the two additional
	    REQs sent by <hostid>A</hostid> before <hostid>B</hostid>
	    started up.  <hostid>B</hostid> then receives the first ACK
	    from <hostid>A</hostid> and enters the
	    <acronym>OPENED</acronym> state.  <hostid>A</hostid>
	    receives the second ACK from <hostid>B</hostid> and goes
	    back to the <acronym>REQ-SENT</acronym> state, sending
	    another (forth) REQ as per the RFC.  It then receives the
	    third ACK and enters the <acronym>OPENED</acronym> state.
	    In the meantime, <hostid>B</hostid> receives the forth REQ
	    from <hostid>A</hostid>, resulting in it reverting to the
	    <acronym>ACK-SENT</acronym> state and sending another
	    (second) REQ and (forth) ACK as per the RFC.
	    <hostid>A</hostid> gets the REQ, goes into
	    <acronym>REQ-SENT</acronym> and sends another REQ.  It
	    immediately receives the following ACK and enters
	    <acronym>OPENED</acronym>.</para>

	  <para>This goes on until one side figures out that they are
	    getting nowhere and gives up.</para>

	  <para>The best way to avoid this is to configure one side to
	    be <literal>passive</literal> &mdash; that is, make one side
	    wait for the other to start negotiating.  This can be done
	    with the following command:</para>

	  <programlisting>set openmode passive</programlisting>

	  <para>Care should be taken with this option.  You should also
	    use this command to limit the amount of time that
	    &man.ppp.8; waits for the peer to begin negotiations:</para>

	  <programlisting>set stopped <replaceable>N</replaceable></programlisting>

	  <para>Alternatively, the following command (where
	    <replaceable>N</replaceable> is the number of seconds to
	    wait before starting negotiations) can be used:</para>

	  <programlisting>set openmode active <replaceable>N</replaceable></programlisting>

	  <para>Check the manual page for details.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-shell-test-lockup">
	  <para>Why does &man.ppp.8; lock up when I shell out to test
	    it?</para>
	</question>

	<answer>
	  <para>When you execute the <command>shell</command> or
	    <command>!</command> command, &man.ppp.8; executes a shell
	    (or if you have passed any arguments, &man.ppp.8; will
	    execute those arguments).  The
	    <application>ppp</application> program will wait for the
	    command to complete before continuing.  If you attempt to
	    use the PPP link while running the command, the link will
	    appear to have frozen.  This is because &man.ppp.8; is
	    waiting for the command to complete.</para>

	  <para>If you wish to execute commands like this, use the
	    <command>!bg</command> command instead.  This will execute
	    the given command in the background, and &man.ppp.8; can
	    continue to service the link.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-null-modem">
	  <para>Why does &man.ppp.8; over a null-modem cable never
	    exit?</para>
	</question>

	<answer>
	  <para>There is no way for &man.ppp.8; to automatically
	    determine that a direct connection has been dropped.  This
	    is due to the lines that are used in a null-modem serial
	    cable.  When using this sort of connection, LQR should
	    always be enabled with the following line:</para>

	  <programlisting>enable lqr</programlisting>

	  <para>LQR is accepted by default if negotiated by the
	    peer.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-auto-noreasondial">
	  <para>Why does &man.ppp.8; dial for no reason in
	    <option>-auto</option> mode?</para>
	</question>

	<answer>
	  <para>If &man.ppp.8; is dialing unexpectedly, you must
	    determine the cause, and set up Dial filters (dfilters) to
	    prevent such dialing.</para>

	  <para>To determine the cause, use the following line:</para>

	  <programlisting>set log +tcp/ip</programlisting>

	  <para>This will log all traffic through the connection.  The
	    next time the line comes up unexpectedly, you will see the
	    reason logged with a convenient timestamp next to it.</para>

	  <para>You can now disable dialing under these circumstances.
	    Usually, this sort of problem arises due to DNS lookups.  To
	    prevent DNS lookups from establishing a connection (this
	    will <emphasis>not</emphasis> prevent &man.ppp.8; from
	    passing the packets through an established connection), use
	    the following:</para>

	  <programlisting>set dfilter 1 deny udp src eq 53
set dfilter 2 deny udp dst eq 53
set dfilter 3 permit 0/0 0/0</programlisting>

	  <para>This is not always suitable, as it will effectively
	    break your demand-dial capabilities &mdash; most programs
	    will need a DNS lookup before doing any other network
	    related things.</para>

	  <para>In the DNS case, you should try to determine what is
	    actually trying to resolve a host name.  A lot of the time,
	    &man.sendmail.8; is the culprit.  You should make sure that
	    you tell <application>sendmail</application> not to do any
	    DNS lookups in its configuration file.  See the section on
	    <ulink
	      url="&url.books.handbook;/smtp-dialup.html">using email with a dialup connection</ulink>
	    in the &os; Handbook for details on how to create your own
	    configuration file and what should go into it.  You may also
	    want to add the following line to your
	    <filename>.mc</filename> file:</para>

	  <programlisting>define(`confDELIVERY_MODE', `d')dnl</programlisting>

	  <para>This will make <application>sendmail</application> queue
	    everything until the queue is run (usually, sendmail is
	    invoked with <option>-bd -q30m</option>, telling it to run
	    the queue every 30 minutes) or until a <command>sendmail
	    <option>-q</option></command> is done (perhaps from your
	    <filename>ppp.linkup</filename> file).</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ccp-errors">
	  <para>What do these CCP errors mean?</para>
	</question>

	<answer>
	  <para>I keep seeing the following errors in my log
	    file:</para>

	  <programlisting>CCP: CcpSendConfigReq
CCP: Received Terminate Ack (1) state = Req-Sent (6)</programlisting>

	  <para>This is because &man.ppp.8; is trying to negotiate
	    Predictor1 compression, and the peer does not want to
	    negotiate any compression at all.  The messages are
	    harmless, but if you wish to remove them, you can disable
	    Predictor1 compression locally too:</para>

	  <programlisting>disable pred1</programlisting>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-connectionspeed">
	  <para>Why does &man.ppp.8; not log my connection speed?</para>
	</question>

	<answer>
	  <para>In order to log all lines of your modem
	    <quote>conversation</quote>, you must enable the
	    following:</para>

	  <programlisting>set log +connect</programlisting>

	  <para>This will make &man.ppp.8; log everything up until the
	    last requested <quote>expect</quote> string.</para>

	  <para>If you wish to see your connect speed and are using PAP
	    or CHAP (and therefore do not have anything to
	    <quote>chat</quote> after the CONNECT in the dial script
	    &mdash; no <literal>set login</literal> script), you must
	    make sure that you instruct &man.ppp.8; to
	    <quote>expect</quote> the whole CONNECT line, something like
	    this:</para>

	  <programlisting>set dial "ABORT BUSY ABORT NO\\sCARRIER TIMEOUT 4 \
  \"\" ATZ OK-ATZ-OK ATDT\\T TIMEOUT 60 CONNECT \\c \\n"</programlisting>

	  <para>Here, we get our CONNECT, send nothing, then expect a
	    line-feed, forcing &man.ppp.8; to read the whole CONNECT
	    response.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-ignores-backslash">
	  <para>Why does &man.ppp.8; ignore the <literal>\</literal>
	    character in my chat script?</para>
	</question>

	<answer>
	  <para>The <application>ppp</application> utility parses each
	    line in your config files so that it can interpret strings
	    such as <literal>set phone "123 456 789"</literal> correctly
	    and realize that the number is actually only
	    <emphasis>one</emphasis> argument.  In order to specify a
	    <literal>&quot;</literal> character, you must escape it
	    using a backslash (<literal>\</literal>).</para>

	  <para>When the chat interpreter parses each argument, it
	    re-interprets the argument in order to find any special
	    escape sequences such as <literal>\P</literal> or
	    <literal>\T</literal> (see the manual page).  As a result of
	    this double-parsing, you must remember to use the correct
	    number of escapes.</para>

	  <para>If you wish to actually send a <literal>\</literal>
	    character to (say) your modem, you would need something
	    like:</para>

	  <programlisting>set dial "\"\" ATZ OK-ATZ-OK AT\\\\X OK"</programlisting>

	  <para>It will result in the following sequence:</para>

	  <programlisting>ATZ
OK
AT\X
OK</programlisting>

	  <para>Or:</para>

	  <programlisting>set phone 1234567
set dial "\"\" ATZ OK ATDT\\T"</programlisting>

	  <para>It will result in the following sequence:</para>

	  <programlisting>ATZ
OK
ATDT1234567</programlisting>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-segfault-nocore">
	  <para>Why does &man.ppp.8; get a <errorname>Segmentation
	    fault</errorname>, but I see no <filename>ppp.core</filename>
	    file?</para>
	</question>

