NFS-Root-Client Mini-HOWTO

Ofer Maor

  v4.1, 02 Feb, 1999
  Revision History
  Revision 4.1 Feb 02, 1999 Revised by: mo

  The purpose of this Mini-Howto is to explain how to create client root
  directories on a server that is using NFS Root mounted clients.
    _________________________________________________________________

  Table of Contents
  1. [1]Copyright

       1.1. [2]Thanks

  2. [3]Preface

       2.1. [4]General Overview

  3. [5]Creating the client's root directory

       3.1. [6]Creating the directory tree
       3.2. [7]Creating the minimal file system needed for boot
       3.3. [8]Building the etc directory and configuring the clients
       3.4. [9]Booting Up

  4. [10]Creating more clients

1. Copyright

  (c) 1996 Ofer Maor (<[11][email protected]>)

  Unless otherwise stated, Linux HOWTO documents are copyrighted by
  their respective authors. Linux HOWTO documents may be reproduced and
  distributed in whole or in part, in any medium physical or electronic,
  as long as this copyright notice is retained on all copies. Commercial
  redistribution is allowed and encouraged; however, the author would
  like to be notified of any such distributions.

  All translations, derivative works, or aggregate works incorporating
  any Linux HOWTO documents must be covered under this copyright notice.
  That is, you may not produce a derivative work from a HOWTO and impose
  additional restrictions on its distribution. Exceptions to these rules
  may be granted under certain conditions; please contact the Linux
  HOWTO coordinator at the address given below.

  In short, we wish to promote dissemination of this information through
  as many channels as possible. However, we do wish to retain copyright
  on the HOWTO documents, and would like to be notified of any plans to
  redistribute the HOWTOs.

  If you have questions, please contact Ofer Maor
  (<[12][email protected]>), the author of this mini-HOWTO, or Greg
  Hankins, the Linux HOWTO coordinator, at <[13][email protected]>
  via email, or at +1 404 853 9989.

  If you have anything to add to this Mini-Howto, please mail the author
  (Ofer Maor, <[14][email protected]>), with the information. Any new
  relevant information would be appreciated.
    _________________________________________________________________

1.1. Thanks

  I would like to express my thanks to the author of the NFS-Root Howto,
  Andreas Kostyrca (<[15][email protected]>). His Mini-Howto
  helped me with the first steps in creating a NFS Root Mounted client.
  My Mini-Howto does not, in any way, try to replace his work, but to
  enhance it using my experiences in this process.

  I would also like to thank Mark Kushinsky (<[16][email protected]>) for
  polishing the english and spelling of this Howto, thus making it much
  more readable.
    _________________________________________________________________

2. Preface

  This Mini-Howto was written in order to help people who want to use
  NFS Root mounting to create their client's directories. Please note
  that there are many ways to accomplish this, depending on your needs
  and intent. If the clients are individual, and each client has its own
  users and administrator, it will be necessary to make significant
  parts of the client dirs not shared with other clients. On the other
  hand, if the client is intended for multiple users, and are all
  administrated by the same person (for instance, a computerclass), make
  as many files as possible shareable in order to make administration
  more manageable. This Howto will focus on the second issue.
    _________________________________________________________________

2.1. General Overview

  When building a client's root directory, and trying to limit ourselves
  to the minimum client size, we mainly focus on which files we can
  share, or mount from the server. In this Howto I will recommend the
  configuration of a client based on my experience. But beforewe begin
  please note:

    * This Mini-Howto does not explain how to do the actual NFS Root
      mounting. Refer to the NFS-Root Mini-Howto if you need more
      information about that issue.
    * I based most of my client's configuration on mounts and symbolic
      links. A lot of those symbolic links can be replaced by hardlinks.
      One should choose according to his personal preference. Putting a
      hardlink over a mount and a symbolic link has its advantages, but
      might cause confusion. A file will not be erased until all its
      hardlinks are removed. Thus, In order to prevent a case in which
      you upgrade a certain file, and the hardlinks still refer to the
      older version, you will have to be very careful and keep track of
      every link you put.
    * While mounting the information from the server, two concepts can
      be used. The first (most common) concept, is to mount the whole
      server root directory under a local directory, and then just
      change the path or link the relevant directories there. I
      personally dislike mounting root partitions of a server on
      clients. Thus, this Howto suggests a way to mount the relevant
      directories of the server to the relevant places on the system.
    * This Howto is based on my experience building client directories
      on a Slackware 3.1 based distribution. Things may be different
      (especially on the rc.* files), for other users, however the
      concepts should still remain the same.
    _________________________________________________________________

3. Creating the client's root directory

3.1. Creating the directory tree

  First of all, you need to create the directory structure itself. I
  created all the clients under /clients/hostname and I will use it for
  my examples listed below. This, however, can be changed to anything
  else. The first stage, then, is to create the relevant directories in
  the root directory. You should create the following directories:

  bin , dev , etc , home , lib , mnt , proc , sbin , server , tmp , usr
  , var

  and any other directories you might want to have on your system.

