The Linux Tips HOWTO
 Paul Anderson, [email protected]
 v3.6, June 1998

 This HOWTO contains those hard to find hints and tweekings that make
 Linux a bit nicer.
 ______________________________________________________________________

 Table of Contents


 1. Introduction

 2. Short Tips

    2.1 Handy Syslog Trick
    2.2 Script to view those compressed HOWTOs.
    2.3 Is there enough free space???
    2.4 Util to clean up your logfiles.
    2.5 Handy Script to Clean Up Corefiles.
    2.6 Moving directories between filesystems.
    2.7 Finding out which directories are the largest.
    2.8 The Linux Gazette
    2.9 Pointer to patch for GNU Make 3.70 to change VPATH behavior.
    2.10 How do I stop my system from fscking on each reboot?
    2.11 How to avoid fscks caused by "device busy" at reboot time.
    2.12 How to find the biggest files on your hard-drive.
    2.13 How to print pages with a margin for hole punching.
    2.14 A way to search through trees of files for a particular regular expression.
    2.15 A script for cleaning up after programs that create autosave and backup files.
    2.16 How to find out what process is eating the most memory.
    2.17 Rigging vi for C programming,
    2.18 Using ctags to ease programming.
    2.19 Why does sendmail hang for 5 minutes on startup with RedHat?
    2.20 How do I configure RedHat for using color-ls?
    2.21 How do I find which library in /usr/lib holds a certain function?
    2.22 I compiled a small test program in C, but when I run it, I get no output!

 3. Detailed Tips

    3.1 Sharing swap partitions between Linux and Windows.
    3.2 Desperate Undelete.
    3.3 How to use the immutable flag.
    3.4 A suggestion for where to put new stuff.
    3.5 Converting all files in a directory to lowercase.
    3.6 How To Upgrade Sendmail
    3.7 Some tips for new sysadmins.
    3.8 How to configure xdm's chooser for host selection.


 ______________________________________________________________________

 1.  Introduction

 Welcome to the Linux Tips HOWTO, a list of neato tricks and
 optimizations that make Linux more fun.  All I have in here right now
 are tips off of the top of my head, and tips from the old Tips-HOWTO
 (Why take out decent tips, right?).  So send all your favorite hints
 and tips to me so I can put them in the next Tips-HOWTO.

 Paul Anderson Maintainer--Linux TIPS HOWTO

 [email protected]



 2.  Short Tips

 2.1.  Handy Syslog Trick Paul Anderson, Tips-HOWTO maintainer

 Edit your /etc/syslog.conf, and put in the following line:


      # Dump everything on tty8
      *.*                                     /dev/tty8



 One caveat: REMEMBER TO USE TABS!  syslog doesn't like spaces...


 2.2.  Script to view those compressed HOWTOs. Didier Juges,  dj@des-
 tin.nfds.net


 From a newbie to another, here is a short script that eases looking
 for and viewing howto documents.  My howto's are in
 /usr/doc/faq/howto/ and are gzipped. The file names are XXX-HOWTO.gz,
 XXX being the subject.  I created the following script called "howto"
 in the /usr/local/sbin directory:


      ______________________________________________________________________
      #!/bin/sh
      if [ "$1" = "" ]; then
          ls /usr/doc/faq/howto | less
      else
          gunzip -c /usr/doc/faq/howto/$1-HOWTO.gz | less
      fi
      ______________________________________________________________________



 When called without argument, it displays a directory of the available
 howto's. Then when entered with the first part of the file name
 (before the hyphen) as an argument, it unzips (keeping the original
 intact) then displays the document.

 For instance, to view the Serial-HOWTO.gz document, enter:

 $ howto Serial



 2.3.  Is there enough free space??? Hans Zoebelein,  zocki@gold-
 fish.cube.net

 Here comes a short script which will check from time to time that
 there is enough free space available on anything which shows up in
 mount (disks, cdrom, floppy...)

 If space runs out, a message is printed every X seconds to the screen
 and 1 mail message per filled device is fired up.