	<answer>
	  <para>The <application>ppp</application> utility (or any other
	    program for that matter) should never dump core.  Because
	    &man.ppp.8; runs with an effective user&nbsp;ID of
	    <literal>0</literal>, the operating system will not write
	    core image of &man.ppp.8; to disk before terminating it.
	    If, however &man.ppp.8; is actually terminating due to a
	    segmentation violation or some other signal that normally
	    causes core to be dumped, <emphasis>and</emphasis> you are
	    sure you are using the latest version (see the start of this
	    section), then you should install the system sources and do
	    the following:</para>

	  <screen>&prompt.root;  <userinput><command>cd</command> <filename class="directory">/usr/src/usr.sbin/ppp</filename></userinput>
&prompt.root; <userinput><command>echo</command> <makevar>STRIP</makevar>= &gt;&gt; <filename>/etc/make.conf</filename></userinput>
&prompt.root; <userinput><command>echo</command> <makevar>CFLAGS</makevar>+=<option>-g</option> &gt;&gt; <filename>/etc/make.conf</filename></userinput>
&prompt.root; <userinput><command>make</command> <maketarget>install</maketarget> <maketarget>clean</maketarget></userinput></screen>

	  <para>You will now have a debuggable version of &man.ppp.8;
	    installed.  You will have to be <username>root</username> to
	    run &man.ppp.8; as all of its privileges have been revoked.
	    When you start &man.ppp.8;, take a careful note of what your
	    current directory was at the time.</para>

	  <para>Now, if and when &man.ppp.8; receives the segmentation
	    violation, it will dump a core file called
	    <filename>ppp.core</filename>.  You should then do the
	    following:</para>

	  <screen>&prompt.user; <userinput>su</userinput>
&prompt.root; <userinput>gdb /usr/sbin/ppp ppp.core</userinput>
<prompt>(gdb)</prompt> <userinput>bt</userinput>
.....
<prompt>(gdb)</prompt> <userinput>f 0</userinput>
....
<prompt>(gdb)</prompt> <userinput>i args</userinput>
....
<prompt>(gdb)</prompt> <userinput>l</userinput>
.....</screen>

	  <para>All of this information should be given alongside your
	    question, making it possible to diagnose the problem.</para>

	  <para>If you are familiar with &man.gdb.1;, you may wish to
	    find out some other bits and pieces such as what actually
	    caused the dump or the addresses and values of the relevant
	    variables.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-autodialprocess-noconnect">
	  <para>Why does the process that forces a dial in
	    <option>-auto</option> mode never connect?</para>
	</question>

	<answer>
	  <para>This was a known problem with &man.ppp.8; set up to
	    negotiate a dynamic local IP number with the peer in
	    <option>-auto</option> mode.  It has been fixed a long time
	    ago &mdash; search the manual page for
	    <literal>iface</literal>.</para>

	  <para>The problem was that when that initial program calls
	    &man.connect.2;, the IP number of the &man.tun.4; interface
	    is assigned to the socket endpoint.  The kernel creates the
	    first outgoing packet and writes it to the &man.tun.4;
	    device.  &man.ppp.8; then reads the packet and establishes a
	    connection.  If, as a result of &man.ppp.8;'s dynamic IP
	    assignment, the interface address is changed, the original
	    socket endpoint will be invalid.  Any subsequent packets
	    sent to the peer will usually be dropped.  Even if they are
	    not, any responses will not route back to the originating
	    machine as the IP number is no longer owned by that
	    machine.</para>

	  <para>There are several theoretical ways to approach this
	    problem.  It would be nicest if the peer would re-assign the
	    same IP number if possible.  The current version of
	    &man.ppp.8; does this, but most other implementations do
	    not.</para>

	  <para>The easiest method from our side would be to never
	    change the &man.tun.4; interface IP number, but instead to
	    change all outgoing packets so that the source IP number is
	    changed from the interface IP to the negotiated IP on the
	    fly.  This is essentially what the
	    <literal>iface-alias</literal> option in the latest version
	    of &man.ppp.8; is doing (with the help of &man.libalias.3;
	    and &man.ppp.8;'s <option>-nat</option> switch) &mdash; it
	    is maintaining all previous interface addresses and NATing
	    them to the last negotiated address.</para>

	  <para>Another alternative (and probably the most reliable)
	    would be to implement a system call that changes all bound
	    sockets from one IP to another.  &man.ppp.8; would use this
	    call to modify the sockets of all existing programs when a
	    new IP number is negotiated.  The same system call could be
	    used by <acronym>DHCP</acronym> clients when they are forced
	    to call the <function>bind()</function> function for their
	    sockets.</para>

	  <para>Yet another possibility is to allow an interface to be
	    brought up without an IP number.  Outgoing packets would be
	    given an IP number of <hostid
	      role="ipaddr">255.255.255.255</hostid> up until the first
	    <literal>SIOCAIFADDR</literal> &man.ioctl.2; is done.  This
	    would result in fully binding the socket.  It would be up to
	    &man.ppp.8; to change the source IP number, but only if it
	    is set to <hostid role="ipaddr">255.255.255.255</hostid>,
	    and only the IP number and IP checksum would need to change.
	    This, however is a bit of a hack as the kernel would be
	    sending bad packets to an improperly configured interface,
	    on the assumption that some other mechanism is capable of
	    fixing things retrospectively.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ppp-nat-games">
	  <para>Why do most games not work with the
	    <option>-nat</option> switch?</para>
	</question>

	<answer>
	  <para>The reason games and the like do not work when
	    &man.libalias.3; is in use is that the machine on the outside
	    will try to open a connection or send (unsolicited) UDP
	    packets to the machine on the inside.  The NAT software does
	    not know that it should send these packets to the interior
	    machine.</para>

	  <para>To make things work, make sure that the only thing
	    running is the software that you are having problems with,
	    then either run &man.tcpdump.1; on the &man.tun.4; interface
	    of the gateway or enable &man.ppp.8; TCP/IP logging
	    (<literal>set log +tcp/ip</literal>) on the gateway.</para>

	  <para>When you start the offending software, you should see
	    packets passing through the gateway machine.  When something
	    comes back from the outside, it will be dropped (that is the
	    problem).  Note the port number of these packets then shut
	    down the offending software.  Do this a few times to see if
	    the port numbers are consistent.  If they are, then the
	    following line in the relevant section of
	    <filename>/etc/ppp/ppp.conf</filename> will make the
	    software functional:</para>

	  <programlisting>nat port <replaceable>proto</replaceable> <replaceable>internalmachine</replaceable>:<replaceable>port</replaceable> <replaceable>port</replaceable></programlisting>

	  <para>where <replaceable>proto</replaceable> is either
	    <literal>tcp</literal> or <literal>udp</literal>,
	    <replaceable>internalmachine</replaceable> is the machine
	    that you want the packets to be sent to and
	    <replaceable>port</replaceable> is the destination port
	    number of the packets.</para>

	  <para>You will not be able to use the software on other
	    machines without changing the above command, and running the
	    software on two internal machines at the same time is out of
	    the question &mdash; after all, the outside world is seeing
	    your entire internal network as being just a single
	    machine.</para>

	  <para>If the port numbers are not consistent, there are three
	    more options:</para>

	  <orderedlist>
	    <listitem>
	      <para>Submit support in &man.libalias.3;.  Examples of
		<quote>special cases</quote> can be found in
		<filename>/usr/src/sys/netinet/libalias/alias_*.c</filename>
		(<filename>alias_ftp.c</filename> is a good prototype).
		This usually involves reading certain recognized
		outgoing packets, identifying the instruction that tells
		the outside machine to initiate a connection back to the
		internal machine on a specific (random) port and setting
		up a <quote>route</quote> in the alias table so that the
		subsequent packets know where to go.</para>

	      <para>This is the most difficult solution, but it is the
		best and will make the software work with multiple
		machines.</para>
	    </listitem>

	    <listitem>
	      <para>Use a proxy.  The application may support
		<literal>socks5</literal> for example, or (as in the
		<command>cvsup</command> case) may have a
		<quote>passive</quote> option that avoids ever
		requesting that the peer open connections back to the
		local machine.</para>
	    </listitem>

	    <listitem>
	      <para>Redirect everything to the internal machine using
		<literal>nat addr</literal>.  This is the sledge-hammer
		approach.</para>
	    </listitem>
	  </orderedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="fcs-errors">
	  <para>What are FCS errors?</para>
	</question>

	<answer>
	  <para>FCS stands for <literal>F</literal>rame
	    <literal>C</literal>heck <literal>S</literal>equence.  Each
	    PPP packet has a checksum attached to ensure that the data
	    being received is the data being sent.  If the FCS of an
	    incoming packet is incorrect, the packet is dropped and the
	    HDLC FCS count is increased.  The HDLC error values can be
	    displayed using the <literal>show hdlc</literal>
	    command.</para>

	  <para>If your link is bad (or if your serial driver is
	    dropping packets), you will see the occasional FCS error.
	    This is not usually worth worrying about although it does
	    slow down the compression protocols substantially.  If you
	    have an external modem, make sure your cable is properly
	    shielded from interference &mdash; this may eradicate the
	    problem.</para>

