#!/usr/bin/perl -w =head1 NAME xen-create-image - Create a new virtual Debian installation for Xen. =head1 SYNOPSIS xen-create-image [options] Help Options: --debug Show useful debugging information. --help Show this scripts help information. --manual Read this scripts manual. --version Show the version number and exit. Size / General options: --boot Boot the new instance after creating it. --cache Cache .deb upon the host for speed improvement. --debootstrap Pass anything named here onto debootstrap. --dist Specify the distribution you wish to install: Sarge/Etch/Sid. --fs Specify the filesystem type to use. --kernel Set the path to the kernel to use for dom U. --memory Setup the amount of memory allocated to the instance. --mirror Setup the mirror to use when installing Sarge. --passwd Ask for a root password during setup. --role Run a role-specific script, post-install. --size Set the size of the primary disk image. --swap Set the size of the swap partition. --ide Use IDE names for virtual devices (hda not sda) Networking options: --dhcp Setup the image to get an IP address via DHCP --gateway Setup the iamge's network gateway. --ip Setup the ip --netmask Setup the netmask Mandatory options: --dir Specify where the output images should go. --volume Specify which LVM volume to save images within. --hostname Set the images hostname. =cut =head1 OPTIONS =over 8 =item B<--boot> Start the new virtual instance as soon as the installation has finished. =item B<--cache> Cache the .deb files in /var/cache/apt/archives upon the host system for a large speed improvement. [Defaults to yes.] =item B<--debug> Show the commands this script executes as an aid to debugging. =item B<--debootstrap> Anything specified after this will be passed onto the debootstrap command executed. =item B<--dhcp> Specify that the virtual image should use DHCP to obtain its networking information. Conflicts with B<--ip>. =item B<--dir> Specify the root directory beneath which the image should be saved. Subdirectories will be created for each virtual image. If you do not wish to use loopback files instead specify an LVM volume with B<--volume>. =item B<--dist> Specify the distribution to install, defaults to 'sarge'. =item B<--fs> Specify the filesystem the image should be given. Valid options are 'ext3', 'xfs', or 'reiserfs'. =item B<--gateway> Specify the gateway address for the virtual image, only useful if DHCP is not used. =item B<--help> Show the brief help information. =item B<--hostname> Set the hostname of the new instance. =item B<--ip> Set the IP address for the virtual image. Conflicts with B<--dhcp> =item B<--kernel> Set the path to the kernel to use for the image. =item B<--manual> Read the manual, with examples. =item B<--memory> Specify the amount of memory the virtual image should be allocated. Defaults to 96Mb. =item B<--mirror> Specify the mirror to use to the installation of Sarge, defaults to http://ftp.us.debian.org/debian =item B<--netmask> Set the netmask the virtual image should use. =item B<--passwd> Setup a password for the root account of the virtual machine. =item B<--role> Run a single, specific, role-script once the image has been setup. =item B<--size> Specify the size of the primary drive to give the virtual image. The size may be suffixed with either Mb, or Gb. =item B<--swap> Specify the size of the virtual swap partition to create. The size may be suffixed with either Mb, or Gb. =item B<--ide> Use IDE style device names for the virtual devices. =item B<--version> Show the version number and exit. =item B<--volume> Show the LVM volume to store images within. If you wish to use loopback files please specify an output directory with B<--dir>. =back =cut =head1 LOOPBACK EXAMPLES The following will create a 2Gb disk image, along with a 128Mb swap file with Debian Sarge setup and running via DHCP. xen-create-image --size=2Gb --swap=128Mb --dhcp \ --dir=/home/xen --hostname=vm01.my.flat This next example sets up a host which has the name 'vm02' and IP address 192.168.1.200, with the gateway address of 192.168.1.1 xen-create-image --size=2Gb --swap=128Mb \ --ip=192.168.1.200 --netmask=255.255.255.0 --gateway=192.168.1.1 \ --dir=/home/xen --hostname=vm02 To save typing these command line options may be specified in the configuration file discussed later. The directory specified for the output will be used to store the files which are produced. To avoid clutter each host will have its images stored beneath the specified directory, named after the hostname. For example the images created above will be stored as: $dir/domains/vm01.my.flat/ $dir/domains/vm01.my.flat/disk.img $dir/domains/vm01.my.flat/swap.img $dir/domains/vm02.my.flat/ $dir/domains/vm02.my.flat/disk.img $dir/domains/vm02.my.flat/swap.img The '/domains/' subdirectory will be created if necessary. =cut =head1 LVM EXAMPLES If you wish to use an LVM volume instead of a pair of loopback images as shown above you can instead use the B<--volume> argument to specify the volume to create the image within. xen-create-image --size=2Gb --swap=128Mb --dhcp \ --volume=myvolume --hostname=vm01.my.flat The given volume will have two new entries created within it: ${hostname}-swap ${hostname}-root =cut =head1 DESCRIPTION xen-create-image is a simple script which allows you to create new Xen instances of Debian GNU/Linux. The new image will be comprised of two seperate images: 1. An image for the systems root disk. 2. An image for the systems swap device. These images will either be created as new LVM volumes, or as local loopback files. The new virtual Debian image will be configured with networking, have OpenSSH installed upon it, and have most of its basic files setup correctly. =head1 CONFIGURATION To reduce the length of the command line each of the options may be specified inside a configuration file. The script will check a global configuration file for options, the file is: /etc/xen-tools/xen-tools.conf The configuration file may contain comments which begin with the hash '#' character. Otherwise the format is 'key = value'. A sample configuration file would look like this: =for example begin # # Output directory. Images are stored beneath this directory, one # subdirectory per hostname. # dir = /home/xen # # LVM users should disable the 'dir' setting above, and instead # specify the name of the volume to use. # # volume = myvolume # # Disk and Sizing options. # size = 2Gb # Disk image size. memory = 128Mb # Memory size swap = 128Mb # Swap size fs = ext3 # use EXT3 filesystems dist = sarge # Default distribution to install. # # Kernel options. # kernel = /boot/vmlinuz-2.6.12-xenU # # Networking options. # gateway = 192.168.1.1 netmask = 255.255.255.0 =for example end Using this configuration file a new image may be created with the command: xen-create-image --hostname='vm03.my.flat' --ip=192.168.1.201 =head1 CACHING Because the virtual systems are installed with the debootstrap tool there can be a lot of network overhead. To minimize this the .deb files which are downloaded into the new instance are cached by default upon the host, in the directory /var/cache/apt/archives. When a new image is created these packages are copied into the new image - before the debootstrap process runs - this should help avoid expensive network reading. If you wish to clean the cache upon the host you may do so with apt-get, as you'd expect: apt-get clean (This feature can be disabled with the command line flag --cache=no, or the configuration file setting 'cache = no'.) =head1 CUSTOMIZATION There are three different types of customization which are available to you. These allow you to change the generated images in several ways. The available options are: =over 8 =item debootstrap customization This allows you to make additions to the debootstrap command which is executed to create the image. =item the hook directory The hook scripts allow you to run customization scripts for each generated image. =item the role directory The role directory allows you to customise a group of images in a special manner. =back =cut =head1 DEBOOTSTRAP CUSTOMIZATION If you wish to add new packages to the image automatically you may take advantage of the '--debootstrap' option which allows you to pass flags to the debootstrap