  The local, proc, and dev directories will be used separately on each
  machine while the rest of the directories will be either partly or
  completely shared with the rest of the clients.
    _________________________________________________________________

3.2. Creating the minimal file system needed for boot

3.2.1. Creating the dev dir.

  Although the dev dir can be shared, it is better to create a separate
  one for each client. You can create your client's dev directory with
  the appropriate MAKEDEV scripts, however in most cases it is simpler
  just to copy it from the server:
bash# cp -a /dev /clients/hostname

  You should keep in mind that /dev/mouse, /dev/cdrom and /dev/modem are
  symbolic links to actually devices, and therefore you should be sure
  that they are linked correctly to fit the client's hardware.
    _________________________________________________________________

3.2.2. Copying the necessary binaries.

  Although we mount everything from the server, there is a minimum that
  we need to copy to each client. First of all, we need "init", our
  system will not be able to run anything before init'ing (as the author
  found out in the hard way ;-). So first, you should copy /sbin/init to
  your client's sbin dir and then so that rc.S will run, you should copy
  /bin/sh to the client's bin directory. Also, in order to mount
  everything you need to copy /sbin/mount to the client's sbin
  directory. This is the minimum, assuming the first line in your rc.S
  is mount -av. However, I recommend copying a few more files: update,
  ls, rm, cp and umount, so that you will have the basic tools in case
  the client has problems mounting. If you choose to leave your swap on
  line before mount, you should also copy the swapon binary.

  Since most of these binaries are by default dynamically linked, you
  will also need to copy a fair part of /lib:
bash# cp -a /lib/ld* /lib/libc.* /lib/libcursses.* /client/hostname/lib

  Hardlinking the binaries themselves, instead of copying them, should
  be considered. Please read my comments on this in [17]Section 2.1 of
  this Howto.

  Please notice, all of the information above assumes that the kernel
  has been given the network parameters while booting up. If you plan to
  use rarp or bootp, you will probably need the relevant binaries for
  these as well.

  Generally, you will need the minimum of files that will enable you to
  configure the network and run rc.S up to the point where it mounts the
  rest of the file system. Make sure you looked into your /etc/init and
  rc.S files, making sure there are no "surprises" in any of them, which
  will require other files to be accessed, before the first mount will
  take place. If you do, however, find such files, you can either copy
  them as well, or remove the relevant parts from your init and your
  rc.S files.
    _________________________________________________________________

3.2.3. The var directory

  The var directory, in most cases, should be separate for each client.
  However, a lot of the data can be shared. Create under the server
  directory a directory called var. We will mount the server's var
  directory there. To create the local var directory, simply type:
bash# cp -a /var /clients/hostname/

  Now, you have a choice as to what you want to separate, and what you
  want to share. Any directory/file that you want to share, simply
  remove it from the client's var dir, and symlink it to the
  /server/var/ directory. However please note that you should either
  symlink it to /server/var or to ../server/var but NOT to
  /clients/hostname/server/var as this will not work when the root
  changes.

  Generally, I would recommend separating /var/run, /var/lock,
  /var/spool, and /var/log.
    _________________________________________________________________

3.2.4. The rest of the directories

    * etc is explained thoroughly in the next section.
    * mnt and proc are for local purposes.
    * usr and home are merely mount points.
    * tmp is up to you. You can create a different tmp directory for
      each client, or create some /clients/tmp directories, and mount it
      for each client under /tmp. I would recommend that you provide
      each client with a separate tmp directory.
    _________________________________________________________________

3.3. Building the etc directory and configuring the clients

  Please Note -

    Note: this section refers to building the etc directory which is
    mostly shared among the clients. If your diskless clients have
    separate system administrators, it's best to set up for each client
    a separate etc directory.
    _________________________________________________________________

3.3.1. Building a clients wide etc directory
    _________________________________________________________________

3.3.2. Creating a client's etc directory

  Although we separate the etc directories of the clients, we still want
  to share a large portion of the files there. Generally, I think
  sharing the etc files with the server's /etc is a bad idea, and
  therefore I recommend creating a /clients/etc directory, which will
  hold the information needed for the clients. To start with, simply
  copy the contents of the server's etc to the /clients/etc directory.

  You should add to this directory all of the non-machine-specific
  configuration files, for instance motd, issue, etc. and not the
  clientspecific ones.(i.e. inittab or fstab)

  The most important changes will be in your rc.d directory. First, you
  should change rc.inet1 to be suitable for your local setup. I pass all
  my network parameters to the kernel through the LILO/Loadlin,
  therefore I remove almost everything from rc.inet1 file. The only
  thing I leave there is the ifconfig and route of the localhost. If you
  use rarp or bootp, you will have to build it accordingly.

  Secondly, you should edit your rc.S. First, remove all the parts that
  are responsible for the fsck check as fsck will occur when the server
  boots up. Then, you should find the line that mounts your fstab. This
  should look something like:

  mount -avt nonfs

  The -t nonfs is there since normal clients first run rc.S and only
  later on use rc.inet1 to configure the Ethernet. As this will cause no
  NFS partitions to be mounted this line should be deleted. Therefore,
  change it to mount -av. If you need to run rarp/bootp to configure
  your network, do it in rc.S (or call the appropriate script from
  rc.S), before the mount, and make sure your physical bin and sbin
  directories have the necessary files available.