	  <para>If your link freezes as soon as you have connected and
	    you see a large number of FCS errors, this may be because your
	    link is not 8-bit clean.  Make sure your modem is not using
	    software flow control (XON/XOFF).  If your datalink
	    <emphasis>must</emphasis> use software flow control, use the
	    command <literal>set accmap 0x000a0000</literal> to tell
	    &man.ppp.8; to escape the <literal>^Q</literal> and
	    <literal>^S</literal> characters.</para>

	  <para>Another reason for seeing too many FCS errors may be
	    that the remote end has stopped talking
	    <acronym>PPP</acronym>.  You may want to enable
	    <literal>async</literal> logging at this point to determine
	    if the incoming data is actually a login or shell prompt.
	    If you have a shell prompt at the remote end, it is possible
	    to terminate &man.ppp.8; without dropping the line by using
	    the <command>close lcp</command> command (a following
	    <command>term</command> command) will reconnect you to the
	    shell on the remote machine.</para>

	  <para>If nothing in your log file indicates why the link might
	    have been terminated, you should ask the remote
	    administrator (your ISP?) why the session was
	    terminated.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="desperation">
	  <para>None of this helps &mdash; I am desperate!  What can I
	    do?</para>
	</question>

	<answer>
	  <para>If all else fails, send as much information as you can,
	    including your config files, how you are starting
	    &man.ppp.8;, the relevant parts of your log file and the
	    output of the <command>netstat -rn</command> command (before
	    and after connecting) to the &a.questions; or the <ulink
	      url="news:comp.unix.bsd.freebsd.misc">comp.unix.bsd.freebsd.misc</ulink>
	    news group, and someone should point you in the right
	    direction.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="serial">
    <title>Serial Communications</title>

    <para>This section answers common questions about serial
      communications with &os;.  PPP and SLIP are covered in the <link
      linkend="networking">Networking</link> section.</para>

    <qandaset>
      <qandaentry>
	<question id="found-serial">
	  <para>How do I tell if &os; found my serial ports?</para>
	</question>

	<answer>
	  <para>As the &os; kernel boots, it will probe for the serial
	    ports in your system for which the kernel was configured.
	    You can either watch your system closely for the messages it
	    prints or run this command after your system is up and
	    running:</para>

	  <screen>&prompt.user; <userinput>dmesg | grep -E "^sio[0-9]"</userinput></screen>

	  <para>Here is some example output from the above
	    command:</para>

	  <programlisting>sio0: &lt;16550A-compatible COM port&gt; port 0x3f8-0x3ff irq 4 flags 0x10 on acpi0
sio0: type 16550A
sio1: &lt;16550A-compatible COM port&gt; port 0x2f8-0x2ff irq 3 on acpi0
sio1: type 16550A</programlisting>

	  <para>This shows two serial ports.  The first is on
	    IRQ&nbsp;4, is using port address <literal>0x3f8</literal>,
	    and has a 16550A-type UART chip.  The second uses the same
	    kind of chip but is on IRQ&nbsp;3 and is at port address
	    <literal>0x2f8</literal>.  Internal modem cards are treated
	    just like serial ports &mdash; except that they always have
	    a modem <quote>attached</quote> to the port.</para>

	  <para>The <filename>GENERIC</filename> kernel includes support
	    for two serial ports using the same IRQ and port address
	    settings in the above example.  If these settings are not
	    right for your system, or if you have added modem cards or
	    have more serial ports than your kernel is configured for,
	    just reconfigure your kernel.  See section <link
	      linkend="make-kernel">about building a kernel</link> for
	    more details.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="found-modem">
	  <para>How do I tell if &os; found my modem cards?</para>
	</question>

	<answer>
	  <para>Refer to the answer to the previous question.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="access-serial-ports">
	  <para>How do I access the serial ports on &os;?</para>
	</question>

	<answer>
	  <para>The third serial port, <devicename>sio2</devicename>
	    (see &man.sio.4;, known as <devicename>COM3</devicename> in
	    DOS), is on <devicename>/dev/cuad2</devicename> for dial-out
	    devices, and on <devicename>/dev/ttyd2</devicename> for
	    dial-in devices.  What is the difference between these two
	    classes of devices?</para>

	  <para>You use
	    <devicename>ttyd<replaceable>X</replaceable></devicename>
	    for dial-ins.  When opening
	    <devicename>/dev/ttyd<replaceable>X</replaceable></devicename>
	    in blocking mode, a process will wait for the corresponding
	    <devicename>cuad<replaceable>X</replaceable></devicename>
	    device to become inactive, and then wait for the carrier
	    detect line to go active.  When you open the
	    <devicename>cuad<replaceable>X</replaceable></devicename>
	    device, it makes sure the serial port is not already in use
	    by the
	    <devicename>ttyd<replaceable>X</replaceable></devicename>
	    device.  If the port is available, it <quote>steals</quote>
	    it from the
	    <devicename>ttyd<replaceable>X</replaceable></devicename>
	    device.  Also, the
	    <devicename>cuad<replaceable>X</replaceable></devicename>
	    device does not care about carrier detect.  With this scheme
	    and an auto-answer modem, you can have remote users log in
	    and you can still dial out with the same modem and the
	    system will take care of all the conflicts.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="enable-multiport-serial">
	  <para>How do I enable support for a multiport serial
	    card?</para>
	</question>

	<answer>
	  <para>Again, the section on kernel configuration provides
	    information about configuring your kernel.  For a multiport
	    serial card, place an &man.sio.4; line for each serial port
	    on the card in the &man.device.hints.5; file.  But place the
	    IRQ specifiers on only one of the entries.  All of the ports
	    on the card should share one IRQ.  For consistency, use the
	    last serial port to specify the IRQ.  Also, specify the
	    following option in the kernel configuration file:</para>

	  <programlisting>options COM_MULTIPORT</programlisting>

	  <para>The following <filename>/boot/device.hints</filename>
	    example is for an AST 4-port serial card on
	    IRQ&nbsp;12:</para>

	  <programlisting>hint.sio.4.at="isa"
hint.sio.4.port="0x2a0"
hint.sio.4.flags="0x701"
hint.sio.5.at="isa"
hint.sio.5.port="0x2a8"
hint.sio.5.flags="0x701"
hint.sio.6.at="isa"
hint.sio.6.port="0x2b0"
hint.sio.6.flags="0x701"
hint.sio.7.at="isa"
hint.sio.7.port="0x2b8"
hint.sio.7.flags="0x701"
hint.sio.7.irq="12"</programlisting>

	  <para>The flags indicate that the master port has minor number
	    <literal>7</literal> (<literal>0x700</literal>), and all the
	    ports share an IRQ (<literal>0x001</literal>).</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="multiport-serial-share-irq">
	  <para>Can &os; handle multiport serial cards sharing
	    IRQs?</para>
	</question>

	<answer>
	  <para>Not yet.  You will have to use a different IRQ for each
	    card.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="default-serial-params">
	  <para>Can I set the default serial parameters for a
	    port?</para>
	</question>

	<answer>
	  <para>See the <ulink
	      url="&url.books.handbook;/serial.html#SERIAL-HW-CONFIG">Serial Communications</ulink>
	    section in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="enable-dialup">
	  <para>How can I enable dialup logins on my modem?</para>
	</question>

	<answer>
	  <para>Please read the section about <ulink
	      url="&url.books.handbook;/dialup.html">Dial-in Services</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="dumb-terminal">
	  <para>How can I connect a dumb terminal to my &os; box?</para>
	</question>

	<answer>
	  <para>You can find this information in the <ulink
	      url="&url.books.handbook;/term.html">Terminals</ulink>
	    section of the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="cannot-tip">
	  <para>Why can I not run <command>tip</command> or
	    <command>cu</command>?</para>
	</question>

	<answer>
	  <para>On your system, the programs &man.tip.1; and &man.cu.1;
	    can only access the <filename class="directory">/var/spool/lock</filename>
	    directory via user <username>uucp</username> and group
	    <groupname>dialer</groupname>.  You can use the group
	    <groupname>dialer</groupname> to control who has access to
	    your modem or remote systems.  Just add yourself to group
	    <groupname>dialer</groupname>.</para>

	  <para>Alternatively, you can let everyone on your system run
	    &man.tip.1; and &man.cu.1; by typing:</para>