command. For example the following command causes three new packages to be added to the base image: xen-create-image --debootstrap='--include=screen,sudo,less' For more details of available options to use with debootstrap please see its manpage. =head1 HOOKS After the image has been installed using debootstrap there is the chance for you to run arbitary scripts upon the new host before it is unmounted. To do this place executable scripts inside the "hook directory" /etc/xen-tools/hook.d/. Each executable script in this directory will be executed in turn and given the name of the mount point the image is available at as a single command line argument. The scripts will also have a complete copy of the configuration options set in its environment. For example the IP address would be set in ${ip}, the mirror in ${mirror}, etc. A script could copy some the kernel modules to the new system, and install a package, for example: =for example start #!/bin/sh prefix=$1 # Copy modules mkdir -p ${prefix}/lib/modules cp -R /lib/modules/2.6.12.6-xen/ ${prefix}/lib/modules # Install the package 'module-init-tools' DEBIAN_FRONTEND=noninteractive chroot $prefix /usr/bin/apt-get --yes --force-yes install module-init-tools =for example cut The hook scripts are used extensively by this script to setup the base system. If you're looking at extending the processing of new images it is recommended you examine the default hooks. =cut =head1 ROLES The scripts which are included in the hook directory are executed for every single image you create. This can allow you to make global changes to the generated image very easily, however it doesn't allow you to do different things for different types of images. Instead the "role" scripts are used for that purpose. If you give an argument "--role=foo" then the script /etc/xen-tools/role.d/foo will be executed once the image has been created, after the hooks have run. This allows you to create different customizations for particular image types. Several role scripts have been included to provide examples of different customizations: =over 8 =item builder Setup the new virtual images with commonly used packages for rebuilding Debian packages from source. =item gdm Install an X11 server, using VNC and GDM =item minimal Customise the generated images to remove some packages. =item xdm Install an X11 server, using VNC and XDM =back =head1 AUTHOR Steve -- http://www.steve.org.uk/ $Id: xen-create-image,v 1.102 2006-03-08 18:37:01 radu Exp $ =cut =head1 CONTRIBUTORS Contributors to this code: =over 8 =item Radu Spineanu =item Justin Azoff =back =head1 LICENSE Copyright (c) 2005 by Steve Kemp. All rights reserved. This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself. The LICENSE file contains the full text of the license. =cut use strict; use diagnostics; use English; use File::Copy; use File::Temp qw/ tempdir /; use Getopt::Long; use IPC::Open3; use Pod::Usage; # # Global configuration options. # # Initially our options are read from the configuration file into this # hash. Later they may be overridden by the command line. # # Command line flags *always* take precedence over the configuration files. # # my %CONFIG; # # Release number. # my $RELEASE = '1.1'; # # The dimensions of the current terminal, used for line-wrapping. # my ( $TERMINAL_WIDTH, $TERMINAL_HEIGHT ) = getTerminalSize(); # # These hashes contain information used for the creation of different # fileystems. # my %FILESYSTEM_BINARY; my %FILESYSTEM_CREATE; my %FILESYSTEM_MOUNT; # # The program to run to create a filesystem - used in the next hash. # $FILESYSTEM_BINARY{'ext3'} = '/sbin/mkfs.ext3'; $FILESYSTEM_BINARY{'xfs'} = '/sbin/mkfs.xfs'; $FILESYSTEM_BINARY{'reiserfs'} = '/sbin/mkfs.reiserfs'; # # The command to use to create a filesystem. The disk image # filename is appended to these commands to generate what is # ultimately executed. # $FILESYSTEM_CREATE{'ext3'} = $FILESYSTEM_BINARY{'ext3'}. ' -F '; $FILESYSTEM_CREATE{'xfs'} = $FILESYSTEM_BINARY{'xfs'}. ' -d name='; $FILESYSTEM_CREATE{'reiserfs'} = $FILESYSTEM_BINARY{'reiserfs'}. ' -f -q '; # # Flags to pass to "mount" to mount