 ______________________________________________________________________
 #!/bin/sh

 #
 # $Id: check_hdspace,v 1.18 1996/12/11 22:33:29 root Exp root $
 #

 #
 # Since I got mysterious error messages during compile when
 # tmp files filled up my disks, I wrote this to get a warning
 # before disks are full.
 #
 # If this stuff saved your servers from exploding,
 # send praising email to [email protected].
 # If your site burns down because of this, sorry but I
 # warned you: no comps.
 # If you really know how to handle sed, please forgive me :)
 #

 #
 # Shoot and forget: Put 'check_hdspace &' in rc.local.
 # Checks for free space on devices every $SLEEPTIME sec.
 # You even might check your floppies or tape drives. :)
 # If free space is below $MINFREE (kb), it will echo a warning
 # and send one mail for each triggering device to $MAIL_TO_ME.
 # If there is more free space than trigger limit again,
 # mail action is also armed again.
 #

 # TODO: Different $MINFREE for each device.
 # Free /*tmp dirs securely from old junk stuff if no more free space.


 DEVICES='/dev/sda2 /dev/sda8 /dev/sda9'         # device; your put disks here
 MINFREE=20480                                   # kb; below this do warning
 SLEEPTIME=10                                    # sec; sleep between checks
 MAIL_TO_ME='root@localhost'                     # fool; to whom mail warning


 # ------- no changes needed below this line (hopefully :) -------

 MINMB=0
 ISFREE=0
 MAILED=""
 let MINMB=$MINFREE/1024         # yep, we are strict :)

 while [ 1 ]; do
         DF="`/bin/df`"
                 for DEVICE in $DEVICES ; do
                 ISFREE=`echo $DF | sed s#.\*$DEVICE" "\*[0-9]\*""\*[0-9]\*" "\*## | sed s#" ".\*##`

                 if [ $ISFREE -le $MINFREE ] ; then
                         let ISMB=$ISFREE/1024
                         echo  "WARNING: $DEVICE only $ISMB mb free." >&2
                         #echo "more stuff here" >&2
                         echo -e "\a\a\a\a"

                         if [ -z  "`echo $MAILED | grep -w $DEVICE`" ] ; then
                                 echo "WARNING: $DEVICE only $ISMB mb free.      (Trigger is set to $MINMB mb)" \
                                 | mail -s "WARNING: $DEVICE only $ISMB mb free!" $MAIL_TO_ME
                                 MAILEDH="$MAILED $DEVICE"
                                 MAILED=$MAILEDH
                                 # put further action here like cleaning
                                 # up */tmp dirs...
                         fi
                         elif [ -n  "`echo $MAILED | grep -w $DEVICE`" ] ; then
                                 # Remove mailed marker if enough disk space
                                 # again. So we are ready for new mailing action.
                                 MAILEDH="`echo $MAILED  | sed s#$DEVICE##`"
                                 MAILED=$MAILEDH
                         fi

                 done
                 sleep $SLEEPTIME

 done
 ______________________________________________________________________



 2.4.  Util to clean up your logfiles. Paul Anderson, Tips-HOWTO Main-
 tainer >

 If you're like me, you have a list with 430 subscribers, plus 100+
 messages per day coming in over UUCP.  Well, what's a hacker to do
 with these huge logs?  Install chklogs, that's what.  Chklogs is
 written by Emilio Grimaldo, [email protected], and the current
 version 1.8 available from
 ftp.iaehv.nl:/pub/users/grimaldo/chklogs-1.8.tar.gz.  It's pretty self
 explanatory to install(you will, of course, check out the info in the
 doc subdirectory).  Once you've got it installed, add a crontab entry
 like this:


      # Run chklogs at 9:00PM daily.
      00 21 * * *       /usr/local/sbin/chklogs -m



 While you're at it, mention to the author how nice a peice of software
 this is:)



 2.5.  [email protected] Handy Script to Clean Up Corefiles. Otto
 Hammersmith,

 Create a file called rmcores(the author calls it handle-cores) with
 the following in it:



 ______________________________________________________________________
 #!/bin/sh
 USAGE="$0 <directory> <message-file>"

 if [ $# != 2 ] ; then
         echo $USAGE
         exit
 fi

  echo Deleting...
 find $1 -name core -atime 7 -print -type f -exec rm {} \;

 echo e-mailing
 for name in `find $1 -name core -exec ls -l {} \; | cut -c16-24`
 do
         echo $name
         cat $2 | mail $name
 done
 ______________________________________________________________________



 And have a cron job run it every so often.