	  <screen>&prompt.root; <userinput>chmod 4511 /usr/bin/cu</userinput>
&prompt.root; <userinput>chmod 4511 /usr/bin/tip</userinput></screen>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="hayes-unsupported">
	  <para>My stock Hayes modem is not supported &mdash; what can I
	    do?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#HAYES-UNSUPPORTED">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="direct-at">
	  <para>How am I expected to enter these AT commands?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#DIRECT-AT">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="gt-failure">
	  <para>Why does the <literal>@</literal> sign for the
	    <literal>pn</literal> capability not
	    work?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#GT-FAILURE">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="dial-command-line">
	  <para>How can I dial a phone number on the command
	    line?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#DIAL-COMMAND-LINE">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="set-bps">
	  <para>Do I have to type in the bps rate every time I do
	    that?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#SET-BPS">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="terminal-server">
	  <para>How can I more easily access a number of hosts through a
	    terminal server?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#TERMINAL-SERVER">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="tip-multiline">
	  <para>Can tip try more than one line for each site?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#TIP-MULTILINE">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="multi-controlp">
	  <para>Why do I have to hit <keycombo
	      action="simul"><keycap>Ctrl</keycap><keycap>P</keycap></keycombo>
	    twice to send <keycombo
	      action="simul"><keycap>Ctrl</keycap><keycap>P</keycap></keycombo>
	    once?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#MULTI-CONTROLP">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="uppercase">
	  <para>Why is everything I type suddenly in UPPER CASE?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#UPPERCASE">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="tip-filetransfer">
	  <para>How can I do file transfers with
	    <command>tip</command>?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#TIP-FILETRANSFER">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="zmodem-tip">
	  <para>How can I run zmodem with
	    <command>tip</command>?</para>
	</question>

	<answer>
	  <para>See <ulink
	      url="&url.books.handbook;/dialout.html#ZMODEM-TIP">this answer</ulink>
	    in the &os; Handbook.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="misc">
    <title>Miscellaneous Questions</title>

    <qandaset>
      <qandaentry>
	<question id="more-swap">
	  <para>&os; uses far more swap space than &linux;.  Why?</para>
	</question>

	<answer>
	  <para>&os; only appears to use more swap than &linux;.  In
	    actual fact, it does not.  The main difference between &os;
	    and &linux; in this regard is that &os; will proactively
	    move entirely idle, unused pages of main memory into swap in
	    order to make more main memory available for active use.
	    &linux; tends to only move pages to swap as a last resort.
	    The perceived heavier use of swap is balanced by the more
	    efficient use of main memory.</para>

	  <para>Note that while &os; is proactive in this regard, it
	    does not arbitrarily decide to swap pages when the system is
	    truly idle.  Thus you will not find your system all paged
	    out when you get up in the morning after leaving it idle
	    overnight.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="top-freemem">
	  <para>Why does <command>top</command> show very little free
	    memory even when I have very few programs running?</para>
	</question>

	<answer>
	  <para>The simple answer is that free memory is wasted memory.
	    Any memory that your programs do not actively allocate is
	    used within the &os; kernel as disk cache.  The values shown
	    by &man.top.1; labeled as <literal>Inact</literal>,
	    <literal>Cache</literal>, and <literal>Buf</literal> are all
	    cached data at different aging levels.  This cached data
	    means the system does not have to access a slow disk again
	    for data it has accessed recently, thus increasing overall
	    performance.  In general, a low value shown for
	    <literal>Free</literal> memory in &man.top.1; is good,
	    provided it is not <emphasis>very</emphasis> low.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="chmod-symlinks">
	  <para>Why will <command>chmod</command> not change the
	    permissions on symlinks?</para>
	</question>

	<answer>
	  <para>Symlinks do not have permissions, and by default,
	    &man.chmod.1; will follow symlinks to change the
	    permissions on the source file, if possible.  So if you have a file,
	    <filename>foo</filename>, and a symlink to that file,
	    <filename>bar</filename>, then this command will always
	    succeed.</para>

	  <screen>&prompt.user; <userinput>chmod g-w bar</userinput></screen>

	  <para>However, the permissions on <filename>bar</filename>
	    will not have changed.</para>

	  <para>When changing modes of the file hierarchies rooted in the
	    files instead of the files themselves,
	    you have to use either <option>-H</option> or
	    <option>-L</option> together with the <option>-R</option>
	    option to make this work.  See the &man.chmod.1; and
	    &man.symlink.7; manual pages for more info.</para>

	  <warning>
	    <para>The <option>-R</option> option does a
	      <emphasis>recursive</emphasis> &man.chmod.1;.  Be careful
	      about specifying directories or symlinks to directories to
	      &man.chmod.1;.  If you want to change the permissions of a
	      directory referenced by a symlink, use &man.chmod.1;
	      without any options and follow the symlink with a trailing
	      slash (<filename class="directory">/</filename>).  For example, if
	      <filename>foo</filename> is a symlink to directory
	      <filename class="directory">bar</filename>, and you want to change the
	      permissions of <filename>foo</filename> (actually
	      <filename class="directory">bar</filename>), you would do something
	      like:</para>

	    <screen>&prompt.user; <userinput>chmod 555 foo/</userinput></screen>

	    <para>With the trailing slash, &man.chmod.1; will follow the
	      symlink, <filename>foo</filename>, to change the
	      permissions of the directory,
	      <filename class="directory">bar</filename>.</para>
	  </warning>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="dos-binaries">
	  <para>Can I run DOS binaries under &os;?</para>
	</question>

	<answer>
	  <para>Yes, you can use <filename
	      role="package">emulators/doscmd</filename>, a DOS
	    emulation program, available in the &os; Ports
	    Collection.</para>

	  <para>If <application>doscmd</application> will not suffice,
	    the add-on utility <filename
	      role="package">emulators/pcemu</filename> emulates an 8088
	    and enough BIOS services to run many DOS text mode
	    applications.  It requires the X Window System.</para>

	  <para>You may also try <filename
	      role="package">emulators/dosbox</filename> from the &os;
	    Ports Collection.  The main focus of this application is
	    emulating old DOS games using the local file system for
	    files.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="translation">
	  <para>What do I need to do to translate a &os; document into
	    my native language?</para>
	</question>

	<answer>
	  <para>See the <ulink
	      url="&url.books.fdp-primer;/translations.html">Translation FAQ</ulink>
	    in the &os; Documentation Project Primer.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="freebsd-mail-bounces">
	  <para>Why does my email to any address at <hostid
	      role="domainname">FreeBSD.org</hostid> bounce?</para>
	</question>

	<answer>
	  <para>The <hostid role="domainname">FreeBSD.org</hostid> mail
	    system implements some of the stricter
	    <application>Postfix</application> checks on incoming mail
	    and rejects mail that is either misconfigured or is
	    potential spam.  Your mail might bounce for one of the
	    following reasons:</para>

	  <itemizedlist>
	    <listitem>
	      <para>The email is being sent from a known spam domain or
		IP block.</para>

	      <para>The &os; mail servers reject email from known spam
		sources.  If you have service through a company or
		domain who generates or relays spam, please switch to a
		service provider who does not.</para>
	    </listitem>

	    <listitem>
	      <para>The body of the email only contains HTML.</para>

	      <para>Mail should be sent in plain text only.  Please
		configure your mail user agent to send plain
		text.</para>
	    </listitem>

	    <listitem>
	      <para>The mailer at <hostid
		  role="domainname">FreeBSD.org</hostid> cannot resolve
		the IP address of the connecting host back to a
		symbolic name.</para>

	      <para>Working reverse DNS is a standard requirement for
		accepting mail from a host.  Set up reverse DNS for your
		mail server's IP address.  Many home services (DSL,
		cable, dialup, etc.) will not give you this option.  In
		this case, relay your email through your service
		provider's mail server.</para>
	    </listitem>

	    <listitem>
	      <para>The hostname given in the EHLO/HELO part of the SMTP
		exchange cannot be resolved to an IP address.</para>

	      <para>A fully qualified, resolvable host name is necessary
		in this part of the SMTP dialogue before mail will be
		accepted.  If you do not have a host name that is
		registered in the DNS, then you should use your service
		provider's mail server to relay your mail.</para>
	    </listitem>

	    <listitem>
	      <para>Your message had a message ID ending with the string
		<quote>localhost</quote>.</para>

	      <para>Some mail user agents generate bad message IDs which
		will not be accepted.  You will need to persuade your mail
		user agent to generate a valid message ID or else
		configure your mail transfer agent to rewrite
		them.</para>
	    </listitem>
	  </itemizedlist>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="free-account">
	  <para>Where can I find a free &os; account?</para>
	</question>

	<answer>
	  <para>While &os; does not provide open access to any of their
	    servers, others do provide open access &unix; systems.  The
	    charge varies and limited services may be available.</para>

	  <para><ulink
	      url="http://www.arbornet.org/">Arbornet, Inc</ulink>,
	    also known as <emphasis>M-Net</emphasis>, has been providing
	    open access to &unix; systems since 1983.  Starting on an
	    Altos running System III, the site switched to BSD/OS in
	    1991.  In June of 2000, the site switched again to &os;.
	    <emphasis>M-Net</emphasis> can be accessed via
	    <application>telnet</application> and
	    <application>SSH</application> and provides basic access to
	    the entire &os; software suite.  However, network access is
	    limited to members and patrons who donate to the system,
	    which is run as a non-profit organization.
	    <emphasis>M-Net</emphasis> also provides an bulletin board
	    system and interactive chat.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="sup-define">
	  <para>What is <command>sup</command>, and how do I use
	    it?</para>
	</question>