our image. Kinda redundent and may # go away - seems to me that just using '-t $CONFIG{'fs'}' is sufficient. # $FILESYSTEM_MOUNT{'ext3'} = '-t ext3'; $FILESYSTEM_MOUNT{'xfs'} = '-t xfs'; $FILESYSTEM_MOUNT{'reiserfs'} = '-t reiserfs'; # # Setup defaults: # # Memory = 96M, Image = 2000Mb, Swap = 128Mb, and filesystem is ext3. # # These may be overriden by settings in the configuration file or by # command line arguments. # $CONFIG{'memory'} = '96Mb'; $CONFIG{'size'} = '2000Mb'; $CONFIG{'swap'} = '128M'; $CONFIG{'fs'} = 'ext3'; $CONFIG{'mirror'} = 'http://ftp.us.debian.org/debian'; $CONFIG{'dist'} = 'sarge'; $CONFIG{'xm'} = '/usr/sbin/xm'; $CONFIG{'kernel'} = '/boot/vmlinuz-2.6.12-xenU'; $CONFIG{'debootstrap'} = ''; $CONFIG{'hook_dir'} = '/etc/xen-tools/hook.d/'; $CONFIG{'role_dir'} = '/etc/xen-tools/role.d/'; $CONFIG{'cache'} = 'yes'; # # Read the global configuration file if it exists. # if ( -e "/etc/xen-tools/xen-tools.conf" ) { readConfigurationFile( "/etc/xen-tools/xen-tools.conf" ); } # # Parse command line arguments. These override the values from the # configuration file. # parseCommandLineArguments(); # # Check that the arguments the user has supplied are both # valid, and complete. # checkArguments(); # # Avoid non-root users after this point. We allowed them up till # now so they could read the manual, see help/version info, etc. # if ( $EFFECTIVE_USER_ID != 0 ) { print < 1 ); my $mount_cmd; if ( $CONFIG{'volume'} ) { $mount_cmd = "mount " . $FILESYSTEM_MOUNT{lc($CONFIG{'fs'})} . " $image $dir"; } else { $mount_cmd = "mount " . $FILESYSTEM_MOUNT{lc($CONFIG{'fs'})} . " -o loop $image $dir"; } runCommand( $mount_cmd ); # # Test that the mount worked # my $mount = runCommand( "/bin/mount" ); if ( ! $mount =~ /$image/) { print "Something went wrong trying to mount the new filesystem\n"; exit; } # # Copy any local .deb files into the right location on the virtual image # so that "debootstrap" will use them - rather than fetching from the network. # # This is a big caching speedup. # if ( $CONFIG{'cache'} eq "yes" ) { print "\nCopying files from host to image.\n"; runCommand( "mkdir -p $dir/var/cache/apt/archives" ); copyDebFiles( "/var/cache/apt/archives", "$dir/var/cache/apt/archives" ); printWideMessage( "\rDone" ); } # # Install the base system - with a simple sense of progress. # print "\n\nRunning debootstrap to install the system. This will take a while!\n"; my $debootstrap = "debootstrap $CONFIG{'debootstrap'} $CONFIG{'dist'} $dir $CONFIG{'mirror'}"; runCommandWithProgress( $debootstrap ); # # Now run any hook scripts post-install. These will be the ones # we ship to setup the new image, or any local additions. # if ( -d $CONFIG{'hook_dir'} ) { print "\n\nRunning hooks from : $CONFIG{'hook_dir'}\n"; runHooks( $CONFIG{'hook_dir'}, $dir ); print "Done\n"; } # # If we have a role to execute then do that now, after the hooks # have finished. # if ( $CONFIG{'role'} ) { print "\n\nRunning role command : $CONFIG{'role'}\n"; my $role_cmd = $CONFIG{'role_dir'} . "/" . $CONFIG{'role'}; runCommand( $role_cmd . " " . $dir ); print "Done\n"; } # # Copy the newly installed files from the virtual image to the host, # these will then be copied back the next time an image is created. # # Big win. # # NOTE: We do this after running any hook or role scripts, to ensure # that any packages installed by either may be cached successfully. # if ( $CONFIG{'cache'} eq "yes" ) { print "\n\nCaching debootstrap files to the host system\n"; copyDebFiles( "$dir/var/cache/apt/archives", "/var/cache/apt/archives/" ); printWideMessage( "\rDone" ); } # # We can not run this through hooks since it's interactive # if ( $CONFIG{'passwd'} ) { print "\nSetting root passwd:\n"; my $passwd_cmd = `chroot $dir /usr/bin/passwd`; } # # Unmount the image; we're done. # runCommand( "umount $dir" ); # # Clear screen before printing success message, unless # we're running under '--debug'. # $TERMINAL_HEIGHT *= 2; while( $TERMINAL_HEIGHT ) { printWideMessage( "\n" ) unless $CONFIG{'debug'}; $TERMINAL_HEIGHT -= 1; } # # Give status message # if ( $CONFIG{'boot'} ) { # # Should we immediately start the new instance? # If so fork() and do it so that we can return to the user, they can # attach to the console via the command : 'xm console $name'. # # print "\n\nBooting newly created virtual image: $CONFIG{'hostname'} in the background.