 2.6.  Moving directories between filesystems. Alan Cox,
 [email protected]

 Quick way to move an entire tree of files from one disk to another


      (cd /source/directory && tar cf - . ) | (cd /dest/directory && tar xvfp -)



 [ Change from cd /source/directory; tar....etc.  to prevent possibil-
 ity of trashing directory in case of disaster.  Thanks to Jim Dennis,
 [email protected], for letting me know. -Maint. ]


 2.7.  [email protected] Finding out which directories are the largest.
 Mick Ghazey,

 Ever wondered which directories are the biggest on your computer?
 Here's how to find out.


      du -S | sort -n



 2.8.  The Linux Gazette

 Kudos go to John Fisk, creator of the Linux Gazette.  This is an
 excellent e-zine plus, it's FREE!!!  Now what more could you ask?
 Check it out at:


      http://www.linuxgazette.com


 BTW, It turns out that (1) LG is now out on a monthly basis, and (2)
 John Fisk no longer maintains it, the fellows at SSC do.



 2.9.  Ted Stern,  [email protected] Pointer to patch for GNU
 Make 3.70 to change VPATH behavior.

 I don't know if many people have this problem, but there is a
 "feature" of GNU make version 3.70 that I don't like. It is that VPATH
 acts funny if you give it an absolute pathname.  There is an extremely
 solid patch that fixes this, which you can get from Paul D. Smith
 <[email protected]>.  He also posts the documentation and patch
 after every revision of GNU make on the newsgroup "gnu.utils.bug"
 Generally, I apply this patch and recompile gmake on every system I
 have access to.


 2.10.  How do I stop my system from fscking on each reboot? Dale Lutz,
 [email protected]

 Q:  How do I stop e2fsck from checking my disk every time I boot up.

 A:  When you rebuild the kernel, the filesystem is marked as 'dirty'
 and so your disk will be checked with each boot.  The fix is to run:

 rdev -R /zImage 1

 This fixes the kernel so that it is no longer convinced that the
 filesystem is dirty.

 Note: If using lilo, then add read-only to your linux setup in your
 lilo config file (Usually /etc/lilo.conf)


 2.11.  How to avoid fscks caused by "device busy" at reboot time. Jon
 Tombs,  [email protected]

 If you often get device busy errors on shutdown that leave the
 filesystem in need of an fsck upon reboot, here is a simple fix:

 To /etc/rc.d/init.d/halt or /etc/rc.d/rc.0, add the line


      mount -o remount,ro /mount.dir



 for all your mounted filesystems except /, before the call to umount
 -a. This means if, for some reason, shutdown fails to kill all pro-
 cesses and umount the disks they will still be clean on reboot. Saves
 a lot of time at reboot for me.



 2.12.  How to find the biggest files on your hard-drive.

 Simon Amor, [email protected]



      ls -l | sort +4n


 Or, for those of you really scrunched for space this takes awhile but
 works great:



      cd /
      ls -lR | sort +4n



 2.13.  How to print pages with a margin for hole punching. Mike
 Dickey,  [email protected]



      ______________________________________________________________________
              #!/bin/sh
              # /usr/local/bin/print
              # a simple formatted printout, to enable someone to
              # 3-hole punch the output and put it in a binder

              cat $1 | pr -t -o 5 -w 85 | lpr
      ______________________________________________________________________



 2.14.  Raul Deluth Miller,  [email protected] A way to search
 through trees of files for a particular regular expression.

 I call this script 'forall'.  Use it like this:


      forall /usr/include grep -i ioctl
      forall /usr/man grep ioctl



 Here's forall:


      ______________________________________________________________________
      #!/bin/sh
      if [ 1 = `expr 2 \> $#` ]
      then
              echo Usage: $0 dir cmd [optargs]
              exit 1
      fi
      dir=$1
      shift
      find $dir -type f -print | xargs "$@"
      ______________________________________________________________________



 2.15.  Barry Tolnas,  [email protected] A script for cleaning
 up after programs that create autosave and backup files.

 Here is a simple two-liner which recursively descends a directory
 hierarchy removing emacs auto-save (#) and backup (~) files, .o files,
 and  TeX .log files. It also compresses .tex files and README files. I
 call it 'squeeze' on my system.


