amd/doc/am-utils.texi

\input texinfo          @c -*-texinfo-*-
@c
@c Copyright (c) 1997-1998 Erez Zadok
@c Copyright (c) 1989 Jan-Simon Pendry
@c Copyright (c) 1989 Imperial College of Science, Technology & Medicine
@c Copyright (c) 1989 The Regents of the University of California.
@c All rights reserved.
@c
@c This code is derived from software contributed to Berkeley by
@c Jan-Simon Pendry at Imperial College, London.
@c
@c Redistribution and use in source and binary forms, with or without
@c modification, are permitted provided that the following conditions
@c are met:
@c 1. Redistributions of source code must retain the above copyright
@c    notice, this list of conditions and the following disclaimer.
@c 2. Redistributions in binary form must reproduce the above copyright
@c    notice, this list of conditions and the following disclaimer in the
@c    documentation and/or other materials provided with the distribution.
@c 3. All advertising materials mentioning features or use of this software
@c    must display the following acknowledgment:
@c      This product includes software developed by the University of
@c      California, Berkeley and its contributors.
@c 4. Neither the name of the University nor the names of its contributors
@c    may be used to endorse or promote products derived from this software
@c    without specific prior written permission.
@c
@c THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
@c ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
@c IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
@c ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
@c FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
@c DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
@c OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
@c HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
@c LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
@c OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
@c
@c      %W% (Berkeley) %G%
@c
@c $Id: am-utils.texi,v 6.0 1997/02/09 15:11:50 ezk beta $
@c
@setfilename am-utils.info

@include version.texi

@c info directory entry
@direntry
* Am-utils: (am-utils).          The Amd automounter suite of utilities
@end direntry

@titlepage
@title Am-utils (4.4BSD Automounter Utilities)
@subtitle For version @value{VERSION}, @value{UPDATED}

@author Erez Zadok
(Originally by Jan-Simon Pendry and Nick Williams)

@page
Copyright @copyright{} 1997-1998 Erez Zadok
@*
Copyright @copyright{} 1989 Jan-Simon Pendry
@*
Copyright @copyright{} 1989 Imperial College of Science, Technology & Medicine
@*
Copyright @copyright{} 1989 The Regents of the University of California.
@sp
All Rights Reserved.
@vskip 1ex
Permission to copy this document, or any portion of it, as
necessary for use of this software is granted provided this
copyright notice and statement of permission are included.
@end titlepage
@page

@c Define a new index for options.
@syncodeindex pg cp
@syncodeindex vr cp

@ifinfo

@c ################################################################
@node Top, License, , (DIR)
Am-utils - The 4.4BSD Automounter Tool Suite
*********************************************

Am-utils is the 4.4BSD Automounter Tool Suite, which includes the Amd
automounter, the Amq query and control program, the Hlfsd daemon, and
other tools.  This Info file describes how to use and understand the
tools within Am-utils.
@end ifinfo

@menu
* License::                  Explains the terms and conditions for using
                             and distributing Am-utils.
* Distrib::                  How to get the latest Am-utils distribution.
* Intro::                    An introduction to Automounting concepts.
* History::                  History of am-utils' development.
* Overview::                 An overview of Amd.
* Supported Platforms::      Machines and Systems supported by Amd.
* Mount Maps::               Details of mount maps
* Amd Command Line Options:: All the Amd command line options explained.
* Filesystem Types::         The different mount types supported by Amd.
* Amd Configuration File::   The amd.conf file syntax and meaning.
* Run-time Administration::  How to start, stop and control Amd.
* FSinfo::                   The FSinfo filesystem management tool.
* Hlfsd::                    The Home-Link Filesystem server.
* Assorted Tools::           Other tools which come with am-utils.
* Examples::                 Some examples showing how Amd might be used.
* Internals::                Implementation details.
* Acknowledgments & Trademarks:: Legal Notes

Indexes
* Index::                    An item for each concept.
@end menu

@iftex
@unnumbered Preface

This manual documents the use of the 4.4BSD automounter tool suite,
which includes @i{Amd}, @i{Amq}, @i{Hlfsd}, and other programs.  This is
primarily a reference manual.  While no tutorial exists, there are
examples available.  @xref{Examples}.

This manual comes in two forms: the published form and the Info form.
The Info form is for on-line perusal with the INFO program which is
distributed along with GNU texinfo package (a version of which is
available for GNU Emacs).@footnote{GNU packages can be found in
@url{ftp://ftp.gnu.org/pub/gnu/}.}  Both forms contain substantially
the same text and are generated from a common source file, which is
distributed with the @i{Am-utils} source.
@end iftex

@c ################################################################
@node License, Distrib, Top, Top
@unnumbered License
@cindex License Information

@i{Am-utils} is not in the public domain; it is copyrighted and there are
restrictions on its distribution.

Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:

@enumerate

@item
Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.

@item
Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.

@item
All advertising materials mentioning features or use of this software
must display the following acknowledgment:

@cartouche
``This product includes software developed by the University of
California, Berkeley and its contributors, as well as the Trustees of
Columbia University.''
@end cartouche

@item
Neither the name of the University nor the names of its contributors may
be used to endorse or promote products derived from this software
without specific prior written permission.

@end enumerate

THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS
BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
THE POSSIBILITY OF SUCH DAMAGE.

@c ################################################################
@node Distrib, Intro, License, Top
@unnumbered Source Distribution
@cindex Source code distribution
@cindex Obtaining the source code

The @i{Am-utils} home page is located in
@example
@url{http://www.cs.columbia.edu/~ezk/am-utils/}
@end example

You can get the latest distribution version of @i{Am-utils} from
@example
@url{ftp://shekel.mcl.cs.columbia.edu/pub/am-utils/am-utils.tar.gz}
@end example

Alpha and beta distributions are available in
@example
@url{ftp://shekel.mcl.cs.columbia.edu/pub/am-utils/}.
@end example

Revision 5.2 was part of the 4.3BSD Reno distribution.

Revision 5.3bsdnet, a late alpha version of 5.3, was part
of the BSD network version 2 distribution

Revision 6.0 was made independently by @email{ezk@@cs.columbia.edu,Erez
Zadok} at the @uref{http://www.cs.columbia.edu/,Computer Science
Department} of @uref{http://www.columbia.edu/,Columbia University}, as
part of his @uref{http://www.cs.columbia.edu/~ezk/research/tp/thesis_proposal.html,PhD thesis work}. @xref{History} for more details.

@unnumberedsec Bug Reports
@cindex Bug reports

Before reporting a bug, see if it is a known one in the
@uref{http://www.cs.columbia.edu/~ezk/am-utils/BUGS.txt,bugs} file.
Send all bug reports to @email{amd-dev@@majordomo.cs.columbia.edu}
quoting the details of the release and your configuration.  These can be
obtained by running the command @samp{amd -v}.  It would greatly help if
you could provide a reproducible procedure for detecting the bug you are
reporting.

Providing working patches is highly encouraged.  Every patch
incorporated, however small, will get its author an honorable mention in
the @uref{http://www.cs.columbia.edu/~ezk/am-utils/AUTHORS.txt,authors
file}.

@unnumberedsec Mailing List
@cindex Mailing list

There are two mailing lists for people interested in keeping up-to-date
with developments.

@c ###############

@enumerate

@item
The older list, @samp{amd-workers} is for general "how to" questions and
announcements.  To subscribe, send a note to
@email{amd-workers-request@@majordomo.glue.umd.edu}.@footnote{Note that
the older address, @email{amd-workers-request@@acl.lanl.gov}, is
defunct.}  To post a message to this list, send mail to
@email{amd-workers@@majordomo.glue.umd.edu}.

@item
The developers only list, @samp{amd-dev} is for

@itemize @minus
@item
announcements of alpha and beta releases of am-utils
@item
reporting of bugs and patches
@item
discussions of new features for am-utils
@item
implementation and porting issues
@end itemize

To subscribe, send a note to @email{majordomo@@majordomo.cs.columbia.edu}
with the single body text line @samp{subscribe amd-dev}.  To post a
message to this list, send mail to
@email{amd-dev@@majordomo.cs.columbia.edu}.  To avoid as much spam as
possible, only subscribers to this list may post to it.

Subscribers of @samp{amd-dev} are most suitable if they have the time
and resources to test new and buggy versions of amd, on as many
different platforms as possible.  They should also be prepared to learn
and use the GNU Autoconf, Automake, and Libtool packages, and of course,
be very familiar with the complex code in the am-utils package.  In
other words, subscribers on this list should be able to contribute
meaningfully to the development of amd.

@end enumerate

@c ################################################################
@node Intro, History, Distrib, Top
@unnumbered Introduction
@cindex Introduction

An @dfn{automounter} maintains a cache of mounted filesystems.
Filesystems are mounted on demand when they are first referenced,
and unmounted after a period of inactivity.

@i{Amd} may be used as a replacement for Sun's automounter.  The choice
of which filesystem to mount can be controlled dynamically with
@dfn{selectors}.  Selectors allow decisions of the form ``hostname is
@var{this},'' or ``architecture is not @var{that}.''  Selectors may be
combined arbitrarily.  @i{Amd} also supports a variety of filesystem
types, including NFS, UFS and the novel @dfn{program} filesystem.  The
combination of selectors and multiple filesystem types allows identical
configuration files to be used on all machines thus reducing the
administrative overhead.

@i{Amd} ensures that it will not hang if a remote server goes down.
Moreover, @i{Amd} can determine when a remote server has become
inaccessible and then mount replacement filesystems as and when they
become available.

@i{Amd} contains no proprietary source code and has been ported to
numerous flavors of Unix.

@c ################################################################
@node History, Overview, Intro, Top
@unnumbered History
@cindex History

The @i{Amd} package has been without an official maintainer since 1992.
Several people have stepped in to maintain it unofficially.  Most
notable were the `upl' (Unofficial Patch Level) releases of @i{Amd},
created by me (@email{ezk@@cs.columbia.edu,Erez Zadok}), and available from
@url{ftp://ftp.cs.columbia.edu/pub/amd/}.  The last such unofficial
release was `upl102'.

Through the process of patching and aging, it was becoming more and more
apparent that @i{Amd} was in much need of revitalizing.  Maintaining
@i{Amd} had become a difficult task.  I took it upon myself to cleanup
the code, so that it would be easier to port to new platforms, add new
features, keep up with the many new feature requests, and deal with the
never ending stream of bug reports.

I have been working on such a release of @i{Amd} on and off since
January of 1996.  The new suite of tools is currently named "am-utils"
(AutoMounter Utilities), in line with GNU naming conventions, befitting
the contents of the package.  In October of 1996 I had received enough
offers to help me with this task that I decided to make a mailing list
for this group of people.  Around the same time, @i{Amd} had become a
necessary part of my PhD thesis work, resulting in more work performed
on am-utils.

Am-utils version 6.0 was numbered with a major new release number to
distinguish it from the last official release of @i{Amd} (5.x).  Many
new features have been added such as a GNU @code{configure} system, NFS
Version 3, Autofs support, a run-time configuration file (`amd.conf'),
many new ports, more scripts and programs, as well as numerous bug
fixes.  Another reason for the new major release number was to alert
users of am-utils that user-visible interfaces may have changed.  In
order to make @i{Amd} work well for the next 10 years, and be easier to
maintain, it was necessary to remove old or unused features, change
various syntax files, etc.  However, great care was taken to ensure the
maximum possible backwards compatibility.

@c ################################################################
@node Overview, Supported Platforms, History, Top
@chapter Overview

@i{Amd} maintains a cache of mounted filesystems.  Filesystems are
@dfn{demand-mounted} when they are first referenced, and unmounted after
a period of inactivity.  @i{Amd} may be used as a replacement for Sun's
@b{automount}(8) program.  It contains no proprietary source code and
has been ported to numerous flavors of Unix.  @xref{Supported
Platforms}.@refill

@i{Amd} was designed as the basis for experimenting with filesystem
layout and management.  Although @i{Amd} has many direct applications it
is loaded with additional features which have little practical use.  At
some point the infrequently used components may be removed to streamline
the production system.

@c @i{Amd} supports the notion of @dfn{replicated} filesystems by evaluating
@c each member of a list of possible filesystem locations in parallel.
@c @i{Amd} checks that each cached mapping remains valid.  Should a mapping be
@c lost -- such as happens when a fileserver goes down -- @i{Amd} automatically
@c selects a replacement should one be available.
@c
@menu
* Fundamentals::
* Filesystems and Volumes::
* Volume Naming::
* Volume Binding::
* Operational Principles::
* Mounting a Volume::
* Automatic Unmounting::
* Keep-alives::
* Non-blocking Operation::
@end menu

@node Fundamentals, Filesystems and Volumes, Overview, Overview
@comment  node-name,  next,  previous,  up
@section Fundamentals
@cindex Automounter fundamentals

The fundamental concept behind @i{Amd} is the ability to separate the
name used to refer to a file from the name used to refer to its physical
storage location.  This allows the same files to be accessed with the
same name regardless of where in the network the name is used.  This is
very different from placing @file{/n/hostname} in front of the pathname
since that includes location dependent information which may change if
files are moved to another machine.

By placing the required mappings in a centrally administered database,
filesystems can be re-organized without requiring changes to
configuration files, shell scripts and so on.

@node Filesystems and Volumes, Volume Naming, Fundamentals, Overview
@comment  node-name,  next,  previous,  up
@section Filesystems and Volumes
@cindex Filesystem
@cindex Volume
@cindex Fileserver
@cindex sublink

@i{Amd} views the world as a set of fileservers, each containing one or
more filesystems where each filesystem contains one or more
@dfn{volumes}.  Here the term @dfn{volume} is used to refer to a
coherent set of files such as a user's home directory or a @TeX{}
distribution.@refill

In order to access the contents of a volume, @i{Amd} must be told in
which filesystem the volume resides and which host owns the filesystem.
By default the host is assumed to be local and the volume is assumed to
be the entire filesystem.  If a filesystem contains more than one
volume, then a @dfn{sublink} is used to refer to the sub-directory
within the filesystem where the volume can be found.

@node Volume Naming, Volume Binding, Filesystems and Volumes, Overview
@comment  node-name,  next,  previous,  up
@section Volume Naming
@cindex Volume names
@cindex Network-wide naming
@cindex Replicated volumes
@cindex Duplicated volumes
@cindex Replacement volumes

Volume names are defined to be unique across the entire network.  A
volume name is the pathname to the volume's root as known by the users
of that volume.  Since this name uniquely identifies the volume
contents, all volumes can be named and accessed from each host, subject
to administrative controls.

Volumes may be replicated or duplicated.  Replicated volumes contain
identical copies of the same data and reside at two or more locations in
the network.  Each of the replicated volumes can be used
interchangeably.  Duplicated volumes each have the same name but contain
different, though functionally identical, data.  For example,
@samp{/vol/tex} might be the name of a @TeX{} distribution which varied
for each machine architecture.@refill

@i{Amd} provides facilities to take advantage of both replicated and
duplicated volumes.  Configuration options allow a single set of
configuration data to be shared across an entire network by taking
advantage of replicated and duplicated volumes.

@i{Amd} can take advantage of replacement volumes by mounting them as
required should an active fileserver become unavailable.

@node Volume Binding, Operational Principles, Volume Naming, Overview
@comment  node-name,  next,  previous,  up
@section Volume Binding
@cindex Volume binding
@cindex Unix namespace
@cindex Namespace
@cindex Binding names to filesystems

Unix implements a namespace of hierarchically mounted filesystems.  Two
forms of binding between names and files are provided.  A @dfn{hard
link} completes the binding when the name is added to the filesystem.  A
@dfn{soft link} delays the binding until the name is accessed.  An
@dfn{automounter} adds a further form in which the binding of name to
filesystem is delayed until the name is accessed.@refill

The target volume, in its general form, is a tuple (host, filesystem,
sublink) which can be used to name the physical location of any volume
in the network.

When a target is referenced, @i{Amd} ignores the sublink element and
determines whether the required filesystem is already mounted.  This is
done by computing the local mount point for the filesystem and checking
for an existing filesystem mounted at the same place.  If such a
filesystem already exists then it is assumed to be functionally
identical to the target filesystem.  By default there is a one-to-one
mapping between the pair (host, filesystem) and the local mount point so
this assumption is valid.

@node Operational Principles, Mounting a Volume, Volume Binding, Overview
@comment  node-name,  next,  previous,  up
@section Operational Principles
@cindex Operational principles

@i{Amd} operates by introducing new mount points into the namespace.
These are called @dfn{automount} points.  The kernel sees these
automount points as NFS filesystems being served by @i{Amd}.  Having
attached itself to the namespace, @i{Amd} is now able to control the
view the rest of the system has of those mount points.  RPC calls are
received from the kernel one at a time.

When a @dfn{lookup} call is received @i{Amd} checks whether the name is
already known.  If it is not, the required volume is mounted.  A
symbolic link pointing to the volume root is then returned.  Once the
symbolic link is returned, the kernel will send all other requests
direct to the mounted filesystem.

If a volume is not yet mounted, @i{Amd} consults a configuration
@dfn{mount-map} corresponding to the automount point.  @i{Amd} then
makes a runtime decision on what and where to mount a filesystem based
on the information obtained from the map.

@i{Amd} does not implement all the NFS requests; only those relevant
to name binding such as @dfn{lookup}, @dfn{readlink} and @dfn{readdir}.
Some other calls are also implemented but most simply return an error
code; for example @dfn{mkdir} always returns ``read-only filesystem''.

@node Mounting a Volume, Automatic Unmounting, Operational Principles, Overview
@comment  node-name,  next,  previous,  up
@section Mounting a Volume
@cindex Mounting a volume
@cindex Location lists
@cindex Alternate locations
@cindex Mount retries
@cindex Background mounts

Each automount point has a corresponding mount map.  The mount map
contains a list of key--value pairs.  The key is the name of the volume
to be mounted.  The value is a list of locations describing where the
filesystem is stored in the network.  In the source for the map the
value would look like

@display
location1  location2  @dots{}  locationN
@end display

@i{Amd} examines each location in turn.  Each location may contain
@dfn{selectors} which control whether @i{Amd} can use that location.
For example, the location may be restricted to use by certain hosts.
Those locations which cannot be used are ignored.

@i{Amd} attempts to mount the filesystem described by each remaining
location until a mount succeeds or @i{Amd} can no longer proceed.  The
latter can occur in three ways:

@itemize @bullet
@item
If none of the locations could be used, or if all of the locations
caused an error, then the last error is returned.

@item
If a location could be used but was being mounted in the background then
@i{Amd} marks that mount as being ``in progress'' and continues with
the next request; no reply is sent to the kernel.

@item
Lastly, one or more of the mounts may have been @dfn{deferred}.  A mount
is deferred if extra information is required before the mount can
proceed.  When the information becomes available the mount will take
place, but in the mean time no reply is sent to the kernel.  If the
mount is deferred, @i{Amd} continues to try any remaining locations.
@end itemize

Once a volume has been mounted, @i{Amd} establishes a @dfn{volume
mapping} which is used to satisfy subsequent requests.@refill

@node Automatic Unmounting, Keep-alives, Mounting a Volume, Overview
@comment  node-name,  next,  previous,  up
@section Automatic Unmounting

To avoid an ever increasing number of filesystem mounts, @i{Amd} removes
volume mappings which have not been used recently.  A time-to-live
interval is associated with each mapping and when that expires the
mapping is removed.  When the last reference to a filesystem is removed,
that filesystem is unmounted.  If the unmount fails, for example the
filesystem is still busy, the mapping is re-instated and its
time-to-live interval is extended.  The global default for this grace
period is controlled by the @code{-w} command-line option (@pxref{-w
Option, -w}) or the @i{amd.conf} parameter @samp{dismount_interval}
(@pxref{dismount_interval Parameter}).  It is also possible to set this
value on a per-mount basis (@pxref{opts Option, opts, opts}).

Filesystems can be forcefully timed out using the @i{Amq} command.
@xref{Run-time Administration}.

@node Keep-alives, Non-blocking Operation, Automatic Unmounting, Overview
@comment  node-name,  next,  previous,  up
@section Keep-alives
@cindex Keep-alives
@cindex Server crashes
@cindex NFS ping

Use of some filesystem types requires the presence of a server on
another machine.  If a machine crashes then it is of no concern to
processes on that machine that the filesystem is unavailable.  However,
to processes on a remote host using that machine as a fileserver this
event is important.  This situation is most widely recognized when an
NFS server crashes and the behavior observed on client machines is that
more and more processes hang.  In order to provide the possibility of
recovery, @i{Amd} implements a @dfn{keep-alive} interval timer for some
filesystem types.  Currently only NFS makes use of this service.

The basis of the NFS keep-alive implementation is the observation that
most sites maintain replicated copies of common system data such as
manual pages, most or all programs, system source code and so on.  If
one of those servers goes down it would be reasonable to mount one of
the others as a replacement.

The first part of the process is to keep track of which fileservers are
up and which are down.  @i{Amd} does this by sending RPC requests to the
servers' NFS @code{NullProc} and checking whether a reply is returned.
While the server state is uncertain the requests are re-transmitted at
three second intervals and if no reply is received after four attempts
the server is marked down.  If a reply is received the fileserver is
marked up and stays in that state for 30 seconds at which time another
NFS ping is sent.

Once a fileserver is marked down, requests continue to be sent every 30
seconds in order to determine when the fileserver comes back up.  During
this time any reference through @i{Amd} to the filesystems on that
server fail with the error ``Operation would block''.  If a replacement
volume is available then it will be mounted, otherwise the error is
returned to the user.

@c @i{Amd} keeps track of which servers are up and which are down.
@c It does this by sending RPC requests to the servers' NFS {\sc NullProc} and
@c checking whether a reply is returned.  If no replies are received after a
@c short period, @i{Amd} marks the fileserver @dfn{down}.
@c RPC requests continue to be sent so that it will notice when a fileserver
@c comes back up.
@c ICMP echo packets \cite{rfc:icmp} are not used because it is the availability
@c of the NFS service that is important, not the existence of a base kernel.
@c Whenever a reference to a fileserver which is down is made via @i{Amd}, an alternate
@c filesystem is mounted if one is available.
@c
Although this action does not protect user files, which are unique on
the network, or processes which do not access files via @i{Amd} or
already have open files on the hung filesystem, it can prevent most new
processes from hanging.

By default, fileserver state is not maintained for NFS/TCP mounts.  The
remote fileserver is always assumed to be up.
@c
@c With a suitable combination of filesystem management and mount-maps,
@c machines can be protected against most server downtime.  This can be
@c enhanced by allocating boot-servers dynamically which allows a diskless
@c workstation to be quickly restarted if necessary.  Once the root filesystem
@c is mounted, @i{Amd} can be started and allowed to mount the remainder of
@c the filesystem from whichever fileservers are available.

@node Non-blocking Operation, , Keep-alives, Overview
@comment  node-name,  next,  previous,  up
@section Non-blocking Operation
@cindex Non-blocking operation
@cindex Multiple-threaded server
@cindex RPC retries

Since there is only one instance of @i{Amd} for each automount point,
and usually only one instance on each machine, it is important that it
is always available to service kernel calls.  @i{Amd} goes to great
lengths to ensure that it does not block in a system call.  As a last
resort @i{Amd} will fork before it attempts a system call that may block
indefinitely, such as mounting an NFS filesystem.  Other tasks such as
obtaining filehandle information for an NFS filesystem, are done using a
purpose built non-blocking RPC library which is integrated with
@i{Amd}'s task scheduler.  This library is also used to implement NFS
keep-alives (@pxref{Keep-alives}).

Whenever a mount is deferred or backgrounded, @i{Amd} must wait for it
to complete before replying to the kernel.  However, this would cause
@i{Amd} to block waiting for a reply to be constructed.  Rather than do
this, @i{Amd} simply @dfn{drops} the call under the assumption that the
kernel RPC mechanism will automatically retry the request.

@c ################################################################
@node Supported Platforms, Mount Maps, Overview, Top
@comment  node-name,  next,  previous,  up
@chapter Supported Platforms
@cindex Supported Platforms
@cindex shared libraries
@cindex NFS V.3 support

@i{Am-utils} has been ported to a wide variety of machines and operating
systems.  @i{Am-utils}'s code works for little-endian and big-endian
machines, as well as 32 bit and 64 bit architectures.  Furthermore, when
@i{Am-utils} ports to an Operating System on one architecture, it is generally
readily portable to the same Operating System on all platforms on which
it is available.

The table below lists those platforms supported by the latest release.
The listing is based on the standard output from GNU's
@code{config.guess} script.  Since significant changes have been made to
am-utils, not all systems listed here have been verified working for all
features.

@multitable {Auto-Configured System Name} {Config} {Compile} {Amd} {NFS3} {Shlib} {Hlfsd}

@item @b{Auto-Configured System Name}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab @b{Config} @tab @b{Compile} @tab @b{Amd} @tab @b{NFS3} @tab @b{Shlib} @tab @b{Hlfsd}

@item @b{alpha-dec-osf2.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab       @tab      @tab

@item @b{alpha-dec-osf4.0}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab      @tab

@item @b{alphaev5-unknown-linux-gnu}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{hppa1.0-hp-hpux11.00}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab no    @tab      @tab

@item @b{hppa1.1-hp-hpux10.10}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab no   @tab

@item @b{hppa1.1-hp-hpux10.20}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab no   @tab

@item @b{hppa1.1-hp-hpux9.01}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{hppa1.1-hp-hpux9.05}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{hppa1.1-hp-hpux9.07}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{i386-pc-bsdi2.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{i386-pc-bsdi3.0}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{i386-pc-bsdi3.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{i386-pc-solaris2.5.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab yes

@item @b{i386-pc-solaris2.6}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab yes

@item @b{i386-unknown-freebsd2.1.0}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{i386-unknown-freebsd2.2.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{i386-unknown-freebsd3.0}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab

@item @b{i386-unknown-netbsd1.2.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab

@item @b{i386-unknown-netbsd1.3}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab

@item @b{i386-unknown-netbsd1.3.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab

@item @b{i386-unknown-openbsd2.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab

@item @b{i486-ncr-sysv4.3.03}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab      @tab yes   @tab      @tab

@item @b{i486-pc-linux-gnulibc1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{i586-pc-linux-gnulinc1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{i686-pc-linux-gnu}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{m68k-hp-hpux9.00}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{m68k-sun-sunos4.1.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{m68k-next-nextstep3}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{mips-dec-ultrix4.3}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{mips-sgi-irix5.2}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab         @tab       @tab      @tab       @tab      @tab

@item @b{mips-sgi-irix5.3}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab      @tab

@item @b{mips-sgi-irix6.2}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab      @tab

@item @b{mips-sgi-irix6.4}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab

@item @b{powerpc-ibm-aix4.1.5.0}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{powerpc-ibm-aix4.2.1.0}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab      @tab

@item @b{rs6000-ibm-aix3.2}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{rs6000-ibm-aix3.2.5}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{rs6000-ibm-aix4.1.4.0}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{rs6000-ibm-aix4.1.5.0}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{sparc-sun-solaris2.3}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab      @tab

@item @b{sparc-sun-solaris2.4}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{sparc-sun-solaris2.5}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab

@item @b{sparc-sun-solaris2.5.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab yes

@item @b{sparc-sun-solaris2.6}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab yes   @tab yes  @tab yes

@item @b{sparc-sun-sunos4.1.1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{sparc-sun-sunos4.1.3}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{sparc-sun-sunos4.1.3C}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{sparc-sun-sunos4.1.3_U1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{sparc-sun-sunos4.1.4}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{sparc-unknown-linux-gnulibc1}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab n/a   @tab yes  @tab

@item @b{sparc-unknown-netbsd1.2E}
@c {Config}  {Compile}  {Amd}     {NFS V.3}  {Shlib}   {Hlfsd}
@tab yes     @tab yes   @tab yes  @tab       @tab      @tab

@end multitable

See the @file{INSTALL} in the distribution for more specific details on
building and/or configuring for some systems.

@c ################################################################
@node Mount Maps, Amd Command Line Options, Supported Platforms, Top
@comment  node-name,  next,  previous,  up
@chapter Mount Maps
@cindex Mount maps
@cindex Automounter configuration maps
@cindex Mount information

@i{Amd} has no built-in knowledge of machines or filesystems.
External @dfn{mount-maps} are used to provide the required information.
Specifically, @i{Amd} needs to know when and under what conditions it
should mount filesystems.

The map entry corresponding to the requested name contains a list of
possible locations from which to resolve the request.  Each location
specifies filesystem type, information required by that filesystem (for
example the block special device in the case of UFS), and some
information describing where to mount the filesystem (@pxref{fs Option}).  A
location may also contain @dfn{selectors} (@pxref{Selectors}).@refill

@menu
* Map Types::
* Key Lookup::
* Location Format::
@end menu

@node Map Types, Key Lookup, Mount Maps, Mount Maps
@comment  node-name,  next,  previous,  up
@section Map Types
@cindex Mount map types
@cindex Map types
@cindex Configuration map types
@cindex Types of mount map
@cindex Types of configuration map
@cindex Determining the map type

A mount-map provides the run-time configuration information to @i{Amd}.
Maps can be implemented in many ways.  Some of the forms supported by
@i{Amd} are regular files, ndbm databases, NIS maps, the @dfn{Hesiod}
name server, and even the password file.

A mount-map @dfn{name} is a sequence of characters.  When an automount
point is created a handle on the mount-map is obtained.  For each map
type configured, @i{Amd} attempts to reference the map of the
appropriate type.  If a map is found, @i{Amd} notes the type for future
use and deletes the reference, for example closing any open file
descriptors.  The available maps are configured when @i{Amd} is built
and can be displayed by running the command @samp{amd -v}.

When using an @i{Amd} configuration file (@pxref{Amd Configuration File})
and the keyword @samp{map_type} (@pxref{map_type Parameter}), you may
force the map used to any type.

By default, @i{Amd} caches data in a mode dependent on the type of map.
This is the same as specifying @samp{cache:=mapdefault} and selects a
suitable default cache mode depending on the map type.  The individual
defaults are described below.  The @var{cache} option can be specified
on automount points to alter the caching behavior (@pxref{Automount
Filesystem}).@refill

The following map types have been implemented, though some are not
available on all machines.  Run the command @samp{amd -v} to obtain a
list of map types configured on your machine.

@menu
* File maps::
* ndbm maps::
* NIS maps::
* NIS+ maps::
* Hesiod maps::
* Password maps::
* Union maps::
* LDAP maps::
@end menu

@node File maps, ndbm maps, Map Types, Map Types
@comment  node-name,  next,  previous,  up
@subsection File maps
@cindex File maps
@cindex Flat file maps
@cindex File map syntactic conventions

When @i{Amd} searches a file for a map entry it does a simple scan of
the file and supports both comments and continuation lines.

Continuation lines are indicated by a backslash character (@samp{\}) as
the last character of a line in the file.  The backslash, newline character
@emph{and any leading white space on the following line} are discarded.  A maximum
line length of 2047 characters is enforced after continuation lines are read
but before comments are stripped.  Each line must end with
a newline character; that is newlines are terminators, not separators.
The following examples illustrate this:

@example
key     valA   valB;   \
          valC
@end example

specifies @emph{three} locations, and is identical to

@example
key     valA   valB;   valC
@end example

However,

@example
key     valA   valB;\
          valC
@end example

specifies only @emph{two} locations, and is identical to

@example
key     valA   valB;valC
@end example

After a complete line has been read from the file, including
continuations, @i{Amd} determines whether there is a comment on the
line.  A comment begins with a hash (``@samp{#}'') character and
continues to the end of the line.  There is no way to escape or change
the comment lead-in character.

Note that continuation lines and comment support @dfn{only} apply to
file maps, or ndbm maps built with the @code{mk-amd-map} program.

When caching is enabled, file maps have a default cache mode of
@code{all} (@pxref{Automount Filesystem}).

@node ndbm maps, NIS maps, File maps, Map Types
@comment  node-name,  next,  previous,  up
@subsection ndbm maps
@cindex ndbm maps

An ndbm map may be used as a fast access form of a file map.  The program,
@code{mk-amd-map}, converts a normal map file into an ndbm database.
This program supports the same continuation and comment conventions that
are provided for file maps.  Note that ndbm format files may @emph{not}
be sharable across machine architectures.  The notion of speed generally
only applies to large maps; a small map, less than a single disk block,
is almost certainly better implemented as a file map.

ndbm maps have a default cache mode of @samp{all} (@pxref{Automount Filesystem}).

@node NIS maps, NIS+ maps, ndbm maps, Map Types
@comment  node-name,  next,  previous,  up
@subsection NIS maps
@cindex NIS (YP) maps

When using NIS (formerly YP), an @i{Amd} map is implemented directly
by the underlying NIS map.  Comments and continuation lines are
@emph{not} supported in the automounter and must be stripped when
constructing the NIS server's database.

NIS maps have a default cache mode of @code{all} (@pxref{Automount
Filesystem}).