\n"; print "\nTo attach to the console run (as root):\n\n"; print "\t\txm console $CONFIG{'hostname'}\n\n"; print "\n"; my $pid = fork(); if ( $pid ) { exit; } else { # # Start Xen instance in our forked process. Avoid annoying # output. # # system( "$CONFIG{'xm'} create $CONFIG{'hostname'}.cfg >/dev/null 2>/dev/null" ); } } else { print <) ) { chomp $line; if ($line =~ s/\\$//) { $line .= ; redo unless eof(FILE); } # Skip lines beginning with comments next if ( $line =~ /^([ \t]*)\#/ ); # Skip blank lines next if ( length( $line ) < 1 ); # Strip trailing comments. if ( $line =~ /(.*)\#(.*)/ ) { $line = $1; } # Find variable settings if ( $line =~ /([^=]+)=([^\n]+)/ ) { my $key = $1; my $val = $2; # Strip leading and trailing whitespace. $key =~ s/^\s+//; $key =~ s/\s+$//; $val =~ s/^\s+//; $val =~ s/\s+$//; # Store value. $CONFIG{ $key } = $val; } } close( FILE ); } =head2 parseCommandLineArguments Parse the arguments specified upon the command line. =cut sub parseCommandLineArguments { my $HELP = 0; my $MANUAL = 0; my $VERSION = 0; # Parse options. # GetOptions( "hostname=s", \$CONFIG{'hostname'}, "ip=s", \$CONFIG{'ip'}, "gateway=s", \$CONFIG{'gateway'}, "netmask=s", \$CONFIG{'netmask'}, "dir=s", \$CONFIG{'dir'}, "volume=s", \$CONFIG{'volume'}, "dhcp", \$CONFIG{'dhcp'}, "mirror=s", \$CONFIG{'mirror'}, "size=s", \$CONFIG{'size'}, "swap=s", \$CONFIG{'swap'}, "memory=s", \$CONFIG{'memory'}, "fs=s", \$CONFIG{'fs'}, "boot", \$CONFIG{'boot'}, "dist=s", \$CONFIG{'dist'}, "debootstrap=s",\$CONFIG{'debootstrap'}, "debug" , \$CONFIG{'debug'}, "kernel=s", \$CONFIG{'kernel'}, "passwd", \$CONFIG{'passwd'}, "role=s", \$CONFIG{'role'}, "cache=s", \$CONFIG{'cache'}, "ide", \$CONFIG{'ide'}, "help", \$HELP, "manual", \$MANUAL, "version", \$VERSION ); pod2usage(1) if $HELP; pod2usage(-verbose => 2 ) if $MANUAL; if ( $VERSION ) { my $REVISION = '$Revision: 1.102 $'; if ( $REVISION =~ /1.([0-9.]+) / ) { $REVISION = $1; } print "xen-create-image release $RELEASE - CVS: $REVISION\n"; exit; } } =head2 checkArguments Check that the arguments the user has specified are complete and make sense. =cut sub checkArguments { if (!defined( $CONFIG{'hostname'} ) ) { print< Mb for the disk image size, and swap size. # if ( $CONFIG{'size'} =~ /^(\d+)Gb*$/i ) { $CONFIG{'size'} = $1 * 1024 . "M"; } if ( $CONFIG{'swap'} =~ /^(\d+)Gb*$/i ) { $CONFIG{'swap'} = $1 * 1024 . "M"; } # # Strip trailing Mb from the memory size. # if ( $CONFIG{'memory'} =~ /^(\d+)Mb*$/i ) { $CONFIG{'memory'} = $1; } # # Convert the image size to k. # # Fails for LVM ? # if ( $CONFIG{'dir'} ) { $CONFIG{'size'} =~ s/Mb*$/k/i; } # # Now strip the trailing 'Mb' from the swap size. # if ( $CONFIG{'swap'} =~ /^(\d+)Mb*$/i ) { $CONFIG{'swap'} = $1; } # # Check mirror format # if (!($CONFIG{'mirror'} =~ /^http/i)) { print "Please enter a valid mirror.\n"; exit; } # # Only one of DHCP / IP is required. # if ( $CONFIG{'ip'} && $CONFIG{'dhcp'}) { print "You've chosen both DHCP and an IP address.\n"; print "Only one is supported\n"; exit; } if ( $CONFIG{'dhcp'} ) { $CONFIG{'gateway'} = ''; $CONFIG{'netmask'} = ''; $CONFIG{'ip'} = ''; } # # Check --cache option # if ( ($CONFIG{'cache'} ne 'yes') && ($CONFIG{'cache'} ne 'no') ) { print "Valid options for --cache are yes/no.\n"; exit; } # # If the user is executing a role-script make sure it exists. # if ( $CONFIG{'role'} ) { if ( ! -e $CONFIG{'role_dir'} . "/" . $CONFIG{'role'} ) { print "The role you've chosen '$CONFIG{'role'}' doesn't have a script.\n"; print "Available roles are:\n\n"; foreach my $file ( glob $CONFIG{'role_dir'} . "/*" ) { if ( $file =~ /(.*)\/(.*)/ ) { $file = $2; } print "\t" . $file . "\n"; } exit; } } } =head2 printWideMessage Print a message, ensuring the width is as wide as the console. =cut sub printWideMessage { my ( $msg ) = ( @_ ); while( length( $msg ) < $TERMINAL_WIDTH ) { $msg .