      ______________________________________________________________________
      #!/bin/sh
      #SQUEEZE removes unnecessary files and compresses .tex and README files
      #By Barry tolnas, [email protected]
      #
      echo squeezing $PWD
      find  $PWD \( -name \*~ -or -name \*.o -or -name \*.log -or -name \*\#\) -exec
      rm -f {} \;
      find $PWD \( -name \*.tex -or -name \*README\* -or -name \*readme\* \) -exec gzip -9 {} \;
      ______________________________________________________________________



 2.16.  [email protected] How to find out what process is eating the
 most memory. Simon Amor,



      ps -aux | sort +4n



 -OR-


      ps -aux | sort +5n



 2.17.  Rigging vi for C programming, Paul Anderson, Tips-HOWTO Main-
 tainer

 I do a lot of C programming in my spare time, and I've taken the time
 to rig vi to be C friendly.  Here's my .exrc:


      ______________________________________________________________________
      set autoindent
      set shiftwidth=4
      set backspace=2
      set ruler
      ______________________________________________________________________



 What does this do?  autoindent causes vi to automatically indent each
 line following the first one indented, shiftwidth sets the distance of
 ^T to 4 spaces, backspace sets the backspace mode, and ruler makes it
 display the line number.  Remember, to go to a specific line number,
 say 20, use:



 ______________________________________________________________________
 vi +20 myfile.c
 ______________________________________________________________________



 2.18.  Using ctags to ease programming.

 Most hackers already have ctags on their computers, but don't use it.
 It can be very handy for editing specific functions.  Suppose you have
 a function, in one of many source files in a directory for a program
 you're writing, and you want to edit this function for updates.  We'll
 call this function foo().  You don't where it is in the source file,
 either.  This is where ctags comes in handy.  When run, ctags produces
 a file named tags in the current dir, which is a listing of all the
 functions, which files they're in and where they are in said files.
 The tags file looks like this:



      ______________________________________________________________________

      ActiveIconManager       iconmgr.c       /^void ActiveIconManager(active)$/
      AddDefaultBindings      add_window.c    /^AddDefaultBindings ()$/
      AddEndResize    resize.c        /^AddEndResize(tmp_win)$/
      AddFuncButton   menus.c /^Bool AddFuncButton (num, cont, mods, func, menu, item)$/
      AddFuncKey      menus.c /^Bool AddFuncKey (name, cont, mods, func, menu, win_name, action)$/
      AddIconManager  iconmgr.c       /^WList *AddIconManager(tmp_win)$/
      AddIconRegion   icons.c /^AddIconRegion(geom, grav1, grav2, stepx, stepy)$/
      AddStartResize  resize.c        /^AddStartResize(tmp_win, x, y, w, h)$/
      AddToClientsList        workmgr.c       /^void AddToClientsList (workspace, client)$/
      AddToList       list.c  /^AddToList(list_head, name, ptr)$/
      ______________________________________________________________________



 To edit, say AddEndResize() in vim, run:



      vim -t AddEndResize



 This will bring the appropriate file up in the editor, with the cursor
 located at the beginning of the function.



 2.19.  [email protected] Why does sendmail hang for 5 minutes on
 startup with RedHat? Paul Anderson,

 This is a fairly common problem, almost to the point of being a FAQ.
 I don't know if RedHat has fixed this bug in their distribution, but
 you can repair it yourself.  If you look in your /etc/hosts file, you
 will find it looks something like:


      127.0.0.1               localhost       yourbox

 When sendmail starts, it does a lookup on your hostname(in this
 example, yourbox).  It then finds that the IP for yourbox is
 127.0.0.1, sendmail doesn't like this, so it does the lookup again.
 It continues with this for a while until it eventually gives up and
 exits.  Fixing the problem is extremely easy, edit your /etc/hosts
 file and change it to something like this:


      127.0.0.1               localhost
      10.56.142.1             yourbox



 2.20.  How do I configure RedHat for using color-ls? Paul Anderson,
 [email protected]

 RedHat's distribution comes with color-ls, however why they don't
 configure it for colour use by default is beyond me.  Here's to fix
 it.