	<answer>
	  <para><ulink
	      url="http://www.FreeBSD.org/cgi/ports.cgi?^sup">SUP</ulink>
	    stands for Software Update Protocol, and was developed by
	    CMU for keeping their development trees in sync.  It was
	    used to keep remote sites in sync with the Project's central
	    development sources.</para>

	  <para>SUP is not bandwidth friendly, and has been retired.
	    The current recommended method to keep your sources up to
	    date is <ulink
	      url="&url.books.handbook;/synching.html#CVSUP">CVSup</ulink>
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="daemon-name">
	  <para>What is the cute little red guy's name?</para>
	</question>

	<answer>
	  <para>He does not have one, and is just called <quote>the BSD
	    daemon</quote>.  If you insist upon using a name, call him
	    <quote>beastie</quote>.  Note that <quote>beastie</quote> is
	    pronounced <quote>BSD</quote>.</para>

	  <para>You can learn more about the BSD daemon on his <ulink
	      url="http://www.mckusick.com/beastie/index.html">home page</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="use-beastie">
	  <para>Can I use the BSD daemon image?</para>
	</question>

	<answer>
	  <para>Perhaps.  The BSD daemon is copyrighted by Marshall Kirk
	    McKusick.  You will want to check his <ulink
	      url="http://www.mckusick.com/beastie/mainpage/copyright.html">Statement on the Use of the BSD Daemon Figure</ulink>
	    for detailed usage terms.</para>

	  <para>In summary, you are free to use the image in a tasteful
	    manner, for personal use, so long as appropriate credit is
	    given.  If you want to use him commercially, you must
	    contact &a.mckusick;.  More details are available on the
	    <ulink
	      url="http://www.mckusick.com/beastie/index.html">BSD Daemon's home page</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="daemon-images">
	  <para>Do you have any BSD daemon images I could use?</para>
	</question>

	<answer>
	  <para>You will find eps and Xfig drawings under
	    <filename class="directory">/usr/share/examples/BSD_daemon/</filename>.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="glossary">
	  <para>I have seen an acronym or other term on the mailing
	    lists and I do not understand what it means.  Where should I
	    look?</para>
	</question>

	<answer>
	  <para>Please see the <ulink
	      url="&url.books.handbook;/freebsd-glossary.html">&os; Glossary</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="bikeshed-painting">
	  <para>Why should I care what color the bikeshed is?</para>
	</question>

	<answer>
	  <para>The really, really short answer is that you should not.
	    The somewhat longer answer is that just because you are
	    capable of building a bikeshed does not mean you should stop
	    others from building one just because you do not like the
	    color they plan to paint it.  This is a metaphor indicating
	    that you need not argue about every little feature just
	    because you know enough to do so.  Some people have
	    commented that the amount of noise generated by a change is
	    inversely proportional to the complexity of the
	    change.</para>

	  <para>The longer and more complete answer is that after a very
	    long argument about whether &man.sleep.1; should take
	    fractional second arguments, &a.phk; posted a long message
	    entitled <quote><ulink
	      url="http://www.FreeBSD.org/cgi/getmsg.cgi?fetch=506636+517178+/usr/local/www/db/text/1999/freebsd-hackers/19991003.freebsd-hackers">A bike shed (any color will do) on greener grass...</ulink></quote>.
	    The appropriate portions of that message are quoted
	    below.</para>

	  <blockquote>
	    <attribution>&a.phk; on &a.hackers.name;, October 2,
	      1999</attribution>

	    <para><quote>What is it about this bike shed?</quote>  Some
	      of you have asked me.</para>

	    <para>It is a long story, or rather it is an old story, but
	      it is quite short actually.  C. Northcote Parkinson wrote
	      a book in the early 1960s, called <quote>Parkinson's
	      Law</quote>, which contains a lot of insight into the
	      dynamics of management.</para>

	    <para><emphasis>[snip a bit of commentary on the
	      book]</emphasis></para>

	    <para>In the specific example involving the bike shed, the
	      other vital component is an atomic power-plant, I guess
	      that illustrates the age of the book.</para>

	    <para>Parkinson shows how you can go into the board of
	      directors and get approval for building a multi-million or
	      even billion dollar atomic power plant, but if you want to
	      build a bike shed you will be tangled up in endless
	      discussions.</para>

	    <para>Parkinson explains that this is because an atomic
	      plant is so vast, so expensive and so complicated that
	      people cannot grasp it, and rather than try, they fall
	      back on the assumption that somebody else checked all the
	      details before it got this far.  Richard P. Feynmann
	      gives a couple of interesting, and very much to the point,
	      examples relating to Los Alamos in his books.</para>

	    <para>A bike shed on the other hand.  Anyone can build one
	      of those over a weekend, and still have time to watch the
	      game on TV.  So no matter how well prepared, no matter how
	      reasonable you are with your proposal, somebody will seize
	      the chance to show that he is doing his job, that he is
	      paying attention, that he is
	      <emphasis>here</emphasis>.</para>

	    <para>In Denmark we call it <quote>setting your
	      fingerprint</quote>.  It is about personal pride and
	      prestige, it is about being able to point somewhere and
	      say <quote>There!  <emphasis>I</emphasis> did
	      that.</quote> It is a strong trait in politicians, but
	      present in most people given the chance.  Just think about
	      footsteps in wet cement.</para>
	  </blockquote>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="funnies">
    <title>The &os; Funnies</title>

    <qandaset>
      <qandaentry>
	<question id="very-very-cool">
	  <para>How cool is &os;?</para>
	</question>

	<answer>
	  <para>Q.  Has anyone done any temperature testing while
	    running &os;?  I know &linux; runs cooler than DOS, but have
	    never seen a mention of &os;.  It seems to run really
	    hot.</para>

	  <para>A.  No, but we have done numerous taste tests on
	    blindfolded volunteers who have also had 250 micrograms of
	    LSD-25 administered beforehand.  35% of the volunteers said
	    that &os; tasted sort of orange, whereas &linux; tasted like
	    purple haze.  Neither group mentioned any significant
	    variances in temperature.  We eventually had to throw the
	    results of this survey out entirely anyway when we found
	    that too many volunteers were wandering out of the room
	    during the tests, thus skewing the results.  We think most
	    of the volunteers are at Apple now, working on their new
	    <quote>scratch and sniff</quote> GUI.  It is a funny old
	    business we are in!</para>

	  <para>Seriously, both &os; and &linux; use the
	    <acronym>HLT</acronym> (halt) instruction when the system is
	    idle thus lowering its energy consumption and therefore the
	    heat it generates.  Also if you have APM (advanced power
	    management) configured, then &os; can also put the CPU into
	    a low power mode.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="letmeoutofhere">
	  <para>Who is scratching in my memory banks??</para>
	</question>

	<answer>
	  <para>Q.  Is there anything <quote>odd</quote> that &os; does
	    when compiling the kernel which would cause the memory to
	    make a scratchy sound?  When compiling (and for a brief
	    moment after recognizing the floppy drive upon startup, as
	    well), a strange scratchy sound emanates from what appears
	    to be the memory banks.</para>

	  <para>A.  Yes!  You will see frequent references to
	    <quote>daemons</quote> in the BSD documentation, and what
	    most people do not know is that this refers to genuine,
	    non-corporeal entities that now possess your computer.  The
	    scratchy sound coming from your memory is actually
	    high-pitched whispering exchanged among the daemons as they
	    best decide how to deal with various system administration
	    tasks.</para>

	  <para>If the noise gets to you, a good
	    <command>fdisk /mbr</command> from DOS will get rid of them,
	    but do not be surprised if they react adversely and try to
	    stop you.  In fact, if at any point during the exercise you
	    hear the satanic voice of Bill Gates coming from the
	    built-in speaker, take off running and do not ever look
	    back!  Freed from the counterbalancing influence of the BSD
	    daemons, the twin demons of DOS and &windows; are often able
	    to re-assert total control over your machine to the eternal
	    damnation of your soul.  Now that you know, given a choice
	    you would probably prefer to get used to the scratchy
	    noises, no?</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="changing-lightbulbs">
	  <para>How many &os; hackers does it take to change a
	    lightbulb?</para>
	</question>

	<answer>
	  <para>One thousand, one hundred and sixty-nine:</para>

	  <para>Twenty-three to complain to -CURRENT about the lights
	    being out;</para>

	  <para>Four to claim that it is a configuration problem, and
	    that such matters really belong on -questions;</para>

	  <para>Three to submit PRs about it, one of which is misfiled
	    under doc and consists only of <quote>it's
	    dark</quote>;</para>

	  <para>One to commit an untested lightbulb which breaks
	    buildworld, then back it out five minutes later;</para>

	  <para>Eight to flame the PR originators for not including
	    patches in their PRs;</para>

	  <para>Five to complain about buildworld being broken;</para>

	  <para>Thirty-one to answer that it works for them, and they
	    must have cvsupped at a bad time;</para>