The following rule illustrates what could be added to your NIS @file{Makefile},
in this case causing the @file{amd.home} map to be rebuilt:
@example
$(YPTSDIR)/amd.home.time: $(ETCDIR)/amd.home
    -@@sed -e "s/#.*$$//" -e "/^$$/d" $(ETCDIR)/amd.home | \
      awk '@{  \
         for (i = 1; i <= NF; i++) \
             if (i == NF) @{ \
             if (substr($$i, length($$i), 1) == "\\") \
                 printf("%s", substr($$i, 1, length($$i) - 1)); \
             else \
                 printf("%s\n", $$i); \
             @} \
             else \
             printf("%s ", $$i); \
         @}' | \
    $(MAKEDBM) - $(YPDBDIR)/amd.home; \
    touch $(YPTSDIR)/amd.home.time; \
    echo "updated amd.home"; \
    if [ ! $(NOPUSH) ]; then \
        $(YPPUSH) amd.home; \
        echo "pushed amd.home"; \
    else \
        : ; \
    fi
@end example

Here @code{$(YPTSDIR)} contains the time stamp files, and @code{$(YPDBDIR)} contains
the dbm format NIS files.

@node NIS+ maps, Hesiod maps, NIS maps, Map Types
@comment  node-name,  next,  previous,  up
@subsection NIS+ maps
@cindex NIS+ maps

NIS+ maps do not support cache mode @samp{all} and, when caching is
enabled, have a default cache mode of @samp{inc}.

XXX: FILL IN WITH AN EXAMPLE.

@node Hesiod maps, Password maps, NIS+ maps, Map Types
@comment  node-name,  next,  previous,  up
@subsection Hesiod maps
@cindex Hesiod maps

When the map name begins with the string @samp{hesiod.} lookups are made
using the @dfn{Hesiod} name server.  The string following the dot is
used as a name qualifier and is prepended with the key being located.
The entire string is then resolved in the @code{automount} context, or
the @i{amd.conf} parameter @samp{hesiod_base} (@pxref{hesiod_base
Parameter}).  For example, if the the key is @samp{jsp} and map name is
@samp{hesiod.homes} then @dfn{Hesiod} is asked to resolve
@samp{jsp.homes.automount}.

Hesiod maps do not support cache mode @samp{all} and, when caching is
enabled, have a default cache mode of @samp{inc} (@pxref{Automount
Filesystem}).

The following is an example of a @dfn{Hesiod} map entry:

@example
jsp.homes.automount HS TXT "rfs:=/home/charm;rhost:=charm;sublink:=jsp"
njw.homes.automount HS TXT "rfs:=/home/dylan/dk2;rhost:=dylan;sublink:=njw"
@end example

@node Password maps, Union maps, Hesiod maps, Map Types
@comment  node-name,  next,  previous,  up
@subsection Password maps
@cindex Password file maps
@cindex /etc/passwd maps
@cindex User maps, automatic generation
@cindex Automatic generation of user maps
@cindex Using the password file as a map

The password map support is unlike the four previous map types.  When
the map name is the string @file{/etc/passwd} @i{Amd} can lookup a user
name in the password file and re-arrange the home directory field to
produce a usable map entry.

@i{Amd} assumes the home directory has the format
`@t{/}@i{anydir}@t{/}@i{dom1}@t{/../}@i{domN}@t{/}@i{login}'.
@c @footnote{This interpretation is not necessarily exactly what you want.}
It breaks this string into a map entry where @code{$@{rfs@}} has the
value `@t{/}@i{anydir}@t{/}@i{domN}', @code{$@{rhost@}} has the value
`@i{domN}@t{.}@i{...}@t{.}@i{dom1}', and @code{$@{sublink@}} has the
value @i{login}.@refill

Thus if the password file entry was

@example
/home/achilles/jsp
@end example

the map entry used by @i{Amd} would be

@example
rfs:=/home/achilles;rhost:=achilles;sublink:=jsp
@end example

Similarly, if the password file entry was

@example
/home/cc/sugar/mjh
@end example

the map entry used by @i{Amd} would be

@example
rfs:=/home/sugar;rhost:=sugar.cc;sublink:=jsp
@end example

@node Union maps, LDAP maps , Password maps, Map Types
@comment  node-name,  next,  previous,  up
@subsection Union maps
@cindex Union file maps

The union map support is provided specifically for use with the union
filesystem, @pxref{Union Filesystem}.

It is identified by the string @samp{union:} which is followed by a
colon separated list of directories.  The directories are read in order,
and the names of all entries are recorded in the map cache.  Later
directories take precedence over earlier ones.  The union filesystem
type then uses the map cache to determine the union of the names in all
the directories.

@node LDAP maps, , Union maps, Map Types
@comment  node-name,  next,  previous,  up
@subsection LDAP maps
@cindex LDAP maps
@cindex Lightweight Directory Access Protocol

LDAP (Lightweight Directory Access Protocol) maps do not support cache
mode @samp{all} and, when caching is enabled, have a default cache mode
of @samp{inc}.

For example, an @i{Amd} map @samp{amd.home} that looks as follows:

@example
/defaults    opts:=rw,intr;type:=link

zing         -rhost:=shekel \
             host==shekel \
             host!=shekel;type:=nfs
@end example
@noindent
when converted to LDAP (@pxref{amd2ldif}), will result in the following
LDAP database:
@example
$ amd2ldif amd.home CUCS < amd.home
dn: cn=amdmap timestamp, CUCS
cn             : amdmap timestamp
objectClass    : amdmapTimestamp
amdmapTimestamp: 873071363

dn: cn=amdmap amd.home[/defaults], CUCS
cn          : amdmap amd.home[/defaults]
objectClass : amdmap
amdmapName  : amd.home
amdmapKey   : /defaults
amdmapValue : opts:=rw,intr;type:=link

dn: cn=amdmap amd.home[], CUCS
cn          : amdmap amd.home[]
objectClass : amdmap
amdmapName  : amd.home
amdmapKey   :
amdmapValue :

dn: cn=amdmap amd.home[zing], CUCS
cn          : amdmap amd.home[zing]
objectClass : amdmap
amdmapName  : amd.home
amdmapKey   : zing
amdmapValue : -rhost:=shekel host==shekel host!=shekel;type:=nfs
@end example

@c subsection Gdbm

@node Key Lookup, Location Format, Map Types, Mount Maps
@comment  node-name,  next,  previous,  up
@section How keys are looked up
@cindex Key lookup
@cindex Map lookup
@cindex Looking up keys
@cindex How keys are looked up
@cindex Wildcards in maps

The key is located in the map whose type was determined when the
automount point was first created.  In general the key is a pathname
component.  In some circumstances this may be modified by variable
expansion (@pxref{Variable Expansion}) and prefixing.  If the automount
point has a prefix, specified by the @var{pref} option, then that is
prepended to the search key before the map is searched.

If the map cache is a @samp{regexp} cache then the key is treated as an
egrep-style regular expression, otherwise a normal string comparison is
made.

If the key cannot be found then a @dfn{wildcard} match is attempted.
@i{Amd} repeatedly strips the basename from the key, appends @samp{/*} and
attempts a lookup.  Finally, @i{Amd} attempts to locate the special key @samp{*}.

For example, the following sequence would be checked if @file{home/dylan/dk2} was
being located:

@example
   home/dylan/dk2
   home/dylan/*
   home/*
   *
@end example

At any point when a wildcard is found, @i{Amd} proceeds as if an exact
match had been found and the value field is then used to resolve the
mount request, otherwise an error code is propagated back to the kernel.
(@pxref{Filesystem Types}).@refill

@node Location Format, , Key Lookup, Mount Maps
@comment  node-name,  next,  previous,  up
@section Location Format
@cindex Location format
@cindex Map entry format
@cindex How locations are parsed

The value field from the lookup provides the information required to
mount a filesystem.  The information is parsed according to the syntax
shown below.

@display
@i{location-list}:
                  @i{location-selection}
                  @i{location-list} @i{white-space} @t{||} @i{white-space} @i{location-selection}
@i{location-selection}:
                  @i{location}
                  @i{location-selection} @i{white-space} @i{location}
@i{location}:
                  @i{location-info}
                  @t{-}@i{location-info}
                  @t{-}
@i{location-info}:
                  @i{sel-or-opt}
                  @i{location-info}@t{;}@i{sel-or-opt}
                  @t{;}
@i{sel-or-opt}:
                  @i{selection}
                  @i{opt-ass}
@i{selection}:
                  selector@t{==}@i{value}
                  selector@t{!=}@i{value}
@i{opt-ass}:
                  option@t{:=}@i{value}
@i{white-space}:
                  space
                  tab
@end display

Note that unquoted whitespace is not allowed in a location description.
White space is only allowed, and is mandatory, where shown with non-terminal
@i{white-space}.

A @dfn{location-selection} is a list of possible volumes with which to
satisfy the request.  @dfn{location-selection}s are separated by the
@samp{||} operator.  The effect of this operator is to prevent use of
location-selections to its right if any of the location-selections on
its left were selected whether or not any of them were successfully
mounted (@pxref{Selectors}).@refill

The location-selection, and singleton @dfn{location-list},
@samp{type:=ufs;dev:=/dev/xd1g} would inform @i{Amd} to mount a UFS
filesystem from the block special device @file{/dev/xd1g}.

The @dfn{sel-or-opt} component is either the name of an option required
by a specific filesystem, or it is the name of a built-in, predefined
selector such as the architecture type.  The value may be quoted with
double quotes @samp{"}, for example
@samp{type:="ufs";dev:="/dev/xd1g"}.  These quotes are stripped when the
value is parsed and there is no way to get a double quote into a value
field.  Double quotes are used to get white space into a value field,
which is needed for the program filesystem (@pxref{Program Filesystem}).@refill

@menu
* Map Defaults::
* Variable Expansion::
* Selectors::
* Map Options::
@end menu

@node Map Defaults, Variable Expansion, Location Format, Location Format
@comment  node-name,  next,  previous,  up
@subsection Map Defaults
@cindex Map defaults
@cindex How to set default map parameters
@cindex Setting default map parameters

A location beginning with a dash @samp{-} is used to specify default
values for subsequent locations.  Any previously specified defaults in
the location-list are discarded.  The default string can be empty in
which case no defaults apply.

The location @samp{-fs:=/mnt;opts:=ro} would set the local mount point
to @file{/mnt} and cause mounts to be read-only by default.  Defaults
specified this way are appended to, and so override, any global map
defaults given with @samp{/defaults}).

@c
@c A @samp{/defaults} value @dfn{gdef} and a location list
@c \begin{quote}
@c $@samp{-}@dfn{def}_a $\verb*+ +$ @dfn{loc}_{a_1} $\verb*+ +$ @dfn{loc}_{a_2} $\verb*+ +$ @samp{-}@dfn{def}_b $\verb*+ +$ @dfn{loc}_{b_1} \ldots$
@c \end{quote}
@c is equivalent to
@c \begin{quote}
@c $@samp{-}@dfn{gdef}@samp{;}@dfn{def}_a $\verb*+ +$ @dfn{loc}_{a_1} $\verb*+ +$ @dfn{loc}_{a_2} $\verb*+ +$ @samp{-}@dfn{gdef}@samp{;}@dfn{def}_b $\verb*+ +$ @dfn{loc}_{b_1} \ldots$
@c \end{quote}
@c which is equivalent to
@c \begin{quote}
@c $@dfn{gdef}@samp{;}@dfn{def}_a@samp{;}@dfn{loc}_{a_1} $\verb*+ +$@dfn{gdef}@samp{;}@dfn{def}_a@samp{;}@dfn{loc}_{a_2} $\verb*+ +$@dfn{gdef}@samp{;}@dfn{def}_b@samp{;}@dfn{loc}_{b_1} \ldots$
@c \end{quote}

@node Variable Expansion, Selectors, Map Defaults, Location Format
@comment  node-name,  next,  previous,  up
@subsection Variable Expansion
@cindex Variable expansion
@cindex How variables are expanded
@cindex Pathname operators
@cindex Domain stripping
@cindex Domainname operators
@cindex Stripping the local domain name
@cindex Environment variables
@cindex How to access environment variables in maps

To allow generic location specifications @i{Amd} does variable expansion
on each location and also on some of the option strings.  Any option or
selector appearing in the form @code{$@dfn{var}} is replaced by the
current value of that option or selector.  For example, if the value of
@code{$@{key@}} was @samp{bin}, @code{$@{autodir@}} was @samp{/a} and
@code{$@{fs@}} was `@t{$@{autodir@}}@t{/local/}@t{$@{key@}}' then
after expansion @code{$@{fs@}} would have the value @samp{/a/local/bin}.
Any environment variable can be accessed in a similar way.@refill

Two pathname operators are available when expanding a variable.  If the
variable name begins with @samp{/} then only the last component of the
pathname is substituted.  For example, if @code{$@{path@}} was
@samp{/foo/bar} then @code{$@{/path@}} would be expanded to @samp{bar}.
Similarly, if the variable name ends with @samp{/} then all but the last
component of the pathname is substituted.  In the previous example,
@code{$@{path/@}} would be expanded to @samp{/foo}.@refill

Two domain name operators are also provided.  If the variable name
begins with @samp{.} then only the domain part of the name is
substituted.  For example, if @code{$@{rhost@}} was
@samp{swan.doc.ic.ac.uk} then @code{$@{.rhost@}} would be expanded to
@samp{doc.ic.ac.uk}.  Similarly, if the variable name ends with @samp{.}
then only the host component is substituted.  In the previous example,
@code{$@{rhost.@}} would be expanded to @samp{swan}.@refill

Variable expansion is a two phase process.  Before a location is parsed,
all references to selectors, @i{eg} @code{$@{path@}}, are expanded.  The
location is then parsed, selections are evaluated and option assignments
recorded.  If there were no selections or they all succeeded the
location is used and the values of the following options are expanded in
the order given: @var{sublink}, @var{rfs}, @var{fs}, @var{opts},
@var{remopts}, @var{mount} and @var{unmount}.

Note that expansion of option values is done after @dfn{all} assignments
have been completed and not in a purely left to right order as is done
by the shell.  This generally has the desired effect but care must be
taken if one of the options references another, in which case the
ordering can become significant.

There are two special cases concerning variable expansion:

@enumerate
@item
before a map is consulted, any selectors in the name received
from the kernel are expanded.  For example, if the request from the
kernel was for `@t{$@{arch@}}@t{.bin}' and the machine architecture
was @samp{vax}, the value given to @code{$@{key@}} would be
@samp{vax.bin}.@refill

@item
the value of @code{$@{rhost@}} is expanded and normalized before the
other options are expanded.  The normalization process strips any local
sub-domain components.  For example, if @code{$@{domain@}} was
@samp{Berkeley.EDU} and @code{$@{rhost@}} was initially
@samp{snow.Berkeley.EDU}, after the normalization it would simply be
@samp{snow}.  Hostname normalization is currently done in a
@emph{case-dependent} manner.@refill
@end enumerate

@c======================================================================
@node Selectors, Map Options, Variable Expansion, Location Format
@comment  node-name,  next,  previous,  up
@subsection Selectors
@cindex Selectors

Selectors are used to control the use of a location.  It is possible to
share a mount map between many machines in such a way that filesystem
location, architecture and operating system differences are hidden from
the users.  A selector of the form @samp{arch==sun3;os==sunos4} would only
apply on Sun-3s running SunOS 4.x.

Selectors can be negated by using @samp{!=} instead of @samp{==}.  For
example to select a location on all non-Vax machines the selector
@samp{arch!=vax} would be used.

Selectors are evaluated left to right.  If a selector fails then that
location is ignored.  Thus the selectors form a conjunction and the
locations form a disjunction.  If all the locations are ignored or
otherwise fail then @i{Amd} uses the @dfn{error} filesystem
(@pxref{Error Filesystem}).  This is equivalent to having a location
@samp{type:=error} at the end of each mount-map entry.@refill

The default value of many of the selectors listed here can be overridden
by an @i{Amd} command line switch or in an @i{Amd} configuration file.
@xref{Amd Configuration File}.

These are the selectors currently implemented.

@menu
* arch Selector Variable::
* autodir Selector Variable::
* byte Selector Variable::
* cluster Selector Variable::
* domain Selector Variable::
* host Selector Variable::
* hostd Selector Variable::
* karch Selector Variable::
* os Selector Variable::
* osver Selector Variable::

* key Selector Variable::
* map Selector Variable::
* netnumber Selector Variable::
* network Selector Variable::
* path Selector Variable::
* wire Selector Variable::

* exists Selector Function::
* false Selector Function::
* netgrp Selector Function::
* in_network Selector Function::
* true Selector Function::
@end menu

@c ----------------------------------------------------------------
@node arch Selector Variable, autodir Selector Variable, Selectors, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection arch Selector Variable
@cindex arch Selector Variable
@cindex arch, mount selector
@cindex Mount selector; arch
@cindex Selector; arch

The machine architecture which was automatically determined at compile
time.  The architecture type can be displayed by running the command
@samp{amd -v}.  @xref{Supported Platforms}.@refill

@c ----------------------------------------------------------------
@node autodir Selector Variable, byte Selector Variable, arch Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection autodir Selector Variable
@cindex autodir Selector Variable
@cindex autodir, mount selector
@cindex Mount selector; autodir
@cindex Selector; autodir

The default directory under which to mount filesystems.  This may be
changed by the @code{-a} command line option.  @xref{fs Option}.

@c ----------------------------------------------------------------
@node byte Selector Variable, cluster Selector Variable, autodir Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection byte Selector Variable
@cindex byte Selector Variable
@cindex byte, mount selector
@cindex Mount selector; byte
@cindex Selector; byte

The machine's byte ordering.  This is either @samp{little}, indicating
little-endian, or @samp{big}, indicating big-endian.  One possible use
is to share @samp{rwho} databases (@pxref{rwho servers}).  Another is to
share ndbm databases, however this use can be considered a courageous
juggling act.

@c ----------------------------------------------------------------
@node cluster Selector Variable, domain Selector Variable, byte Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection cluster Selector Variable
@cindex cluster Selector Variable
@cindex cluster, mount selector
@cindex Mount selector; cluster
@cindex Selector; cluster

This is provided as a hook for the name of the local cluster.  This can
be used to decide which servers to use for copies of replicated
filesystems.  @code{$@{cluster@}} defaults to the value of
@code{$@{domain@}} unless a different value is set with the @code{-C}
command line option.

@c ----------------------------------------------------------------
@node domain Selector Variable, host Selector Variable, cluster Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection domain Selector Variable
@cindex domain Selector Variable
@cindex domain, mount selector
@cindex Mount selector; domain
@cindex Selector; domain

The local domain name as specified by the @code{-d} command line option.
@xref{host Selector Variable}.

@c ----------------------------------------------------------------
@node host Selector Variable, hostd Selector Variable, domain Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection host Selector Variable
@cindex host Selector Variable
@cindex host, mount selector
@cindex Mount selector; host
@cindex Selector; host

The local hostname as determined by @b{gethostname}(2).  If no domain
name was specified on the command line and the hostname contains a
period @samp{.} then the string before the period is used as the host
name, and the string after the period is assigned to @code{$@{domain@}}.
For example, if the hostname is @samp{styx.doc.ic.ac.uk} then
@code{host} would be @samp{styx} and @code{domain} would be
@samp{doc.ic.ac.uk}.  @code{hostd} would be
@samp{styx.doc.ic.ac.uk}.@refill

@c ----------------------------------------------------------------
@node hostd Selector Variable, karch Selector Variable, host Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection hostd Selector Variable
@cindex hostd Selector Variable
@cindex hostd, mount selector
@cindex Mount selector; hostd
@cindex Selector; hostd

This resolves to the @code{$@{host@}} and @code{$@{domain@}}
concatenated with a @samp{.} inserted between them if required.  If
@code{$@{domain@}} is an empty string then @code{$@{host@}} and
@code{$@{hostd@}} will be identical.

@c ----------------------------------------------------------------
@node karch Selector Variable, os Selector Variable, hostd Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection karch Selector Variable
@cindex karch Selector Variable
@cindex karch, mount selector
@cindex Mount selector; karch
@cindex Selector; karch

This is provided as a hook for the kernel architecture.  This is used on
SunOS 4 and SunOS 5, for example, to distinguish between different
@samp{/usr/kvm} volumes.  @code{$@{karch@}} defaults to the ``machine''
value gotten from @b{uname}(2).  If the @b{uname}(2) system call is not
available, the value of @code{$@{karch@}} defaults to that of
@code{$@{arch@}}.  Finally, a different value can be set with the @code{-k}
command line option.

@c ----------------------------------------------------------------
@node os Selector Variable, osver Selector Variable, karch Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection os Selector Variable
@cindex os Selector Variable
@cindex os, mount selector
@cindex Mount selector; os
@cindex Selector; os

The operating system.  Like the machine architecture, this is
automatically determined at compile time.  The operating system name can
be displayed by running the command @samp{amd -v}.  @xref{Supported
Platforms}.@refill

@c ----------------------------------------------------------------
@node osver Selector Variable, key Selector Variable, os Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection osver Selector Variable
@cindex osver Selector Variable
@cindex osver, mount selector
@cindex Mount selector; osver
@cindex Selector; osver

The operating system version.  Like the machine architecture, this is
automatically determined at compile time.  The operating system name can
be displayed by running the command @samp{amd -v}.  @xref{Supported
Platforms}.@refill

@c ----------------------------------------------------------------
@ifhtml
<HR>
@end ifhtml
@sp 3
The following selectors are also provided.  Unlike the other selectors,
they vary for each lookup.  Note that when the name from the kernel is
expanded prior to a map lookup, these selectors are all defined as empty
strings.

@c ----------------------------------------------------------------
@node key Selector Variable, map Selector Variable, osver Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection key Selector Variable
@cindex key Selector Variable
@cindex key, mount selector
@cindex Mount selector; key
@cindex Selector; key

The name being resolved.  For example, if @file{/home} is an automount
point, then accessing @file{/home/foo} would set @code{$@{key@}} to the
string @samp{foo}.  The key is prefixed by the @var{pref} option set in
the parent mount point.  The default prefix is an empty string.  If the
prefix was @file{blah/} then @code{$@{key@}} would be set to
@file{blah/foo}.@refill

@c ----------------------------------------------------------------
@node map Selector Variable, netnumber Selector Variable, key Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection map Selector Variable
@cindex map Selector Variable
@cindex map, mount selector
@cindex Mount selector; map
@cindex Selector; map

The name of the mount map being used.

@c ----------------------------------------------------------------
@node netnumber Selector Variable, network Selector Variable, map Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection netnumber Selector Variable
@cindex netnumber Selector Variable
@cindex netnumber, mount selector
@cindex Mount selector; netnumber
@cindex Selector; netnumber

This selector is identical to the @samp{in_network} selector function,
see @ref{in_network Selector Function}.  It will match either the name
or number of @i{any} network interface on which this host is connected
to.  The names and numbers of all attached interfaces are available from
the output of @samp{amd -v}.

@c ----------------------------------------------------------------
@node network Selector Variable, path Selector Variable, netnumber Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection network Selector Variable
@cindex network Selector Variable
@cindex network, mount selector
@cindex Mount selector; network
@cindex Selector; network

This selector is identical to the @samp{in_network} selector function,
see @ref{in_network Selector Function}.  It will match either the name
or number of @i{any} network interface on which this host is connected
to.  The names and numbers of all attached interfaces are available from
the output of @samp{amd -v}.

@c ----------------------------------------------------------------
@node path Selector Variable, wire Selector Variable, network Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection path Selector Variable
@cindex path Selector Variable
@cindex path, mount selector
@cindex Mount selector; path
@cindex Selector; path

The full pathname of the name being resolved.  For example
@file{/home/foo} in the example above.

@c ----------------------------------------------------------------
@node wire Selector Variable, exists Selector Function, path Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection wire Selector Variable
@cindex wire Selector Variable
@cindex wire, mount selector
@cindex Mount selector; wire
@cindex Selector; wire

This selector is identical to the @samp{in_network} selector function,
see @ref{in_network Selector Function}.  It will match either the name
or number of @i{any} network interface on which this host is connected
to.  The names and numbers of all attached interfaces are available from
the output of @samp{amd -v}.

@c ----------------------------------------------------------------
@ifhtml
<HR>
@end ifhtml
@sp 2
The following boolean functions are selectors which take an argument
@i{ARG}.  They return a value of true or false, and thus do not need to
be compared with a value.  Each of these may be negated by prepending
@samp{!} to their name.

@c ----------------------------------------------------------------
@node exists Selector Function, false Selector Function, wire Selector Variable, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection exists Selector Function
@cindex exists Selector Function
@cindex exists, boolean mount selector
@cindex !exists, boolean mount selector
@cindex Mount selector; exists
@cindex Selector; exists

If the file listed by @i{ARG} exists (via @b{lstat}(2)), this function
evaluates to true.  Otherwise it evaluates to false.

@c ----------------------------------------------------------------
@node false Selector Function, netgrp Selector Function, exists Selector Function, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection false Selector Function
@cindex false Selector Function
@cindex false, boolean mount selector
@cindex !false, boolean mount selector
@cindex Mount selector; false
@cindex Selector; false

Always evaluates to false.  @i{ARG} is ignored.

@c ----------------------------------------------------------------
@node netgrp Selector Function, in_network Selector Function, false Selector Function, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection netgrp Selector Function
@cindex netgrp Selector Function
@cindex netgrp, boolean mount selector
@cindex !netgrp, boolean mount selector
@cindex Mount selector; netgrp
@cindex Selector; netgrp

If the current host as determined by the value of @code{$@{host@}} is a
member of the netgroup @i{ARG}, this selector evaluates to true.
Otherwise it evaluates to false.

For example, suppose you have a netgroup @samp{ppp-hosts}, and for
reasons of performance, these have a local @file{/home} partition, while
all other clients on the faster network can access a shared home
directory.  A common map to use for both might look like the following:

@example
home/*  netgrp(ppp-hosts);type:=link;fs:=/local/$@{key@} \
        !netgrp(ppp-hosts);type:=nfs;rhost=serv1;rfs:=/remote/$@{key@}
@end example

@c ----------------------------------------------------------------
@node in_network Selector Function, true Selector Function, netgrp Selector Function, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection in_network Selector Function
@cindex in_network Selector Function
@cindex in_network, boolean mount selector
@cindex !in_network, boolean mount selector
@cindex Mount selector; in_network
@cindex Selector; in_network

If the current host has any network interface that is locally attached
to the network specified in @i{ARG} (either via name or number), this
selector evaluates to true.  Otherwise it evaluates to false.

For example, suppose you have two servers that have an exportable
@file{/opt} that smaller clients can NFS mount.  The two servers are
say, @samp{serv1} on network @samp{foo-net.site.com} and @samp{serv2} on
network @samp{123.4.5.0}.  You can write a map to be used by all clients
that will attempt to mount the closest one as follows:

@example
opt in_network(foo-net.site.com);rhost:=serv1;rfs:=/opt \
    in_network(123.4.5.0);rhost:=serv2;rfs:=/opt \
    rhost:=fallback-server
@end example

@c ----------------------------------------------------------------
@node true Selector Function, , in_network Selector Function, Selectors
@comment  node-name,  next,  previous,  up
@subsubsection true Selector Function
@cindex true Selector Function
@cindex true, boolean mount selector
@cindex !true, boolean mount selector
@cindex Mount selector; true
@cindex Selector; true

Always evaluates to true.  @i{ARG} is ignored.

@c ================================================================
@node Map Options,  , Selectors, Location Format
@comment  node-name,  next,  previous,  up
@subsection Map Options
@cindex Map options
@cindex Setting map options

Options are parsed concurrently with selectors.  The difference is that
when an option is seen the string following the @samp{:=} is
recorded for later use.  As a minimum the @var{type} option must be
specified.  Each filesystem type has other options which must also be
specified.  @xref{Filesystem Types}, for details on the filesystem
specific options.@refill

Superfluous option specifications are ignored and are not reported
as errors.

The following options apply to more than one filesystem type.

@menu
* addopts Option::
* delay Option::
* fs Option::
* opts Option::
* remopts Option::
* sublink Option::
* type Option::
@end menu

@node addopts Option, delay Option, Map Options, Map Options
@comment  node-name,  next,  previous,  up
@subsubsection addopts Option
@cindex Setting additional options on a mount location
@cindex Overriding or adding options to a mount
@cindex addopts, mount option
@cindex Mount option; addopts

This option adds additional options to default options normally
specified in the @samp{/defaults} entry or the defaults of the key entry
being processed (@xref{opts Option}).  Normally when you specify
@samp{opts} in both the @samp{/defaults} and the map entry, the latter
overrides the former completely.  But with @samp{addopts} it will
append the options and override any conflicting ones.

Options which start with @samp{no} will override those with the same
name that do not start with @samp{no} and vice verse.  Special handling
is given to inverted options such as @samp{soft} and @samp{hard},
@samp{bg} and @samp{fg}, @samp{ro} and @samp{rw}, etc.

For example, if the default options specified were
@example
opts:=rw,nosuid,intr,rsize=1024,wsize=1024,quota,posix
@end example

and the ones specified in a map entry were

@example
addopts:=grpid,suid,ro,rsize=2048,quota,nointr
@end example

then the actual options used would be

@example
wsize=1024,posix,grpid,suid,ro,rsize=2048,quota,nointr
@end example

@node delay Option, fs Option, addopts Option, Map Options
@comment  node-name,  next,  previous,  up
@subsubsection delay Option
@cindex Setting a delay on a mount location
@cindex Delaying mounts from specific locations
@cindex Primary server
@cindex Secondary server
@cindex delay, mount option
@cindex Mount option; delay

The delay, in seconds, before an attempt will be made to mount from the
current location.  Auxiliary data, such as network address, file handles
and so on are computed regardless of this value.

A delay can be used to implement the notion of primary and secondary
file servers.  The secondary servers would have a delay of a few
seconds, thus giving the primary servers a chance to respond first.

@node fs Option, opts Option, delay Option, Map Options
@comment  node-name,  next,  previous,  up
@subsubsection fs Option
@cindex Setting the local mount point
@cindex Overriding the default mount point
@cindex fs, mount option
@cindex Mount option; fs

The local mount point.  The semantics of this option vary between
filesystems.

For NFS and UFS filesystems the value of @code{$@{fs@}} is used as the
local mount point.  For other filesystem types it has other meanings
which are described in the section describing the respective filesystem
type.  It is important that this string uniquely identifies the
filesystem being mounted.  To satisfy this requirement, it should
contain the name of the host on which the filesystem is resident and the
pathname of the filesystem on the local or remote host.

The reason for requiring the hostname is clear if replicated filesystems
are considered.  If a fileserver goes down and a replacement filesystem
is mounted then the @dfn{local} mount point @dfn{must} be different from
that of the filesystem which is hung.  Some encoding of the filesystem
name is required if more than one filesystem is to be mounted from any
given host.

If the hostname is first in the path then all mounts from a particular
host will be gathered below a single directory.  If that server goes
down then the hung mount points are less likely to be accidentally
referenced, for example when @b{getcwd}(3) traverses the namespace to
find the pathname of the current directory.

The @samp{fs} option defaults to
@code{$@{autodir@}/$@{rhost@}$@{rfs@}}.  In addition,
@samp{rhost} defaults to the local host name (@code{$@{host@}}) and
@samp{rfs} defaults to the value of @code{$@{path@}}, which is the full
path of the requested file; @samp{/home/foo} in the example above
(@pxref{Selectors}).  @code{$@{autodir@}} defaults to @samp{/a} but may
be changed with the @code{-a} command line option.  Sun's automounter
defaults to @samp{/tmp_mnt}.  Note that there is no @samp{/} between
the @code{$@{rhost@}} and @code{$@{rfs@}} since @code{$@{rfs@}} begins
with a @samp{/}.@refill

@node opts Option, remopts Option, fs Option, Map Options
@comment  node-name,  next,  previous,  up
@subsubsection opts Option
@cindex Setting system mount options
@cindex Passing parameters to the mount system call
@cindex mount system call
@cindex mount system call flags
@cindex The mount system call
@cindex opts, mount option
@cindex Mount option; opts

The options to pass to the mount system call.  A leading @samp{-} is
silently ignored.  The mount options supported generally correspond to
those used by @b{mount}(8) and are listed below.  Some additional
pseudo-options are interpreted by @i{Amd} and are also listed.

Unless specifically overridden, each of the system default mount options
applies.  Any options not recognized are ignored.  If no options list is
supplied the string @samp{rw,defaults} is used and all the system
default mount options apply.  Options which are not applicable for a
particular operating system are silently ignored.  For example, only 4.4BSD
is known to implement the @code{compress} and @code{spongy} options.

@table @code

@item acdirmax=@var{n}
@cindex Mount flags; acdirmax
Set the maximum directory attribute cache timeout to @var{n}.

@item acdirmin=@var{n}
@cindex Mount flags; acdirmin
Set the minimum directory attribute cache timeout to @var{n}.

@item acregmax=@var{n}
@cindex Mount flags; acregmax
Set the maximum file attribute cache timeout to @var{n}.

@item acregmin=@var{n}
@cindex Mount flags; acregmin
Set the minimum file attribute cache timeout to @var{n}.

@item actimeo=@var{n}
@cindex Mount flags; actimeo
Set the overall attribute cache timeout to @var{n}.

@item auto
@cindex Mount flags; auto
@itemx ignore
@cindex Mount flags; ignore
Ignore this mount by @b{df}(1).

@item cache
@cindex Mount flags; cache
Allow data to be cached from a remote server for this mount.