 First, type eval `DIRCOLORS`

 Next, alias ls='ls --color=auto'

 And put the 'alias.....' in your /etc/bashrc



 2.21.  [email protected] How do I find which library in
 /usr/lib holds a certain function? Pawel Veselow,

 What if you're compiling and you've missed a library that needed
 linking in?  All gcc reports are function names...  Here's a simple
 command that'll find what you're looking for:


      for i in *; do echo $i:;nm $i|grep tgetnum 2>/dev/null;done



 Where tgetnum is the name of the function you're looking for.



 2.22.  I compiled a small test program in C, but when I run it, I get
 no output!

 You probably compiled the program into a binary named test, didn't
 you?  Linux has a program called test, which tests if a certain
 condition is true, it never produces any output on the screen.
 Instead of just typing test, try: ./test



 3.  Detailed Tips

 3.1.  Sharing swap partitions between Linux and Windows. Tony Acero,
 [email protected]


 1. Format the partition as a dos partition, and create the Windows
    swap file on it, but don't run windows yet. (You want to keep the
    swap file completely empty for now, so that it compresses well).

 2. Boot linux and save the partition into a file.  For example if the
    partition was /dev/hda8:


      dd if=/dev/hda8 of=/etc/dosswap



 3. Compress the dosswap file; since it is virtually all 0's it will
    compress very well


      gzip -9 /etc/dosswap



 4. Add the following to the /etc/rc file to prepare and install the
    swap space under Linux:

    XXXXX is the number of blocks in the swap partition


      mkswap /dev/hda8 XXXXX
      swapon -av



 Make sure you add an entry for the swap partition in your /etc/fstab
 file

 5. If your init/reboot package supports /etc/brc or /sbin/brc add the
    following to /etc/brc, else do this by hand when you want to boot
    to dos|os/2 and you want to convert the swap partition back to the
    dos/windows version:



      swapoff -av
      zcat /etc/dosswap.gz | dd of=/dev/hda8 bs=1k count=100



 # Note that this only writes the first 100 blocks back to the parti-
 tion. I've found empirically that this is sufficient

 >>  What are the pros and cons of doing this?

 Pros: you save a substantial amount of disk space.

 Cons: if step 5 is not automatic, you have to remember to do it by
 hand, and it slows the reboot process by a nanosecond :-)


 3.2.  Desperate Undelete. Michael Hamilton,  [email protected]

 Here's a trick I've had to use a few times.


 Desperate person's text file undelete.

 If you accidentally remove a text file, for example, some email, or
 the results of a late night programming session, all may not be lost.
 If the file ever made it to disk, ie it was around for more than 30
 seconds, its contents may still be in the disk partition.

 You can use the grep command to search the raw disk partition for the
 contents of file.

 For example, recently, I accidentally deleted a piece of email.  So I
 immediately ceased any activity that could modify that partition: in
 this case I just refrained from saving any files or doing any compiles
 etc.  On other occasions, I've actually gone to the trouble of bring
 the system down to single user mode, and unmounted the filesystem.

 I then used the egrep command on the disk partition:  in my case the
 email message was in /usr/local/home/michael/, so from the output from
 df, I could see this was in /dev/hdb5


        sputnik3:~ % df
          Filesystem         1024-blocks  Used Available Capacity Mounted on
          /dev/hda3              18621    9759     7901     55%   /
          /dev/hdb3             308852  258443    34458     88%   /usr
          /dev/hdb5             466896  407062    35720     92%   /usr/local

          sputnik3:~ % su
          Password:
          [michael@sputnik3 michael]# egrep -50 'ftp.+COL' /dev/hdb5 > /tmp/x



 Now I'm ultra careful when fooling around with disk partitions, so I
 paused to make sure I understood the command syntax BEFORE pressing
 return.  In this case the email contained the word 'ftp' followed by
 some text followed by the word 'COL'.  The message was about 20 lines
 long, so I used -50 to get all the lines around the phrase.  In the
 past I've used -3000 to make sure I got all the lines of some source
 code.  I directed the output from the egrep to a different disk parti-
 tion - this prevented it from over writing the message I was looking
 for.