	  <para>One to post a patch for a new lightbulb to
	    -hackers;</para>

	  <para>One to complain that he had patches for this three years
	    ago, but when he sent them to -CURRENT they were just
	    ignored, and he has had bad experiences with the PR system;
	    besides, the proposed new lightbulb is non-reflexive;</para>

	  <para>Thirty-seven to scream that lightbulbs do not belong in
	    the base system, that committers have no right to do things
	    like this without consulting the Community, and WHAT IS
	    -CORE DOING ABOUT IT!?</para>

	  <para>Two hundred to complain about the color of the bicycle
	    shed;</para>

	  <para>Three to point out that the patch breaks
	    &man.style.9;;</para>

	  <para>Seventeen to complain that the proposed new lightbulb is
	    under GPL;</para>

	  <para>Five hundred and eighty-six to engage in a flame war
	    about the comparative advantages of the GPL, the BSD
	    license, the MIT license, the NPL, and the personal hygiene
	    of unnamed FSF founders;</para>

	  <para>Seven to move various portions of the thread to -chat
	    and -advocacy;</para>

	  <para>One to commit the suggested lightbulb, even though it
	    shines dimmer than the old one;</para>

	  <para>Two to back it out with a furious flame of a commit
	    message, arguing that &os; is better off in the dark than
	    with a dim lightbulb;</para>

	  <para>Forty-six to argue vociferously about the backing out of
	    the dim lightbulb and demanding a statement from
	    -core;</para>

	  <para>Eleven to request a smaller lightbulb so it will fit
	    their Tamagotchi if we ever decide to port &os; to that
	    platform;</para>

	  <para>Seventy-three to complain about the SNR on -hackers and
	    -chat and unsubscribe in protest;</para>

	  <para>Thirteen to post <quote>unsubscribe</quote>, <quote>How
	    do I unsubscribe?</quote>, or <quote>Please remove me from
	    the list</quote>, followed by the usual footer;</para>

	  <para>One to commit a working lightbulb while everybody is too
	    busy flaming everybody else to notice;</para>

	  <para>Thirty-one to point out that the new lightbulb would
	    shine 0.364% brighter if compiled with TenDRA (although it
	    will have to be reshaped into a cube), and that &os; should
	    therefore switch to TenDRA instead of GCC;</para>

	  <para>One to complain that the new lightbulb lacks
	    fairings;</para>

	  <para>Nine (including the PR originators) to ask <quote>what
	    is MFC?</quote>;</para>

	  <para>Fifty-seven to complain about the lights being out two
	    weeks after the bulb has been changed.</para>

	  <para><emphasis>&a.nik; adds:</emphasis></para>

	  <para><emphasis>I was laughing quite hard at
	    this.</emphasis></para>

	  <para><emphasis>And then I thought, <quote>Hang on, shouldn't
	    there be '1 to document it.' in that list
	    somewhere?</quote></emphasis></para>

	  <para><emphasis>And then I was enlightened
	    :-)</emphasis></para>

	  <para><emphasis>&a.tabthorpe;</emphasis> says: <quote>None,
	    <emphasis>real</emphasis> &os; hackers are not afraid of the
	    dark!</quote></para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="dev-null">
	  <para>Where does data written to
	    <filename>/dev/null</filename> go?</para>
	</question>

	<answer>
	  <para>It goes into a special data sink in the CPU where it is
	    converted to heat which is vented through the heatsink / fan
	    assembly.  This is why CPU cooling is increasingly
	    important; as people get used to faster processors, they
	    become careless with their data and more and more of it ends
	    up in <filename>/dev/null</filename>, overheating their
	    CPUs.  If you delete <filename>/dev/null</filename> (which
	    effectively disables the CPU data sink) your CPU may run
	    cooler but your system will quickly become constipated with
	    all that excess data and start to behave erratically.  If
	    you have a fast network connection you can cool down your
	    CPU by reading data out of <filename>/dev/random</filename>
	    and sending it off somewhere; however you run the risk of
	    overheating your network connection and
	    <filename class="directory">/</filename> or angering your ISP, as most of the
	    data will end up getting converted to heat by their
	    equipment, but they generally have good cooling, so if you
	    do not overdo it you should be OK.</para>

	  <para><emphasis>Paul Robinson adds:</emphasis></para>

	  <para>There are other methods.  As every good sysadmin knows,
	    it is part of standard practice to send data to the screen
	    of interesting variety to keep all the pixies that make up
	    your picture happy.  Screen pixies (commonly mis-typed or
	    re-named as <quote>pixels</quote>) are categorized by the
	    type of hat they wear (red, green or blue) and will hide or
	    appear (thereby showing the color of their hat) whenever
	    they receive a little piece of food.  Video cards turn data
	    into pixie-food, and then send them to the pixies &mdash;
	    the more expensive the card, the better the food, so the
	    better behaved the pixies are.  They also need constant
	    stimulation &mdash; this is why screen savers exist.</para>

	  <para>To take your suggestions further, you could just throw
	    the random data to console, thereby letting the pixies
	    consume it.  This causes no heat to be produced at all,
	    keeps the pixies happy and gets rid of your data quite
	    quickly, even if it does make things look a bit messy on
	    your screen.</para>

	  <para>Incidentally, as an ex-admin of a large ISP who
	    experienced many problems attempting to maintain a stable
	    temperature in a server room, I would strongly discourage
	    people sending the data they do not want out to the network.
	    The fairies who do the packet switching and routing get
	    annoyed by it as well.</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="advanced">
    <title>Advanced Topics</title>

    <qandaset>
      <qandaentry>
	<question id="learn-advanced">
	  <para>How can I learn more about &os;'s internals?</para>
	</question>

	<answer>
	  <para>At this time, there is only one book on &os;-specific OS
	    internals, namely <quote>The Design and Implementation of
	    the &os; Operating System</quote> by Marshall Kirk McKusick
	    and George V. Neville-Neil, ISBN 0-201-70245-2, which
	    focuses on version 5.<replaceable>X</replaceable> of
	    &os;.</para>

	  <para>Additionally, much general &unix; knowledge is directly
	    applicable to &os;.</para>

	  <para>For a list of relevant books, please check the
	    Handbook's <ulink
	      url="&url.books.handbook;/bibliography-osinternals.html">Operating System Internals Bibliography</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="how-to-contribute">
	  <para>How can I contribute to &os;?</para>
	</question>

	<answer>
	  <para>Please see the article on <ulink
	      url="&url.articles.contributing;/article.html">Contributing to &os;</ulink>
	    for specific advice on how to do this.  Assistance is more
	    than welcome!</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="define-snap-release">
	  <para>What are snapshots and releases?</para>
	</question>

	<answer>
	  <para>There are currently four active/semi-active branches in
	    the &os; <ulink
	      url="http://www.FreeBSD.org/cgi/cvsweb.cgi">CVS Repository</ulink>.
	    (Earlier branches are only changed very rarely, which is why
	    there are only four active branches of development):</para>

	  <itemizedlist>
	    <listitem>
	      <para>&rel3.releng; AKA
		&rel3.stable;</para>
	    </listitem>

	    <listitem>
	      <para>&rel2.releng; AKA
		&rel2.stable;</para>
	    </listitem>

	    <listitem>
	      <para>&rel.releng; AKA
		&rel.stable;</para>
	    </listitem>

	    <listitem>
	      <para>&rel.head.releng; AKA
		<emphasis>-CURRENT</emphasis> AKA
		&rel.head;</para>
	    </listitem>
	  </itemizedlist>

	  <para><literal>HEAD</literal> is not an actual branch tag,
	    like the others; it is simply a symbolic constant for
	    <quote><emphasis>the current, non-branched development
	    stream</emphasis></quote> which we simply refer to as
	    <emphasis>-CURRENT</emphasis>.</para>

	  <para>Right now, <emphasis>-CURRENT</emphasis> is the
	    &rel.head.relx; development stream; the
	    &rel.stable; branch,
	    &rel.releng;, forked off from
	    <emphasis>-CURRENT</emphasis> in &rel.relengdate; and the
	    &rel2.stable; branch,
	    &rel2.releng;, forked off from
	    <emphasis>-CURRENT</emphasis> in &rel2.relengdate;.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="custrel">
	  <para>How do I make my own custom release?</para>
	</question>

	<answer>
	  <para>Please see the <ulink
	      url="&url.articles.releng;/article.html">Release Engineering</ulink>
	    article.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="makeworld-clobbers">
	  <para>Why does
	    <command>make <maketarget>world</maketarget></command>
	    clobber my existing installed binaries?</para>
	</question>

	<answer>
	  <para>Yes, this is the general idea; as its name might
	    suggest,
	    <command>make <maketarget>world</maketarget></command>
	    rebuilds every system binary from scratch, so you can be
	    certain of having a clean and consistent environment at the
	    end (which is why it takes so long).</para>