@item compress
@cindex Mount flags; compress
Use NFS compression protocol.

@item defperm
@cindex Mount flags; defperm
Ignore the permission mode bits, and default file permissions to 0555,
UID 0, and GID 0.  Useful for CD-ROMs formatted as ISO-9660.

@item dev
@cindex Mount flags; dev
Allow local special devices on this filesystem.

@item dumbtimr
@cindex Mount flags; dumbtimr
(XXX: a dumb timer?)

@item fsid
@cindex Mount flags; fsid
Set ID of filesystem.

@item grpid
@cindex Mount flags; grpid
Use BSD directory group-id semantics.

@item int
@cindex Mount flags; int
@itemx intr
@cindex Mount flags; intr
Allow keyboard interrupts on hard mounts.

@item multi
@cindex Mount flags; multi
Perform multi-component lookup on files.

@item maxgroups
@cindex Mount flags; maxgroups
Set the maximum number of groups to allow for this mount.

@item nfsv3
@cindex Mount flags; nfsv3
Use NFS Version 3 for this mount.

@item noac
@cindex Mount flags; noac
Turn off the attribute cache.

@item noauto
@cindex Mount flags; noauto
(XXX: No automatic what?)

@item nocache
@cindex Mount flags; nocache
Do not allow data to be cached from a remote server for this
mount.

@item noconn
@cindex Mount flags; noconn
Don't make a connection on datagram transports.

@item nocto
@cindex Mount flags; nocto
No close-to-open consistency.

@item nodefperm
@cindex Mount flags; nodefperm
Do not ignore the permission mode bits.  Useful for CD-ROMS formatted as
ISO-9660.

@item nodev
@cindex Mount flags; nodev
@itemx nodevs
@cindex Mount flags; nodevs
Don't allow local special devices on this filesystem.

@item noint
@cindex Mount flags; noint
Do not allow keyboard interrupts for this mount

@item nosub
@cindex Mount flags; nosub
Disallow mounts beneath this mount.

@item nosuid
@cindex Mount flags; nosuid
Don't allow set-uid or set-gid executables on this filesystem.

@item noversion
@cindex Mount flags; noversion
Strip the extension @samp{;#} from the version string of files recorded
on an ISO-9660 CD-ROM.

@item overlay
@cindex Mount flags; overlay
Overlay this mount on top of an existing mount, if any.

@item pgthresh=@var{n}
@cindex Mount flags; pgthresh
Set the paging threshold to @var{n} kilobytes.

@item port=@var{n}
@cindex Mount flags; port
Set the NFS port to @var{n}.

@item posix
@cindex Mount flags; posix
Turn on POSIX static pathconf for mounts.

@item proto=@var{s}
@cindex Mount flags; proto
Use transport protocol @var{s} for NFS (can be @code{"tcp"} or @code{"udp"}).

@item quota
@cindex Mount flags; quota
Enable quota checking on this mount.

@item rdonly
@cindex Mount flags; rdonly
@itemx ro
@cindex Mount flags; ro
Mount this filesystem readonly.

@item resvport
@cindex Mount flags; resvport
Use a reserved port (smaller than 1024) for remote NFS mounts.  Most
systems assume that, but some allow for mounts to occur on non-reserved
ports.   This causes problems when such a system tries to NFS mount one
that requires reserved ports.  It is recommended that this option always
be on.

@item retrans=@i{n}
@cindex Mount flags; retrans
The number of NFS retransmits made before a user error is generated by a
@samp{soft} mounted filesystem, and before a @samp{hard} mounted
filesystem reports @samp{NFS server @dfn{yoyo} not responding still
trying}.

@item retry
@cindex Mount flags; retry
Set the NFS retry counter.

@item rrip
@cindex Mount flags; rrip
Uses the Rock Ridge Interchange Protocol (RRIP) extensions to ISO-9660.

@item rsize=@var{n}
@cindex Mount flags; rsize
The NFS read packet size.  You may need to set this if you are using
NFS/UDP through a gateway or a slow link.

@item rw
@cindex Mount flags; rw
Allow reads and writes on this filesystem.

@item soft
@cindex Mount flags; soft
Give up after @dfn{retrans} retransmissions.

@item spongy
@cindex Mount flags; spongy
Like @samp{soft} for status requests, and @samp{hard} for data transfers.

@item suid
@cindex Mount flags; suid
Allow set-uid programs on this mount.

@item symttl
@cindex Mount flags; symttl
Turn of the symbolic link cache time-to-live.

@item sync
@cindex Mount flags; sync
Perform synchronous filesystem operations on this mount.

@item tcp
@cindex Mount flags; tcp
Use TCP/IP instead of UDP/IP, ignored if the NFS implementation does not
support TCP/IP mounts.

@item timeo=@var{n}
@cindex Mount flags; timeo
The NFS timeout, in tenth-seconds, before a request is retransmitted.

@item vers=@var{n}
@cindex Mount flags; vers
 Use NFS protocol version number @var{n} (can be 2 or 3).

@item wsize=@var{n}
@cindex Mount flags; wsize
The NFS write packet size.  You may need to set this if you are using
NFS/UDP through a gateway or a slow link.

@end table

The following options are implemented by @i{Amd}, rather than being
passed to the kernel.

@table @code

@item nounmount
@cindex Mount flags; nounmount
Configures the mount so that its time-to-live will
never expire.  This is also the default for some filesystem types.
@c
@c Implementation broken:

@item ping=@var{n}
@cindex Mount flags; ping
The interval, in seconds, between keep-alive pings.  When four
consecutive pings have failed the mount point is marked as hung.  This
interval defaults to 30 seconds.  If the ping interval is less than zero,
no pings are sent and the host is assumed to be always
up.  By default, pings are not sent for an NFS/TCP mount.

@item retry=@var{n}
@cindex Mount flags; retry=@var{n}
The number of times to retry the mount system call.

@item utimeout=@var{n}
@cindex Mount flags; utimeout=@var{n}
The interval, in seconds, by which the mount's
time-to-live is extended after an unmount attempt
has failed.  In fact the interval is extended before the unmount is
attempted to avoid thrashing.  The default value is 120 seconds (two
minutes) or as set by the @code{-w} command line option.

@end table

@node remopts Option, sublink Option, opts Option, Map Options
@comment  node-name,  next,  previous,  up
@subsubsection remopts Option
@cindex Setting system mount options for non-local networks
@cindex remopts, mount option
@cindex Mount option; remopts

This option has the same use as @code{$@{opts@}} but applies only when
the remote host is on a non-local network.  For example, when using NFS
across a gateway it is often necessary to use smaller values for the
data read and write sizes.  This can simply be done by specifying the
small values in @var{remopts}.  When a non-local host is accessed, the
smaller sizes will automatically be used.

@i{Amd} determines whether a host is local by examining the network
interface configuration at startup.  Any interface changes made after
@i{Amd} has been started will not be noticed.  The likely effect will
be that a host may incorrectly be declared non-local.

Unless otherwise set, the value of @code{$@{remopts@}} is the same as
the value of @code{$@{opts@}}.

@node sublink Option, type Option, remopts Option, Map Options
@comment  node-name,  next,  previous,  up
@subsubsection sublink Option
@cindex Setting the sublink option
@cindex sublink, mount option
@cindex Mount option; sublink

The subdirectory within the mounted filesystem to which the reference
should point.  This can be used to prevent duplicate mounts in cases
where multiple directories in the same mounted filesystem are used.

@node type Option, , sublink Option, Map Options
@comment  node-name,  next,  previous,  up
@subsubsection type Option
@cindex Setting the filesystem type option
@cindex type, mount option
@cindex Mount option; type

The filesystem type to be used.  @xref{Filesystem Types}, for a full
description of each type.@refill

@c ################################################################
@node Amd Command Line Options, Filesystem Types, Mount Maps, Top
@comment  node-name,  next,  previous,  up
@chapter @i{Amd} Command Line Options
@cindex Command line options, Amd
@cindex Amd command line options
@cindex Overriding defaults on the command line

Many of @i{Amd}'s parameters can be set from the command line.  The
command line is also used to specify automount points and maps.

The general format of a command line is

@example
amd [@i{options}] [@{ @i{directory} @i{map-name} [-@i{map-options}] @} ...]
@end example

For each directory and map-name given or specified in the
@file{amd.conf} file, @i{Amd} establishes an automount point.  The
@dfn{map-options} may be any sequence of options or
selectors---@pxref{Location Format}.  The @dfn{map-options} apply only
to @i{Amd}'s mount point.

@samp{type:=toplvl;cache:=mapdefault;fs:=$@{map@}} is the default value for the
map options.  Default options for a map are read from a special entry in
the map whose key is the string @samp{/defaults}.  When default options
are given they are prepended to any options specified in the mount-map
locations as explained in @ref{Map Defaults}.

The @dfn{options} are any combination of those listed below.

Once the command line has been parsed, the automount points are mounted.
The mount points are created if they do not already exist, in which case they
will be removed when @i{Amd} exits.
Finally, @i{Amd} disassociates itself from its controlling terminal and
forks into the background.

Note: Even if @i{Amd} has been built with @samp{-DDEBUG} (via
@code{configure --enable-debug}), it will still background itself and
disassociate itself from the controlling terminal.  To use a debugger it
is necessary to specify @samp{-D nodaemon} on the command line.
However, even with all of this, mounts and unmounts are performed in the
background, and @i{Amd} will always fork before doing them.  Therefore,
debugging what happens closely during un/mounts is more challenging.

@emph{All} of @i{Amd}'s command options (save @code{-F} and @code{-T})
can be specified in the @file{amd.conf} file. @xref{Amd Configuration
File}.  If @i{Amd} is invoked without any command line options, it will
default to using the configuration file @file{/etc/amd.conf}, if one
exists.

@menu
* -a Option::   Automount directory.
* -c Option::   Cache timeout interval.
* -d Option::   Domain name.
* -k Option::   Kernel architecture.
* -l Option::   Log file.
* -n Option::   Hostname normalization.
* -o Option::   Operating system version.
* -p Option::   Output process id.
* -r Option::   Restart existing mounts.
* -t Option::   Kernel RPC timeout.
* -v Option::   Version information.
* -w Option::   Wait interval after failed unmount.
* -x Option::   Log options.
* -y Option::   NIS domain.
* -C-Option::   Cluster name.
* -D-Option::   Debug flags.
* -F Option::   Amd configuration file.
* -H Option::   Show brief help.
* -O-Option::   Operating system name.
* -S Option::   Lock executable pages in memory.
* -T-Option::   Set tag for configuration file.
@end menu

@c ----------------------------------------------------------------
@node -a Option, -c Option, Amd Command Line Options, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-a} @var{directory}
@cindex Automount directory
@cindex Setting the default mount directory

Specifies the default mount directory.  This option changes the variable
@code{$@{autodir@}} which otherwise defaults to @file{/a}.  For example,
some sites prefer @file{/amd} or @file{/n}.

@example
amd -a /amd ...
@end example

@c ----------------------------------------------------------------
@node -c Option, -d Option, -a Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-c} @var{cache-interval}
@cindex Cache interval
@cindex Interval before a filesystem times out
@cindex Setting the interval before a filesystem times out
@cindex Changing the interval before a filesystem times out

Selects the period, in seconds, for which a name is cached by @i{Amd}.
If no reference is made to the volume in this period, @i{Amd} discards
the volume name to filesystem mapping.

Once the last reference to a filesystem has been removed, @i{Amd}
attempts to unmount the filesystem.  If the unmount fails the interval
is extended by a further period as specified by the @samp{-w} command
line option or by the @samp{utimeout} mount option.

The default @dfn{cache-interval} is 300 seconds (five minutes).

@c ----------------------------------------------------------------
@node -d Option, -k Option, -c Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-d} @var{domain}
@cindex Domain name
@cindex Setting the local domain name
@cindex Overriding the local domain name

Specifies the host's domain.  This sets the internal variable
@code{$@{domain@}} and affects the @code{$@{hostd@}} variable.

If this option is not specified and the hostname already contains the
local domain then that is used, otherwise the default value of
@code{$@{domain@}} is @samp{unknown.domain}.

For example, if the local domain was @samp{doc.ic.ac.uk}, @i{Amd} could
be started as follows:

@example
amd -d doc.ic.ac.uk ...
@end example

@c ----------------------------------------------------------------
@node -k Option, -l Option, -d Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-k} @var{kernel-architecture}
@cindex Setting the Kernel architecture

Specifies the kernel architecture of the system.  This is usually the
output of @samp{uname -m} (the ``machine'' value gotten from
@b{uname}(2)).  If the @b{uname}(2) system call is not available, the
value of @code{$@{karch@}} defaults to that of @code{$@{arch@}}.

The only effect of this option is to set the variable @code{$@{karch@}}.

This option would be used as follows:

@example
amd -k `arch -k` ...
@end example

@c ----------------------------------------------------------------
@node -l Option, -n Option, -k Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-l} @var{log-option}
@cindex Log filename
@cindex Setting the log file
@cindex Using syslog to log errors
@cindex syslog

Selects the form of logging to be made.  Several special @dfn{log-options}
are recognized.

@enumerate
@item
If @dfn{log-option} is the string @samp{syslog}, @i{Amd} will use the
@b{syslog}(3) mechanism.  If your system supports syslog facilities, then
the default facility used is @samp{LOG_DAEMON}.

@item
@cindex syslog facility; specifying an alternate
When using syslog, if you wish to change the facility, append its name
to the log option name, delimited by a single colon.  For example, if
@dfn{log-options} is the string @samp{syslog:local7} then @b{Amd} will
log messages via @b{syslog}(3) using the @samp{LOG_LOCAL7} facility.  If
the facility name specified is not recognized, @i{Amd} will default to
@samp{LOG_DAEMON}.  Note: while you can use any syslog facility
available on your system, it is generally a bad idea to use those
reserved for other services such as @samp{kern}, @samp{lpr},
@samp{cron}, etc.

@item
If @dfn{log-option} is the string @samp{/dev/stderr}, @i{Amd} will use
standard error, which is also the default target for log messages.  To
implement this, @i{Amd} simulates the effect of the @samp{/dev/fd}
driver.
@end enumerate

Any other string is taken as a filename to use for logging.  Log
messages are appended to the file if it already exists, otherwise a new
file is created.  The file is opened once and then held open, rather
than being re-opened for each message.

Normally, when long-running daemons hold an open file descriptor on a
log file, it is impossible to ``rotate'' the log file and compress older
logs on a daily basis.  The daemon needs to be told to discard (via
@b{close}(2)) its file handle, and re-open the log file.  This is done
using @code{amq -l} @i{log-option}. @xref{Amq -l option}.

If the @samp{syslog} option is specified but the system does not support
syslog or if the named file cannot be opened or created, @i{Amd} will
use standard error.  Error messages generated before @i{Amd} has
finished parsing the command line are printed on standard error.

Since @i{Amd} tends to generate a lot of logging information (especially
if debugging was turned on), and due to it being an important program
running on the system, it is usually best to log to a separate disk
file.  In that case @i{Amd} would be started as follows:

@example
amd -l /var/log/amd ...
@end example

@c ----------------------------------------------------------------
@node -n Option, -o Option, -l Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-n}
@cindex Hostname normalization
@cindex Aliased hostnames
@cindex Resolving aliased hostnames
@cindex Normalizing hostnames

Normalizes the remote hostname before using it.  Normalization is done
by replacing the value of @code{$@{rhost@}} with the (generally fully
qualified) primary name returned by a hostname lookup.

This option should be used if several names are used to refer to a
single host in a mount map.

@c ----------------------------------------------------------------
@node -o Option, -p Option, -n Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-o} @var{op-sys-ver}
@cindex Operating System version
@cindex Setting the Operating System version

Override the compiled-in version number of the operating system, with
@var{op-sys-ver}.  Useful when the built-in version is not desired for
backward compatibility reasons.  For example, if the built-in version is
@samp{2.5.1}, you can override it to @samp{5.5.1}, and use older maps
that were written with the latter in mind.

@c ----------------------------------------------------------------
@node -p Option, -r Option, -o Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-p}
@cindex Process id
@cindex Displaying the process id
@cindex process id of Amd daemon
@cindex pid file, creating with -p option
@cindex Creating a pid file

Causes @i{Amd}'s process id to be printed on standard output.
This can be redirected to a suitable file for use with kill:

@example
amd -p > /var/run/amd.pid ...
@end example

This option only has an affect if @i{Amd} is running in daemon mode.
If @i{Amd} is started with the @code{-D nodaemon} debug flag, this
option is ignored.

@c ----------------------------------------------------------------
@node -r Option, -t Option, -p Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-r}
@cindex Restarting existing mounts
@cindex Picking up existing mounts

Tells @i{Amd} to restart existing mounts (@pxref{Inheritance Filesystem}).
@c @dfn{This option will be made the default in the next release.}

@c ----------------------------------------------------------------
@node -t Option, -v Option, -r Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-t} @var{timeout.retransmit}
@cindex Setting Amd's RPC parameters

Specifies the RPC @dfn{timeout} and @dfn{retransmit} intervals used by
the kernel to communicate to @i{Amd}.  These are used to set the
@samp{timeo} and @samp{retrans} mount options.

@i{Amd} relies on the kernel RPC retransmit mechanism to trigger mount
retries.  The value of this parameter changes the retry interval.  Too
long an interval gives poor interactive response, too short an interval
causes excessive retries.

@c ----------------------------------------------------------------
@node -v Option, -w Option, -t Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-v}
@cindex Version information
@cindex Discovering version information
@cindex How to discover your version of Amd

Print version information on standard error and then exit.  The output
is of the form:

@example
Copyright (c) 1997-1998 Erez Zadok
Copyright (c) 1990 Jan-Simon Pendry
Copyright (c) 1990 Imperial College of Science, Technology & Medicine
Copyright (c) 1990 The Regents of the University of California.
am-utils version 6.0a15 (build 61).
Built by ezk@@cs.columbia.edu on date Wed Oct 22 15:21:03 EDT 1997.
cpu=sparc (big-endian), arch=sun4, karch=sun4u.
full_os=solaris2.5.1, os=sos5, osver=5.5.1, vendor=sun.
Map support for: root, passwd, union, nisplus, nis, ndbm, file, error.
AMFS: nfs, link, nfsx, nfsl, host, linkx, program, union, inherit,
      ufs, lofs, hsfs, pcfs, auto, direct, toplvl, error.
FS: autofs, cachefs, cdfs, lofs, nfs, nfs3, pcfs, tfs, tmpfs, ufs.
Network 1: wire="mcl-lab-net.cs.columbia.edu" (netnumber=128.59.13).
Network 2: wire="14-net.cs.columbia.edu" (netnumber=128.59.14).
Network 3: wire="old-net.cs.columbia.edu" (netnumber=128.59.16).
@end example

The information includes the version number, number of times @i{Amd} was
compiled on the local system, release date and name of the release.
Following come the cpu type, byte ordering, and the architecture and
kernel architecture as @code{$@{arch@}} and @code{$@{karch@}},
respectively.  The next line lists the full name of the system, the
variables @code{$@{os@}} and @code{$@{osver@}}, and the vendor's
name. @xref{Supported Platforms}.

Then come a list of map types supported, filesystems internally
supported by @i{Amd} (AMFS), and generic filesystems available (FS).
Finally all known networks (if any) of this host are listed by name
and number.  They are available via the variables
@code{$@{wire@}} or @code{$@{network@}}, and
@code{$@{netnumber@}} (@pxref{Selectors}) or the @samp{in_network}
selector function (@pxref{in_network Selector Function}).

@c ----------------------------------------------------------------
@node -w Option, -x Option, -v Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-w} @var{wait-timeout}
@cindex Setting the interval between unmount attempts
@cindex unmount attempt backoff interval

Selects the interval in seconds between unmount attempts after the
initial time-to-live has expired.

This defaults to 120 seconds (two minutes).

@c ----------------------------------------------------------------
@node -x Option, -y Option, -w Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-x} @var{opts}
@cindex Log message selection
@cindex Selecting specific log messages
@cindex How to select log messages
@cindex syslog priorities

Specifies the type and verbosity of log messages.  @dfn{opts} is
a comma separated list selected from the following options:

@table @code
@item fatal
Fatal errors
@item error
Non-fatal errors
@item user
Non-fatal user errors
@item warn
Recoverable errors
@item warning
Alias for @code{warn}
@item info
Information messages
@item map
Mount map usage
@item stats
Additional statistics
@item all
All of the above
@end table

Initially a set of default logging flags is enabled.  This is as if
@samp{-x all,nomap,nostats} had been selected.  The command line is
parsed and logging is controlled by the @code{-x} option.  The very first
set of logging flags is saved and can not be subsequently disabled using
@i{Amq}.  This default set of options is useful for general production
use.@refill

The @samp{info} messages include details of what is mounted and
unmounted and when filesystems have timed out.  If you want to have the
default set of messages without the @samp{info} messages then you simply
need @samp{-x noinfo}.  The messages given by @samp{user} relate to
errors in the mount maps, so these are useful when new maps are
installed.  The following table lists the syslog priorities used for each
of the message types.@refill

@table @code
@item fatal
@samp{LOG_CRIT}
@item error
@samp{LOG_ERR}
@item user
@samp{LOG_WARNING}
@item warning
@samp{LOG_WARNING}
@item info
@samp{LOG_INFO}
@item debug
@samp{LOG_DEBUG}
@item map
@samp{LOG_DEBUG}
@item stats
@samp{LOG_INFO}
@end table


The options can be prefixed by the string @samp{no} to indicate
that this option should be turned off.  For example, to obtain all
but @samp{info} messages the option @samp{-x all,noinfo} would be used.

If @i{Amd} was built with debugging enabled the @code{debug} option is
automatically enabled regardless of the command line options.

@c ----------------------------------------------------------------
@node -y Option, -C-Option, -x Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-y} @var{NIS-domain}
@cindex NIS (YP) domain name
@cindex Overriding the NIS (YP) domain name
@cindex Setting the NIS (YP) domain name
@cindex YP domain name

Selects an alternate NIS domain.  This is useful for debugging and
cross-domain shared mounting.  If this flag is specified, @i{Amd}
immediately attempts to bind to a server for this domain.
@c @i{Amd} refers to NIS maps when it starts, unless the @code{-m} option
@c is specified, and whenever required in a mount map.

@c ----------------------------------------------------------------
@node -C-Option, -D-Option, -y Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-C} @var{cluster-name}
@cindex Cluster names
@cindex Setting the cluster name

Specifies the name of the cluster of which the local machine is a member.
The only effect is to set the variable @code{$@{cluster@}}.
The @dfn{cluster-name} is will usually obtained by running another command which uses
a database to map the local hostname into a cluster name.
@code{$@{cluster@}} can then be used as a selector to restrict mounting of
replicated data.
If this option is not given, @code{$@{cluster@}} has the same value as @code{$@{domain@}}.
This would be used as follows:

@example
amd -C `clustername` ...
@end example

@c ----------------------------------------------------------------
@node -D-Option, -F Option, -C-Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-D} @var{opts}
@cindex Debug options
@cindex Setting debug flags

Controls the verbosity and coverage of the debugging trace; @dfn{opts}
is a comma separated list of debugging options.  The @code{-D} option is
only available if @i{Amd} was compiled with @samp{-DDEBUG}, or
configured with @code{configure --enable-debug}.  The memory debugging
facilities (@samp{mem}) are only available if @i{Amd} was compiled with
@samp{-DDEBUG_MEM} (in addition to @samp{-DDEBUG}), or configured with
@code{configure --enable-debug=mem}.

The most common options to use are @samp{-D trace} and @samp{-D test}
(which turns on all the useful debug options).  As usual, every option
can be prefixed with @samp{no} to turn it off.

@table @code
@item all
all options
@item amq
register for amq
@item daemon
enter daemon mode
@item fork
fork server
@item full
program trace
@item info
@cindex debugging hesiod resolver service
@cindex Hesiod: turning on RES_DEBUG
info service specific debugging (hesiod, nis, etc.)  In the case of
hesiod maps, turns on the hesiod RES_DEBUG internal debugging option.
@item mem
trace memory allocations
@item mtab
use local @file{./mtab} file
@item str
debug string munging
@item test
full debug but no daemon
@item trace
protocol trace
@end table

You may also refer to the program source for a more detailed explanation
of the available options.

@c ----------------------------------------------------------------
@node -F Option, -H Option, -D-Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-F} @var{conf-file}
@cindex Amd configuration file; specifying name
@cindex Amd configuration file
@cindex amd.conf file

Specify an @i{Amd} configuration file @var{conf-file} to use.  For a
description of the format and syntax, @pxref{Amd Configuration File}.
This configuration file is used to specify any options in lieu of typing
many of them on the command line.  The @file{amd.conf} file includes
directives for every command line option @i{Amd} has, and many more that
are only available via the configuration file facility.  The
configuration file specified by this option is processed after all other
options had been processed, regardless of the actual location of this
option on the command line.

@c ----------------------------------------------------------------
@node -H Option, -O-Option, -F Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-H}
@cindex Displaying brief help
@cindex Help; showing from Amd

Print a brief help and usage string.

@c ----------------------------------------------------------------
@node -O-Option, -S Option, -H Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-O} @var{op-sys-name}
@cindex Operating System name
@cindex Setting the Operating System name

Override the compiled-in name of the operating system, with
@var{op-sys-name}.  Useful when the built-in name is not desired for
backward compatibility reasons.  For example, if the build in name is
@samp{sunos5}, you can override it to the old name @samp{sos5}, and use
older maps which were written with the latter in mind.

@c ----------------------------------------------------------------
@node -S Option, -T-Option, -O-Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-S}
@cindex plock; using
@cindex locking executable pages in memory

Do @emph{not} lock the running executable pages of @i{Amd} into memory.
To improve @i{Amd}'s performance, systems that support the @b{plock}(3)
call lock the @i{Amd} process into memory.  This way there is less
chance the operating system will schedule, page out, and swap the
@i{Amd} process as needed.  This tends to improve @i{Amd}'s performance,
at the cost of reserving the memory used by the @i{Amd} process (making
it unavailable for other processes).  If this behavior is not desired,
use the @code{-S} option.

@c ----------------------------------------------------------------
@node -T-Option, , -S Option, Amd Command Line Options
@comment  node-name,  next,  previous,  up
@section @code{-T} @var{tag}
@cindex Tags for Amd configuration file
@cindex Configuration file; tags

Specify a tag to use with @file{amd.conf}.  All map entries tagged with
@var{tag} will be processed.  Map entries that are not tagged are always
processed.  Map entries that are tagged with a tag other than @var{tag}
will not be processed.

@c ################################################################
@node Filesystem Types, Amd Configuration File, Amd Command Line Options, Top
@comment  node-name,  next,  previous,  up
@chapter Filesystem Types
@cindex Filesystem types
@cindex Mount types
@cindex Types of filesystem

To mount a volume, @i{Amd} must be told the type of filesystem to be
used.  Each filesystem type typically requires additional information
such as the fileserver name for NFS.

From the point of view of @i{Amd}, a @dfn{filesystem} is anything that
can resolve an incoming name lookup.  An important feature is support
for multiple filesystem types.  Some of these filesystems are
implemented in the local kernel and some on remote fileservers, whilst
the others are implemented internally by @i{Amd}.@refill

The two common filesystem types are UFS and NFS.  Four other user
accessible filesystems (@samp{link}, @samp{program}, @samp{auto} and
@samp{direct}) are also implemented internally by @i{Amd} and these are
described below.  There are two additional filesystem types internal to
@i{Amd} which are not directly accessible to the user (@samp{inherit}
and @samp{error}).  Their use is described since they may still have an
effect visible to the user.@refill

@menu
* Network Filesystem::          A single NFS filesystem.
* Network Host Filesystem::     NFS mount a host's entire export tree.
* Network Filesystem Group::    An atomic group of NFS filesystems.
* Unix Filesystem::             Native disk filesystem.
* Caching Filesystem::          Caching from remote server filesystem.
* CD-ROM Filesystem::           ISO9660 CD ROM.
* Loopback Filesystem::         Local loopback-mount filesystem.
* Memory/RAM Filesystem::       A memory or RAM-based filesystem.
* Null Filesystem::             4.4BSD's loopback-mount filesystem.
* Floppy Filesystem::           MS-DOS Floppy filesystem.
* Translucent Filesystem::      The directory merging filesystem.
* Shared Memory+Swap Filesystem:: Sun's tmpfs filesystem.
* User ID Mapping Filesystem::  4.4BSD's umapfs filesystem.
* Program Filesystem::          Generic Program mounts.
* Symbolic Link Filesystem::    Local link.
* Symbolic Link Filesystem II:: Local link referencing existing filesystem.
* NFS-Link Filesystem::         Link if path exists, NFS otherwise.
* Automount Filesystem::
* Direct Automount Filesystem::
* Union Filesystem::
* Error Filesystem::
* Top-level Filesystem::
* Autofs Filesystem::           Sun's kernel-based automounter filesystem.
* Root Filesystem::
* Inheritance Filesystem::
@end menu

@c ----------------------------------------------------------------
@node Network Filesystem, Network Host Filesystem, Filesystem Types, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Network Filesystem (@samp{nfs})
@cindex NFS
@cindex Mounting an NFS filesystem
@cindex How to mount and NFS filesystem
@cindex nfs, filesystem type
@cindex Filesystem type; nfs

The @dfn{nfs} (@samp{type:=nfs}) filesystem type provides access to Sun's NFS.

@noindent
The following options must be specified:

@table @code
@cindex rhost, mount option
@cindex Mount option; rhost
@item rhost
the remote fileserver.  This must be an entry in the hosts database.  IP
addresses are not accepted.  The default value is taken
from the local host name (@code{$@{host@}}) if no other value is
specified.

@cindex rfs, mount option
@cindex Mount option; rfs
@item rfs
the remote filesystem.
If no value is specified for this option, an internal default of
@code{$@{path@}} is used.
@end table

NFS mounts require a two stage process.  First, the @dfn{file handle} of
the remote file system must be obtained from the server.  Then a mount
system call must be done on the local system.  @i{Amd} keeps a cache
of file handles for remote file systems.  The cache entries have a
lifetime of a few minutes.

If a required file handle is not in the cache, @i{Amd} sends a request
to the remote server to obtain it.  @i{Amd} @dfn{does not} wait for
a response; it notes that one of the locations needs retrying, but
continues with any remaining locations.  When the file handle becomes
available, and assuming none of the other locations was successfully
mounted, @i{Amd} will retry the mount.  This mechanism allows several
NFS filesystems to be mounted in parallel.
@c @footnote{The mechanism
@c is general, however NFS is the only filesystem
@c for which the required hooks have been written.}
The first one which responds with a valid file handle will be used.

@noindent
An NFS entry might be:

@example
jsp  host!=charm;type:=nfs;rhost:=charm;rfs:=/home/charm;sublink:=jsp
@end example

The mount system call and any unmount attempts are always done
in a new task to avoid the possibility of blocking @i{Amd}.

@c ----------------------------------------------------------------
@node Network Host Filesystem, Network Filesystem Group, Network Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Network Host Filesystem (@samp{host})
@cindex Network host filesystem
@cindex Mounting entire export trees
@cindex How to mount all NFS exported filesystems
@cindex host, filesystem type
@cindex Filesystem type; host

@c NOTE: the current implementation of the @dfn{host} filesystem type
@c sometimes fails to maintain a consistent view of the remote mount tree.
@c This happens when the mount times out and only some of the remote mounts
@c are successfully unmounted.  To prevent this from occurring, use the
@c @samp{nounmount} mount option.

The @dfn{host} (@samp{type:=host}) filesystem allows access to the entire export tree of an
NFS server.  The implementation is layered above the @samp{nfs}
implementation so keep-alives work in the same way.  The only option
which needs to be specified is @samp{rhost} which is the name of the
fileserver to mount.

The @samp{host} filesystem type works by querying the mount daemon on
the given fileserver to obtain its export list.  @i{Amd} then obtains
filehandles for each of the exported filesystems.  Any errors at this
stage cause that particular filesystem to be ignored.  Finally each
filesystem is mounted.  Again, errors are logged but ignored.  One
common reason for mounts to fail is that the mount point does not exist.
Although @i{Amd} attempts to automatically create the mount point, it
may be on a remote filesystem to which @i{Amd} does not have write
permission.

When an attempt to unmount a @samp{host} filesystem mount fails, @i{Amd}
remounts any filesystems which had successfully been unmounted.  To do
this @i{Amd} queries the mount daemon again and obtains a fresh copy of
the export list.  @i{Amd} then tries to mount any exported filesystems
which are not currently mounted.

Sun's automounter provides a special @samp{-hosts} map.  To achieve the
same effect with @i{Amd} requires two steps.  First a mount map must
be created as follows:

@example
*       type:=host;rhost:=$@{key@};fs:=$@{autodir@}/$@{rhost@}/root
@end example

@noindent
and then start @i{Amd} with the following command

@example
amd /net net.map
@end example

@noindent
where @samp{net.map} is the name of map described above.  Note that the
value of @code{$@{fs@}} is overridden in the map.  This is done to avoid
a clash between the mount tree and any other filesystem already mounted
from the same fileserver.

If different mount options are needed for different hosts then
additional entries can be added to the map, for example

@example
host2       opts:=ro,nosuid,soft
@end example

@noindent
would soft mount @samp{host2} read-only.