 I then used strings to help me inspect the output


         strings /tmp/x | less



 Sure enough the email was in there.

 This method can't be relied on, all, or some, of the disk space may
 have already been re-used.

 This trick is probably only useful on single user systems.  On multi-
 users systems with high disk activity, the space you free'ed up may
 have already been reused.  And most of use can't just rip the box out
 from under our users when ever we need to recover a file.

 On my home system this trick has come in handy on about three
 occasions in the past few years - usually when I accidentally trash
 some of the days work.  If what I'm working survives to a point where
 I feel I made significant progress, it get's backed up onto floppy, so
 I haven't needed this trick very often.



 3.3.  How to use the immutable flag. Jim Dennis,  [email protected]

 Use the Immutable Flag

 Right after you install and configure your system go through the /bin,
 /sbin/, /usr/bin, /usr/sbin and /usr/lib (and a few of the other usual
 suspects and make liberal use of the 'chattr +i command'.  Also add
 that to the the kernel files in root.  Now 'mkdir /etc/.dist/' copy
 everything from /etc/ on down (I do this in two steps using
 /tmp/etcdist.tar to avoid recursion) into that directory.  (Optionally
 you can just create /etc/.dist.tar.gz) -- and mark that as immutable.

 The reason for all of this is to limit the damage that you can do when
 logged in as root.  You won't overwrite files with a stray redirection
 operator, and you won't make the system unusable with a stray space in
 an 'rm -fr' command (you might still do alot of damage to your data --
 but your libs and bins will be safer.

 This also makes a variety of security and denial of service exploits
 either impossible or more difficult (since many of them rely on
 overwriting a file through the actions of some SUID program that
 *isn't providing an arbitrary shell command*).

 The only inconvenience of this is when building and doing your 'make
 install' on various sorts of system binaries.  On the other hand it
 also prevents the 'make install' from over-writing the files.  When
 you forget to read the Makefile and chattr -i the files that are to be
 overwritten (and the directories to which you want to add files) --
 the make fails, you just use the chattr command and rerun it.  You can
 also take that opportunity to move your old bin's, libs, or whatever
 into a .old/ directory or rename or tar them or whatever.


 3.4.  Jim Dennis,  [email protected] A suggestion for where to put
 new stuff.

 All new stuff starts under /usr/local! or /usr/local/`hostname`

 If your distribution is one that leaves /usr/local empty then just
 create your /usr/local/src, /usr/local/bin etc and use that.  If your
 distribution puts things in the /usr/local tree than you may want to
 'mkdir /usr/local/`hostname`' and give the 'wheel' group +w to it (I
 also make it SUID and SGID to insure that each member of the wheel
 group can only mess with their own files thereunder, and that all
 files created will belong to the 'wheel' group.

 Now discipline yourself to *ALWAYS! ALWAYS! ALWAYS!* put new packages
 under /usr/local/src/.from/$WHEREVER_I_GOT_IT/ (for the .tar or
 whatever files) and build them  under /usr/local/src (or
 .../$HOSTNAME/src).  Make sure that it installs under the local
 hierarchy.  If it *absolutely must* be installed back in /bin or
 /usr/bin or somewhere else -- put a symlink from the local heirarchy
 to each element that when anywhere else.

 The reason for this -- even though it's more work -- is that it helps
 isolate what has to be backed up and restored or reinstalled in the
 event of a full re-install from the distribution medio (usually CD
 these days).  By using a /usr/local/.from directory you also keep an
 informal log of where your sources are coming from -- which helps when
 you're looking for new updates -- and may be critical when monitoring
 the security announcement lists.

 One of my systems at home (the one I'm calling from) was put together
 before I adopted these policies for myself.  I still don't "know" all
 the ways that it differs from the stock "as installed" system.  This
 is despite the fact that I've done very little with my home system's
 configuration and I'm the *only* person who ever uses it.