	  <para>If the environment variable <envar>DESTDIR</envar>
	    is defined while running
	    <command>make <maketarget>world</maketarget></command> or
	    <command>make <maketarget>install</maketarget></command>,
	    the newly-created binaries will be deposited in a directory
	    tree identical to the installed one, rooted at
	    <literal>${DESTDIR}</literal>.  Some random combination of
	    shared libraries modifications and program rebuilds can
	    cause this to fail in
	    <command>make <maketarget>world</maketarget></command>
	    however.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="cvsup-round-robin">
	  <para>Why isn't <hostid role="fqdn">cvsup.FreeBSD.org</hostid>
	    a round robin DNS entry to share the load amongst the various
	    <application>CVSup</application> servers?</para>
	</question>

	<answer>
	  <para>While <application>CVSup</application> mirrors update
	    from the master <application>CVSup</application> server
	    hourly, this update might happen at any time during the
	    hour.  This means that some servers have newer code than
	    others, even though all servers have code that is less than
	    an hour old.  If <hostid
	      role="fqdn">cvsup.FreeBSD.org</hostid> was a round robin
	    DNS entry that simply redirected users to a random
	    <application>CVSup</application> server, running
	    <application>CVSup</application> twice in a row could
	    download code older than the code already on the
	    system.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="ctm">
	  <para>Can I follow <emphasis>-CURRENT</emphasis> with limited
	    Internet access?</para>
	</question>

	<answer>
	  <para>Yes, you can do this <emphasis>without</emphasis>
	    downloading the whole source tree by using the <ulink
	      url="&url.books.handbook;/synching.html#CTM">CTM facility</ulink>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="submitting-kernel-extensions">
	  <para>I have written a kernel extension, who do I send it
	    to?</para>
	</question>

	<answer>
	  <para>Please take a look at the article on <ulink
	      url="&url.articles.contributing;/article.html">Contributing to &os;</ulink>
	    to learn how to submit code.</para>

	  <para>And thanks for the thought!</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="pnp-initialize">
	  <para>How are Plug&nbsp;N&nbsp;Play ISA cards detected and
	    initialized?</para>
	</question>

	<answer>
	  <para>By: Frank Durda IV
	    <email>uhclem@nemesis.lonestar.org</email></para>

	  <para>In a nutshell, there a few I/O ports that all of the PnP
	    boards respond to when the host asks if anyone is out there.
	    So when the PnP probe routine starts, it asks if there are
	    any PnP boards present, and all the PnP boards respond with
	    their model # to a I/O read of the same port, so the probe
	    routine gets a wired-OR <quote>yes</quote> to that question.
	    At least one bit will be on in that reply.  Then the probe
	    code is able to cause boards with board model IDs (assigned
	    by &microsoft;/&intel;) lower than <literal>X</literal> to
	    go <quote>off-line</quote>.  It then looks to see if any
	    boards are still responding to the query.  If the answer was
	    <literal>0</literal>, then there are no boards with IDs
	    above <literal>X</literal>.  Probe will then ask for boards
	    below <literal>X</literal>.  Finally, probe requests boards
	    greater than
	    <literal>X&nbsp;-&nbsp;(limit&nbsp;/&nbsp;4)</literal> to go
	    off-line.  It then repeats this query.  By repeating this
	    semi-binary search of IDs-in-range enough times, the probing
	    code will eventually identify all PnP boards present in a
	    given machine with a number of iterations that is much lower
	    than what 2<superscript>64</superscript> would take.</para>

	  <para>The IDs are two 32-bit fields (hence
	    2<superscript>64</superscript>) + 8-bit checksum.  The first
	    32&nbsp;bits are a vendor identifier.  They never come out
	    and say it, but it appears to be assumed that different
	    types of boards from the same vendor could have different
	    32-bit vendor IDs.  The idea of needing 32&nbsp;bits just
	    for unique manufacturers is a bit excessive.</para>

	  <para>The lower 32&nbsp;bits are a serial #, or something else
	    that makes this one board unique.  The vendor must never
	    produce a second board that has the same lower 32&nbsp;bits
	    unless the upper 32&nbsp;bits are also different.  So you
	    can have multiple boards of the same type in the machine and
	    the full 64&nbsp;bits will still be unique.</para>

	  <para>The 32-bit groups can never be all zero.  This
	    allows the wired-OR to show non-zero bits during the initial
	    binary search.</para>

	  <para>Once the system has identified all the board IDs
	    present, it will reactivate each board, one at a time (via the
	    same I/O ports), and find out what resources the given board
	    needs, what interrupt choices are available, etc.  A scan is
	    made over all the boards to collect this information.</para>

	  <para>This info is then combined with info from any ECU files
	    on the hard disk or wired into the MLB BIOS.  The ECU and
	    BIOS PnP support for hardware on the MLB is usually
	    synthetic, and the peripherals do not really do genuine PnP.
	    However by examining the BIOS info plus the ECU info, the
	    probe routines can cause the devices that are PnP to avoid
	    those devices the probe code cannot relocate.</para>

	  <para>Then the PnP devices are visited once more and given
	    their I/O, DMA, IRQ and Memory-map address assignments.  The
	    devices will then appear at those locations and remain there
	    until the next reboot, although there is nothing that says
	    you cannot move them around whenever you want.</para>

	  <para>There is a lot of oversimplification above, but you
	    should get the general idea.</para>

	  <para>&microsoft; took over some of the primary printer status
	    ports to do PnP, on the logic that no boards decoded those
	    addresses for the opposing I/O cycles.  I found a genuine
	    IBM printer board that did decode writes of the status port
	    during the early PnP proposal review period, but &microsoft;
	    said <quote>tough</quote>.  So they do a write to the
	    printer status port for setting addresses, plus that use
	    that address + <literal>0x800</literal>, and a third I/O
	    port for reading that can be located anywhere between
	    <literal>0x200</literal> and <literal>0x3ff</literal>.
	    </para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="major-numbers">
	  <para>Can you assign a major number for a device driver I have
	    written?</para>
	</question>

	<answer>
	  <para>&os; releases after February 2003 has a facility for
	    dynamically and automatically allocating major numbers for
	    device drivers at runtime (see &man.devfs.5;), so there is
	    no need for this.</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="alternate-directory-layout">
	  <para>What about alternative layout policies for
	    directories?</para>
	</question>

	<answer>
	  <para>In answer to the question of alternative layout policies
	    for directories, the scheme that is currently in use is
	    unchanged from what I wrote in 1983.  I wrote that policy
	    for the original fast file system, and never revisited it.
	    It works well at keeping cylinder groups from filling up.
	    As several of you have noted, it works poorly for find.
	    Most file systems are created from archives that were
	    created by a depth first search (aka ftw).  These
	    directories end up being striped across the cylinder groups
	    thus creating a worst possible scenario for future depth
	    first searches.  If one knew the total number of directories
	    to be created, the solution would be to create
	    <literal>(total&nbsp;/&nbsp;fs_ncg)</literal> per cylinder
	    group before moving on.  Obviously, one would have to create
	    some heuristic to guess at this number.  Even using a small
	    fixed number like say 10 would make an order of magnitude
	    improvement.  To differentiate restores from normal
	    operation (when the current algorithm is probably more
	    sensible), you could use the clustering of up to 10 if they
	    were all done within a ten second window.  Anyway, my
	    conclusion is that this is an area ripe for
	    experimentation.</para>

	  <para>&a.mckusick;, September 1998</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="kernel-panic-troubleshooting">
	  <para>How can I make the most of the data I see when my kernel
	    panics?</para>
	</question>

	<answer>
	  <para>Here is typical kernel panic:</para>

	  <programlisting>Fatal trap 12: page fault while in kernel mode
fault virtual address   = 0x40
fault code              = supervisor read, page not present
instruction pointer     = 0x8:0xf014a7e5
stack pointer           = 0x10:0xf4ed6f24
frame pointer           = 0x10:0xf4ed6f28
code segment            = base 0x0, limit 0xfffff, type 0x1b
                        = DPL 0, pres 1, def32 1, gran 1
processor eflags        = interrupt enabled, resume, IOPL = 0
current process         = 80 (mount)
interrupt mask          =
trap number             = 12
panic: page fault</programlisting>

	  <para>When you see a message like this, it is not enough to
	    just reproduce it and send it in.  The instruction pointer
	    value is important;
	    unfortunately, it is also configuration dependent.  In other
	    words, the value varies depending on the exact kernel image
	    that you are using.  If you are using a
	    <filename>GENERIC</filename> kernel image from one of the
	    snapshots, then it is possible for somebody else to track
	    down the offending function, but if you are running a custom
	    kernel then only <emphasis>you</emphasis> can tell us where
	    the fault occurred.</para>

	  <para>What you should do is this:</para>

	  <procedure>
	    <step>
	      <para>Write down the instruction pointer value.  Note
		that the <literal>0x8:</literal> part at the beginning
		is not significant in this case: it is the
		<literal>0xf0xxxxxx</literal> part that we
		want.</para>
	    </step>