@c ----------------------------------------------------------------
@node Network Filesystem Group, Unix Filesystem, Network Host Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Network Filesystem Group (@samp{nfsx})
@cindex Network filesystem group
@cindex Atomic NFS mounts
@cindex Mounting an atomic group of NFS filesystems
@cindex How to mount an atomic group of NFS filesystems
@cindex nfsx, filesystem type
@cindex Filesystem type; nfsx

The @dfn{nfsx} (@samp{type:=nfsx}) filesystem allows a group of filesystems to be mounted
from a single NFS server.  The implementation is layered above the
@samp{nfs} implementation so keep-alives work in the same way.

The options are the same as for the @samp{nfs} filesystem with one
difference.

@noindent
The following options must be specified:

@table @code
@item rhost
the remote fileserver.  This must be an entry in the hosts database.  IP
addresses are not accepted.  The default value is taken from the local
host name (@code{$@{host@}}) if no other value is specified.

@item rfs
as a list of filesystems to mount.  The list is in the form of a comma
separated strings.
@end table

@noindent
For example:

@example
pub  type:=nfsx;rhost:=gould;\
     rfs:=/public,/,graphics,usenet;fs:=$@{autodir@}/$@{rhost@}/root
@end example

The first string defines the root of the tree, and is applied as a
prefix to the remaining members of the list which define the individual
filesystems.  The first string is @emph{not} used as a filesystem name.
A parallel operation is used to determine the local mount points to
ensure a consistent layout of a tree of mounts.

Here, the @emph{three} filesystems, @samp{/public},
@samp{/public/graphics} and @samp{/public/usenet}, would be mounted.@refill

A local mount point, @code{$@{fs@}}, @emph{must} be specified.  The
default local mount point will not work correctly in the general case.
A suggestion is to use @samp{fs:=$@{autodir@}/$@{rhost@}/root}.@refill

@c ----------------------------------------------------------------
@node Unix Filesystem, Caching Filesystem, Network Filesystem Group, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Unix Filesystem (@samp{ufs}, @samp{xfs}, or @samp{efs})
@cindex Unix filesystem
@cindex UFS
@cindex XFS
@cindex EFS
@cindex Mounting a UFS filesystem
@cindex Mounting a local disk
@cindex How to mount a UFS filesystems
@cindex How to mount a local disk
@cindex Disk filesystems
@cindex ufs, filesystem type
@cindex Filesystem type; ufs
@cindex xfs, filesystem type
@cindex Filesystem type; xfs
@cindex efs, filesystem type
@cindex Filesystem type; efs

The @dfn{ufs} (@samp{type:=ufs}) filesystem type provides access to the system's standard
disk filesystem---usually a derivative of the Berkeley Fast Filesystem.

@noindent
The following option must be specified:

@table @code
@cindex dev, mount option
@cindex Mount option; dev
@item dev
the block special device to be mounted.
@end table

A UFS entry might be:

@example
jsp   host==charm;type:=ufs;dev:=/dev/sd0d;sublink:=jsp
@end example

UFS is the default Unix disk-based file system, which Am-utils picks up
during the autoconfiguration phase.  Some systems have more than one
type, such as IRIX, that comes with EFS (Extent File System) and XFS
(Extended File System).  In those cases, you may explicitly set the file
system type, by using entries such:

@example
ez1   type:=efs;dev:=/dev/xd0a
ez2   type:=xfs;dev:=/dev/sd3c
@end example

@c ----------------------------------------------------------------
@node Caching Filesystem, CD-ROM Filesystem, Unix Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Caching Filesystem (@samp{cachefs})
@cindex Caching Filesystem
@cindex cachefs, filesystem type
@cindex Filesystem type; cachefs

The @dfn{cachefs} (@samp{type:=cachefs}) filesystem caches files from
one location onto another, presumably providing faster access.  It is
particularly useful to cache from a larger and remote (slower) NFS
partition to a smaller and local (faster) UFS directory.

@noindent
The following options must be specified:

@table @code
@cindex cachedir, mount option
@cindex Mount option; cachedir
@item cachedir
the directory where the cache is stored.
@item rfs
the path name to the ``back file system'' to be cached from.
@item fs
the ``front file system'' mount point to the cached files, where @i{Amd}
will set a symbolic link pointing to.
@end table

A CacheFS entry for, say, the @file{/import} @i{Amd} mount point, might
be:

@example
copt  type:=cachefs;cachedir:=/cache;rfs:=/import/opt;fs:=/n/import/copt
@end example

Access to the pathname @file{/import/copt} will follow a symbolic link
to @file{/n/import/copt}.  The latter is the mount point for a caching
file system, that caches from @file{/import/opt} to @file{/cache}.

@b{Caveats}:
@enumerate
@item This file system is currently only implemented for Solaris 2.x!
@item Before being used for the first time, the cache directory @i{must} be
initialized with @samp{cfsadmin -c @var{cachedir}}.  See the manual page for
@b{cfsadmin}(1M) for more information.
@item The ``back file system'' mounted must be a complete file system, not
a subdirectory thereof; otherwise you will get an error ``Invalid Argument''.
@item If @i{Amd} aborts abnormally, the state of the cache may be
inconsistent, requiring running the command @file{fsck -F cachefs
@var{cachedir}}.  Otherwise you will get the error ``No Space Left on Device''.
@end enumerate

@c ----------------------------------------------------------------
@node CD-ROM Filesystem, Loopback Filesystem, Caching Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section CD-ROM Filesystem (@samp{cdfs})
@cindex CD-ROM Filesystem
@cindex cdfs, filesystem type
@cindex Filesystem type; cdfs

The @dfn{cdfs} (@samp{type:=cdfs}) filesystem mounts a CD-ROM with an
ISO9660 format filesystem on it.

@noindent
The following option must be specified:

@table @code
@cindex dev, mount option
@cindex Mount option; dev
@item dev
the block special device to be mounted.
@end table

A cdfs entry might be:

@example
cdfs      os==sunos4;type:=cdfs;dev:=/dev/sr0 \
          os==sunos5;type:=cdfs;dev:=/dev/dsk/c0t6d0s2
@end example

@c ----------------------------------------------------------------
@node Loopback Filesystem, Memory/RAM Filesystem, CD-ROM Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Loopback Filesystem (@samp{lofs})
@cindex Loopback Filesystem
@cindex lofs, filesystem type
@cindex Filesystem type; lofs

The @dfn{lofs} (@samp{type:=lofs}) filesystem is also called the
loopback filesystem.  It mounts a local directory on another, thus
providing mount-time binding to another location (unlike symbolic
links).

The loopback filesystem is particularly useful within the context of a
chroot-ed directory (via @b{chroot}(2)), to provide access to
directories otherwise inaccessible.

@noindent
The following option must be specified:

@table @code
@cindex rfs, mount option
@cindex Mount option; rfs
@item rfs
the pathname to be mounted on top of @code{$@{fs@}}.
@end table

Usually, the FTP server runs in a chroot-ed environment, for security
reasons.  In this example, lofs is used to provide a subdirectory within
a user's home directory, also available for public ftp.

@example
lofs      type:=lofs;rfs:=/home/ezk/myftpdir;fs:=/usr/ftp/pub/ezk
@end example

@c ----------------------------------------------------------------
@node Memory/RAM Filesystem, Null Filesystem, Loopback Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Memory/RAM Filesystem (@samp{mfs})
@cindex Memory/RAM Filesystem
@cindex mfs, filesystem type
@cindex Filesystem type; mfs

The @dfn{mfs} (@samp{type:=mfs}) filesystem is available in 4.4BSD,
Linux, and other systems.  It creates a filesystem in a portion of the
system's memory, thus providing very fast file (volatile) access.

XXX: THIS FILESYSTEM IS NOT IMPLEMENTED YET!

@c ----------------------------------------------------------------
@node Null Filesystem, Floppy Filesystem, Memory/RAM Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Null Filesystem (@samp{nullfs})
@cindex Null Filesystem
@cindex nullfs, filesystem type
@cindex Filesystem type; nullfs

The @dfn{nullfs} (@samp{type:=nullfs}) filesystem is available from 4.4BSD,
and is very similar to the loopback filesystem, @dfn{lofs}.

XXX: THIS FILESYSTEM IS NOT IMPLEMENTED YET!

@c ----------------------------------------------------------------
@node Floppy Filesystem, Translucent Filesystem, Null Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Floppy Filesystem (@samp{pcfs})
@cindex Floppy Filesystem
@cindex pcfs, filesystem type
@cindex Filesystem type; pcfs

The @dfn{pcfs} (@samp{type:=pcfs}) filesystem mounts a floppy previously
formatted for the MS-DOS format.

@noindent
The following option must be specified:

@table @code
@cindex dev, mount option
@cindex Mount option; dev
@item dev
the block special device to be mounted.
@end table

A pcfs entry might be:

@example
pcfs      os==sunos4;type:=pcfs;dev:=/dev/fd0 \
          os==sunos5;type:=pcfs;dev:=/dev/diskette
@end example

@c ----------------------------------------------------------------
@node Translucent Filesystem, Shared Memory+Swap Filesystem, Floppy Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Translucent Filesystem (@samp{tfs})
@cindex Translucent Filesystem
@cindex tfs, filesystem type
@cindex Filesystem type; tfs

The @dfn{tfs} (@samp{type:=tfs}) filesystem is an older version of the
4.4BSD @dfn{unionfs}.

XXX: THIS FILESYSTEM IS NOT IMPLEMENTED YET!

@c ----------------------------------------------------------------
@node Shared Memory+Swap Filesystem, User ID Mapping Filesystem, Translucent Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Shared Memory+Swap Filesystem (@samp{tmpfs})
@cindex Shared Memory and Swap Filesystem
@cindex tmpfs, filesystem type
@cindex Filesystem type; tmpfs

The @dfn{tmpfs} (@samp{type:=tmpfs}) filesystem shares memory between a
the swap device and the rest of the system.  It is generally used to
provide a fast access @file{/tmp} directory, one that uses memory that
is otherwise unused.  This filesystem is available in SunOS 4.x and 5.x.

XXX: THIS FILESYSTEM IS NOT IMPLEMENTED YET!

@c ----------------------------------------------------------------
@node User ID Mapping Filesystem, Program Filesystem, Shared Memory+Swap Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section User ID Mapping Filesystem (@samp{umapfs})
@cindex User ID Mapping Filesystem
@cindex umapfs, filesystem type
@cindex Filesystem type; umapfs

The @dfn{umapfs} (@samp{type:=umapfs}) filesystem maps User IDs of file
ownership, and is available from 4.4BSD.

XXX: THIS FILESYSTEM IS NOT IMPLEMENTED YET!

@c ----------------------------------------------------------------
@node Program Filesystem, Symbolic Link Filesystem, User ID Mapping Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Program Filesystem (@samp{program})
@cindex Program filesystem
@cindex Mount a filesystem under program control
@cindex program, filesystem type
@cindex Filesystem type; program

The @dfn{program} (@samp{type:=program}) filesystem type allows a program to be run whenever a
mount or unmount is required.  This allows easy addition of support for
other filesystem types, such as MIT's Remote Virtual Disk (RVD)
which has a programmatic interface via the commands
@samp{rvdmount} and @samp{rvdunmount}.

@noindent
The following options must be specified:

@table @code
@cindex mount, mount option
@cindex Mount option; mount
@item mount
the program which will perform the mount.

@cindex unmount, mount option
@cindex Mount option; unmount
@item unmount
the program which will perform the unmount.
@end table

The exit code from these two programs is interpreted as a Unix error
code.  As usual, exit code zero indicates success.  To execute the
program @i{Amd} splits the string on whitespace to create an array of
substrings.  Single quotes @samp{'} can be used to quote whitespace
if that is required in an argument.  There is no way to escape or change
the quote character.

To run the program @samp{rvdmount} with a host name and filesystem as
arguments would be specified by @samp{mount:="/etc/rvdmount rvdmount
fserver $@{path@}"}.

The first element in the array is taken as the pathname of the program
to execute.  The other members of the array form the argument vector to
be passed to the program, @dfn{including argument zero}.  This means
that the split string must have at least two elements.  The program is
directly executed by @i{Amd}, not via a shell.  This means that scripts
must begin with a @code{#!} interpreter specification.

If a filesystem type is to be heavily used, it may be worthwhile adding
a new filesystem type into @i{Amd}, but for most uses the program
filesystem should suffice.

When the program is run, standard input and standard error are inherited
from the current values used by @i{Amd}.  Standard output is a
duplicate of standard error.  The value specified with the @code{-l}
command line option has no effect on standard error.

@c ----------------------------------------------------------------
@node Symbolic Link Filesystem, Symbolic Link Filesystem II, Program Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Symbolic Link Filesystem (@samp{link})
@cindex Symbolic link filesystem
@cindex Referencing part of the local name space
@cindex Mounting part of the local name space
@cindex How to reference part of the local name space
@cindex link, filesystem type
@cindex symlink, link filesystem type
@cindex Filesystem type; link

Each filesystem type creates a symbolic link to point from the volume
name to the physical mount point.  The @samp{link} filesystem does the
same without any other side effects.  This allows any part of the
machines name space to be accessed via @i{Amd}.

One common use for the symlink filesystem is @file{/homes} which can be
made to contain an entry for each user which points to their
(auto-mounted) home directory.  Although this may seem rather expensive,
it provides a great deal of administrative flexibility.

@noindent
The following option must be defined:

@table @code
@item fs
The value of @var{fs} option specifies the destination of the link, as
modified by the @var{sublink} option.  If @var{sublink} is non-null, it
is appended to @code{$@{fs@}}@code{/} and the resulting string is used
as the target.
@end table

The @samp{link} filesystem can be thought of as identical to the
@samp{ufs} filesystem but without actually mounting anything.

An example entry might be:

@example
jsp   host==charm;type:=link;fs:=/home/charm;sublink:=jsp
@end example
which would return a symbolic link pointing to @file{/home/charm/jsp}.

@c ----------------------------------------------------------------
@node Symbolic Link Filesystem II, NFS-Link Filesystem, Symbolic Link Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Symbolic Link Filesystem II (@samp{linkx})
@cindex Symbolic link filesystem II
@cindex Referencing an existing part of the local name space
@cindex Mounting an existing part of the local name space
@cindex How to reference an existing part of the local name space
@cindex linkx, filesystem type
@cindex symlink, linkx filesystem type
@cindex Filesystem type; linkx

The @dfn{linkx} (@samp{type:=linkx}) filesystem type is identical to @samp{link} with the
exception that the target of the link must exist.  Existence is checked
with the @b{lstat}(2) system call.

The @samp{linkx} filesystem type is particularly useful for wildcard map
entries.  In this case, a list of possible targets can be given and
@i{Amd} will choose the first one which exists on the local machine.

@c ----------------------------------------------------------------
@node NFS-Link Filesystem, Automount Filesystem, Symbolic Link Filesystem II, Filesystem Types
@comment  node-name,  next,  previous,  up
@section NFS-Link Filesystem (@samp{nfsl})
@cindex NFS-Link filesystem II
@cindex Referencing an existing part of the name space if target exists
@cindex Mounting a remote part of the name space if target is missing
@cindex Symlink if target exists, NFS otherwise
@cindex nfsl, filesystem type
@cindex symlink, nfsl filesystem type
@cindex Filesystem type; nfsl

The @dfn{nfsl} (@samp{type:=nfsl}) filesystem type is a combination of two others:
@samp{link} and @samp{nfs}.  If the local host name is equal to the
value of @code{$@{rhost@}}, or if the target pathname listed in
@code{$@{fs@}} exists, @samp{nfsl} will behave exactly as
@samp{type:=link}, and refer to the target as a symbolic link.  If the
local host name is not equal to the value of @code{$@{rhost@}}, or if
the target of the link does not exist, @i{Amd} will treat it as
@samp{type:=nfs}, and will mount a remote pathname for it.

The @samp{nfsl} filesystem type is particularly useful as a shorthand
for the more cumbersome and yet one of the most popular @i{Amd}
entries.  For example, you can simplify all map entries that look like:

@example
zing    -fs:=/n/shekel/u/zing \
        host!=shekel;type:=nfs;rhost:=shekel;rfs:=$@{fs@} \
        host==shekel;type:=link
@end example

or

@example
zing    -fs:=/n/shekel/u/zing \
        exists($@{fs@});type:=link \
        !exists($@{fs@});type:=nfs;rhost:=shekel;rfs:=$@{fs@}
@end example

into a shorter form

@example
zing    type:=nfsl;fs:=/n/shekel/u/zing;rhost:=shekel;rfs:=$@{fs@}
@end example

Not just does it make the maps smaller and simpler, but it avoids
possible mistakes that often happen when forgetting to set up the two
entries (one for @samp{type:=nfs} and the other for @samp{type:=link})
necessary to perform transparent mounts of existing or remote mounts.

@c ----------------------------------------------------------------
@node Automount Filesystem, Direct Automount Filesystem, NFS-Link Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Automount Filesystem (@samp{auto})
@cindex Automount filesystem
@cindex Map cache types
@cindex Setting map cache parameters
@cindex How to set map cache parameters
@cindex How to start an indirect automount point
@cindex auto, filesystem type
@cindex Filesystem type; auto
@cindex SIGHUP signal
@cindex Map cache synchronizing
@cindex Synchronizing the map cache
@cindex Map cache options
@cindex Regular expressions in maps

The @dfn{auto} (@samp{type:=auto}) filesystem type creates a new automount point below an
existing automount point.  Top-level automount points appear as system
mount points.  An automount mount point can also appear as a
sub-directory of an existing automount point.  This allows some
additional structure to be added, for example to mimic the mount tree of
another machine.

The following options may be specified:

@table @code
@cindex cache, mount option
@cindex Mount option; cache
@item cache
specifies whether the data in this mount-map should be
cached.  The default value is @samp{none}, in which case
no caching is done in order to conserve memory.
However, better performance and reliability can be obtained by caching
some or all of a mount-map.

If the cache option specifies @samp{all},
the entire map is enumerated when the mount point is created.

If the cache option specifies @samp{inc}, caching is done incrementally
as and when data is required.
Some map types do not support cache mode @samp{all}, in which case @samp{inc}
is used whenever @samp{all} is requested.

Caching can be entirely disabled by using cache mode @samp{none}.

If the cache option specifies @samp{regexp} then the entire map will be
enumerated and each key will be treated as an egrep-style regular
expression.  The order in which a cached map is searched does not
correspond to the ordering in the source map so the regular expressions
should be mutually exclusive to avoid confusion.

Each mount map type has a default cache type, usually @samp{inc}, which
can be selected by specifying @samp{mapdefault}.

The cache mode for a mount map can only be selected on the command line.
Starting @i{Amd} with the command:

@example
amd /homes hesiod.homes -cache:=inc
@end example

will cause @samp{/homes} to be automounted using the @dfn{Hesiod} name
server with local incremental caching of all successfully resolved names.

All cached data is forgotten whenever @i{Amd} receives a @samp{SIGHUP}
signal and, if cache @samp{all} mode was selected, the cache will be
reloaded.  This can be used to inform @i{Amd} that a map has been
updated.  In addition, whenever a cache lookup fails and @i{Amd} needs
to examine a map, the map's modify time is examined.  If the cache is
out of date with respect to the map then it is flushed as if a
@samp{SIGHUP} had been received.

An additional option (@samp{sync}) may be specified to force @i{Amd} to
check the map's modify time whenever a cached entry is being used.  For
example, an incremental, synchronized cache would be created by the
following command:

@example
amd /homes hesiod.homes -cache:=inc,sync
@end example

@item fs
specifies the name of the mount map to use for the new mount point.

Arguably this should have been specified with the @code{$@{rfs@}} option but
we are now stuck with it due to historical accident.

@c %If the string @samp{.} is used then the same map is used;
@c %in addition the lookup prefix is set to the name of the mount point followed
@c %by a slash @samp{/}.
@c %This is the same as specifying @samp{fs:=\$@{map@};pref:=\$@{key@}/}.
@c

@item pref
alters the name that is looked up in the mount map.  If
@code{$@{pref@}}, the @dfn{prefix}, is non-null then it is prepended to
the name requested by the kernel @dfn{before} the map is searched.
@end table

The server @samp{dylan.doc.ic.ac.uk} has two user disks:
@samp{/dev/dsk/2s0} and @samp{/dev/dsk/5s0}.  These are accessed as
@samp{/home/dylan/dk2} and @samp{/home/dylan/dk5} respectively.  Since
@samp{/home} is already an automount point, this naming is achieved with
the following map entries:@refill

@example
dylan        type:=auto;fs:=$@{map@};pref:=$@{key@}/
dylan/dk2    type:=ufs;dev:=/dev/dsk/2s0
dylan/dk5    type:=ufs;dev:=/dev/dsk/5s0
@end example

@c ----------------------------------------------------------------
@node Direct Automount Filesystem, Union Filesystem, Automount Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Direct Automount Filesystem (@samp{direct})
@cindex Direct automount filesystem
@cindex How to start a direct automount point
@cindex direct, filesystem type
@cindex Filesystem type; direct

The @dfn{direct} (@samp{type:=direct}) filesystem is almost identical to the automount
filesystem.  Instead of appearing to be a directory of mount points, it
appears as a symbolic link to a mounted filesystem.  The mount is done
at the time the link is accessed.  @xref{Automount Filesystem} for a
list of required options.

Direct automount points are created by specifying the @samp{direct}
filesystem type on the command line:

@example
amd ... /usr/man auto.direct -type:=direct
@end example

where @samp{auto.direct} would contain an entry such as:

@example
usr/man    -type:=nfs;rfs:=/usr/man \
           rhost:=man-server1  rhost:=man-server2
@end example

In this example, @samp{man-server1} and @samp{man-server2} are file
servers which export copies of the manual pages.  Note that the key
which is looked up is the name of the automount point without the
leading @samp{/}.

@c ----------------------------------------------------------------
@node Union Filesystem, Error Filesystem, Direct Automount Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Union Filesystem (@samp{union})
@cindex Union filesystem
@cindex union, filesystem type
@cindex Filesystem type; union

The @dfn{union} (@samp{type:=union}) filesystem type allows the contents of several
directories to be merged and made visible in a single directory.  This
can be used to overcome one of the major limitations of the Unix mount
mechanism which only allows complete directories to be mounted.

For example, supposing @file{/tmp} and @file{/var/tmp} were to be merged
into a new directory called @file{/mtmp}, with files in @file{/var/tmp}
taking precedence.  The following command could be used to achieve this
effect:

@example
amd ... /mtmp union:/tmp:/var/tmp -type:=union
@end example

Currently, the unioned directories must @emph{not} be automounted.  That
would cause a deadlock.  This seriously limits the current usefulness of
this filesystem type and the problem will be addressed in a future
release of @i{Amd}.

Files created in the union directory are actually created in the last
named directory.  This is done by creating a wildcard entry which points
to the correct directory.  The wildcard entry is visible if the union
directory is listed, so allowing you to see which directory has
priority.

The files visible in the union directory are computed at the time
@i{Amd} is started, and are not kept up-to-date with respect to the
underlying directories.  Similarly, if a link is removed, for example
with the @samp{rm} command, it will be lost forever.

@c ----------------------------------------------------------------
@node Error Filesystem, Top-level Filesystem, Union Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Error Filesystem (@samp{error})
@cindex Error filesystem
@cindex error, filesystem type
@cindex Filesystem type; error

The @dfn{error} (@samp{type:=error}) filesystem type is used internally as a catch-all in the
case where none of the other filesystems was selected, or some other
error occurred.  Lookups and mounts always fail with ``No such file or
directory''.  All other operations trivially succeed.

The error filesystem is not directly accessible.

@c ----------------------------------------------------------------
@node Top-level Filesystem, Autofs Filesystem, Error Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Top-level Filesystem (@samp{toplvl})
@cindex Top level filesystem
@cindex toplvl, filesystem type
@cindex Filesystem type; toplvl

The @dfn{toplvl} (@samp{type:=toplvl}) filesystems is derived from the @samp{auto} filesystem
and is used to mount the top-level automount nodes.  Requests of this
type are automatically generated from the command line arguments and can
also be passed in by using the @code{-M} option of the @dfn{Amq} command.
That option is insecure, and is unavailable unless am-utils was
configured with @samp{--with-amq-mount}.

@c ----------------------------------------------------------------
@node Root Filesystem, Inheritance Filesystem, Autofs Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Root Filesystem (@samp{root})
@cindex Root filesystem
@cindex root, filesystem type
@cindex Filesystem type; root

The @dfn{root} (@samp{type:=root}) filesystem type acts as an internal
placeholder onto which @i{Amd} can pin @samp{toplvl} mounts.  Only one
node of this type need ever exist and one is created automatically
during startup.  The effect of having more than one root node is
undefined.

The root filesystem is not directly accessible.

@c ----------------------------------------------------------------
@node Autofs Filesystem, Root Filesystem, Top-level Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Autofs Filesystem (@samp{autofs})
@cindex Autofs filesystem
@cindex autofs, filesystem type
@cindex Filesystem type; autofs

The @dfn{autofs} (@samp{type:=autofs}) filesystem uses Sun's kernel-based automounter
supporting filesystem for @i{Amd}'s mount points.  Hence it is another
type of top level filesystem.

The autofs filesystem is not directly accessible from @i{Amd} maps, but
only from the @file{amd.conf} file (@pxref{mount_type Parameter}).

Note that Autofs support is still very early.  See the distribution file
@file{README.autofs} for detail of what works and what does not.

@c ----------------------------------------------------------------
@node Inheritance Filesystem, , Root Filesystem, Filesystem Types
@comment  node-name,  next,  previous,  up
@section Inheritance Filesystem (@samp{inherit})
@cindex Inheritance filesystem
@cindex Nodes generated on a restart
@cindex inherit, filesystem type
@cindex Filesystem type; inherit

The @dfn{inheritance} (@samp{type:=inherit}) filesystem is not directly
accessible.  Instead, internal mount nodes of this type are
automatically generated when @i{Amd} is started with the @code{-r} option.
At this time the system mount table is scanned to locate any filesystems
which are already mounted.  If any reference to these filesystems is
made through @i{Amd} then instead of attempting to mount it, @i{Amd}
simulates the mount and @dfn{inherits} the filesystem.  This allows a
new version of @i{Amd} to be installed on a live system simply by
killing the old daemon with @samp{SIGTERM} and starting the new one.@refill

This filesystem type is not generally visible externally, but it is
possible that the output from @samp{amq -m} may list @samp{inherit} as
the filesystem type.  This happens when an inherit operation cannot
be completed for some reason, usually because a fileserver is down.

@c ################################################################
@node Amd Configuration File, Run-time Administration, Filesystem Types, Top
@comment  node-name,  next,  previous,  up
@chapter Amd Configuration File
@cindex  Amd Configuration File
@cindex amd.conf

The @samp{amd.conf} file is the configuration file for @i{Amd}, as part
of the am-utils suite.  This file contains runtime configuration
information for the @i{Amd} automounter program.

@menu
* File Format::
* The Global Section::
* Regular Map Sections::
* Common Parameters::
* Global Parameters::
* Regular Map Parameters::
* amd.conf Examples::
@end menu

@c ================================================================
@node File Format, The Global Section, Amd Configuration File, Amd Configuration File
@comment  node-name,  next,  previous,  up
@section File Format
@cindex amd.conf file format

The @samp{amd.conf} file consists of sections and parameters.  A section
begins with the name of the section in square brackets @samp{[]} and
continues until the next section begins or the end of the file is reached.
Sections contain parameters of the form @samp{name = value}.

The file is line-based --- that is, each newline-terminated line
represents either a comment, a section name or a parameter.  No
line-continuation syntax is available.

Section names, parameter names and their values are case sensitive.

Only the first equals sign in a parameter is significant.  Whitespace
before or after the first equals sign is discarded.  Leading, trailing
and internal whitespace in section and parameter names is irrelevant.
Leading and trailing whitespace in a parameter value is discarded.
Internal whitespace within a parameter value is not allowed, unless the
whole parameter value is quoted with double quotes as in @samp{name =
"some value"}.

Any line beginning with a pound sign @samp{#} is ignored, as are lines
containing only whitespace.

The values following the equals sign in parameters are all either a
string (no quotes needed if string does not include spaces) or a
boolean, which may be given as @samp{yes}/@samp{no}.  Case is significant in all
values.  Some items such as cache timeouts are numeric.

@c ================================================================
@node The Global Section, Regular Map Sections, File Format, Amd Configuration File
@comment  node-name,  next,  previous,  up
@section The Global Section
@cindex amd.conf global section

The global section must be specified as @samp{[global]}.  Parameters in
this section either apply to @i{Amd} as a whole, or to all other regular map
sections which follow.  There should be only one global section defined
in one configuration file.

It is highly recommended that this section be specified first in the
configuration file.  If it is not, then regular map sections which
precede it will not use global values defined later.

@c ================================================================
@node Regular Map Sections, Common Parameters, The Global Section, Amd Configuration File
@comment  node-name,  next,  previous,  up
@section Regular Map Sections
@cindex amd.conf regular map sections

Parameters in regular (non-global) sections apply to a single map entry.
For example, if the map section @samp{[/homes]} is defined, then all
parameters following it will be applied to the @file{/homes}
@i{Amd}-managed mount point.

@c ================================================================
@node Common Parameters, Global Parameters, Regular Map Sections, Amd Configuration File
@comment  node-name,  next,  previous,  up
@section Common Parameters
@cindex amd.conf common parameters

These parameters can be specified either in the global or a map-specific
section.  Entries specified in a map-specific section override the default
value or one defined in the global section.   If such a common parameter is
specified only in the global section, it is applicable to all regular map
sections that follow.

@menu
* browsable_dirs Parameter::
* map_options Parameter::
* map_type Parameter::
* mount_type Parameter::
* search_path Parameter::
@end menu

@c ----------------------------------------------------------------
@node browsable_dirs Parameter, map_options Parameter, Common Parameters, Common Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{browsable_dirs} Parameter
@cindex browsable_dirs Parameter

(type=string, default=@samp{no}).  If @samp{yes}, then @i{Amd}'s top-level
mount points will be browsable to @b{readdir}(3) calls.  This means you
could run for example @b{ls}(1) and see what keys are available to mount
in that directory.  Not all entries are made visible to @b{readdir}(3):
the @samp{/defaults} entry, wildcard entries, and those with a @file{/}
in them are not included.  If you specify @samp{full} to this option,
all but the @samp{/defaults} entry will be visible.  Note that if you run
a command which will attempt to @b{stat}(2) the entries, such as often
done by @samp{ls -l} or @samp{ls -F}, @i{Amd} will attempt to mount
@i{every} entry in that map.  This is often called a ``mount storm''.

@c ----------------------------------------------------------------
@node map_options Parameter, map_type Parameter, browsable_dirs Parameter, Common Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{map_options} Parameter
@cindex map_options Parameter

(type=string, default no options).  This option is the same as
specifying map options on the command line to @i{Amd}, such as
@samp{cache:=all}.

@c ----------------------------------------------------------------
@node map_type Parameter, mount_type Parameter, map_options Parameter, Common Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{map_type} Parameter
@cindex map_type Parameter

(type=string, default search all map types).  If specified, @i{Amd} will
initialize the map only for the type given.  This is useful to avoid the
default map search type used by @i{Amd} which takes longer and can have
undesired side-effects such as initializing NIS even if not used.
Possible values are

@table @samp
@item file
plain files
@item hesiod
Hesiod name service from MIT
@item ldap
Lightweight Directory Access Protocol
@item ndbm
(New) dbm style hash files
@item nis
Network Information Services (version 2)
@item nisplus
Network Information Services Plus (version 3)
@item passwd
local password files
@item union
union maps
@end table

@c ----------------------------------------------------------------
@node mount_type Parameter, search_path Parameter, map_type Parameter, Common Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{mount_type} Parameter
@cindex mount_type Parameter

(type=string, default=@samp{nfs}).  All @i{Amd} mount types default to NFS.
That is, @i{Amd} is an NFS server on the map mount points, for the local
host it is running on.  If @samp{autofs} is specified, @i{Amd} will be
an autofs server for those mount points.

@c ----------------------------------------------------------------
@node search_path Parameter, , mount_type Parameter, Common Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{search_path} Parameter
@cindex search_path Parameter

(type=string, default no search path).  This provides a
(colon-delimited) search path for file maps.  Using a search path,
sites can allow for local map customizations and overrides, and can
distributed maps in several locations as needed.

@c ================================================================
@node Global Parameters, Regular Map Parameters, Common Parameters, Amd Configuration File
@comment  node-name,  next,  previous,  up
@section Global Parameters
@cindex amd.conf global parameters

The following parameters are applicable to the @samp{[global]} section only.