= " "; } print $msg; } =head2 runCommand Run a command, and if debugging is turned on then display it to the user along with output. Otherwise hide all output. =cut sub runCommand { my ( $cmd ) = ( @_ ); # # Header. # $CONFIG{'debug'} && print "Executing : $cmd\n"; # # Hide output unless running with --debug. # $cmd .= " >/dev/null 2>/dev/null" unless $CONFIG{'debug'}; # # Run it. # my $output = `$cmd`; # # All done. # $CONFIG{'debug'} && print "Finished : $cmd\n"; return( $output ); } =head2 runCommandWithProgress Run a command whilst immediately writing the output to the console. This is a cheap hack to give a sense of 'progress'. =cut sub runCommandWithProgress { my ( $cmd ) = ( @_ ); $CONFIG{'debug'} && print "Executing : $cmd\n"; my $pid = open3(undef, \*READ,0, $cmd ); # # A failure to run debootstrap is pretty much fatal to us. # # Since without it there will be no installed filesystem. # # Abort, after unmounting the directory we're using. # if ( ! $pid ) { print "Error executing command : '$cmd' - $!"; runCommand( "umount $dir" ); exit; } my $output =''; while(1) { # # Wait for input. # select(undef,undef,undef,.01); # # Read output from the command, max 1k. # if( sysread \*READ,$output,1024 ) { # # Remove newlines to avoid weirdness. $output =~ s/\n//g; # # If there is more output than our terminal width then # truncate it. # if ( length( $output ) > $TERMINAL_WIDTH ) { $output = substr( $output, 0, ( $TERMINAL_WIDTH - 5 ) ); } # # Pad to exactly terminal width. while( length( $output ) < $TERMINAL_WIDTH ) { $output .= " " ; } # # Now rewind cursor to start of line and display # the text. # print STDERR "\r"; print STDERR $output; } else { # # command finished. # my $over = "\rFinished"; while( length( $over ) < $TERMINAL_WIDTH ) { $over .= " "; } print STDERR $over . "\n"; return; } } } =head2 getTerminalSize Find and return the size and width of the current terminal. This function makes use of the Term::Size module if available, if it is not installed then we return the standard size of 80x25. =cut sub getTerminalSize { my $testModule = "use Term::Size;"; my $width = 80; my $height = 25; # # Test loading the size module. If this fails # then we will use the defaults sizes. # eval( $testModule ); if ( $@ ) { } else { # # Term::Size is available, so use it to find # the current terminal size. # ($width, $height ) = Term::Size::chars(); } return( $width, $height ); } =head2 runHooks When the image has been created, but before the temporary image is unmounted, each executable script inside the hook directory will be executed. (The scripts are executed "in order" which might be useful for users who wish to ensure some actions occur before others.) The scripts will be given single argument: the name of the directory within which the image is mounted. The rest of the configuration variables will be passed via environmental variables. =cut sub runHooks { my ( $HOOK_DIR, $prefix ) = ( @_ ); # # Setup the environment for the child processes. # foreach my $key ( keys %CONFIG ) { if ( defined( $CONFIG{$key} ) ) { $ENV{$key} = $CONFIG{$key}; } } $ENV{'image_vdb'} = $image_vbd; $ENV{'swap_vdb'} = $swap_vbd; # # Make sure that our scripts run in sorted order, as # the user would expect. # foreach my $file ( sort( glob( $HOOK_DIR . "*" ) ) ) { # # Only run executable files. # if ( ( -x $file ) && ( -f $file ) ) { # # Just display the name - no need to see the full path. # my $name = $file; if ( $file =~ /(.*)\/(.*)/ ) { $name = $2; } print " Running hook: $name"; runCommand( $file . " " . $prefix ); print " - done.\n"; } } } =head2 copyDebFiles This function will copy all the .deb files from one directory to another, giving a "progress indicator" of sorts. =cut sub copyDebFiles { my ( $source, $dest ) = ( @_ ); my @files = glob( $source . "/*.deb" ); my $count = 1; my $total = $#files+1; foreach my $file ( sort @files ) { my $name = $file; if ( $name =~ /(.*)\/(.*)/ ) { $name = $2; } my $t = "\r[$count/$total] : $name"; # # Print the status message and do the copy. # printWideMessage( $t ); # # Only copy if the file doesn't exist. # if ( ! ( -e $dest . "/" . $name ) ) { File::Copy::cp( $file, $dest ); } $count += 1; } }