	    <step>
	      <para>When the system reboots, do the following:</para>

	      <screen>&prompt.user; <userinput><command>nm</command> <option>-n</option> <replaceable>kernel.that.caused.the.panic</replaceable> | <command>grep</command> f0xxxxxx</userinput></screen>

	      <para>where <literal>f0xxxxxx</literal> is the
		instruction pointer value.  The odds are you will not
		get an exact match since the symbols in the kernel
		symbol table are for the entry points of functions and
		the instruction pointer address will be somewhere
		inside a function, not at the start.  If you do not
		get an exact match, omit the last digit from the
		instruction pointer value and try again, i.e.:</para>

	      <screen>&prompt.user; <userinput><command>nm</command> <option>-n</option> <replaceable>kernel.that.caused.the.panic</replaceable> | <command>grep</command> f0xxxxx</userinput></screen>

	      <para>If that does not yield any results, chop off another
		digit.  Repeat until you get some sort of output.  The
		result will be a possible list of functions which caused
		the panic.  This is a less than exact mechanism for
		tracking down the point of failure, but it is better
		than nothing.</para>
	    </step>
	  </procedure>

	  <para>However, the best way to track down the cause of a panic
	    is by capturing a crash dump, then using &man.kgdb.1; to
	    generate a stack trace on the crash dump.</para>

	  <para>In any case, the method is this:</para>

	  <procedure>
	    <step>
	      <para>Make sure that the following line is included in
		your kernel configuration file
		(<filename>/usr/src/sys/<replaceable>arch</replaceable>/conf/<replaceable>MYKERNEL</replaceable></filename>):</para>

	      <programlisting>makeoptions     DEBUG=-g          # Build kernel with gdb(1) debug symbols</programlisting>
	    </step>

	    <step>
	      <para>Change to the <filename
		  class="directory">/usr/src</filename>
		directory:</para>

	      <screen>&prompt.root; <userinput><command>cd</command> <filename class="directory">/usr/src</filename></userinput></screen>
	    </step>

	    <step>
	      <para>Compile the kernel:</para>

	      <screen>&prompt.root; <userinput><command>make</command> <maketarget>buildkernel</maketarget> <makevar>KERNCONF</makevar>=<replaceable>MYKERNEL</replaceable></userinput></screen>
	    </step>

	    <step>
	      <para>Wait for &man.make.1; to finish compiling.</para>
	    </step>

	    <step>
	      <screen>&prompt.root; <userinput><command>make</command> <maketarget>installkernel</maketarget> <makevar>KERNCONF</makevar>=<replaceable>MYKERNEL</replaceable></userinput></screen>
	    </step>

	    <step>
	      <para>Reboot.</para>
	    </step>
	  </procedure>

	  <note>
	    <para>If you do not use the <makevar>KERNCONF</makevar>
	      make variable a <filename>GENERIC</filename> kernel will
	      be built and installed.</para>
	  </note>

	  <para>The &man.make.1; process will have built two kernels.
	    <filename>/usr/obj/usr/src/sys/<replaceable>MYKERNEL</replaceable>/kernel</filename>
	    and
	    <filename>/usr/obj/usr/src/sys/<replaceable>MYKERNEL</replaceable>/kernel.debug</filename>.
	    <filename>kernel</filename> was installed as
	    <filename>/boot/kernel/kernel</filename>, while
	    <filename>kernel.debug</filename> can be used as the source
	    of debugging symbols for &man.kgdb.1;.</para>

	  <para>To make sure you capture a crash dump, you need edit
	    <filename>/etc/rc.conf</filename> and set
	    <literal>dumpdev</literal> to point to your swap partition
	    (or <literal>AUTO</literal>).  This will cause the
	    &man.rc.8; scripts to use the &man.dumpon.8; command to
	    enable crash dumps.  You can also run &man.dumpon.8;
	    manually.  After a panic, the crash dump can be recovered
	    using &man.savecore.8;; if <literal>dumpdev</literal> is set
	    in <filename>/etc/rc.conf</filename>, the &man.rc.8; scripts
	    will run &man.savecore.8; automatically and put the crash
	    dump in <filename class="directory">/var/crash</filename>.</para>

	  <note>
	    <para>&os; crash dumps are usually the same size as the
	      physical RAM size of your machine.  That is, if you have
	      512&nbsp;MB of RAM, you will get a 512&nbsp;MB crash dump.
	      Therefore you must make sure there is enough space in
	      <filename class="directory">/var/crash</filename> to hold the dump.
	      Alternatively, you run &man.savecore.8; manually and have
	      it recover the crash dump to another directory where you
	      have more room.  It is possible to limit the size of the
	      crash dump by using <literal>options
	      MAXMEM=<replaceable>N</replaceable></literal> where
	      <replaceable>N</replaceable> is the size of kernel's
	      memory usage in KBs.  For example, if you have 1&nbsp;GB
	      of RAM, you can limit the kernel's memory usage to
	      128&nbsp;MB by this way, so that your crash dump size will
	      be 128&nbsp;MB instead of 1&nbsp;GB.</para>
	  </note>

	  <para>Once you have recovered the crash dump, you can get a
	    stack trace with &man.kgdb.1; as follows:</para>

	  <screen>&prompt.user; <userinput><command>kgdb</command> <filename>/usr/obj/usr/src/sys/<replaceable>MYKERNEL</replaceable>/kernel.debug</filename> <filename class="directory">/var/crash/<replaceable>vmcore.0</replaceable></filename></userinput>
<prompt>(kgdb)</prompt> <userinput>backtrace</userinput></screen>

	  <para>Note that there may be several screens worth of
	    information; ideally you should use &man.script.1; to
	    capture all of them.  Using the unstripped kernel image with
	    all the debug symbols should show the exact line of kernel
	    source code where the panic occurred.  Usually you have to
	    read the stack trace from the bottom up in order to trace
	    the exact sequence of events that lead to the crash.  You
	    can also use &man.kgdb.1; to print out the contents of
	    various variables or structures in order to examine the
	    system state at the time of the crash.</para>

	  <tip>
	    <para>Now, if you are really insane and have a second
	      computer, you can also configure &man.kgdb.1; to do remote
	      debugging such that you can use &man.kgdb.1; on one system
	      to debug the kernel on another system, including setting
	      breakpoints, single-stepping through the kernel code, just
	      like you can do with a normal user-mode program.</para>
	  </tip>

	  <note>
	    <para>If you have <literal>DDB</literal> enabled and the
	      kernel drops into the debugger, you can force a panic (and
	      a crash dump) just by typing <literal>panic</literal> at
	      the <literal>ddb</literal> prompt.  It may stop in the
	      debugger again during the panic phase.  If it does, type
	      <literal>continue</literal> and it will finish the crash
	      dump.</para>
	  </note>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="dlsym-failure">
	  <para>Why has <function>dlsym()</function> stopped working for
	    ELF executables?</para>
	</question>

	<answer>
	  <para>The ELF toolchain does not, by default, make the symbols
	    defined in an executable visible to the dynamic linker.
	    Consequently <function>dlsym()</function> searches on
	    handles obtained from calls to <function>dlopen(NULL,
	    flags)</function> will fail to find such symbols.</para>

	  <para>If you want to search, using
	    <function>dlsym()</function>, for symbols present in the
	    main executable of a process, you need to link the
	    executable using the <option>--export-dynamic</option>
	    option to the ELF linker (&man.ld.1;).</para>
	</answer>
      </qandaentry>

      <qandaentry>
	<question id="change-kernel-address-space">
	  <para>How can I increase or reduce the kernel address space on
	    i386?</para>
	</question>

	<answer>
	  <para>By default, the kernel address space is 1&nbsp;GB
	    (2&nbsp;GB for PAE) for i386.  If you run a
	    network-intensive server (e.g. a large FTP or HTTP server),
	    or you want to use ZFS, you might find that is not
	    enough.</para>

	  <para>Add the following line to your kernel configuration file
	    to increase available space and rebuild your kernel:</para>

	  <programlisting>options KVA_PAGES=<replaceable>N</replaceable></programlisting>

	  <para>To find the correct value of
	    <replaceable>N</replaceable>, divide the desired address
	    space size (in megabytes) by four.  (For example, it is
	    <literal>512</literal> for 2&nbsp;GB.)</para>
	</answer>
      </qandaentry>
    </qandaset>
  </chapter>

  <chapter id="acknowledgments">
    <title>Acknowledgments</title>

    <para>This innocent little Frequently Asked Questions document has
      been written, rewritten, edited, folded, spindled, mutilated,
      eviscerated, contemplated, discombobulated, cogitated,
      regurgitated, rebuilt, castigated, and reinvigorated over the last
      decade, by a cast of hundreds if not thousands.
      Repeatedly.</para>

    <para>We wish to thank every one of the people responsible, and we
      encourage you to to <ulink
	url="&url.articles.contributing;/article.html">join them</ulink>
      in making this FAQ even better.</para>
  </chapter>

  &bibliography;
</book>