@menu
* arch Parameter::
* auto_dir Parameter::
* cache_duration Parameter::
* cluster Parameter::
* debug_options Parameter::
* dismount_interval Parameter::
* fully_qualified_hosts Parameter::
* hesiod_base Parameter::
* karch Parameter::
* ldap_base Parameter::
* ldap_cache_maxmem Parameter::
* ldap_cache_seconds Parameter::
* ldap_hostports Parameter::
* local_domain Parameter::
* log_file Parameter::
* log_options Parameter::
* nfs_retransmit_counter Parameter::
* nfs_retry_interval Parameter::
* nis_domain Parameter::
* normalize_hostnames Parameter::
* os Parameter::
* osver Parameter::
* pid_file Parameter::
* plock Parameter::
* portmap_program Parameter::
* print_pid Parameter::
* print_version Parameter::
* restart_mounts Parameter::
* selectors_on_default Parameter::
* show_statfs_entries Parameter::
* unmount_on_exit Parameter::
@end menu

@c ----------------------------------------------------------------
@node arch Parameter, auto_dir Parameter, Global Parameters, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{arch} Parameter
@cindex arch Parameter

(type=string, default to compiled in value).  Allows you to override the
value of the @i{arch} @i{Amd} variable.

@c ----------------------------------------------------------------
@node auto_dir Parameter, cache_duration Parameter, arch Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{auto_dir} Parameter
@cindex auto_dir Parameter

(type=string, default=@samp{/a}).  Same as the @code{-a} option to @i{Amd}.
This sets the private directory where @i{Amd} will create
sub-directories for its real mount points.

@c ----------------------------------------------------------------
@node cache_duration Parameter, cluster Parameter, auto_dir Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{cache_duration} Parameter
@cindex cache_duration Parameter

(type=numeric, default=300).  Same as the @code{-c} option to
@i{Amd}.  Sets the duration in seconds that looked up map entries remain
in the cache.

@c ----------------------------------------------------------------
@node cluster Parameter, debug_options Parameter, cache_duration Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{cluster} Parameter
@cindex cluster Parameter

(type=string, default no cluster).  Same as the @code{-C} option to
@i{Amd}.  Specifies the alternate HP-UX cluster to use.

@c ----------------------------------------------------------------
@node debug_options Parameter, dismount_interval Parameter, cluster Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{debug_options} Parameter
@cindex debug_options Parameter

(type=string, default no debug options).  Same as the @code{-D}
option to @i{Amd}.  Specify any debugging options for @i{Amd}.  Works
only if am-utils was configured for debugging using the
@code{--enable-debug} option.  The @samp{mem} option alone can be turned
on via @code{--enable-debug=mem}.  Otherwise debugging options are
ignored.  Options are comma delimited, and can be preceded by the string
@samp{no} to negate their meaning.  You can get the list of supported
debugging options by running @code{amd -v}.  Possible values are:

@table @samp
@item all
all options
@item amq
register for amq
@item daemon
enter daemon mode
@item fork
fork server
@item full
program trace
@item mem
trace memory allocations
@item mtab
use local @file{./mtab} file
@item str
debug string munging
@item test
full debug but no daemon
@item trace
protocol trace
@end table

@c ----------------------------------------------------------------
@node dismount_interval Parameter, fully_qualified_hosts Parameter, debug_options Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{dismount_interval} Parameter
@cindex dismount_interval Parameter

(type=numeric, default=120).  Same as the @code{-w} option to
@i{Amd}.  Specify in seconds, the time between attempts to dismount file
systems that have exceeded their cached times.

@c ----------------------------------------------------------------
@node fully_qualified_hosts Parameter, hesiod_base Parameter, dismount_interval Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{fully_qualified_hosts} Parameter
@cindex fully_qualified_hosts Parameter

(type=string, default=@samp{no}).  If @samp{yes}, @i{Amd} will perform RPC
authentication using fully-qualified host names.  This is necessary for
some systems, and especially when performing cross-domain mounting.  For
this function to work, the @i{Amd} variable @samp{$@{hostd@}} is used,
requiring that @samp{$@{domain@}} not be null.

@c ----------------------------------------------------------------
@node hesiod_base Parameter, karch Parameter, fully_qualified_hosts Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{hesiod_base} Parameter
@cindex hesiod_base Parameter

(type=string, default=@samp{automount}).  Specify the base name for
hesiod maps.

@c ----------------------------------------------------------------
@node karch Parameter, ldap_base Parameter, hesiod_base Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{karch} Parameter
@cindex karch Parameter

(type=string, default to karch of the system).  Same as the @code{-k}
option to @i{Amd}.  Allows you to override the kernel-architecture of
your system.  Useful for example on Sun (Sparc) machines, where you can
build one @i{Amd} binary, and run it on multiple machines, yet you want
each one to get the correct @i{karch} variable set (for example, sun4c,
sun4m, sun4u, etc.)  Note that if not specified, @i{Amd} will use
@b{uname}(2) to figure out the kernel architecture of the machine.

@c ----------------------------------------------------------------
@node ldap_base Parameter, ldap_cache_maxmem Parameter, karch Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{ldap_base} Parameter
@cindex ldap_base Parameter

(type=string, default not set).  Specify the base name for
LDAP.

@c ----------------------------------------------------------------
@node ldap_cache_maxmem Parameter, ldap_cache_seconds Parameter, ldap_base Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{ldap_cache_maxmem} Parameter
@cindex ldap_cache_maxmem Parameter

(type=numeric, default=131072).  Specify the maximum memory @i{Amd}
should use to cache LDAP entries.

@c ----------------------------------------------------------------
@node ldap_cache_seconds Parameter, ldap_hostports Parameter, ldap_cache_maxmem Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{ldap_cache_seconds} Parameter
@cindex ldap_cache_seconds Parameter

(type=numeric, default=0).  Specify the number of seconds to keep
entries in the cache.

@c ----------------------------------------------------------------
@node ldap_hostports Parameter, local_domain Parameter, ldap_cache_seconds Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{ldap_hostports} Parameter
@cindex ldap_hostports Parameter

(type=string, default not set).  Specify
LDAP-specific values such as country and organization.

@c ----------------------------------------------------------------
@node local_domain Parameter, log_file Parameter, ldap_hostports Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{local_domain} Parameter
@cindex local_domain Parameter

(type=string, default no sub-domain).  Same as the @code{-d} option
to @i{Amd}.  Specify the local domain name.  If this option is not given
the domain name is determined from the hostname, by removing the first
component of the fully-qualified host name.

@c ----------------------------------------------------------------
@node log_file Parameter, log_options Parameter, local_domain Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{log_file} Parameter
@cindex log_file Parameter

(type=string, default=@samp{stderr}).  Same as the @code{-l} option to
@i{Amd}.  Specify a file name to log @i{Amd} events to.
If the string @samp{/dev/stderr} is specified,
@i{Amd} will send its events to the standard error file descriptor.

If the string @samp{syslog} is given, @i{Amd} will record its events
with the system logger @b{syslogd}(8).  If your system supports syslog
facilities, then the default facility used is @samp{LOG_DAEMON}.

When using syslog, if you wish to change the facility, append its name
to the option name, delimited by a single colon.  For example, if it is
the string @samp{syslog:local7} then @i{Amd} will log messages via
@b{syslog}(3) using the @samp{LOG_LOCAL7} facility.  If the facility
name specified is not recognized, @i{Amd} will default to @samp{LOG_DAEMON}.
Note: while you can use any syslog facility available on your system, it
is generally a bad idea to use those reserved for other services such as
@samp{kern}, @samp{lpr}, @samp{cron}, etc.

@c ----------------------------------------------------------------
@node log_options Parameter, nfs_retransmit_counter Parameter, log_file Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{log_options} Parameter
@cindex log_options Parameter

(type=string, default no logging options).  Same as the @code{-x}
option to @i{Amd}.  Specify any logging options for @i{Amd}.  Options
are comma delimited, and can be preceded by the string @samp{no} to
negate their meaning.  The @samp{debug} logging option is only available
if am-utils was configured with @code{--enable-debug}.  You can get the
list of supported debugging options by running @code{amd -v}.  Possible
values are:

@table @samp
@item all
all messages
@item debug
debug messages
@item error
non-fatal system errors
@item fatal
fatal errors
@item info
information
@item map
map errors
@item stats
additional statistical information
@item user
non-fatal user errors
@item warn
warnings
@item warning
warnings
@end table

@c ----------------------------------------------------------------
@node nfs_retransmit_counter Parameter, nfs_retry_interval Parameter, log_options Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{nfs_retransmit_counter} Parameter
@cindex nfs_retransmit_counter Parameter

(type=numeric, default=110).  Same as the @i{counter} part of the
@code{-t} @i{interval.counter} option to @i{Amd}.  Specifies the
retransmit counter's value in @emph{tenths} of seconds.

@c ----------------------------------------------------------------
@node nfs_retry_interval Parameter, nis_domain Parameter, nfs_retransmit_counter Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{nfs_retry_interval} Parameter
@cindex nfs_retry_interval Parameter

(type=numeric, default=8).  Same as the @i{interval} part of the
@code{-t} @i{interval.counter} option to @i{Amd}.  Specifies the
interval in @emph{tenths} of seconds, between NFS/RPC/UDP retries.

@c ----------------------------------------------------------------
@node nis_domain Parameter, normalize_hostnames Parameter, nfs_retry_interval Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{nis_domain} Parameter
@cindex nis_domain Parameter

(type=string, default to local NIS domain name).  Same as the
@code{-y} option to @i{Amd}.  Specify an alternative NIS domain from
which to fetch the NIS maps.  The default is the system domain name.
This option is ignored if NIS support is not available.

@c ----------------------------------------------------------------
@node normalize_hostnames Parameter, os Parameter, nis_domain Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{normalize_hostnames} Parameter
@cindex normalize_hostnames Parameter

(type=boolean, default=@samp{no}).  Same as the @code{-n} option to @i{Amd}.
If @samp{yes}, then the name referred to by @code{$@{rhost@}} is normalized
relative to the host database before being used.  The effect is to
translate aliases into ``official'' names.

@c ----------------------------------------------------------------
@node os Parameter, osver Parameter, normalize_hostnames Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{os} Parameter
@cindex os Parameter

(type=string, default to compiled in value).  Same as the @code{-O}
option to @i{Amd}.  Allows you to override the compiled-in name of the
operating system.  Useful when the built-in name is not desired for
backward compatibility reasons.  For example, if the built-in name is
@samp{sunos5}, you can override it to @samp{sos5}, and use older maps
which were written with the latter in mind.

@c ----------------------------------------------------------------
@node osver Parameter, pid_file Parameter, os Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{osver} Parameter
@cindex osver Parameter

(type=string, default to compiled in value).  Same as the @code{-o}
option to @i{Amd}.  Allows you to override the compiled-in version
number of the operating system.  Useful when the built-in version is not
desired for backward compatibility reasons.  For example, if the build
in version is @samp{2.5.1}, you can override it to @samp{5.5.1}, and use
older maps that were written with the latter in mind.

@c ----------------------------------------------------------------
@node pid_file Parameter, plock Parameter, osver Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{pid_file} Parameter
@cindex pid_file Parameter

(type=string, default=@samp{/dev/stdout}).  Specify a file to store the process
ID of the running daemon into.  If not specified, @i{Amd} will print its
process id onto the standard output.  Useful for killing @i{Amd} after
it had run.  Note that the PID of a running @i{Amd} can also be
retrieved via @i{Amq} (@pxref{Amq -p option}).

This file is used only if the @samp{print_pid} option is on
(@pxref{print_pid Parameter}).

@c ----------------------------------------------------------------
@node plock Parameter, portmap_program Parameter, pid_file Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{plock} Parameter
@cindex plock Parameter

(type=boolean, default=@samp{yes}).  Same as the @code{-S} option to @i{Amd}.
If @samp{yes}, lock the running executable pages of @i{Amd} into memory.
To improve @i{Amd}'s performance, systems that support the @b{plock}(3)
call can lock the @i{Amd} process into memory.  This way there is less
chance the operating system will schedule, page out, and swap the
@i{Amd} process as needed.  This improves @i{Amd}'s performance, at the
cost of reserving the memory used by the @i{Amd} process (making it
unavailable for other processes).

@c ----------------------------------------------------------------
@node portmap_program Parameter, print_pid Parameter, plock Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{portmap_program} Parameter
@cindex portmap_program Parameter

(type=numeric, default=300019).  Specify an alternate Port-mapper RPC
program number, other than the official number.  This is useful when
running multiple @i{Amd} processes.  For example, you can run another
@i{Amd} in ``test'' mode, without affecting the primary @i{Amd} process
in any way.  For safety reasons, the alternate program numbers that can
be specified must be in the range 300019-300029, inclusive.  @i{Amq} has
an option @code{-P} which can be used to specify an alternate program
number of an @i{Amd} to contact.  In this way, amq can fully control any
number of @i{Amd} processes running on the same host.

@c ----------------------------------------------------------------
@node print_pid Parameter, print_version Parameter, portmap_program Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{print_pid} Parameter
@cindex print_pid Parameter

(type=boolean, default=@samp{no}).  Same as the @code{-p} option to @i{Amd}.
If @samp{yes}, @i{Amd} will print its process ID upon starting.

@c ----------------------------------------------------------------
@node print_version Parameter, restart_mounts Parameter, print_pid Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{print_version} Parameter
@cindex print_version Parameter

(type=boolean, default=@samp{no}).  Same as the @code{-v} option to @i{Amd},
but the version prints and @i{Amd} continues to run.  If @samp{yes}, @i{Amd}
will print its version information string, which includes some
configuration and compilation values.

@c ----------------------------------------------------------------
@node restart_mounts Parameter, selectors_on_default Parameter, print_version Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{restart_mounts} Parameter
@cindex restart_mounts Parameter

(type=boolean, default=@samp{no}).  Same as the @code{-r} option to @i{Amd}.
If @samp{yes} @i{Amd} will scan the mount table to determine which file
systems are currently mounted.  Whenever one of these would have been
auto-mounted, @i{Amd} inherits it.

@c ----------------------------------------------------------------
@node selectors_on_default Parameter, show_statfs_entries Parameter, restart_mounts Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{selectors_on_default} Parameter
@cindex selectors_on_default Parameter

(type=boolean, default=@samp{no}).  If @samp{yes}, then the @samp{/defaults} entry of
maps will be looked for and any selectors processed before setting defaults
for all other keys in that map.  Useful when you want to set different
options for a complete map based on some parameters.  For example, you
may want to better the NFS performance over slow slip-based networks as
follows:

@example
/defaults \
    wire==slip-net;opts:=intr,rsize=1024,wsize=1024 \
    wire!=slip-net;opts:=intr,rsize=8192,wsize=8192
@end example

@c ----------------------------------------------------------------
@node show_statfs_entries Parameter, unmount_on_exit Parameter , selectors_on_default Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{show_statfs_entries} Parameter
@cindex show_statfs_entries Parameter

(type=boolean), default=@samp{no}).  If @samp{yes}, then all maps which are
browsable will also show the number of entries (keys) they have when
@b{df}(1) runs. (This is accomplished by returning non-zero values to
the @b{statfs}(2) system call).

@c ----------------------------------------------------------------
@node unmount_on_exit Parameter, , show_statfs_entries Parameter, Global Parameters
@comment  node-name,  next,  previous,  up
@subsection @t{unmount_on_exit} Parameter
@cindex unmount_on_exit Parameter

(type=boolean), default=@samp{no}).  If @samp{yes}, then @i{Amd} will attempt
to unmount all file systems which it knows about.  Normally it leaves
all (esp. NFS) mounted file systems intact.  Note that @i{Amd} does not
know about file systems mounted before it starts up, unless the
@samp{restart_mounts} option is used (@pxref{restart_mounts Parameter}).

@c ================================================================
@node Regular Map Parameters, amd.conf Examples, Global Parameters, Amd Configuration File
@comment  node-name,  next,  previous,  up
@section Regular Map Parameters
@cindex amd.conf regular map parameters

The following parameters are applicable only to regular map sections.

@menu
* map_name Parameter::
* tag Parameter::
@end menu

@c ----------------------------------------------------------------
@node map_name Parameter, tag Parameter, Regular Map Parameters, Regular Map Parameters
@comment  node-name,  next,  previous,  up
@subsection map_name Parameter
@cindex map_name Parameter

(type=string, must be specified).  Name of the map where the keys are
located.

@c ----------------------------------------------------------------
@node tag Parameter, , map_name Parameter, Regular Map Parameters
@comment  node-name,  next,  previous,  up
@subsection tag Parameter
@cindex tag Parameter

(type=string, default no tag).  Each map entry in the configuration file
can be tagged.  If no tag is specified, that map section will always be
processed by @i{Amd}.  If it is specified, then @i{Amd} will process the map
if the @code{-T} option was given to @i{Amd}, and the value given to that
command-line option matches that in the map section.

@c ================================================================
@node amd.conf Examples, , Regular Map Parameters, Amd Configuration File
@comment  node-name,  next,  previous,  up
@section amd.conf Examples
@cindex amd.conf examples

The following is the actual @code{amd.conf} file I use at the
Computer Science Department of Columbia University.

@example
# GLOBAL OPTIONS SECTION
[ global ]
normalize_hostnames =    no
print_pid =              no
#pid_file =              /var/run/amd.pid
restart_mounts =         yes
#unmount_on_exit =       yes
auto_dir =               /n
log_file =               /var/log/amd
log_options =            all
#debug_options =         all
plock =                  no
selectors_on_default =   yes
# config.guess picks up "sunos5" and I don't want to edit my maps yet
os =                     sos5
# if you print_version after setting up "os", it will show it.
print_version =          no
map_type =               file
search_path =            /etc/amdmaps:/usr/lib/amd:/usr/local/AMD/lib
browsable_dirs =         yes
fully_qualified_hosts =  no

# DEFINE AN AMD MOUNT POINT
[ /u ]
map_name =               amd.u

[ /proj ]
map_name =               amd.proj

[ /src ]
map_name =               amd.src

[ /misc ]
map_name =               amd.misc

[ /import ]
map_name =               amd.import

[ /tftpboot/.amd ]
tag =                    tftpboot
map_name =               amd.tftpboot
@end example

@c ################################################################
@node Run-time Administration, FSinfo, Amd Configuration File, Top
@comment  node-name,  next,  previous,  up
@chapter Run-time Administration
@cindex Run-time administration
@cindex Amq command

@menu
* Starting Amd::
* Stopping Amd::
* Restarting Amd::
* Controlling Amd::
@end menu

@node Starting Amd, Stopping Amd, Run-time Administration, Run-time Administration
@comment  node-name,  next,  previous,  up
@section Starting @i{Amd}
@cindex Starting Amd
@cindex Additions to /etc/rc.local
@cindex /etc/rc.local additions
@cindex ctl-amd

@i{Amd} is best started from @samp{/etc/rc.local} on BSD systems, or
from the appropriate start-level script in @samp{/etc/init.d} on System V
systems.

@example
if [ -f /usr/local/sbin/ctl-amd ]; then
    /usr/local/sbin/ctl-amd start; (echo -n ' amd') > /dev/console
fi
@end example

@noindent
The shell script, @samp{ctl-amd} is used to start, stop, or restart
@i{Amd}.  It is a relatively generic script.  All options you want to
set should not be made in this script, but rather updated in the
@file{amd.conf} file. @xref{Amd Configuration File}.

If you do not wish to use an @i{Amd} configuration file, you may start
@i{Amd} manually.  For example, getting the map entries via NIS:

@example
amd -r -l /var/log/amd `ypcat -k auto.master`
@end example

@node Stopping Amd, Restarting Amd, Starting Amd, Run-time Administration
@comment  node-name,  next,  previous,  up
@section Stopping @i{Amd}
@cindex Stopping Amd
@cindex SIGTERM signal
@cindex SIGINT signal

@i{Amd} stops in response to two signals.

@table @samp
@item SIGTERM
causes the top-level automount points to be unmounted and then @i{Amd}
to exit.  Any automounted filesystems are left mounted.  They can be
recovered by restarting @i{Amd} with the @code{-r} command line option.@refill

@item SIGINT
causes @i{Amd} to attempt to unmount any filesystems which it has
automounted, in addition to the actions of @samp{SIGTERM}.  This signal
is primarily used for debugging.@refill
@end table

Actions taken for other signals are undefined.

The easiest and safest way to stop @i{Amd}, without having to find its
process ID by hand, is to use the @file{ctl-amd} script, as with:

@example
ctl-amd stop
@end example

@node Restarting Amd, Controlling Amd, Stopping Amd, Run-time Administration
@comment  node-name,  next,  previous,  up
@section Restarting @i{Amd}
@cindex Restarting Amd
@cindex Killing and starting Amd

Before @i{Amd} can be started, it is vital to ensure that no other
@i{Amd} processes are managing any of the mount points, and that the
previous process(es) have terminated cleanly.  When a terminating signal
is set to @i{Amd}, the automounter does @emph{not} terminate right then.
Rather, it starts by unmounting all of its managed mount mounts in the
background, and then terminates.  It usually takes a few seconds for
this process to happen, but it can take an arbitrarily longer time.  If
two or more @i{Amd} processes attempt to manage the same mount point, it
usually will result in a system lockup.

The easiest and safest way to restart @i{Amd}, without having to find
its process ID by hand, sending it the @samp{SIGTERM} signal, waiting for @i{Amd}
to die cleanly, and verifying so, is to use the @file{ctl-amd} script,
as with:

@example
ctl-amd restart
@end example

The script will locate the process ID of @i{Amd}, kill it, and wait for
it to die cleanly before starting a new instance of the automounter.
@file{ctl-amd} will wait for a total of 30 seconds for @i{Amd} to die,
and will check once every 5 seconds if it had.

@node Controlling Amd, , Restarting Amd, Run-time Administration
@comment  node-name,  next,  previous,  up
@section Controlling @i{Amd}
@cindex Controlling Amd
@cindex Discovering what is going on at run-time
@cindex Listing currently mounted filesystems

It is sometimes desirable or necessary to exercise external control
over some of @i{Amd}'s internal state.  To support this requirement,
@i{Amd} implements an RPC interface which is used by the @dfn{Amq} program.
A variety of information is available.

@i{Amq} generally applies an operation, specified by a single letter option,
to a list of mount points.  The default operation is to obtain statistics
about each mount point.  This is similar to the output shown above
but includes information about the number and type of accesses to each
mount point.

@menu
* Amq default::       Default command behavior.
* Amq -f option::     Flushing the map cache.
* Amq -h option::     Controlling a non-local host.
* Amq -l option::     Controlling the log file.
* Amq -m option::     Obtaining mount statistics.
* Amq -M-option::     Mounting a volume.
* Amq -p option::     Getting Amd's process ID.
* Amq -P-option::     Contacting alternate Amd processes.
* Amq -s option::     Obtaining global statistics.
* Amq -T option::     Use TCP transport.
* Amq -U-option::     Use UDP transport.
* Amq -u option::     Forcing volumes to time out.
* Amq -v option::     Version information.
* Other Amq options:: Three other special options.
@end menu

@c ----------------------------------------------------------------
@node Amq default, Amq -f option, Controlling Amd, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} default information

With no arguments, @dfn{Amq} obtains a brief list of all existing
mounts created by @i{Amd}.  This is different from the list displayed by
@b{df}(1) since the latter only includes system mount points.

@noindent
The output from this option includes the following information:

@itemize @bullet
@item
the automount point,
@item
the filesystem type,
@item
the mount map or mount information,
@item
the internal, or system mount point.
@end itemize

@noindent
For example:

@example
/            root   "root"                    sky:(pid75)
/homes       toplvl /usr/local/etc/amd.homes  /homes
/home        toplvl /usr/local/etc/amd.home   /home
/homes/jsp   nfs    charm:/home/charm         /a/charm/home/charm/jsp
/homes/phjk  nfs    toytown:/home/toytown     /a/toytown/home/toytown/ai/phjk
@end example

@noindent
If an argument is given then statistics for that volume name will
be output.  For example:

@example
What         Uid   Getattr Lookup RdDir   RdLnk   Statfs Mounted@@
/homes       0     1196    512    22      0       30     90/09/14 12:32:55
/homes/jsp   0     0       0      0       1180    0      90/10/13 12:56:58
@end example

@table @code
@item What
the volume name.

@item Uid
ignored.

@item Getattr
the count of NFS @dfn{getattr} requests on this node.  This should only be
non-zero for directory nodes.

@item Lookup
the count of NFS @dfn{lookup} requests on this node.  This should only be
non-zero for directory nodes.

@item RdDir
the count of NFS @dfn{readdir} requests on this node.  This should only
be non-zero for directory nodes.

@item RdLnk
the count of NFS @dfn{readlink} requests on this node.  This should be
zero for directory nodes.

@item Statfs
the count of NFS @dfn{statfs} requests on this node.  This should only
be non-zero for top-level automount points.

@item Mounted@@
the date and time the volume name was first referenced.
@end table

@c ----------------------------------------------------------------
@node Amq -f option, Amq -h option, Amq default, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-f} option
@cindex Flushing the map cache
@cindex Map cache, flushing

The @code{-f} option causes @i{Amd} to flush the internal mount map cache.
This is useful for example in Hesiod maps since @i{Amd} will not
automatically notice when they have been updated.  The map cache can
also be synchronized with the map source by using the @samp{sync} option
(@pxref{Automount Filesystem}).@refill

@c ----------------------------------------------------------------
@node Amq -l option, Amq -m option, Amq -h option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-l} option
@cindex Resetting the Amd log file
@cindex Setting the Amd log file via Amq
@cindex Log file, resetting

Tell @i{Amd} to use @i{log_file} as the log file name.  For security
reasons, this @emph{must} be the same log file which @i{Amd} used when
started.  This option is therefore only useful to refresh @i{Amd}'s open
file handle on the log file, so that it can be rotated and compressed
via daily cron jobs.

@c ----------------------------------------------------------------
@node Amq -h option, Amq -l option, Amq -f option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-h} option
@cindex Querying an alternate host

By default the local host is used.  In an HP-UX cluster the root server
is used since that is the only place in the cluster where @i{Amd} will
be running.  To query @i{Amd} on another host the @code{-h} option should
be used.

@c ----------------------------------------------------------------
@node Amq -m option, Amq -M-option, Amq -l option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-m} option

The @code{-m} option displays similar information about mounted
filesystems, rather than automount points.  The output includes the
following information:

@itemize @bullet
@item
the mount information,
@item
the mount point,
@item
the filesystem type,
@item
the number of references to this filesystem,
@item
the server hostname,
@item
the state of the file server,
@item
any error which has occurred.
@end itemize

For example:

@example
"root"           truth:(pid602)     root   1 localhost is up
hesiod.home      /home              toplvl 1 localhost is up
hesiod.vol       /vol               toplvl 1 localhost is up
hesiod.homes     /homes             toplvl 1 localhost is up
amy:/home/amy    /a/amy/home/amy    nfs    5 amy is up
swan:/home/swan  /a/swan/home/swan  nfs    0 swan is up (Permission denied)
ex:/home/ex      /a/ex/home/ex      nfs    0 ex is down
@end example

When the reference count is zero the filesystem is not mounted but
the mount point and server information is still being maintained
by @i{Amd}.

@c ----------------------------------------------------------------
@node Amq -M-option, Amq -p option, Amq -m option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-M} option

The @code{-M} option passes a new map entry to @i{Amd} and waits for it to
be evaluated, possibly causing a mount.  For example, the following
command would cause @samp{/home/toytown} on host @samp{toytown} to be
mounted locally on @samp{/mnt/toytown}.

@example
amq -M '/mnt/toytown type:=nfs;rfs:=/home/toytown;rhost:=toytown;fs:=$@{key@}'
@end example

@i{Amd} applies some simple security checks before allowing this
operation.  The check tests whether the incoming request is from a
privileged UDP port on the local machine.  ``Permission denied'' is
returned if the check fails.

This option is very insecure as it is vulnerable to attacks such as IP
Spoofing.  In other words, it is relatively easy for an attacker who
really wants to, to make your @i{Amd} process mount any filesystem from
the Internet!  Therefore, the @emph{complete} code which supports the
@code{-M} option in @i{Amd} and @i{Amq} is turned off by default.  To turn
it on, you have to reconfigure am-utils with @code{configure
--enable-amq-mount}.  Think twice before doing so, and use this option
only if you absolutely need to.

A future release of @i{Amd} will include code to allow the @b{mount}(8)
command to mount automount points:

@example
mount -t amd /vol hesiod.vol
@end example

This will then allow @i{Amd} to be controlled from the standard system
filesystem mount list.

@c ----------------------------------------------------------------
@node Amq -p option, Amq -P-option, Amq -M-option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-p} option
@cindex Process ID; Amd
@cindex Amd's process ID
@cindex Amd's PID
@cindex PID; Amd

Return the process ID of the remote or locally running @i{Amd}.  Useful
when you need to send a signal to the local @i{Amd} process, and would
rather not have to search through the process table.  This option is
used in the @file{ctl-amd} script.

@c ----------------------------------------------------------------
@node Amq -P-option, Amq -s option, Amq -p option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-P} option
@cindex Multiple Amd processes
@cindex Running multiple Amd
@cindex Debugging a new Amd configuration
@cindex RPC Program numbers; Amd

Contact an alternate running @i{Amd} that had registered itself on a
different RPC @var{program_number} and apply all other operations to
that instance of the automounter.  This is useful when you run multiple
copies of @i{Amd}, and need to manage each one separately.  If not
specified, @i{Amq} will use the default program number for @i{Amd}, 300019.
For security reasons, the only alternate program numbers @i{Amd} can use
range from 300019 to 300029, inclusive.

For example, to kill an alternate running @i{Amd}:

@example
kill `amq -p -P 300020`
@end example

@c ----------------------------------------------------------------
@node Amq -s option, Amq -T option, Amq -P-option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-s} option
@cindex Global statistics
@cindex Statistics

The @code{-s} option displays global statistics.  If any other options are specified
or any filesystems named then this option is ignored.  For example:

@example
requests  stale     mount     mount     unmount
deferred  fhandles  ok        failed    failed
1054      1         487       290       7017
@end example

@table @samp
@item Deferred requests
are those for which an immediate reply could not be constructed.  For
example, this would happen if a background mount was required.

@item Stale filehandles
counts the number of times the kernel passes a stale filehandle to @i{Amd}.
Large numbers indicate problems.

@item Mount ok
counts the number of automounts which were successful.

@item Mount failed
counts the number of automounts which failed.

@item Unmount failed
counts the number of times a filesystem could not be unmounted.  Very
large numbers here indicate that the time between unmount attempts
should be increased.
@end table

@c ----------------------------------------------------------------
@node Amq -T option, Amq -U-option, Amq -s option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-T} option
@cindex Forcing Amq to use a TCP transport
@cindex TCP; using with Amq

The @code{-T} option causes the @i{Amq} to contact @i{Amd} using the TCP
transport only (connection oriented).  Normally, @i{Amq} will use TCP
first, and if that failed, will try UDP.

@c ----------------------------------------------------------------
@node Amq -U-option, Amq -u option, Amq -T option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-U} option
@cindex Forcing Amq to use a UDP transport
@cindex UDP; using with Amq

The @code{-U} option causes the @i{Amq} to contact @i{Amd} using the UDP
transport only (connectionless).  Normally, @i{Amq} will use TCP first,
and if that failed, will try UDP.

@c ----------------------------------------------------------------
@node Amq -u option, Amq -v option, Amq -U-option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-u} option
@cindex Forcing filesystem to time out
@cindex Unmounting a filesystem

The @code{-u} option causes the time-to-live interval of the named mount
points to be expired, thus causing an unmount attempt.  This is the only
safe way to unmount an automounted filesystem.  It is not possible to
unmount a filesystem which has been mounted with the @samp{nounmount}
flag.

@c The @code{-H} option informs @i{Amd} that the specified mount point has hung -
@c as if its keepalive timer had expired.

@c ----------------------------------------------------------------
@node Amq -v option, Other Amq options, Amq -u option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection @i{Amq} @code{-v} option
@cindex Version information at run-time

The @code{-v} option displays the version of @i{Amd} in a similar way to
@i{Amd}'s @code{-v} option.

@c ----------------------------------------------------------------
@node Other Amq options, , Amq -v option, Controlling Amd
@comment  node-name,  next,  previous,  up
@subsection Other @i{Amq} options
@cindex Logging options via Amq
@cindex Debugging options via Amq

Two other operations are implemented.  These modify the state of @i{Amd}
as a whole, rather than any particular filesystem.  The @code{-x} and
@code{-D} options have exactly the same effect as @i{Amd}'s corresponding
command line options.

When @i{Amd} receives a @code{-x} flag it limits the log options being
modified to those which were not enabled at startup.  This prevents a
user turning @emph{off} any logging option which was specified at
startup, though any which have been turned on since then can still be
turned off.  The @code{-D} option has a similar behavior.

@c ################################################################
@node FSinfo, Hlfsd, Run-time Administration, Top
@comment  node-name,  next,  previous,  up
@chapter FSinfo
@cindex FSinfo
@cindex Filesystem info package

XXX: this chapter should be reviewed by someone knowledgeable with
fsinfo.

@menu
* FSinfo Overview::                 Introduction to FSinfo.
* Using FSinfo::                    Basic concepts.
* FSinfo Grammar::                  Language syntax, semantics and examples.
* FSinfo host definitions::         Defining a new host.
* FSinfo host attributes::          Definable host attributes.
* FSinfo filesystems::              Defining locally attached filesystems.
* FSinfo static mounts::            Defining additional static mounts.
* FSinfo automount definitions::
* FSinfo Command Line Options::
* FSinfo errors::
@end menu

@node FSinfo Overview, Using FSinfo, FSinfo, FSinfo
@comment  node-name,  next,  previous,  up
@section @i{FSinfo} overview
@cindex FSinfo overview

@i{FSinfo} is a filesystem management tool.  It has been designed to
work with @i{Amd} to help system administrators keep track of the ever
increasing filesystem namespace under their control.