 By contrast the systems I've set up at work (when I was thrust into
 the role of system administrator there) have all been configured this
 way -- have been administered by many contractors and other MIS
 people, and have had a large number of upgrades and package
 installations.  Nonetheless I have a very good idea which precise
 elements were put in *after* the initial installation and
 configuration.


 3.5.  Converting all files in a directory to lowercase. Justin Dossey,
 [email protected]


 I noticed a few overly difficult or unnecessary procedures recommended
 in the 2c tips section of Issue 12.  Since there is more than one, I'm
 sending it to you:


      ______________________________________________________________________
      #!/bin/sh
               # lowerit
               # convert all file names in the current directory to lower case
               # only operates on plain files--does not change the name of directories
               # will ask for verification before overwriting an existing file
               for x in `ls`
                 do
                 if [ ! -f $x ]; then
                   continue
                   fi
                 lc=`echo $x  | tr '[A-Z]' '[a-z]'`
                 if [ $lc != $x ]; then
                   mv -i $x $lc
                 fi
                 done
      ______________________________________________________________________



 Wow.  That's a long script.  I wouldn't write a script to do that;
 instead, I would use this command:


      for i in * ; do [ -f $i ] && mv -i $i `echo $i | tr '[A-Z]' '[a-z]'`;
      done;



 on the command line.

 The contributor says he wrote the script how he did for
 understandability (see below).

 On the next tip, this one about adding and removing users, Geoff is
 doing fine until that last step.  Reboot?  Boy, I hope he doesn't
 reboot every time he removes a user.  All you have to do is the first
 two steps.  What sort of processes would that user have going, anyway?
 An irc bot?  Killing the processes with a simple


      kill -9 `ps -aux |grep ^<username> |tr -s " " |cut -d " " -f2`



 Example, username is foo


      kill -9 `ps -aux |grep ^foo |tr -s " " |cut -d " " -f2`



 That taken care of, let us move to the forgotten root password.

 The solution given in the Gazette is the most universal one, but not
 the easiest one.  With both LILO and loadlin, one may provide the boot
 parameter "single" to boot directly into the default shell with no
 login or password prompt.  From there, one may change or remove any
 passwords before typing "init 3" to start multiuser mode.  Number of
 reboots: 1 The other way Number of reboots: 2


 Justin Dossey



 3.6.  Paul Anderson,  [email protected] How To Upgrade Sendmail

 We're starting from raw, clean source.  First, obtain the sendmail
 source code.  I've d/led version 8.9.0, which is, as you will notice,
 bleeding edge.  I grabbed it from
 ftp.sendmail.org:/pub/sendmail/sendmail.8.9.0.tar.gz

 It's about 1Meg, and considering I'm running 8.7.6, I think it's worth
 the effort.  If this works, you'll undoubtedly hear about it, elsewise
 I can't get the new HOWTO versions out without e-mail:)

 Now, once you've got the source d/led, unpack it.  It'll create a dir
 called sendmail-8.9.0 in the current directory.  Change into that
 directory, read the README and RELEASE_NOTES files(and be amazed at
 the updates they've done).  Now, cd in src.  This is where most of
 your work will be done.

 A quick note: Sendmail is a small, powerful and well-written program.
 The sendmail binary itself compiled in less than 5 minutes on my 5x86
 133 with 32Megs RAM!  The entire compile and install(sans config) took
 under 15 minutes!

 I don't normally run BIND on my system, so I found the lines:



      ______________________________________________________________________
      # ifndef NAMED_BIND
      #  define NAMED_BIND    1       /* use Berkeley Internet Domain Server */
      # endif
      ______________________________________________________________________



 and changed the 1 to a 0, ala:



      ______________________________________________________________________
      # ifndef NAMED_BIND
      #  define NAMED_BIND    0       /* use Berkeley Internet Domain Server */
      # endif
      ______________________________________________________________________



 On Debian 1.3.1, db.h is by default installed in /usr/include/db,
 instead of /usr/include, where sendmail hopes to find it.  Change to
 the src, mailstats, makemap, praliases, rmail and smrsh directories
 and execute the following command:



       ./Build -I/usr/include/db



 Once you've done that, cd .. and type make install.  There!  Sendmail
 version 8.9.0 should now be installed!  This is, of course, assuming
 you already have your original configuration.  For everything to work
 smoothly on my system, since I host free mailing lists for people
 using majordomo, I had to add the following to the beginning of my
 /etc/sendmail.cf:



      ______________________________________________________________________
      O DontBlameSendmail=forwardfileinunsafedirpath, forwardfileinunsafedirpathsafe
      ______________________________________________________________________



 Sendmail 8.9.0 is rather pedantic about directory and file permissions
 these days, and will complain about dirs and files in aliases or
 .forward files that are group or world writeable.  While it's not a
 good idea to disable this pedantry, I am only running with a single
 person at the console and I felt it was okay to allow this minor
 security hole.  YMMV.



 3.7.  Jim Dennis,  [email protected] Some tips for new sysadmins.

 Create and maintain a /README.`hostname` and/or a
 /etc/README.`hostname` [Or possibly /usr/local/etc/README.`hostname`
 -Maint. ]

 Absolutely, from *day one* of administering a system take notes in an
 online log file.  You might make "vi /README.$(hostname)" a line in
 root's  /bash_logout.  Another way to do this is to write an su or a
 sudo script that does something like:



                 function exit \
                         { unset exit; exit; \
                           cat ~/tmp/session.$(date +%y%m%d) \
                           >> /README.$(hostname) && \
                           vi /README.$(hostname)
                           }
                 script -a ~/tmp/session.$(date +%y%m%d)
                 /bin/su.org -



 (use the typescript command to create a session log and create a
 function to automate appending and updating the log).

 I'll admit that I haven't implemented this automation of policy --
 I've just relied on self-discipline so far.  However I have been
 toying with the idea (even to the point of prototyping the scripts and
 shell functions as you see them).  One thing that holds me back on
 this is the 'script' command itself.  I think I'll have to grab the
 sources and add a couple of command line parameters (to pause/stop the
 script recording from the command line) before I commit to using
 this).

 My last suggestion (for this round):

 Root's path should consist of 'PATH= /bin'

 That's it.  Nothing else on root's path.  Everything root does is
 provided by a symlink from  /bin or by an alias or shell function, or
 is a script or binary in  /bin, or is typed out with an explicit path.

 This makes anyone running as root aware (sometimes painfully so) of
 how he or she is trusting binaries.  The wise admin of a multi-user
 host will periodically look through his or here  /bin and  /.*history
 files to look for patterns and loopholes.

 The really motivated admin will spot sequences that can be automated,
 places where sanity checks can be inserted, and tasks for which "root"
 privileges should be temporarily eschewed (launching editors, MTA's
 and other large interactive programs with elaborate scripting features
 that *might* be embedded in transparent or data files -- like the
 infamous vi ./.exrc and emacs ./.emacs and the even more insidous
 $EXINIT and the embedded header/footer macros).  Naturally those sorts
 of commands can be run with something like:


                      cp $data $some_users_home/tmp
                      su -c $origcommand $whatever_switches
                      cp $some_users_home/tmp $data



 (...where the specifics depend on the command).

 Mostly these last sorts of precautions are overboard for the home or
 "single" user workstation -- but they are very good policy the admin
 of a multi-user -- particular a publicly exposed system (like the
 one's at netcom).



 3.8.  How to configure xdm's chooser for host selection. Arrigo Tri-
 ulzi,  [email protected]


 1. Edit the file that launches xdm most likely /etc/rc/rc.6 or
    /etc/rc.local) so that it contains the following lines in the xdm
    startup section.



      /usr/bin/X11/xdm
      exec /usr/bin/X11/X -indirect hostname



 2. Edit /usr/lib/X11/xdm/Xservers and comment out the line which
    starts the server on the local machine (i.e. starting 0:)

 3. Reboot the machine and you're home and away.

 I add this because when I was, desperately, trying to set it up for my
 own subnet over here it took me about a week to suss out all the
 problems.

 Caveat: with old SLS (1.1.1) for some reason you can leave a -nodaemon
 after the xdm line -- this does NOT work for later releases.