  After the mount -av is performed, you will have a working file system.
  Build a general fstab, so that you can later copy it to each client.
  Your fstab should look something like this:

  Table 1. fstab
  server:/clients/hostname / nfs default 1 1
  server:/bin /bin nfs default 1 1
  server:/usr /usr nfs default 1 1
  server:/sbin /sbin nfs default 1 1
  erver:/home /home nfs default 1 1
  server:/lib /lib nfs default 1 1
  server:/clients/etc /server/etc nfs default 1 1
  server:/clients/var /server/var nfs default 1 1
  none /proc proc default 1 1

  Please notice, that the keyword default might not work on all versions
  of mount. You might change it to rw or ro or remove all of the default
  1 1 part.

  Also, make sure your server's /etc/exports looks like this:

  Table 2. /etc/exports
  /clients/hostname hostname.domainname(rw,no_root_squash)
  /clients/etc hostname.domainname(ro,no_root_squash)
  /clients/var hostname.domainname(ro,no_root_squash)
  /usr hostname.domainname(ro,no_root_squash)
  /sbin hostname.domainname(ro,no_root_squash)
  /bin hostname.domainname(ro,no_root_squash)
  /lib hostname.domainname(ro,no_root_squash)
  /home hostname.domainname(rw,no_root_squash)

  Other than the first line, which should be separate for each host, the
  rest of the lines can be replaced with a hostmask to fit all your
  hosts (like pc*.domain - keep in mind though, that * will substitue
  only strings without a dot in them). I suggest that you make most of
  the directories read only, but this is up to you. The no_root_squash
  will make sure root users on the clients have actual root permissions
  on the nfsd as well. Check out man exports(5). If you want users to be
  able to run passwd from the clients also, make sure the /etc has rw
  and not ro permissions. However, this is not advisable.

  Please note another thing concerning the rc.S file. In Slackware, by
  default, it creates a new /etc/issue and /etc/motd every time it runs.
  This function MUST be disabled if these files are mounted ro from the
  server, and I would recommend that they should be disabled in any
  case.

  Lastly, if you want to have the same userbase on the server as on the
  clients, you should choose between 1), using NIS (Yellow Pages - check
  the yp-howto), and then each client will have a separate /etc/passwd
  and /etc/group as it receives it from the NIS server. 2) In most
  cases, a simple symbolic link will suffice. Therefore, you will need
  to either hardlink /clients/etc/passwd to /etc/passwd, or if you
  prefer a symlink, link /etc/passwd to /clients/etc/passwd (and not the
  other way around, since the clients do not mount the server's etc
  directory). Do the same for /etc/group.
    _________________________________________________________________

3.3.3. Creating a client's etc directory

  Generally, most of the files in the client's etc should be symlinked
  to the /server/etc directory. However, some files are different for
  each machine, and some just have to be there when the kernel loads.
  The minimum you need from the etc dir is as follows:

  resolv.conf
  hosts
  inittab
  rc.d/rc.S
  fstab

  Since these 5 files can be identical on all clients, you can simply
  hardlink them or copy them again. However, with the rc.S and fstab
  file it is advised to keep a separate copy for each client. You will
  also need a separate etc/HOSTNAME for each client. I personally
  recommend having all of the rc.d files separate for each client, as
  configuration and hardware might vary from one to another.

  For each client, add to the fstab the proper swap line:

  Table 3. fstab
  /dev/swap_partition swap swap default 1 1

  The rest of the /etc files of the client, you can either hardlink to
  the /clients/etc/* files, or symlink them to the /server/etc (which is
  the mount point of /clients/etc/).

  Make sure your machine can resolve properly, either through a named or
  through etc/hosts. It is not a bad idea to keep the server's IP in the
  etc/hosts, instead of counting on resolving. If you will count only on
  named resolving, a problem in the named will prevent your clients from
  booting up.
    _________________________________________________________________

3.4. Booting Up

  Now, all you have to do is to boot up your machine, cross your fingers
  and hope everything works as it should :-).
    _________________________________________________________________

4. Creating more clients

  If you have followed my instructions so far this should be simple - cd
  to /clients/ and type:
bash# cp -a hostname1 hostname2

  and then make sure you check these points:

    * rc.d/* files matches the hardware and wanted software
      configuration
    * etc/HOSTNAME is correct
    * fstab's swap line is correct
    * the symbolic links of dev/mouse, dev/modem and dev/cdromare right.

  Good Luck....

References

  1. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#COPYRIGHT
  2. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#AEN28
  3. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#PREFACE
  4. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#OVERVIEW
  5. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#CLIENTROOT
  6. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#AEN52
  7. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#AEN74
  8. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#AEN158
  9. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#AEN371
 10. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#MORECLIENTS
 11. mailto:[email protected]
 12. mailto:[email protected]
 13. mailto:[email protected]
 14. mailto:[email protected]
 15. mailto:[email protected]
 16. mailto:[email protected]
 17. file://localhost/export/sunsite/users/gferg/howto/00_NFS-Root-Client-mini-HOWTO.html#OVERVIEW

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.