The purpose of @i{FSinfo} is to generate all the important standard
filesystem data files from a single set of input data.  Starting with a
single data source guarantees that all the generated files are
self-consistent.  One of the possible output data formats is a set of
@i{Amd} maps which can be used amongst the set of hosts described in the
input data.

@i{FSinfo} implements a declarative language.  This language is
specifically designed for describing filesystem namespace and physical
layouts.  The basic declaration defines a mounted filesystem including
its device name, mount point, and all the volumes and access
permissions.  @i{FSinfo} reads this information and builds an internal
map of the entire network of hosts.  Using this map, many different data
formats can be produced including @file{/etc/fstab},
@file{/etc/exports}, @i{Amd} mount maps and
@file{/etc/bootparams}.@refill

@node Using FSinfo, FSinfo Grammar, FSinfo Overview, FSinfo
@comment  node-name,  next,  previous,  up
@section Using @i{FSinfo}
@cindex Using FSinfo

The basic strategy when using @i{FSinfo} is to gather all the
information about all disks on all machines into one set of
declarations.  For each machine being managed, the following data is
required:

@itemize @bullet
@item
Hostname
@item
List of all filesystems and, optionally, their mount points.
@item
Names of volumes stored on each filesystem.
@item
NFS export information for each volume.
@item
The list of static filesystem mounts.
@end itemize

The following information can also be entered into the same
configuration files so that all data can be kept in one place.

@itemize @bullet
@item
List of network interfaces
@item
IP address of each interface
@item
Hardware address of each interface
@item
Dumpset to which each filesystem belongs
@item
and more @dots{}
@end itemize

To generate @i{Amd} mount maps, the automount tree must also be defined
(@pxref{FSinfo automount definitions}).  This will have been designed at
the time the volume names were allocated.  Some volume names will not be
automounted, so @i{FSinfo} needs an explicit list of which volumes
should be automounted.@refill

Hostnames are required at several places in the @i{FSinfo} language.  It
is important to stick to either fully qualified names or unqualified
names.  Using a mixture of the two will inevitably result in confusion.

Sometimes volumes need to be referenced which are not defined in the set
of hosts being managed with @i{FSinfo}.  The required action is to add a
dummy set of definitions for the host and volume names required.  Since
the files generated for those particular hosts will not be used on them,
the exact values used is not critical.

@node FSinfo Grammar, FSinfo host definitions, Using FSinfo, FSinfo
@comment  node-name,  next,  previous,  up
@section @i{FSinfo} grammar
@cindex FSinfo grammar
@cindex Grammar, FSinfo

@i{FSinfo} has a relatively simple grammar.  Distinct syntactic
constructs exist for each of the different types of data, though they
share a common flavor.  Several conventions are used in the grammar
fragments below.

The notation, @i{list(}@t{xxx}@i{)}, indicates a list of zero or more
@t{xxx}'s.  The notation, @i{opt(}@t{xxx}@i{)}, indicates zero or one
@t{xxx}.  Items in double quotes, @i{eg} @t{"host"}, represent input
tokens.  Items in angle brackets, @i{eg} @var{<hostname>}, represent
strings in the input.  Strings need not be in double quotes, except to
differentiate them from reserved words.  Quoted strings may include the
usual set of C ``@t{\}'' escape sequences with one exception: a
backslash-newline-whitespace sequence is squashed into a single space
character.  To defeat this feature, put a further backslash at the start
of the second line.

At the outermost level of the grammar, the input consists of a
sequence of host and automount declarations.  These declarations are
all parsed before they are analyzed.  This means they can appear in
any order and cyclic host references are possible.

@example
fsinfo      : @i{list(}fsinfo_attr@i{)} ;

fsinfo_attr : host | automount ;
@end example

@menu
* FSinfo host definitions::
* FSinfo automount definitions::
@end menu

@node FSinfo host definitions, FSinfo host attributes, FSinfo Grammar, FSinfo
@comment  node-name,  next,  previous,  up
@section @i{FSinfo} host definitions
@cindex FSinfo host definitions
@cindex Defining a host, FSinfo

A host declaration consists of three parts: a set of machine attribute
data, a list of filesystems physically attached to the machine, and a
list of additional statically mounted filesystems.

@example
host        : "host" host_data @i{list(}filesystem@i{@i{)}} @i{list(}mount@i{@i{)}} ;
@end example

Each host must be declared in this way exactly once.  Such things as the
hardware address, the architecture and operating system types and the
cluster name are all specified within the @dfn{host data}.

All the disks the machine has should then be described in the @dfn{list
of filesystems}.  When describing disks, you can specify what
@dfn{volname} the disk/partition should have and all such entries are
built up into a dictionary which can then be used for building the
automounter maps.

The @dfn{list of mounts} specifies all the filesystems that should be
statically mounted on the machine.

@menu
* FSinfo host attributes::
* FSinfo filesystems::
* FSinfo static mounts::
@end menu

@node FSinfo host attributes, FSinfo filesystems, FSinfo host definitions , FSinfo host definitions
@comment  node-name,  next,  previous,  up
@section @i{FSinfo} host attributes
@cindex FSinfo host attributes
@cindex Defining host attributes, FSinfo

The host data, @dfn{host_data}, always includes the @dfn{hostname}.  In
addition, several other host attributes can be given.

@example
host_data   : @var{<hostname>}
            | "@{" @i{list(}host_attrs@i{)} "@}" @var{<hostname>}
            ;

host_attrs  : host_attr "=" @var{<string>}
            | netif
            ;

host_attr   : "config"
            | "arch"
            | "os"
            | "cluster"
            ;
@end example

The @dfn{hostname} is, typically, the fully qualified hostname of the
machine.

Examples:

@example
host dylan.doc.ic.ac.uk

host @{
    os = hpux
    arch = hp300
@} dougal.doc.ic.ac.uk
@end example

The options that can be given as host attributes are shown below.

@menu
* netif Option:         FSinfo host netif:
* config Option:        FSinfo host config:
* arch Option:          FSinfo host arch:
* os Option:            FSinfo host os:
* cluster Option:       FSinfo host cluster:
@end menu

@node FSinfo host netif, FSinfo host config, , FSinfo host attributes
@comment  node-name,  next,  previous,  up
@subsection netif Option

This defines the set of network interfaces configured on the machine.
The interface attributes collected by @i{FSinfo} are the IP address,
subnet mask and hardware address.  Multiple interfaces may be defined
for hosts with several interfaces by an entry for each interface.  The
values given are sanity checked, but are currently unused for anything
else.

@example
netif       : "netif" @var{<string>} "@{" @i{list(}netif_attrs@i{)} "@}" ;

netif_attrs : netif_attr "=" @var{<string>} ;

netif_attr  : "inaddr" | "netmask" | "hwaddr" ;
@end example

Examples:

@example
netif ie0 @{
    inaddr  = 129.31.81.37
    netmask = 0xfffffe00
    hwaddr  = "08:00:20:01:a6:a5"
@}

netif ec0 @{ @}
@end example

@node FSinfo host config, FSinfo host arch, FSinfo host netif, FSinfo host attributes
@comment  node-name,  next,  previous,  up
@subsection config Option
@cindex FSinfo config host attribute
@cindex config, FSinfo host attribute

This option allows you to specify configuration variables for the
startup scripts (@file{rc} scripts).  A simple string should immediately
follow the keyword.

Example:

@example
config "NFS_SERVER=true"
config "ZEPHYR=true"
@end example

This option is currently unsupported.

@node FSinfo host arch, FSinfo host os, FSinfo host config, FSinfo host attributes
@comment  node-name,  next,  previous,  up
@subsection arch Option
@cindex FSinfo arch host attribute
@cindex arch, FSinfo host attribute

This defines the architecture of the machine.  For example:

@example
arch = hp300
@end example

This is intended to be of use when building architecture specific
mountmaps, however, the option is currently unsupported.

@node FSinfo host os, FSinfo host cluster, FSinfo host arch, FSinfo host attributes
@comment  node-name,  next,  previous,  up
@subsection os Option
@cindex FSinfo os host attribute
@cindex os, FSinfo host attribute

This defines the operating system type of the host.  For example:

@example
os = hpux
@end example

This information is used when creating the @file{fstab} files, for
example in choosing which format to use for the @file{fstab} entries
within the file.

@node FSinfo host cluster, , FSinfo host os, FSinfo host attributes
@comment  node-name,  next,  previous,  up
@subsection cluster Option
@cindex FSinfo cluster host attribute
@cindex cluster, FSinfo host attribute

This is used for specifying in which cluster the machine belongs.  For
example:

@example
cluster = "theory"
@end example

The cluster is intended to be used when generating the automount maps,
although it is currently unsupported.

@node FSinfo filesystems, FSinfo static mounts, FSinfo host attributes, FSinfo host definitions
@comment  node-name,  next,  previous,  up
@section @i{FSinfo} filesystems
@cindex FSinfo filesystems

The list of physically attached filesystems follows the machine
attributes.  These should define all the filesystems available from this
machine, whether exported or not.  In addition to the device name,
filesystems have several attributes, such as filesystem type, mount
options, and @samp{fsck} pass number which are needed to generate
@file{fstab} entries.

@example
filesystem  : "fs" @var{<device>} "@{" @i{list(}fs_data@i{)} "@}" ;

fs_data     : fs_data_attr "=" @var{<string>}
            | mount
            ;

fs_data_attr
            : "fstype" | "opts" | "passno"
            | "freq" | "dumpset" | "log"
            ;
@end example

Here, @var{<device>} is the device name of the disk (for example,
@file{/dev/dsk/2s0}).  The device name is used for building the mount
maps and for the @file{fstab} file.  The attributes that can be
specified are shown in the following section.

The @i{FSinfo} configuration file for @code{dylan.doc.ic.ac.uk} is listed below.

@example
host dylan.doc.ic.ac.uk

fs /dev/dsk/0s0 @{
        fstype = swap
@}

fs /dev/dsk/0s0 @{
        fstype = hfs
        opts = rw,noquota,grpid
        passno = 0;
        freq = 1;
        mount / @{ @}
@}

fs /dev/dsk/1s0 @{
        fstype = hfs
        opts = defaults
        passno = 1;
        freq = 1;
        mount /usr @{
                local @{
                        exportfs "dougal eden dylan zebedee brian"
                        volname /nfs/hp300/local
                @}
        @}
@}

fs /dev/dsk/2s0 @{
        fstype = hfs
        opts = defaults
        passno = 1;
        freq = 1;
        mount default @{
                exportfs "toytown_clients hangers_on"
                volname /home/dylan/dk2
        @}
@}

fs /dev/dsk/3s0 @{
        fstype = hfs
        opts = defaults
        passno = 1;
        freq = 1;
        mount default @{
                exportfs "toytown_clients hangers_on"
                volname /home/dylan/dk3
        @}
@}

fs /dev/dsk/5s0 @{
        fstype = hfs
        opts = defaults
        passno = 1;
        freq = 1;
        mount default @{
                exportfs "toytown_clients hangers_on"
                volname /home/dylan/dk5
        @}
@}
@end example

@menu
* fstype Option:        FSinfo filesystems fstype:
* opts Option:          FSinfo filesystems opts:
* passno Option:        FSinfo filesystems passno:
* freq Option:          FSinfo filesystems freq:
* mount Option:         FSinfo filesystems mount:
* dumpset Option:       FSinfo filesystems dumpset:
* log Option:           FSinfo filesystems log:
@end menu

@node FSinfo filesystems fstype, FSinfo filesystems opts, , FSinfo filesystems
@comment  node-name,  next,  previous,  up
@subsection fstype Option
@cindex FSinfo fstype filesystems option
@cindex fstype, FSinfo filesystems option
@cindex export, FSinfo special fstype

This specifies the type of filesystem being declared and will be placed
into the @file{fstab} file as is.  The value of this option will be
handed to @code{mount} as the filesystem type---it should have such
values as @code{4.2}, @code{nfs} or @code{swap}.  The value is not
examined for correctness.

There is one special case.  If the filesystem type is specified as
@samp{export} then the filesystem information will not be added to the
host's @file{fstab} information, but it will still be visible on the
network.  This is useful for defining hosts which contain referenced
volumes but which are not under full control of @i{FSinfo}.

Example:

@example
fstype = swap
@end example

@node FSinfo filesystems opts, FSinfo filesystems passno, FSinfo filesystems fstype, FSinfo filesystems
@comment  node-name,  next,  previous,  up
@subsection opts Option
@cindex FSinfo opts filesystems option
@cindex opts, FSinfo filesystems option

This defines any options that should be given to @b{mount}(8) in the
@file{fstab} file.  For example:

@example
opts = rw,nosuid,grpid
@end example

@node FSinfo filesystems passno, FSinfo filesystems freq, FSinfo filesystems opts, FSinfo filesystems
@comment  node-name,  next,  previous,  up
@subsection passno Option
@cindex FSinfo passno filesystems option
@cindex passno, FSinfo filesystems option

This defines the @b{fsck}(8) pass number in which to check the
filesystem.  This value will be placed into the @file{fstab} file.

Example:

@example
passno = 1
@end example

@node FSinfo filesystems freq, FSinfo filesystems mount, FSinfo filesystems passno, FSinfo filesystems
@comment  node-name,  next,  previous,  up
@subsection freq Option
@cindex FSinfo freq filesystems option
@cindex freq, FSinfo filesystems option

This defines the interval (in days) between dumps.  The value is placed
as is into the @file{fstab} file.

Example:

@example
freq = 3
@end example

@node FSinfo filesystems mount, FSinfo filesystems dumpset, FSinfo filesystems freq, FSinfo filesystems
@comment  node-name,  next,  previous,  up
@subsection mount Option
@cindex FSinfo mount filesystems option
@cindex mount, FSinfo filesystems option
@cindex exportfs, FSinfo mount option
@cindex volname, FSinfo mount option
@cindex sel, FSinfo mount option

This defines the mountpoint at which to place the filesystem.  If the
mountpoint of the filesystem is specified as @code{default}, then the
filesystem will be mounted in the automounter's tree under its volume
name and the mount will automatically be inherited by the automounter.

Following the mountpoint, namespace information for the filesystem may
be described.  The options that can be given here are @code{exportfs},
@code{volname} and @code{sel}.

The format is:

@example
mount       : "mount" vol_tree ;

vol_tree    : @i{list(}vol_tree_attr@i{)} ;

vol_tree_attr
            :  @var{<string>} "@{" @i{list(}vol_tree_info@i{)} vol_tree "@}" ;

vol_tree_info
            : "exportfs" @var{<export-data>}
            | "volname" @var{<volname>}
            | "sel" @var{<selector-list>}
            ;
@end example

Example:

@example
mount default @{
    exportfs "dylan dougal florence zebedee"
    volname /vol/andrew
@}
@end example

In the above example, the filesystem currently being declared will have
an entry placed into the @file{exports} file allowing the filesystem to
be exported to the machines @code{dylan}, @code{dougal}, @code{florence}
and @code{zebedee}.  The volume name by which the filesystem will be
referred to remotely, is @file{/vol/andrew}.  By declaring the
mountpoint to be @code{default}, the filesystem will be mounted on the
local machine in the automounter tree, where @i{Amd} will automatically
inherit the mount as @file{/vol/andrew}.@refill

@table @samp
@item exportfs
a string defining which machines the filesystem may be exported to.
This is copied, as is, into the @file{exports} file---no sanity checking
is performed on this string.@refill

@item volname
a string which declares the remote name by which to reference the
filesystem.  The string is entered into a dictionary and allows you to
refer to this filesystem in other places by this volume name.@refill

@item sel
a string which is placed into the automounter maps as a selector for the
filesystem.@refill

@end table

@node FSinfo filesystems dumpset, FSinfo filesystems log, FSinfo filesystems mount, FSinfo filesystems
@comment  node-name,  next,  previous,  up
@subsection dumpset Option
@cindex FSinfo dumpset filesystems option
@cindex dumpset, FSinfo filesystems option

This provides support for Imperial College's local file backup tools and
is not documented further here.

@node FSinfo filesystems log, , FSinfo filesystems dumpset, FSinfo filesystems
@comment  node-name,  next,  previous,  up
@subsection log Option
@cindex FSinfo log filesystems option
@cindex log, FSinfo filesystems option

Specifies the log device for the current filesystem. This is ignored if
not required by the particular filesystem type.

@node FSinfo static mounts, FSinfo automount definitions , FSinfo filesystems, FSinfo host definitions
@comment  node-name,  next,  previous,  up
@section @i{FSinfo} static mounts
@cindex FSinfo static mounts
@cindex Statically mounts filesystems, FSinfo

Each host may also have a number of statically mounted filesystems.  For
example, the host may be a diskless workstation in which case it will
have no @code{fs} declarations.  In this case the @code{mount}
declaration is used to determine from where its filesystems will be
mounted.  In addition to being added to the @file{fstab} file, this
information can also be used to generate a suitable @file{bootparams}
file.@refill

@example
mount       : "mount" @var{<volname>} @i{list(}localinfo@i{)} ;

localinfo   : localinfo_attr @var{<string>} ;

localinfo_attr
            : "as"
            | "from"
            | "fstype"
            | "opts"
            ;
@end example

The filesystem specified to be mounted will be searched for in the
dictionary of volume names built when scanning the list of hosts'
definitions.

The attributes have the following semantics:
@table @samp
@item from @var{machine}
mount the filesystem from the machine with the hostname of
@dfn{machine}.@refill

@item as @var{mountpoint}
mount the filesystem locally as the name given, in case this is
different from the advertised volume name of the filesystem.

@item opts @var{options}
native @b{mount}(8) options.

@item fstype @var{type}
type of filesystem to be mounted.
@end table

An example:

@example
mount /export/exec/hp300/local as /usr/local
@end example

If the mountpoint specified is either @file{/} or @file{swap}, the
machine will be considered to be booting off the net and this will be
noted for use in generating a @file{bootparams} file for the host which
owns the filesystems.

@node FSinfo automount definitions, FSinfo Command Line Options, FSinfo static mounts, FSinfo
@comment  node-name,  next,  previous,  up
@section Defining an @i{Amd} Mount Map in @i{FSinfo}
@cindex FSinfo automount definitions
@cindex Defining an Amd mount map, FSinfo

The maps used by @i{Amd} can be constructed from @i{FSinfo} by defining
all the automount trees.  @i{FSinfo} takes all the definitions found and
builds one map for each top level tree.

The automount tree is usually defined last.  A single automount
configuration will usually apply to an entire management domain.  One
@code{automount} declaration is needed for each @i{Amd} automount point.
@i{FSinfo} determines whether the automount point is @dfn{direct}
(@pxref{Direct Automount Filesystem}) or @dfn{indirect}
(@pxref{Top-level Filesystem}).  Direct automount points are
distinguished by the fact that there is no underlying
@dfn{automount_tree}.@refill

@example
automount   : "automount" @i{opt(}auto_opts@i{)} automount_tree ;

auto_opts   : "opts" @var{<mount-options>} ;

automount_tree
            : @i{list(}automount_attr@i{)}
            ;

automount_attr
            : @var{<string>} "=" @var{<volname>}
            | @var{<string>} "->" @var{<symlink>}
            | @var{<string>} "@{" automount_tree "@}"
            ;
@end example

If @var{<mount-options>} is given, then it is the string to be placed in
the maps for @i{Amd} for the @code{opts} option.

A @dfn{map} is typically a tree of filesystems, for example @file{home}
normally contains a tree of filesystems representing other machines in
the network.

A map can either be given as a name representing an already defined
volume name, or it can be a tree.  A tree is represented by placing
braces after the name.  For example, to define a tree @file{/vol}, the
following map would be defined:

@example
automount /vol @{ @}
@end example

Within a tree, the only items that can appear are more maps.
For example:

@example
automount /vol @{
    andrew @{ @}
    X11 @{ @}
@}
@end example

In this case, @i{FSinfo} will look for volumes named @file{/vol/andrew}
and @file{/vol/X11} and a map entry will be generated for each.  If the
volumes are defined more than once, then @i{FSinfo} will generate
a series of alternate entries for them in the maps.@refill

Instead of a tree, either a link (@var{name} @code{->}
@var{destination}) or a reference can be specified (@var{name} @code{=}
@var{destination}).  A link creates a symbolic link to the string
specified, without further processing the entry.  A reference will
examine the destination filesystem and optimize the reference.  For
example, to create an entry for @code{njw} in the @file{/homes} map,
either of the two forms can be used:@refill

@example
automount /homes @{
    njw -> /home/dylan/njw
@}
@end example

or

@example
automount /homes @{
    njw = /home/dylan/njw
@}
@end example

In the first example, when @file{/homes/njw} is referenced from @i{Amd},
a link will be created leading to @file{/home/dylan/njw} and the
automounter will be referenced a second time to resolve this filename.
The map entry would be:

@example
njw type:=link;fs:=/home/dylan/njw
@end example

In the second example, the destination directory is analyzed and found
to be in the filesystem @file{/home/dylan} which has previously been
defined in the maps. Hence the map entry will look like:

@example
njw rhost:=dylan;rfs:=/home/dylan;sublink:=njw
@end example

Creating only one symbolic link, and one access to @i{Amd}.

@node FSinfo Command Line Options, FSinfo errors, FSinfo automount definitions, FSinfo
@comment  node-name,  next,  previous,  up
@section @i{FSinfo} Command Line Options
@cindex FSinfo command line options
@cindex Command line options, FSinfo

@i{FSinfo} is started from the command line by using the command:

@example
fsinfo [@i{options}] @i{files} ...
@end example

The input to @i{FSinfo} is a single set of definitions of machines and
automount maps.  If multiple files are given on the command-line, then
the files are concatenated together to form the input source.  The files
are passed individually through the C pre-processor before being parsed.

Several options define a prefix for the name of an output file.  If the
prefix is not specified no output of that type is produced.  The suffix
used will correspond either to the hostname to which a file belongs, or
to the type of output if only one file is produced.  Dumpsets and the
@file{bootparams} file are in the latter class.  To put the output into
a subdirectory simply put a @file{/} at the end of the prefix, making
sure that the directory has already been made before running
@i{Fsinfo}.

@menu
* -a FSinfo Option::    Amd automount directory:
* -b FSinfo Option::    Prefix for bootparams files.
* -d FSinfo Option::    Prefix for dumpset data files.
* -e FSinfo Option::    Prefix for exports files.
* -f FSinfo Option::    Prefix for fstab files.
* -h FSinfo Option::    Local hostname.
* -m FSinfo Option::    Prefix for automount maps.
* -q FSinfo Option::    Ultra quiet mode.
* -v FSinfo Option::    Verbose mode.
* -I FSinfo Option::    Define new #include directory.
* -D-FSinfo Option::    Define macro.
* -U FSinfo Option::    Undefine macro.
@end menu

@node -a FSinfo Option, -b FSinfo Option, FSinfo Command Line Options, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-a} @var{autodir}

Specifies the directory name in which to place the automounter's
mountpoints.  This defaults to @file{/a}.  Some sites have the autodir set
to be @file{/amd}, and this would be achieved by:

@example
fsinfo -a /amd ...
@end example

@node -b FSinfo Option, -d FSinfo Option, -a FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-b} @var{bootparams}
@cindex bootparams, FSinfo prefix

This specifies the prefix for the @file{bootparams} filename.  If it is
not given, then the file will not be generated.  The @file{bootparams}
file will be constructed for the destination machine and will be placed
into a file named @file{bootparams} and prefixed by this string.  The
file generated contains a list of entries describing each diskless
client that can boot from the destination machine.

As an example, to create a @file{bootparams} file in the directory
@file{generic}, the following would be used:

@example
fsinfo -b generic/ ...
@end example

@node -d FSinfo Option, -e FSinfo Option, -b FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-d} @var{dumpsets}
@cindex dumpset, FSinfo prefix

This specifies the prefix for the @file{dumpsets} file.  If it is not
specified, then the file will not be generated.  The file will be for
the destination machine and will be placed into a filename
@file{dumpsets}, prefixed by this string.  The @file{dumpsets} file is
for use by Imperial College's local backup system.

For example, to create a @file{dumpsets} file in the directory @file{generic},
then you would use the following:

@example
fsinfo -d generic/ ...
@end example

@node -e FSinfo Option, -f FSinfo Option, -d FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-e} @var{exportfs}
@cindex exports, FSinfo prefix

Defines the prefix for the @file{exports} files.  If it is not given,
then the file will not be generated.  For each machine defined in the
configuration files as having disks, an @file{exports} file is
constructed and given a filename determined by the name of the machine,
prefixed with this string.  If a machine is defined as diskless, then no
@file{exports} file will be created for it.  The files contain entries
for directories on the machine that may be exported to clients.

Example: To create the @file{exports} files for each diskfull machine
and place them into the directory @file{exports}:

@example
fsinfo -e exports/ ...
@end example

@node -f FSinfo Option, -h FSinfo Option, -e FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-f} @var{fstab}
@cindex fstab, FSinfo prefix

This defines the prefix for the @file{fstab} files.  The files will only
be created if this prefix is defined.  For each machine defined in the
configuration files, a @file{fstab} file is created with the filename
determined by prefixing this string with the name of the machine.  These
files contain entries for filesystems and partitions to mount at boot
time.

Example, to create the files in the directory @file{fstabs}:

@example
fsinfo -f fstabs/ ...
@end example

@node -h FSinfo Option, -m FSinfo Option, -f FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-h} @var{hostname}
@cindex hostname, FSinfo command line option

Defines the hostname of the destination machine to process for.  If this
is not specified, it defaults to the local machine name, as returned by
@b{gethostname}(2).

Example:

@example
fsinfo -h dylan.doc.ic.ac.uk ...
@end example

@node -m FSinfo Option, -q FSinfo Option, -h FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-m} @var{mount-maps}
@cindex maps, FSinfo command line option

Defines the prefix for the automounter files.  The maps will only be
produced if this prefix is defined.  The mount maps suitable for the
network defined by the configuration files will be placed into files
with names calculated by prefixing this string to the name of each map.

For example, to create the automounter maps and place them in the
directory @file{automaps}:

@example
fsinfo -m automaps/ ...
@end example

@node -q FSinfo Option, -v FSinfo Option, -m FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-q}
@cindex quiet, FSinfo command line option

Selects quiet mode.  @i{FSinfo} suppress the ``running commentary'' and
only outputs any error messages which are generated.

@node -v FSinfo Option, -D-FSinfo Option, -q FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-v}
@cindex verbose, FSinfo command line option

Selects verbose mode.  When this is activated, the program will display
more messages, and display all the information discovered when
performing the semantic analysis phase.  Each verbose message is output
to @file{stdout} on a line starting with a @samp{#} character.

@node -D-FSinfo Option, -I FSinfo Option, -v FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-D} @var{name[=defn]}

Defines a symbol @dfn{name} for the preprocessor when reading the
configuration files.  Equivalent to @code{#define} directive.

@node -I FSinfo Option, -U FSinfo Option, -D-FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-I} @var{directory}

This option is passed into the preprocessor for the configuration files.
It specifies directories in which to find include files

@node -U FSinfo Option, , -I FSinfo Option, FSinfo Command Line Options
@comment  node-name,  next,  previous,  up
@subsection @code{-U} @var{name}

Removes any initial definition of the symbol @dfn{name}.  Inverse of the
@code{-D} option.

@node FSinfo errors, , FSinfo Command Line Options, FSinfo
@comment  node-name,  next,  previous,  up
@section Errors produced by @i{FSinfo}
@cindex FSinfo error messages

The following table documents the errors and warnings which @i{FSinfo} may produce.

@table @t

@item " expected
Occurs if an unescaped newline is found in a quoted string.

@item ambiguous mount: @var{volume} is a replicated filesystem
If several filesystems are declared as having the same volume name, they
will be considered replicated filesystems.  To mount a replicated
filesystem statically, a specific host will need to be named, to say
which particular copy to try and mount, else this error will
result.

@item can't open @var{filename} for writing
Occurs if any errors are encountered when opening an output file.

@item cannot determine localname since volname @var{volume} is not uniquely defined
If a volume is replicated and an attempt is made to mount the filesystem
statically without specifying a local mountpoint, @i{FSinfo} cannot
calculate a mountpoint, as the desired pathname would be
ambiguous.

@item @var{device} has duplicate exportfs data
Produced if the @samp{exportfs} option is used multiple times within the
same branch of a filesystem definition. For example, if you attempt to
set the @samp{exportfs} data at different levels of the mountpoint
directory tree.

@item dump frequency for @var{host}:@var{device} is non-zero
Occurs if @var{device} has its @samp{fstype} declared to be @samp{swap}
or @samp{export} and the @samp{dump} option is set to a value greater
than zero.  Swap devices should not be dumped.

@item duplicate host @var{hostname}!
If a host has more than one definition.

@item end of file within comment
A comment was unterminated before the end of one of the configuration
files.

@item @var{filename}: cannot open for reading
If a file specified on the command line as containing configuration data
could not be opened.

@item @var{filesystem} has a volname but no exportfs data
Occurs when a volume name is declared for a file system, but the string
specifying what machines the filesystem can be exported to is
missing.

@item fs field "@var{field-name}" already set
Occurs when multiple definitions are given for one of the attributes of a
host's filesystem.

@item host field "@var{field-name}" already set
If duplicate definitions are given for any of the fields with a host
definition.

@item @var{host}:@var{device} has more than one mount point
Occurs if the mount option for a host's filesystem specifies multiple
trees at which to place the mountpoint.

@item @var{host}:@var{device} has no mount point
Occurs if the @samp{mount} option is not specified for a host's
filesystem.

@item @var{host}:@var{device} needs field "@var{field-name}"
Occurs when a filesystem is missing a required field. @var{field-name} could
be one of @samp{fstype}, @samp{opts}, @samp{passno} or
@samp{mount}.

@item @var{host}:mount field specified for swap partition
Occurs if a mountpoint is given for a filesystem whose type is declared
to be @samp{swap}.

@item malformed IP dotted quad: @var{address}
If the Internet address of an interface is incorrectly specified.  An
Internet address definition is handled to @b{inet_addr}(3N) to see if it
can cope.  If not, then this message will be displayed.

@item malformed netmask: @var{netmask}
If the netmask cannot be decoded as though it were a hexadecimal number,
then this message will be displayed.  It will typically be caused by
incorrect characters in the @var{netmask} value.

@item mount field "@var{field-name}" already set
Occurs when a static mount has multiple definitions of the same field.

@item mount tree field "@var{field-name}" already set
Occurs when the @var{field-name} is defined more than once during the
definition of a filesystems mountpoint.

@item netif field @var{field-name} already set
Occurs if you attempt to define an attribute of an interface more than
once.

@item network booting requires both root and swap areas
Occurs if a machine has mount declarations for either the root partition
or the swap area, but not both.  You cannot define a machine to only
partially boot via the network.

@item no disk mounts on @var{hostname}
If there are no static mounts, nor local disk mounts specified for a
machine, this message will be displayed.

@item no volname given for @var{host}:@var{device}
Occurs when a filesystem is defined to be mounted on @file{default}, but
no volume name is given for the file system, then the mountpoint cannot
be determined.

@item not allowed '/' in a directory name
Occurs when a pathname with multiple directory elements is specified as
the name for an automounter tree.  A tree should only have one name at
each level.

@item pass number for @var{host}:@var{device} is non-zero
Occurs if @var{device} has its @samp{fstype} declared to be @samp{swap}
or @samp{export} and the @b{fsck}(8) pass number is set. Swap devices should not be
fsck'd.  @xref{FSinfo filesystems fstype}.

@item sub-directory @var{directory} of @var{directory-tree} starts with '/'
Within the filesystem specification for a host, if an element
@var{directory} of the mountpoint begins with a @samp{/} and it is not
the start of the tree.

@item sub-directory of @var{directory-tree} is named "default"
@samp{default} is a keyword used to specify if a mountpoint should be
automatically calculated by @i{FSinfo}.  If you attempt to specify a
directory name as this, it will use the filename of @file{default} but
will produce this warning.

@item unknown \ sequence
Occurs if an unknown escape sequence is found inside a string.  Within a
string, you can give the standard C escape sequences for strings, such
as newlines and tab characters.

@item unknown directory attribute
If an unknown keyword is found while reading the definition of a host's
filesystem mount option.

@item unknown filesystem attribute
Occurs if an unrecognized keyword is used when defining a host's
filesystems.

@item unknown host attribute
Occurs if an unrecognized keyword is used when defining a host.

@item unknown mount attribute
Occurs if an unrecognized keyword is found while parsing the list of
static mounts.

@item unknown volname @var{volume} automounted @i{[} on @i{name} @i{]}
Occurs if @var{volume} is used in a definition of an automount map but the volume
name has not been declared during the host filesystem definitions.

@item volname @var{volume} is unknown
Occurs if an attempt is made to mount or reference a volume name which
has not been declared during the host filesystem definitions.

@item volname @var{volume} not exported from @var{machine}
Occurs if you attempt to mount the volume @var{volume} from a machine
which has not declared itself to have such a filesystem
available.

@end table

@c ################################################################
@node Hlfsd, Assorted Tools, FSinfo, Top
@comment  node-name,  next,  previous,  up
@chapter Hlfsd
@pindex Hlfsd
@cindex Home-Link Filesystem

@i{Hlfsd} is a daemon which implements a filesystem containing a
symbolic link to subdirectory within a user's home directory, depending
on the user which accessed that link.  It was primarily designed to
redirect incoming mail to users' home directories, so that it can be read
from anywhere.  It was designed and implemented by
@email{ezk@@cs.columbia.edu,Erez Zadok} and
@email{dupuy@@cs.columbia.edu,Alexander Dupuy}, at the
@uref{http://www.cs.columbia.edu/,Computer Science Department} of
@uref{http://www.columbia.edu/,Columbia University}.  A
@uref{http://www.cs.columbia.edu/~ezk/research/hlfsd/hlfsd.html,paper}
on @i{Hlfsd} was presented at the Usenix LISA VII conference in 1993.

@i{Hlfsd} operates by mounting itself as an NFS server for the directory
containing @i{linkname}, which defaults to @file{/hlfs/home}.  Lookups
within that directory are handled by @i{Hlfsd}, which uses the
password map to determine how to resolve the lookup.  The directory will
be created if it doesn't already exist.  The symbolic link will be to
the accessing user's home directory, with @i{subdir} appended to it.  If
not specified, @i{subdir} defaults to @file{.hlfsdir}.  This directory
will also be created if it does not already exist.

A @samp{SIGTERM} sent to @i{Hlfsd} will cause it to shutdown.  A @samp{SIGHUP} will
flush the internal caches, and reload the password map.  It will also
close and reopen the log file, to enable the original log file to be
removed or rotated.  A @samp{SIGUSR1} will cause it to dump its internal table
of user IDs and home directories to the file @file{/tmp/hlfsddump}.

@menu
* Introduction to Hlfsd::
* Background to Mail Delivery::
* Using Hlfsd::
@end menu

@c ================================================================
@node Introduction to Hlfsd, Background to Mail Delivery, Hlfsd, Hlfsd
@comment  node-name,  next,  previous,  up
@section Introduction to Hlfsd
@cindex Introduction to Hlfsd
@cindex Hlfsd; introduction

Electronic mail has become one of the major applications for many
computer networks, and use of this service is expected to increase over
time, as networks proliferate and become faster.  Providing a convenient
environment for users to read, compose, and send electronic mail has
become a requirement for systems administrators (SAs).

Widely used methods for handling mail usually require users to be logged
into a designated ``home'' machine, where their mailbox files reside.
Only on that one machine can they read newly arrived mail.  Since users
have to be logged into that system to read their mail, they often find
it convenient to run all of their other processes on that system as
well, including memory and CPU-intensive jobs.  For example, in our
department, we have allocated and configured several multi-processor
servers to handle such demanding CPU/memory applications, but these were
underutilized, in large part due to the inconvenience of not being able
to read mail on those machines.  (No home directories were located on
these designated CPU-servers, since we did not want NFS service for
users' home directories to have to compete with CPU-intensive jobs.  At the
same time, we discouraged users from running demanding applications on
their home machines.)

Many different solutions have been proposed to allow users to read their
mail on any host.  However, all of these solutions fail in one or more
of several ways:

@itemize @bullet

@item
they introduce new single points of failure

@item
they require using different mail transfer agents (MTAs) or user agents
(UAs)

@item
they do not solve the problem for all cases, i.e.  the solution is only
partially successful for a particular environment.

@end itemize

We have designed a simple filesystem, called the @dfn{Home-Link File
System}, to provide the ability to deliver mail to users' home
directories, without modification to mail-related applications. We have
endeavored to make it as stable as possible.  Of great importance to us
was to make sure the HLFS daemon, @file{hlfsd} , would not hang under
any circumstances, and would take the next-best action when faced with
problems.  Compared to alternative methods, @i{Hlfsd} is a stable, more
general solution, and easier to install/use.  In fact, in some ways, we
have even managed to improve the reliability and security of mail
service.

Our server implements a small filesystem containing a symbolic link
to a subdirectory of the invoking user's home directory, and named symbolic
links to users' mailbox files.

The @i{Hlfsd} server finds out the @var{uid} of the process that is
accessing its mount point, and resolves the pathname component @samp{home} as a
symbolic link to a subdirectory within the home directory given by the
@var{uid}'s entry in the password file.  If the @var{gid} of the process
that attempts to access a mailbox file is a special one (called
HLFS_GID), then the server maps the name of the @emph{next} pathname
component directly to the user's mailbox.  This is necessary so that
access to a mailbox file by users other than the owner can succeed.  The
server has safety features in case of failures such as hung filesystems
or home directory filesystems that are inaccessible or full.

On most of our machines, mail gets delivered to the directory
@file{/var/spool/mail}. Many programs, including UAs, depend on that
path.  @i{Hlfsd} creates a directory @file{/mail}, and mounts itself on
top of that directory.  @i{Hlfsd} implements the path name component
called @samp{home}, pointing to a subdirectory of the user's home directory.
We have made @file{/var/spool/mail} a symbolic link to
@file{/mail/home}, so that accessing @file{/var/spool/mail} actually
causes access to a subdirectory within a user's home directory.

The following table shows an example of how resolving the pathname
@file{/var/mail/@i{NAME}} to @file{/users/ezk/.mailspool/@i{NAME}} proceeds.

@multitable {Resolving Component} {Pathname left to resolve} {Value if symbolic link}

@item @b{Resolving Component}
@tab @b{Pathname left to resolve}
@tab @b{Value if symbolic link}

@item @t{/}
@tab @t{var/mail/}@i{NAME}

@item @t{var/}
@tab @t{mail/}@i{NAME}

@item @t{mail}@@
@tab @t{/mail/home/}@i{NAME}
@tab @t{mail}@@ -> @t{/mail/home}

@item @t{/}
@tab @t{mail/home/}@i{NAME}

@item @t{mail/}
@tab @t{home/}@i{NAME}

@item @t{home}@@
@tab @i{NAME}
@tab @t{home}@@ -> @t{/users/ezk/.mailspool}

@item @t{/}
@tab @t{users/ezk/.mailspool/}@i{NAME}

@item @t{users/}
@tab @t{ezk/.mailspool/}@i{NAME}

@item @t{ezk/}
@tab @t{.mailspool/}@i{NAME}

@item @t{.mailspool/}
@tab @i{NAME}

@item @i{NAME}

@end multitable

@c ================================================================
@node Background to Mail Delivery, Using Hlfsd, Introduction to Hlfsd, Hlfsd
@comment  node-name,  next,  previous,  up
@section Background to Mail Delivery
@cindex Background to Mail Delivery
@cindex Hlfsd; background

This section provides an in-depth discussion of why available methods
for delivering mail to home directories are not as good as the one used
by @i{Hlfsd}.

@menu
* Single-Host Mail Spool Directory::
* Centralized Mail Spool Directory::
* Distributed Mail Spool Service::
* Why Deliver Into the Home Directory?::
@end menu

@c ----------------------------------------------------------------
@node Single-Host Mail Spool Directory, Centralized Mail Spool Directory, Background to Mail Delivery, Background to Mail Delivery
@comment  node-name,  next,  previous,  up
@subsection Single-Host Mail Spool Directory
@cindex Single-Host Mail Spool Directory

The most common method for mail delivery is for mail to be appended to a
mailbox file in a standard spool directory on the designated ``mail
home'' machine of the user. The greatest advantage of this method is
that it is the default method most vendors provide with their systems,
thus very little (if any) configuration is required on the SA's part.
All they need to set up are mail aliases directing mail to the host on
which the user's mailbox file is assigned.  (Otherwise, mail is
delivered locally, and users find mailboxes on many machines.)

As users become more sophisticated, and aided by windowing systems, they
find themselves logging in on multiple hosts at once, performing several
tasks concurrently.  They ask to be able to read their mail on any host
on the network, not just the one designated as their ``mail home''.

@c ----------------------------------------------------------------
@node Centralized Mail Spool Directory, Distributed Mail Spool Service, Single-Host Mail Spool Directory, Background to Mail Delivery
@comment  node-name,  next,  previous,  up
@subsection Centralized Mail Spool Directory
@cindex Centralized Mail Spool Directory

A popular method for providing mail readability from any host is to have
all mail delivered to a mail spool directory on a designated
``mail-server'' which is exported via NFS to all of the hosts on the
network.  Configuring such a system is relatively easy.  On most
systems, the bulk of the work is a one-time addition to one or two
configuration files in @file{/etc}.  The file-server's spool directory
is then hard-mounted across every machine on the local network.  In
small environments with only a handful of hosts this can be an
acceptable solution.  In our department, with a couple of hundred active
hosts and thousands of mail messages processed daily, this was deemed
completely unacceptable, as it introduced several types of problems:

@table @b

@item Scalability and Performance

As more and more machines get added to the network, more mail traffic
has to go over NFS to and from the mail-server. Users like to run
mail-watchers, and read their mail often. The stress on the shared
infrastructure increases with every user and host added; loads on the
mail server would most certainly be high since all mail delivery goes
through that one machine.@footnote{ Delivery via NFS-mounted filesystems
may require usage of @samp{rpc.lockd} and @samp{rpc.statd} to provide
distributed file-locking, both of which are widely regarded as unstable
and unreliable.  Furthermore, this will degrade performance, as local
processes as well as remote @samp{nfsd} processes are kept busy.}  This
leads to lower reliability and performance.  To reduce the number of
concurrent connections between clients and the server host, some SAs
have resorted to automounting the mail-spool directory.  But this
solution only makes things worse: since users often run mail watchers,
and many popular applications such as @samp{trn}, @samp{emacs},
@samp{csh} or @samp{ksh} check periodically for new mail, the
automounted directory would be effectively permanently mounted.  If it
gets unmounted automatically by the automounter program, it is most
likely to get mounted shortly afterwards, consuming more I/O resources
by the constant cycle of mount and umount calls.

@item Reliability

The mail-server host and its network connectivity must be very reliable.
Worse, since the spool directory has to be hard-mounted,@footnote{No SA
in their right minds would soft-mount read/write partitions --- the
chances for data loss are too great.} many processes which access the
spool directory (various shells, @samp{login}, @samp{emacs}, etc.)
would be hung as long as connectivity to the mail-server is severed. To
improve reliability, SAs may choose to backup the mail-server's spool
partition several times a day.  This may make things worse since reading
or delivering mail while backups are in progress may cause backups to be
inconsistent; more backups consume more backup-media resources, and
increase the load on the mail-server host.

@end table

@c ----------------------------------------------------------------
@node Distributed Mail Spool Service, Why Deliver Into the Home Directory?, Centralized Mail Spool Directory, Background to Mail Delivery
@comment  node-name,  next,  previous,  up
@subsection Distributed Mail Spool Service
@cindex Distributed Mail Spool Service

Despite the existence of a few systems that support delivery to users'
home directories, mail delivery to home directories hasn't caught on.
We believe the main reason is that there are too many programs that
``know'' where mailbox files reside.  Besides the obvious (the delivery
program @file{/bin/mail} and mail readers like @file{/usr/ucb/Mail},
@samp{mush}, @samp{mm}, etc.), other programs that know mailbox location
are login, from, almost every shell, @samp{xbiff}, @samp{xmailbox}, and
even some programs not directly related to mail, such as @samp{emacs}
and @samp{trn}.  Although some of these programs can be configured to
look in different directories with the use of environment variables and
other resources, many of them cannot.  The overall porting work is
significant.

Other methods that have yet to catch on require the use of a special
mail-reading server, such as IMAP or POP.  The main disadvantage of
these systems is that UAs need to be modified to use these services ---
a long and involved task.  That is why they are not popular at this
time.

Several other ideas have been proposed and even used in various
environments.  None of them is robust.  They are mostly very
specialized, inflexible, and do not extend to the general case.  Some of
the ideas are plain bad, potentially leading to lost or corrupt mail:

@table @b

@item automounters

Using an automounter such as @i{Amd} to provide a set of symbolic links
from the normal spool directory to user home directories is not
sufficient.  UAs rename, unlink, and recreate the mailbox as a regular
file, therefore it must be a real file, not a symbolic link.
Furthermore, it must reside in a real directory which is writable by the
UAs and MTAs.  This method may also require populating
@file{/var/spool/mail} with symbolic links and making sure they are
updated.  Making @i{Amd} manage that directory directly fails, since
many various lock files need to be managed as well.  Also, @i{Amd} does
not provide all of the NFS operations which are required to write mail
such as write, create, remove, and unlink.

@item @code{$MAIL}

Setting this variable to an automounted directory pointing to the user's
mail spool host only solves the problem for those programs which know
and use @code{$MAIL}.  Many programs don't, therefore this solution is partial
and of limited flexibility.  Also, it requires the SAs or the users to
set it themselves --- an added level of inconvenience and possible
failures.

@item @t{/bin/mail}

Using a different mail delivery agent could be the solution.  One such
example is @samp{hdmail}.  However, @samp{hdmail} still requires
modifying all UAs, the MTA's configuration, installing new daemons, and
changing login scripts.  This makes the system less upgradable or
compatible with others, and adds one more complicated system for SAs to
deal with.  It is not a complete solution because it still requires each
user have their @code{$MAIL} variable setup correctly, and that every program
use this variable.

@end table

@c ----------------------------------------------------------------
@node Why Deliver Into the Home Directory?, , Distributed Mail Spool Service, Background to Mail Delivery
@comment  node-name,  next,  previous,  up
@subsection Why Deliver Into the Home Directory?
@cindex Why Deliver Into the Home Directory?
@cindex Hlfsd; Why Deliver Into the Home Directory?

There are several major reasons why SAs might want to deliver mail
directly into the users' home directories:

@table @b

@item Location

Many mail readers need to move mail from the spool directory to the
user's home directory.  It speeds up this operation if the two are on
the same filesystem.  If for some reason the user's home directory is
inaccessible, it isn't that useful to be able to read mail, since there
is no place to move it to.  In some cases, trying to move mail to a
non-existent or hung filesystem may result in mail loss.

@item Distribution

Having all mail spool directories spread among the many more filesystems
minimizes the chances that complete environments will grind to a halt
when a single server is down.  It does increase the chance that there
will be someone who is not able to read their mail when a machine is
down, but that is usually preferred to having no one be able to read
their mail because a centralized mail server is down.  The problem of
losing some mail due to the (presumably) higher chances that a user's
machine is down is minimized in HLFS.

@item Security

Delivering mail to users' home directories has another advantage ---
enhanced security and privacy.  Since a shared system mail spool
directory has to be world-readable and searchable, any user can see
whether other users have mail, when they last received new mail, or when
they last read their mail.  Programs such as @samp{finger} display this
information, which some consider an infringement of privacy.  While it
is possible to disable this feature of @samp{finger} so that remote
users cannot see a mailbox file's status, this doesn't prevent local
users from getting the information.  Furthermore, there are more
programs which make use of this information.  In shared environments,
disabling such programs has to be done on a system-wide basis, but with
mail delivered to users' home directories, users less concerned with
privacy who do want to let others know when they last received or read
mail can easily do so using file protection bits.

@c Lastly, on systems that do not export their NFS filesystem with
@c @t{anon=0}, superusers are less likely to snoop around others' mail, as
@c they become ``nobodies'' across NFS.

@end table

In summary, delivering mail to home directories provides users the
functionality sought, and also avoids most of the problems just
discussed.

@c ================================================================
@node Using Hlfsd, , Background to Mail Delivery, Hlfsd
@comment  node-name,  next,  previous,  up
@section Using Hlfsd
@cindex Using Hlfsd
@cindex Hlfsd; using

@menu
* Controlling Hlfsd::
* Hlfsd Options::
* Hlfsd Files::
@end menu

@c ----------------------------------------------------------------
@node Controlling Hlfsd, Hlfsd Options, Using Hlfsd, Using Hlfsd
@comment  node-name,  next,  previous,  up
@subsection Controlling Hlfsd
@cindex Controlling Hlfsd
@cindex Hlfsd; controlling
@pindex ctl-hlfsd

Much the same way @i{Amd} is controlled by @file{ctl-amd}, so does
@i{Hlfsd} get controlled by the @file{ctl-hlfsd} script:

@table @t

@item ctl-hlfsd start
Start a new @i{Hlfsd}.

@item ctl-hlfsd stop
Stop a running @i{Hlfsd}.

@item ctl-hlfsd restart
Stop a running @i{Hlfsd}, wait for 10 seconds, and then start a new
one.  It is hoped that within 10 seconds, the previously running
@i{Hlfsd} terminate properly; otherwise, starting a second one could
cause system lockup.

@end table

For example, on our systems, we start @i{Hlfsd} within @file{ctl-hlfsd}
as follows on Solaris 2 systems:

@example
hlfsd -a /var/alt_mail -x all -l /var/log/hlfsd /mail/home .mailspool
@end example

The directory @file{/var/alt_mail} is a directory in the root partition
where alternate mail will be delivered into, when it cannot be delivered
into the user's home directory.

Normal mail gets delivered into @file{/var/mail}, but on our systems,
that is a symbolic link to @file{/mail/home}.  @file{/mail} is managed
by @i{Hlfsd}, which creates a dynamic symlink named @samp{home},
pointing to the subdirectory @file{.mailspool} @emph{within} the
accessing user's home directory.  This results in mail which normally
should go to @file{/var/mail/@code{$USER}}, to go to
@file{@code{$HOME}/.mailspool/@code{$USER}}.

@i{Hlfsd} does not create the @file{/var/mail} symlink.  This needs to
be created (manually) once on each host, by the system administrators,
as follows:

@example
mv /var/mail /var/alt_mail
ln -s /mail/home /var/mail
@end example

@c ----------------------------------------------------------------
@node Hlfsd Options, Hlfsd Files, Controlling Hlfsd, Using Hlfsd
@comment  node-name,  next,  previous,  up
@subsection Hlfsd Options
@cindex Hlfsd Options
@cindex Hlfsd; Options

@table @t

@item -a @var{alt_dir}
Alternate directory.  The name of the directory to which the symbolic
link returned by @i{Hlfsd} will point, if it cannot access the home
directory of the user.  This defaults to @file{/var/hlfs}.  This
directory will be created if it doesn't exist.  It is expected that
either users will read these files, or the system administrators will
run a script to resend this ``lost mail'' to its owner.

@item -c @var{cache-interval}
Caching interval.  @i{Hlfsd} will cache the validity of home directories
for this interval, in seconds.  Entries which have been verified within
the last @var{cache-interval} seconds will not be verified again, since
the operation could be expensive, and the entries are most likely still
valid.  After the interval has expired, @i{Hlfsd} will re-verify the
validity of the user's home directory, and reset the cache time-counter.
The default value for @var{cache-interval} is 300 seconds (5 minutes).

@item -f
Force fast startup.  This option tells @i{Hlfsd} to skip startup-time
consistency checks such as existence of mount directory, alternate spool
directory, symlink to be hidden under the mount directory, their
permissions and validity.

@item -g @var{group}
Set the special group HLFS_GID to @var{group}.  Programs such as
@file{/usr/ucb/from} or @file{/usr/sbin/in.comsat}, which access the
mailboxes of other users, must be setgid @samp{HLFS_GID} to work properly.  The
default group is @samp{hlfs}.  If no group is provided, and there is no
group @samp{hlfs}, this feature is disabled.

@item -h
Help.  Print a brief help message, and exit.

@item -i @var{reload-interval}
Map-reloading interval.  Each @var{reload-interval} seconds, @i{Hlfsd}
will reload the password map.  @i{Hlfsd} needs the password map for the
UIDs and home directory pathnames.  @i{Hlfsd} schedules a @samp{SIGALRM} to
reload the password maps.  A @samp{SIGHUP} sent to @i{Hlfsd} will force it to
reload the maps immediately.  The default value for
@var{reload-interval} is 900 seconds (15 minutes.)

@item -l @var{logfile}
Specify a log file to which @i{Hlfsd} will record events.  If
@var{logfile} is the string @samp{syslog} then the log messages will be
sent to the system log daemon by @b{syslog}(3), using the @samp{LOG_DAEMON}
facility.  This is also the default.

@item -n
No verify.  @i{Hlfsd} will not verify the validity of the symbolic link
it will be returning, or that the user's home directory contains
sufficient disk-space for spooling.  This can speed up @i{Hlfsd} at the
cost of possibly returning symbolic links to home directories which are
not currently accessible or are full.  By default, @i{Hlfsd} validates
the symbolic-link in the background.  The @code{-n} option overrides the
meaning of the @code{-c} option, since no caching is necessary.

@item -o @var{mount-options}
Mount options which @i{Hlfsd} will use to mount itself on top of
@var{dirname}.  By default, @var{mount-options} is set to @samp{ro}.  If
the system supports symbolic-link caching, default options are set
to @samp{ro,nocache}.

@item -p
Print PID.  Outputs the process-id of @i{Hlfsd} to standard output where
it can be saved into a file.

@item -v
Version.  Displays version information to standard error.

@item -x @var{log-options}
Specify run-time logging options.  The options are a comma separated
list chosen from: @samp{fatal}, @samp{error}, @samp{user}, @samp{warn}, @samp{info}, @samp{map}, @samp{stats}, @samp{all}.

@item -C
Force @i{Hlfsd} to run on systems that cannot turn off the NFS
attribute-cache.  Use of this option on those systems is discouraged, as
it may result in loss or misdelivery of mail.  The option is ignored on
systems that can turn off the attribute-cache.

@item -D @var{log-options}
Select from a variety of debugging options.  Prefixing an option with
the string @samp{no} reverses the effect of that option.  Options are
cumulative.  The most useful option is @samp{all}.  Since this option is
only used for debugging other options are not documented here.  A fuller
description is available in the program source.  A @samp{SIGUSR1} sent
to @i{Hlfsd} will cause it to dump its internal password map to the file
@file{/usr/tmp/hlfsd.dump.XXXXXX}, where @samp{XXXXXX} will be replaced
by a random string generated by @b{mktemp}(3) or (the more secure)
@b{mkstemp}(3).

@item -P @var{password-file}
Read the user-name, user-id, and home directory information from the
file @var{password-file}.  Normally, @i{Hlfsd} will use @b{getpwent}(3)
to read the password database.  This option allows you to override the
default database, and is useful if you want to map users' mail files to
a directory other than their home directory.  Only the username, uid,
and home-directory fields of the file @var{password-file} are read and
checked.  All other fields are ignored.  The file @var{password-file}
must otherwise be compliant with Unix Version 7 colon-delimited format
@b{passwd}(4).

@end table

@c ----------------------------------------------------------------
@node Hlfsd Files, , Hlfsd Options, Using Hlfsd
@comment  node-name,  next,  previous,  up
@subsection Hlfsd Files
@cindex Hlfsd Files
@cindex Hlfsd; Files

The following files are used by @i{Hlfsd}:

@table @file

@item /hlfs
directory under which @i{Hlfsd} mounts itself and manages the symbolic
link @file{home}.

@item .hlfsdir
default sub-directory in the user's home directory, to which the
@file{home} symbolic link returned by @i{Hlfsd} points.

@item /var/hlfs
directory to which @file{home} symbolic link returned by @i{Hlfsd}
points if it is unable to verify the that user's home directory is
accessible.

@end table

For discussion on other files used by @i{Hlfsd}, see @ref{lostaltmail} and
@ref{lostaltmail.conf-sample}.

@c ################################################################
@node Assorted Tools, Examples, Hlfsd, Top
@comment  node-name,  next,  previous,  up
@chapter Assorted Tools
@cindex  Assorted Tools

The following are additional utilities and scripts included with
am-utils, and get installed.

@menu
* am-eject::
* amd.conf-sample::
* amd2ldif::
* amd2sun::
* ctl-amd::
* ctl-hlfsd::
* expn::
* fix-amd-map::
* fixmount::
* fixrmtab::
* lostaltmail::
* lostaltmail.conf-sample::
* mk-amd-map::
* pawd::
* wait4amd::
* wait4amd2die::
* wire-test::
@end menu

@c ----------------------------------------------------------------
@node am-eject, amd.conf-sample, Assorted Tools, Assorted Tools
@comment  node-name,  next,  previous,  up
@section am-eject
@pindex am-eject

A shell script unmounts a floppy or CD-ROM that is automounted, and
then attempts to eject the removable device.

@c ----------------------------------------------------------------
@node amd.conf-sample, amd2ldif, am-eject, Assorted Tools
@comment  node-name,  next,  previous,  up
@section amd.conf-sample
@pindex amd.conf-sample

A sample @i{Amd} configuration file. @xref{Amd Configuration File}.

@c ----------------------------------------------------------------
@node amd2ldif, amd2sun, amd.conf-sample, Assorted Tools
@comment  node-name,  next,  previous,  up
@section amd2ldif
@pindex amd2ldif

A script to convert @i{Amd} maps to LDAP input files.  Use it as follows:

@example
amd2ldif @i{mapname} @i{base} < @i{amd.mapfile} > @i{mapfile.ldif}
@end example

@c ----------------------------------------------------------------
@node amd2sun, ctl-amd, amd2ldif, Assorted Tools
@comment  node-name,  next,  previous,  up
@section amd2sun
@pindex amd2sun

A script to convert @i{Amd} maps to Sun Automounter maps.  Use it as
follows

@example
amd2sun < @i{amd.mapfile} > @i{auto_mapfile}
@end example

@c ----------------------------------------------------------------
@node ctl-amd, ctl-hlfsd, amd2sun, Assorted Tools
@comment  node-name,  next,  previous,  up
@section ctl-amd
@pindex ctl-amd

A script to start, stop, or restart @i{Amd}.  Use it as follows:

@table @t
@item ctl-amd start
Start a new @i{Amd} process.
@item ctl-amd stop
Stop the running @i{Amd}.
@item ctl-amd restart
Stop the running @i{Amd} (if any), safely wait for it to terminate, and
then start a new process --- only if the previous one died cleanly.
@end table

@xref{Run-time Administration} for more details.

@c ----------------------------------------------------------------
@node ctl-hlfsd, expn, ctl-amd, Assorted Tools
@comment  node-name,  next,  previous,  up
@section ctl-hlfsd
@pindex ctl-hlfsd

A script for controlling @i{Hlfsd}, much the same way @file{ctl-amd}
controls @i{Amd}.  Use it as follows:

@table @t
@item ctl-hlfsd start
Start a new @i{Hlfsd} process.
@item ctl-hlfsd stop
Stop the running @i{Hlfsd}.
@item ctl-hlfsd restart
Stop the running @i{Hlfsd} (if any), wait for 10 seconds for it to
terminate, and then start a new process --- only if the previous one
died cleanly.
@end table

@xref{Hlfsd} for more details.

@c ----------------------------------------------------------------
@node expn, fix-amd-map, ctl-hlfsd, Assorted Tools
@comment  node-name,  next,  previous,  up
@section expn
@pindex expn

A script to expand email addresses into their full name.  It is
generally useful when using with the @file{lostaltmail} script, but is a
useful tools otherwise.

@example
$ expn -v ezk@@cs.columbia.edu
ezk@@cs.columbia.edu ->
        ezk@@shekel.mcl.cs.columbia.edu
ezk@@shekel.mcl.cs.columbia.edu ->
        Erez Zadok <"| /usr/local/mh/lib/slocal -user ezk || exit 75>
        Erez Zadok <\ezk>
        Erez Zadok </u/zing/ezk/.mailspool/backup>
@end example

@c ----------------------------------------------------------------
@node fix-amd-map, fixmount, expn, Assorted Tools
@comment  node-name,  next,  previous,  up
@section fix-amd-map
@pindex fix-amd-map

Am-utils changed some of the syntax and default values of some
variables.  For example, the default value for @samp{$@{os@}} for
Solaris 2.x (aka SunOS 5.x) systems used to be @samp{sos5}, it is now
more automatically generated from @file{config.guess} and its value is
@samp{sunos5}.

This script converts older @i{Amd} maps to new ones.  Use it as follows:

@example
fix-amd-map < @i{old.map} > @i{new.map}
@end example

@c ----------------------------------------------------------------
@node fixmount, fixrmtab, fix-amd-map, Assorted Tools
@comment  node-name,  next,  previous,  up
@section fixmount
@pindex fixmount

@samp{fixmount} is a variant of @b{showmount}(8) that can delete bogus
mount entries in remote @b{mountd}(8) daemons.  This is useful to
cleanup otherwise ever-accumulating ``junk''.  Use it for example:

@example
fixmount -r @i{host}
@end example

See the online manual page for @samp{fixmount} for more details of its
usage.

@c ----------------------------------------------------------------
@node fixrmtab, lostaltmail, fixmount, Assorted Tools
@comment  node-name,  next,  previous,  up
@section fixrmtab
@pindex fixrmtab

A script to invalidate @file{/etc/rmtab} entries for hosts named.  Also
restart mountd for changes to take effect.  Use it for example:

@example
fixrmtab @i{host1} @i{host2} @i{...}
@end example

@c ----------------------------------------------------------------
@node lostaltmail, lostaltmail.conf-sample, fixrmtab, Assorted Tools
@comment  node-name,  next,  previous,  up
@section lostaltmail
@pindex lostaltmail

A script used with @i{Hlfsd} to resend any ``lost'' mail.  @i{Hlfsd}
redirects mail which cannot be written into the user's home directory to
an alternate directory.  This is useful to continue delivering mail,
even if the user's file system was unavailable, full, or over quota.
But, the mail which gets delivered to  the alternate directory needs to
be resent to its respective users.  This is what the @samp{lostaltmail}
script does.

Use it as follows:

@example
lostaltmail
@end example

This script needs a configuration file @samp{lostaltmail.conf} set up
with the right parameters to properly work.  @xref{Hlfsd} for more
details.

@c ----------------------------------------------------------------
@node lostaltmail.conf-sample, mk-amd-map, lostaltmail, Assorted Tools
@comment  node-name,  next,  previous,  up
@section lostaltmail.conf-sample
@pindex lostaltmail.conf-sample
@cindex lostaltmail; configuration file

This is a text file with configuration parameters needed for the
@samp{lostaltmail} script.  The script includes comments explaining each
of the configuration variables.  See it for more information.  Also
@pxref{Hlfsd} for general information.

@c ----------------------------------------------------------------
@node mk-amd-map, pawd, lostaltmail.conf-sample, Assorted Tools
@comment  node-name,  next,  previous,  up
@section mk-amd-map
@pindex mk-amd-map

This program converts a normal @i{Amd} map file into an ndbm database
with the same prefix as the named file.  Use it as follows:

@example
mk-amd-map @i{mapname}
@end example

@c ----------------------------------------------------------------
@node pawd, wait4amd, mk-amd-map, Assorted Tools
@comment  node-name,  next,  previous,  up
@section pawd
@pindex pawd

@i{Pawd} is used to print the current working directory, adjusted to
reflect proper paths that can be reused to go through the automounter
for the shortest possible path.  In particular, the path printed back
does not include any of @i{Amd}'s local mount points.  Using them is
unsafe, because @i{Amd} may unmount managed file systems from the mount
points, and thus including them in paths may not always find the files
within.

Without any arguments, @i{Pawd} will print the automounter adjusted
current working directory.  With any number of arguments, it will print
the adjusted path of each one of the arguments.

@c ----------------------------------------------------------------
@node wait4amd, wait4amd2die, pawd, Assorted Tools
@comment  node-name,  next,  previous,  up
@section wait4amd
@pindex wait4amd

A script to wait for @i{Amd} to start on a particular host before
performing an arbitrary command.  The command is executed repeatedly,
with 1 second intervals in between.  You may interrupt the script using
@samp{^C} (or whatever keyboard sequence your terminal's @samp{intr} function
is bound to).

Examples:

@table @t
@item wait4amd saturn amq -p -h saturn
When @i{Amd} is up on host @samp{saturn}, get the process ID of that
running @i{Amd}.
@item wait4amd pluto rlogin pluto
Remote login to host @samp{pluto} when @i{Amd} is up on that host.  It
is generally necessary to wait for @i{Amd} to properly start and
initialize on a remote host before logging in to it, because otherwise
user home directories may not be accessible across the network.
@item wait4amd pluto
A short-hand version of the previous command, since the most useful
reason for this script is to login to a remote host.  I use it very
often when testing out new versions of @i{Amd}, and need to reboot hung
hosts.
@end table

@c ----------------------------------------------------------------
@node wait4amd2die, wire-test, wait4amd, Assorted Tools
@comment  node-name,  next,  previous,  up
@section wait4amd2die
@pindex wait4amd2die

This script is used internally by @samp{ctl-amd} when used to restart
@i{Amd}.  It waits for @i{Amd} to terminate.  If it detected that
@i{Amd} terminated cleanly, this script will return an exist status of
zero.  Otherwise, it will return a non-zero exit status.

The script tests for @i{Amd}'s existence once every 5 seconds, six
times, for a total of 30 seconds.  It will return a zero exist status as
soon as it detects that @i{Amd} dies.

@c ----------------------------------------------------------------
@node wire-test, , wait4amd2die, Assorted Tools
@comment  node-name,  next,  previous,  up
@section wire-test
@pindex wire-test

A simple program to test if some of the most basic networking functions
in am-util's library @file{libamu} work.  It also tests the combination
of NFS protocol and version number that are supported from the current
host, to a remote one.

For example, in this test a machine which only supports NFS Version 2 is
contacting a remote host that can support the same version, but using
both UDP and TCP.  If no host name is specified, @samp{wire-test} will
try @file{localhost}.

@example
$ wire-test moisil
Network name is "mcl-lab-net.cs.columbia.edu"
Network number is "128.59.13"
Network name is "old-net.cs.columbia.edu"
Network number is "128.59.16"
My IP address is 0x7f000001.
NFS Version and protocol tests to host "moisil"...
        testing vers=2, proto="udp" -> found version 2.
        testing vers=3, proto="udp" -> failed!
        testing vers=2, proto="tcp" -> found version 2.
        testing vers=3, proto="tcp" -> failed!
@end example

@c ################################################################
@node Examples, Internals, Assorted Tools, Top
@comment  node-name,  next,  previous,  up
@chapter Examples

@menu
* User Filesystems::
* Home Directories::
* Architecture Sharing::
* Wildcard Names::
* rwho servers::
* /vol::
* /defaults with selectors::
* /tftpboot in a chroot-ed environment::

@end menu

@node User Filesystems, Home Directories, Examples, Examples
@comment  node-name,  next,  previous,  up
@section User Filesystems
@cindex User filesystems
@cindex Mounting user filesystems

With more than one fileserver, the directories most frequently
cross-mounted are those containing user home directories.  A common
convention used at Imperial College is to mount the user disks under
@t{/home/}@i{machine}.

Typically, the @samp{/etc/fstab} file contained a long list of entries
such as:

@example
@i{machine}:/home/@i{machine} /home/@i{machine} nfs ...
@end example

for each fileserver on the network.

There are numerous problems with this system.  The mount list can become
quite large and some of the machines may be down when a system is
booted.  When a new fileserver is installed, @samp{/etc/fstab} must be
updated on every machine, the mount directory created and the filesystem
mounted.

In many environments most people use the same few workstations, but
it is convenient to go to a colleague's machine and access your own
files.  When a server goes down, it can cause a process on a client
machine to hang.  By minimizing the mounted filesystems to only include
those actively being used, there is less chance that a filesystem will
be mounted when a server goes down.

The following is a short extract from a map taken from a research fileserver
at Imperial College.

Note the entry for @samp{localhost} which is used for users such as
the operator (@samp{opr}) who have a home directory on most machine as
@samp{/home/localhost/opr}.

@example
/defaults       opts:=rw,intr,grpid,nosuid
charm           host!=$@{key@};type:=nfs;rhost:=$@{key@};rfs:=/home/$@{key@} \
                host==$@{key@};type:=ufs;dev:=/dev/xd0g
#
...

#
localhost       type:=link;fs:=$@{host@}
...
#
# dylan has two user disks so have a
# top directory in which to mount them.
#
dylan           type:=auto;fs:=$@{map@};pref:=$@{key@}/
#
dylan/dk2       host!=dylan;type:=nfs;rhost:=dylan;rfs:=/home/$@{key@} \
                host==dylan;type:=ufs;dev:=/dev/dsk/2s0
#
dylan/dk5       host!=dylan;type:=nfs;rhost:=dylan;rfs:=/home/$@{key@} \
                host==dylan;type:=ufs;dev:=/dev/dsk/5s0
...
#
toytown         host!=$@{key@};type:=nfs;rhost:=$@{key@};rfs:=/home/$@{key@} \
                host==$@{key@};type:=ufs;dev:=/dev/xy1g
...
#
zebedee         host!=$@{key@};type:=nfs;rhost:=$@{key@};rfs:=/home/$@{key@} \
                host==$@{key@};type:=ufs;dev:=/dev/dsk/1s0
#
# Just for access...
#
gould           type:=auto;fs:=$@{map@};pref:=$@{key@}/
gould/staff     host!=gould;type:=nfs;rhost:=gould;rfs:=/home/$@{key@}
#
gummo           host!=$@{key@};type:=nfs;rhost:=$@{key@};rfs:=/home/$@{key@}
...
@end example

This map is shared by most of the machines listed so on those
systems any of the user disks is accessible via a consistent name.
@i{Amd} is started with the following command

@example
amd /home amd.home
@end example

Note that when mounting a remote filesystem, the @dfn{automounted}
mount point is referenced, so that the filesystem will be mounted if
it is not yet (at the time the remote @samp{mountd} obtains the file handle).

@node Home Directories, Architecture Sharing, User Filesystems, Examples
@comment  node-name,  next,  previous,  up
@section Home Directories
@cindex Home directories
@cindex Example of mounting home directories
@cindex Mount home directories

One convention for home directories is to locate them in @samp{/homes}
so user @samp{jsp}'s home directory is @samp{/homes/jsp}.  With more
than a single fileserver it is convenient to spread user files across
several machines.  All that is required is a mount-map which converts
login names to an automounted directory.

Such a map might be started by the command:

@example
amd /homes amd.homes
@end example

where the map @samp{amd.homes} contained the entries:

@example
/defaults   type:=link   # All the entries are of type:=link
jsp         fs:=/home/charm/jsp
njw         fs:=/home/dylan/dk5/njw
...
phjk        fs:=/home/toytown/ai/phjk
sjv         fs:=/home/ganymede/sjv
@end example

Whenever a login name is accessed in @samp{/homes} a symbolic link
appears pointing to the real location of that user's home directory.  In
this example, @samp{/homes/jsp} would appear to be a symbolic link
pointing to @samp{/home/charm/jsp}.  Of course, @samp{/home} would also
be an automount point.

This system causes an extra level of symbolic links to be used.
Although that turns out to be relatively inexpensive, an alternative is
to directly mount the required filesystems in the @samp{/homes}
map.  The required map is simple, but long, and its creation is best automated.
The entry for @samp{jsp} could be:

@example
jsp   -sublink:=$@{key@};rfs:=/home/charm \
               host==charm;type:=ufs;dev:=/dev/xd0g \
               host!=charm;type:=nfs;rhost:=charm
@end example

This map can become quite big if it contains a large number of entries.
By combining two other features of @i{Amd} it can be greatly simplified.

First the UFS partitions should be mounted under the control of
@samp{/etc/fstab}, taking care that they are mounted in the same place
that @i{Amd} would have automounted them.  In most cases this would be
something like @samp{/a/@dfn{host}/home/@dfn{host}} and
@samp{/etc/fstab} on host @samp{charm} would have a line:@refill

@example
/dev/xy0g /a/charm/home/charm 4.2 rw,nosuid,grpid 1 5
@end example

The map can then be changed to:

@example
/defaults    type:=nfs;sublink:=$@{key@};opts:=rw,intr,nosuid,grpid
jsp          rhost:=charm;rfs:=/home/charm
njw          rhost:=dylan;rfs:=/home/dylan/dk5
...
phjk         rhost:=toytown;rfs:=/home/toytown;sublink:=ai/$@{key@}
sjv          rhost:=ganymede;rfs:=/home/ganymede
@end example

This map operates as usual on a remote machine (@i{ie} @code{$@{host@}}
not equal to @code{$@{rhost@}}).  On the machine where the filesystem is
stored (@i{ie} @code{$@{host@}} equal to @code{$@{rhost@}}), @i{Amd}
will construct a local filesystem mount point which corresponds to the
name of the locally mounted UFS partition.  If @i{Amd} is started with
the @code{-r} option then instead of attempting an NFS mount, @i{Amd} will
simply inherit the UFS mount (@pxref{Inheritance Filesystem}).  If
@code{-r} is not used then a loopback NFS mount will be made.  This type of
mount is known to cause a deadlock on many systems.

@node Architecture Sharing, Wildcard Names, Home Directories, Examples
@comment  node-name,  next,  previous,  up
@section Architecture Sharing
@cindex Architecture sharing
@cindex Sharing a fileserver between architectures
@cindex Architecture dependent volumes

@c %At the moment some of the research machines have sets of software
@c %mounted in @samp{/vol}.  This contains subdirectories for \TeX,
@c %system sources, local sources, prolog libraries and so on.
Often a filesystem will be shared by machines of different architectures.
Separate trees can be maintained for the executable images for each
architecture, but it may be more convenient to have a shared tree,
with distinct subdirectories.

A shared tree might have the following structure on the fileserver (called
@samp{fserver} in the example):

@example
local/tex
local/tex/fonts
local/tex/lib
local/tex/bin
local/tex/bin/sun3
local/tex/bin/sun4
local/tex/bin/hp9000
...
@end example

In this example, the subdirectories of @samp{local/tex/bin} should be
hidden when accessed via the automount point (conventionally @samp{/vol}).
A mount-map for @samp{/vol} to achieve this would look like:

@example
/defaults   sublink:=$@{/key@};rhost:=fserver;type:=link
tex         type:=auto;fs:=$@{map@};pref:=$@{key@}/
tex/fonts   host!=fserver;type:=nfs;rfs:=/vol/tex \
            host==fserver;fs:=/usr/local/tex
tex/lib     host!=fserver;type:=nfs;rfs:=/vol/tex \
            host==fserver;fs:=/usr/local/tex
tex/bin     -sublink:=$@{/key@}/$@{arch@} \
            host!=fserver;type:=nfs;rfs:=/vol/tex \
            host:=fserver;fs:=/usr/local/tex
@end example

When @samp{/vol/tex/bin} is referenced, the current machine architecture
is automatically appended to the path by the @code{$@{sublink@}}
variable.  This means that users can have @samp{/vol/tex/bin} in their
@samp{PATH} without concern for architecture dependencies.

@node Wildcard Names, rwho servers, Architecture Sharing, Examples
@comment  node-name,  next,  previous,  up
@section Wildcard Names & Replicated Servers

By using the wildcard facility, @i{Amd} can @dfn{overlay} an existing
directory with additional entries.
The system files are usually mounted under @samp{/usr}.  If instead,
@i{Amd} is mounted on @samp{/usr}, additional
names can be overlayed to augment or replace names in the ``master'' @samp{/usr}.
A map to do this would have the form:

@example
local  type:=auto;fs:=local-map
share  type:=auto;fs:=share-map
*      -type:=nfs;rfs:=/export/exec/$@{arch@};sublink:="$@{key@}" \
        rhost:=fserv1  rhost:=fserv2  rhost:=fserv3
@end example

Note that the assignment to @code{$@{sublink@}} is surrounded by double
quotes to prevent the incoming key from causing the map to be
misinterpreted.  This map has the effect of directing any access to
@samp{/usr/local} or @samp{/usr/share} to another automount point.

In this example, it is assumed that the @samp{/usr} files are replicated
on three fileservers: @samp{fserv1}, @samp{fserv2} and @samp{fserv3}.
For any references other than to @samp{local} and @samp{share} one of
the servers is used and a symbolic link to
@t{$@{autodir@}/$@{rhost@}/export/exec/$@{arch@}/@i{whatever}} is
returned once an appropriate filesystem has been mounted.@refill

@node rwho servers, /vol, Wildcard Names, Examples
@comment  node-name,  next,  previous,  up
@section @samp{rwho} servers
@cindex rwho servers
@cindex Architecture specific mounts
@cindex Example of architecture specific mounts

The @samp{/usr/spool/rwho} directory is a good candidate for automounting.
For efficiency reasons it is best to capture the rwho data on a small
number of machines and then mount that information onto a large number
of clients.  The data written into the rwho files is byte order dependent
so only servers with the correct byte ordering can be used by a client:

@example
/defaults         type:=nfs
usr/spool/rwho    -byte==little;rfs:=/usr/spool/rwho \
                      rhost:=vaxA  rhost:=vaxB \
                  || -rfs:=/usr/spool/rwho \
                      rhost:=sun4  rhost:=hp300
@end example

@node /vol, /defaults with selectors, rwho servers, Examples
@comment  node-name,  next,  previous,  up
@section @samp{/vol}
@cindex /vol
@cindex Catch-all mount point
@cindex Generic volume name

@samp{/vol} is used as a catch-all for volumes which do not have other
conventional names.

Below is part of the @samp{/vol} map for the domain @samp{doc.ic.ac.uk}.
The @samp{r+d} tree is used for new or experimental software that needs
to be available everywhere without installing it on all the fileservers.
Users wishing to try out the new software then simply include
@samp{/vol/r+d/@{bin,ucb@}} in their path.@refill

The main tree resides on one host @samp{gould.doc.ic.ac.uk}, which has
different @samp{bin}, @samp{etc}, @samp{lib} and @samp{ucb}
sub-directories for each machine architecture.  For example,
@samp{/vol/r+d/bin} for a Sun-4 would be stored in the sub-directory
@samp{bin/sun4} of the filesystem @samp{/usr/r+d}.  When it was accessed
a symbolic link pointing to @samp{/a/gould/usr/r+d/bin/sun4} would be
returned.@refill

@example
/defaults    type:=nfs;opts:=rw,grpid,nosuid,intr,soft
wp           -opts:=rw,grpid,nosuid;rhost:=charm \
             host==charm;type:=link;fs:=/usr/local/wp \
             host!=charm;type:=nfs;rfs:=/vol/wp
...
#
src          -opts:=rw,grpid,nosuid;rhost:=charm \
             host==charm;type:=link;fs:=/usr/src \
             host!=charm;type:=nfs;rfs:=/vol/src
#
r+d          type:=auto;fs:=$@{map@};pref:=r+d/
# per architecture bin,etc,lib&ucb...
r+d/bin      rhost:=gould.doc.ic.ac.uk;rfs:=/usr/r+d;sublink:=$@{/key@}/$@{arch@}
r+d/etc      rhost:=gould.doc.ic.ac.uk;rfs:=/usr/r+d;sublink:=$@{/key@}/$@{arch@}
r+d/include  rhost:=gould.doc.ic.ac.uk;rfs:=/usr/r+d;sublink:=$@{/key@}
r+d/lib      rhost:=gould.doc.ic.ac.uk;rfs:=/usr/r+d;sublink:=$@{/key@}/$@{arch@}
r+d/man      rhost:=gould.doc.ic.ac.uk;rfs:=/usr/r+d;sublink:=$@{/key@}
r+d/src      rhost:=gould.doc.ic.ac.uk;rfs:=/usr/r+d;sublink:=$@{/key@}
r+d/ucb      rhost:=gould.doc.ic.ac.uk;rfs:=/usr/r+d;sublink:=$@{/key@}/$@{arch@}
# hades pictures
pictures     -opts:=rw,grpid,nosuid;rhost:=thpfs \
             host==thpfs;type:=link;fs:=/nbsd/pictures \
             host!=thpfs;type:=nfs;rfs:=/nbsd;sublink:=pictures
# hades tools
hades        -opts:=rw,grpid,nosuid;rhost:=thpfs \
             host==thpfs;type:=link;fs:=/nbsd/hades \
             host!=thpfs;type:=nfs;rfs:=/nbsd;sublink:=hades
# bsd tools for hp.
bsd          -opts:=rw,grpid,nosuid;arch==hp9000;rhost:=thpfs \
             host==thpfs;type:=link;fs:=/nbsd/bsd \
             host!=thpfs;type:=nfs;rfs:=/nbsd;sublink:=bsd
@end example

@node /defaults with selectors, /tftpboot in a chroot-ed environment, /vol, Examples
@comment  node-name,  next,  previous,  up
@section @samp{/defaults} with selectors
@cindex /defaults with selectors
@cindex selectors on default

It is sometimes useful to have different defaults for a given map.  To
achieve this, the @samp{/defaults} entry must be able to process normal
selectors.  This feature is turned on by setting
@samp{selectors_on_default = yes} in the @file{amd.conf} file.
@xref{selectors_on_default Parameter}.

In this example, I set different default NFS mount options for hosts
which are running over a slower network link.  By setting a smaller size
for the NFS read and write buffer sizes, you can greatly improve remote
file service performance.

@example
/defaults \
  wire==slip-net;opts:=rw,intr,rsize=1024,wsize=1024,timeo=20,retrans=10 \
  wire!=slip-net;opts:=rw,intr
@end example

@node /tftpboot in a chroot-ed environment, , /defaults with selectors, Examples
@comment  node-name,  next,  previous,  up
@section @samp{/tftpboot} in a chroot-ed environment
@cindex /tftpboot in a chroot-ed environment
@cindex chroot: /tftpboot example

In this complex example, we attempt to run an @i{Amd} process
@emph{inside} a chroot-ed environment.  @samp{tftpd} (Trivial FTP) is
used to trivially retrieve files used to boot X-Terminals, Network
Printers, Network routers, diskless workstations, and other such
devices.  For security reasons, @samp{tftpd} (and also @samp{ftpd})
processes are run using the @b{chroot}(2) system call.  This provides an
environment for these processes, where access to any files outside the
directory where the chroot-ed process runs is denied.

For example, if you start @samp{tftpd} on your system with

@example
chroot /tftpboot /usr/sbin/tftpd
@end example

@noindent
then the @samp{tftpd} process will not be able to access any files
outside @file{/tftpboot}.  This ensures that no one can retrieve files
such as @file{/etc/passwd} and run password crackers on it.

Since the TFTP service works by broadcast, it is necessary to have at
least one TFTP server running on each subnet.  If you have lots of files
that you need to make available for @samp{tftp}, and many subnets, it
could take significant amounts of disk space on each host serving them.

A solution we implemented at Columbia University was to have every host
run @samp{tftpd}, but have those servers retrieve the boot files from
two replicated servers.  Those replicated servers have special
partitions dedicated to the many network boot files.

We start @i{Amd} as follows:

@example
amd /tftpboot/.amd amd.tftpboot
@end example

That is, @i{Amd} is serving the directory @file{/tftpboot/.amd}.  The
@samp{tftp} server runs inside @file{/tftpboot} and is chroot-ed in that
directory too.  The @file{amd.tftpboot} map looks like:

@example
#
# Amd /tftpboot directory -> host map
#

/defaults  opts:=nosuid,ro,intr,soft;fs:=/tftpboot/import;type:=nfs

tp         host==lol;rfs:=/n/lol/import/tftpboot;type:=lofs \
           host==ober;rfs:=/n/ober/misc/win/tftpboot;type:=lofs \
           rhost:=ober;rfs:=/n/ober/misc/win/tftpboot \
           rhost:=lol;rfs:=/n/lol/import/tftpboot
@end example

To help understand this example, I list a few of the file entries that
are created inside @file{/tftpboot}:

@example
$ ls -la /tftpboot
dr-xr-xr-x   2 root   512 Aug 30 23:11 .amd
drwxrwsr-x  12 root   512 Aug 30 08:00 import
lrwxrwxrwx   1 root    33 Feb 27  1997 adminpr.cfg -> ./.amd/tp/hplj/adminpr.cfg
lrwxrwxrwx   1 root    22 Dec  5  1996 tekxp -> ./.amd/tp/xterms/tekxp
lrwxrwxrwx   1 root     1 Dec  5  1996 tftpboot -> .
@end example

Here is an explanation of each of the entries listed above:

@table @code

@item .amd
This is the @i{Amd} mount point.  Note that you do not need to run a
separate @i{Amd} process for the TFTP service.  The @b{chroot}(2) system
call only protects against file access, but the same process can still
serve files and directories inside and outside the chroot-ed
environment, because @i{Amd} itself was not run in chroot-ed mode.

@item import
This is the mount point where @i{Amd} will mount the directories
containing the boot files.  The map is designed so that remote
directories will be NFS mounted (even if they are already mounted
elsewhere), and local directories are loopback mounted (since they are
not accessible outside the chroot-ed @file{/tftpboot} directory).

@item adminpr.cfg
@itemx tekxp
Two manually created symbolic links to directories @emph{inside} the
@i{Amd}-managed directory.  The crossing of the component @file{tp} will
cause @i{Amd} to automount one of the remote replicas.  Once crossed,
access to files inside proceeds as usual.  The @samp{adminpr.cfg} is a
configuration file for an HP Laser-Jet 4si printer, and the @samp{tekxp}
is a directory for Tektronix X-Terminal boot files.

@item tftpboot
This innocent looking symlink is important.  Usually, when devices boot
via the TFTP service, they perform the @samp{get file} command to
retrieve @var{file}.  However, some devices assume that @samp{tftpd}
does not run in a chroot-ed environment, but rather ``unprotected'', and
thus use a full pathname for files to retrieve, as in @samp{get
/tftpboot/file}.  This symlink effectively strips out the leading
@file{/tftpboot/}.

@end table

@c ################################################################
@node Internals, Acknowledgments & Trademarks, Examples, Top
@comment  node-name,  next,  previous,  up
@chapter Internals

Note that there are more error and logging messages possible than are
listed here.  Most of them are self-explanatory.  Refer to the program
sources for more details on the rest.

@menu
* Log Messages::
@end menu

@node Log Messages, , Internals, Internals
@comment  node-name,  next,  previous,  up
@section Log Messages

In the following sections a brief explanation is given of some of the
log messages made by @i{Amd}.  Where the message is in @samp{typewriter}
font, it corresponds exactly to the message produced by @i{Amd}.  Words
in @dfn{italic} are replaced by an appropriate string.  Variables,
@code{$@{@i{var}@}}, indicate that the value of the appropriate variable is
output.

Log messages are either sent directly to a file,
or logged via the @b{syslog}(3) mechanism.  @xref{log_file Parameter}
In either case, entries in the file are of the form:
@example
@i{date-string}  @i{hostname} @t{amd[}@i{pid}@t{]}  @i{message}
@end example

@menu
* Fatal errors::
* Info messages::
@end menu

@node Fatal errors, Info messages, Log Messages, Log Messages
@comment  node-name,  next,  previous,  up
@subsection Fatal errors

@i{Amd} attempts to deal with unusual events.  Whenever it is not
possible to deal with such an error, @i{Amd} will log an appropriate
message and, if it cannot possibly continue, will either exit or abort.
These messages are selected by @samp{-x fatal} on the command line.
When @b{syslog}(3) is being used, they are logged with level
@samp{LOG_FATAL}.  Even if @i{Amd} continues to operate it is likely to
remain in a precarious state and should be restarted at the earliest
opportunity.

@table @t

@item Attempting to inherit not-a-filesystem
The prototype mount point created during a filesystem restart did not
contain a reference to the restarted filesystem.  This error ``should
never happen''.

@item Can't bind to domain "@i{NIS-domain}"
A specific NIS domain was requested on the command line, but no server
for that domain is available on the local net.

@item Can't determine IP address of this host (@i{hostname})
When @i{Amd} starts it determines its own IP address.  If this lookup
fails then @i{Amd} cannot continue.  The hostname it looks up is that
obtained returned by @b{gethostname}(2) system call.

@item Can't find root file handle for @i{automount point}
@i{Amd} creates its own file handles for the automount points.  When it
mounts itself as a server, it must pass these file handles to the local
kernel.  If the filehandle is not obtainable the mount point is ignored.
This error ``should never happen''.

@item Must be root to mount filesystems (euid = @i{euid})
To prevent embarrassment, @i{Amd} makes sure it has appropriate system
privileges.  This amounts to having an euid of 0.  The check is made
after argument processing complete to give non-root users a chance to
access the @code{-v} option.

@item No work to do - quitting
No automount points were given on the command line and so there is no
work to do.

@item Out of memory
While attempting to malloc some memory, the memory space available to
@i{Amd} was exhausted.  This is an unrecoverable error.

@item Out of memory in realloc
While attempting to realloc some memory, the memory space available to
@i{Amd} was exhausted.  This is an unrecoverable error.

@item cannot create rpc/udp service
Either the NFS or AMQ endpoint could not be created.

@item gethostname: @i{description}
The @b{gethostname}(2) system call failed during startup.

@item host name is not set
The @b{gethostname}(2) system call returned a zero length host name.
This can happen if @i{Amd} is started in single user mode just after
booting the system.

@item ifs_match called!
An internal error occurred while restarting a pre-mounted filesystem.
This error ``should never happen''.

@item mount_afs: @i{description}
An error occurred while @i{Amd} was mounting itself.

@item run_rpc failed
Somehow the main NFS server loop failed.  This error ``should never
happen''.

@item unable to free rpc arguments in amqprog_1
The incoming arguments to the AMQ server could not be free'ed.

@item unable to free rpc arguments in nfs_program_1
The incoming arguments to the NFS server could not be free'ed.

@item unable to register (AMQ_PROGRAM, AMQ_VERSION, udp)
The AMQ server could not be registered with the local portmapper or the
internal RPC dispatcher.

@item unable to register (NFS_PROGRAM, NFS_VERSION, 0)
The NFS server could not be registered with the internal RPC dispatcher.

@end table

XXX: This section needs to be updated

@node Info messages, , Fatal errors, Log Messages
@comment  node-name,  next,  previous,  up
@subsection Info messages

@i{Amd} generates information messages to record state changes.  These
messages are selected by @samp{-x info} on the command line.  When
@b{syslog}(3) is being used, they are logged with level @samp{LOG_INFO}.

The messages listed below can be generated and are in a format suitable
for simple statistical analysis.  @dfn{mount-info} is the string
that is displayed by @dfn{Amq} in its mount information column and
placed in the system mount table.

@table @t

@item "@t{$@{@i{path}@}}" forcibly timed out
An automount point has been timed out by the @i{Amq} command.

@item "@t{$@{@i{path}@}}" has timed out
No access to the automount point has been made within the timeout
period.

@item Filehandle denied for "$@{@i{rhost}@}:$@{@i{rfs}@}"
The mount daemon refused to return a file handle for the requested filesystem.

@item Filehandle error for "$@{@i{rhost}@}:$@{@i{rfs}@}": @i{description}
The mount daemon gave some other error for the requested filesystem.

@item Finishing with status @i{exit-status}
@i{Amd} is about to exit with the given exit status.

@item Re-synchronizing cache for map @t{$@{@i{map}@}}
The named map has been modified and the internal cache is being re-synchronized.

@item file server @t{$@{@i{rhost}@}} is down - timeout of "@t{$@{@i{path}@}}" ignored
An automount point has timed out, but the corresponding file server is
known to be down.  This message is only produced once for each mount
point for which the server is down.

@item file server @t{$@{@i{rhost}@}} type nfs is down
An NFS file server that was previously up is now down.

@item file server @t{$@{@i{rhost}@}} type nfs is up
An NFS file server that was previously down is now up.

@item file server @t{$@{@i{rhost}@}} type nfs starts down
A new NFS file server has been referenced and is known to be down.

@item file server @t{$@{@i{rhost}@}} type nfs starts up
A new NFS file server has been referenced and is known to be up.

@item mount of "@t{$@{@i{path}@}}" on @t{$@{@i{fs}@}} timed out
Attempts to mount a filesystem for the given automount point have failed
to complete within 30 seconds.

@item @i{mount-info} mounted fstype @t{$@{@i{type}@}} on @t{$@{@i{fs}@}}
A new file system has been mounted.

@item @i{mount-info} restarted fstype @t{$@{@i{type}@}} on @t{$@{@i{fs}@}}
@i{Amd} is using a pre-mounted filesystem to satisfy a mount request.

@item @i{mount-info} unmounted fstype @t{$@{@i{type}@}} from @t{$@{@i{fs}@}}
A file system has been unmounted.

@item @i{mount-info} unmounted fstype @t{$@{@i{type}@}} from @t{$@{@i{fs}@}} link @t{$@{@i{fs}@}}/@t{$@{@i{sublink}@}}
A file system of which only a sub-directory was in use has been unmounted.

@item restarting @i{mount-info} on @t{$@{@i{fs}@}}
A pre-mounted file system has been noted.

@end table

XXX: This section needs to be updated

@c ################################################################
@node Acknowledgments & Trademarks, Index, Internals, Top
@comment  node-name,  next,  previous,  up
@unnumbered Acknowledgments & Trademarks

Many thanks to the @email{amd-dev@@majordomo.cs.columbia.edu,Amd
Developers} mailing list through the months developing am-utils.  These
members have contributed to the discussions, ideas, code and
documentation, and subjected their systems to alpha quality code.
Special thanks go to those
@uref{http://www.cs.columbia.edu/~ezk/am-utils/AUTHORS.txt,authors} who
have submitted patches.

Thanks to the Formal Methods Group at Imperial College for suffering
patiently while @i{Amd} was being developed on their machines.

Thanks to the many people who have helped with the development of
@i{Amd}, especially Piete Brooks at the Cambridge University Computing
Lab for many hours of testing, experimentation and discussion.

Thanks to the @email{amd-workers@@majordomo.glue.umd.edu,Amd Workers}
mailing list members for many suggestions and bug reports to @i{Amd}.

@itemize @bullet
@item
@b{DEC}, @b{VAX} and @b{Ultrix} are registered trademarks of Digital
Equipment Corporation.
@item
@b{AIX} and @b{IBM} are registered trademarks of International Business
Machines Corporation.
@item
@b{Sun}, @b{NFS} and @b{SunOS} are registered trademarks of Sun
Microsystems, Inc.
@item
@b{UNIX} is a registered trademark in the USA and other countries,
exclusively licensed through X/Open Company, Ltd.
@item
All other registered trademarks are owned by their respective owners.
@end itemize

@c ################################################################
@node Index, , Acknowledgments & Trademarks, Top
@comment  node-name,  next,  previous,  up
@unnumbered Index

@printindex cp

@contents
@bye

@c ====================================================================
@c ISPELL LOCAL WORDS:
@c LocalWords:  setfilename amdref overfullrule settitle titlepage titlefont nz
@c LocalWords:  authorfont vskip ifinfo iftex cindex unnumberedsec dfn xref vol
@c LocalWords:  locationN pxref jpo nott concentrix Sjoerd sjoerd cwi Eitan vuw
@c LocalWords:  Mizrotsky eitan shumuji dgux fpx scp hcx metcalf masala hlh OTS
@c LocalWords:  Presnell srp cgl Trost trost ogi pyrOSx OSx tubsibr riscix iX
@c LocalWords:  Piete pb Lindblad cjl ai umax utek xinu Mitchum D'Souza dsouza
@c LocalWords:  mrc apu alliant aviion AViiON fps macII multimax tahoe vax emph
@c LocalWords:  mapdefault valA valB valC YPTSDIR ETCDIR substr MAKEDBM YPDBDIR
@c LocalWords:  NOPUSH njw dylan dk dylan njw anydir domN achilles mjh pref sel
@c LocalWords:  gdef loc loc loc ldots autodir remopts rwho rwho styx styx yoyo
@c LocalWords:  noindent gould rvdmount rvdunmount fserver mtmp unioned logfile
@c LocalWords:  dmn esac phjk toytown toytown toytown toytown phjk RdDir RdLnk
@c LocalWords:  volname attrs netif dougal inaddr hwaddr ec mountmaps passno xy
@c LocalWords:  freq dumpset hfs brian florence localinfo fstabs automaps defn
@c LocalWords:  localname fsck'd opr gummo sjv ganymede sjv fserv fserv fserv
@c LocalWords:  vaxA vaxB wp thpfs nbsd asis ifs amqprog free'ed printindex gov
@c LocalWords:  LocalWords syncodeindex Distrib bsdnet lanl AutoMounter acis ic
@c LocalWords:  ac uk aix bsd Mullender nl il DG lcs hpux irix ucsf NeXT cse cl
@c LocalWords:  mt FX hp ibm mips utils def def Domainname eg hostd getwd tmp
@c LocalWords:  subsubsection rw grpid intr noconn nocto nodevs nosuid retrans
@c LocalWords:  rsize tcp timeo nounmount utimeout DDEBUG nodaemon fd hostnames
@c LocalWords:  pid Amd's pendry vangogh nfsx backoff stats nomap nostats CRIT
@c LocalWords:  noinfo clustername RVD dsk dsk amq hostports osver statfs str
@c LocalWords:  ou counter's amdmaps proj src tftpboot sh mv cd sbin ypcat inet
@c LocalWords:  Getattr getattr localhost fhandles netmask fstype noquota addr
@c LocalWords:  exportfs Dumpsets dumpsets pindex ldif fixmount fixrmtab euid
@c LocalWords:  lostaltmail realloc netnumber itemx primnetnum primnetname ARG
@c LocalWords:  subsnetname subsnetnum netgrp netgroup multitable Shlib dec osf
@c LocalWords:  hppa pc bsdi freebsd netbsd openbsd ncr sysv rs acdirmax fsid
@c LocalWords:  acdirmin acregmax acregmin actimeo dumbtimr nfsv noac noauto sd
@c LocalWords:  nocache nodev noint nosub pgthresh posix rdonly suid symttl mfs
@c LocalWords:  AMFS umapfs myftpdir unionfs es mapname mapfile mapfile slocal
@c LocalWords:  mailspool saturn saturn notknown lol ober dr xr xr drwxrwsr cfg
@c LocalWords:  lrwxrwxrwx adminpr hplj adminpr cfg tekxp xterms tekxp Dupuy tp
@c LocalWords:  linkname hlfsddump dirname rmtab pluto rlogin direntry pg vr dn
@c LocalWords:  maxmem hlfsdir xmailbox showmount cn amdmap amdmapName resvport
@c LocalWords:  objectClass amdmapKey amdmapValue ln powerpc amdmapTimestamp ez
@c LocalWords:  moisil FSinfo Libtool Unmounting sublink fileservers NullProc
@c LocalWords:  gethostname mount's unmounts linkx remounts unmounting UAs SA's
@c LocalWords:  mountpoint mountpoints unescaped UIDs util's overlayed uref EFS
@c LocalWords:  serv maxgroups nfsl cachedir copt cfsadmin efs addopts fg
@c LocalWords:  nointr