habe das berühmte Script "mysql-check" aus dem hiesigen "FAQ" bei mir laufen. Jetzt möchte ich, dass mir das aktuelle Backup täglich automatisch per Mail zugesandt wird.
Bei meiner Suche im Forum fand ich nur eine passende variante mit "mutt":
mutt -a /root/backup/mysql/databasebackup-$date.sql.bz2 -s Backup-MySQL-Database-$date email@domain.tld < /root/backup/mailbody.conf
Da ich aber nur "pine" installiert habe, möchte ich wissen, wie ich das Script entsprechend umstelle (auf "pine").
Oder kennt jemand eine andere (bessere) Methode für den Zweck?
Geht das auch mit anderen "Bordmitteln" des 1&1-Rootservers?
Habe hier in der "Suche" nichts passendes gefunden.
Kann mir jemand von euch vielleicht helfen? :roll:
==============================================
#!/bin/sh
# script to email files as attachments.
# ------------------------------------
# Additional documentation for this script, including a brief introdcution
# to MIME can be found at: http://home.clara.net/dwotton/unix/mail_files.htm
# Written: Dave Wotton, July 1998, (Cambridge UK)
# This script comes with no warranty or support. You are
# free to modify it as you wish, but please retain an
# acknowledgement of my original authorship.
# Usage: mail_files [-t] mailid [ -c mailid ] [ -s subject ] [ -f mailid ]
# [-n file_list] [-u file_list] [-b file_list] file_list
#
# -f : The mailid of the sender ( defaults to your userid )
# Only userids that have been defined as "trusted" in the sendmail
# config file can make use of the -f option. For non-trusted users
# any value specified by this parameter will be ignored by
# sendmail.
# -t : The mailid of the recipient. Mandatory, no default
# multiple mailids can be specified, separated by commas.
# -c : The mailid of any carbon-copy recipients. Optional.
# multiple mailids can be specified, separated by commas.
# -s : The subject string. Optional, default = "Not specified".
# Enclose in quotes.
# -n : no-encode: indicates a list of files which are NOT to be base64
# or uuencode encoded. Multiple files may be enclosed in double
# quotes. Usual wildcard notation can be used. This option is
# for completeness and can be omitted because the default action
# is not to encode the file-list.
# -b : base64 encoding: indicates a list of files which are to be
# base64 encoded. Multiple files may be enclosed in double quotes.
# Usual wildcard notation can be used.
# -u : uuencode encoding: indicates a list of files which are to be
# uuencode encoded. Multiple files may be enclosed in double
# quotes. Usual wildcard notation can be used.
# file_list : The list of files to send as attachments with no-encoding
# (same as -n option, but the file list does not need to be
# enclosed in quotes if more than one file specified).
# Usual wildcard notation can be used.
# The program will also prompt for text to be supplied on standard input
# as the main text of the message.
# eg.
# 1) mail_files Dave.Wotton -b file9.gif t*.htm < /dev/null
#
# email file9.gif as a base64 encoded attachment and the t*.htm
# files unencoded.
#
# 2) mail_files Dave.Wotton -s "my test" -b "file1.gif file2.gif"
# < /dev/null
#
# email file1.gif and file2.gif as base64 encoded attachments.
# The script makes use of perl's MIME package to perform the base-64
# encoding/decoding.
# Note that files destined for Windows environments should have a name of
# the form aaaa.bbb where aaaa is up to 8 characters long, and bbb is a
# 3 character sufix. The suffix determines which program is used to
# display/process the data at the remote end.
# Simple text files can be emailed unencoded. Binary files, or text files
# with long lines ( ie > 1000 chars ) should use the base64 or uuencode
# encoding procedures. Base64 is preferred because it is more universally
# supported. In particular, most PC mail-clients can automatically decode
# base64 encoded attachments. Note that simple text files with short lines
# which are destined for PC environments should not be base64 encoded.
# This is because PCs use a different line-break character to Unix.
# If the text is base64 encoded, the line-breaks are not converted
# automatically and so the data arrives at the remote end without
# line-breaks.
# set up a 'usage' routine
# ------------------------
usage()
{
[ "$1" ] && ( echo $* ; echo "" )
cat <<!
Usage: mail_files [-t] mailid [ -c mailid ] [ -s subject ] [ -f mailid ]
[-n file_list] [-u file_list] [-b file_list] file_list
!
exit 4
}
# Initialise main variables ...
# -------------------------
FROM=$LOGNAME
SUBJ=${SUBJ:-"Not specified"}
TO="" ; CC="" ; SUBJ="" ; NOENC="" ; BASE64="" ; UUE=""
# First parse the command line options. Using getopts means the parameters
# can be supplied in any order. But first we handle the first parameter,
# which may be a recipient, without a -t flag...
case "$1" in
-* ) : ;; # ignore it, let getopts handle flags
* ) TO=$1 ; shift ;;
esac
while getopts f:s:t:c:n:b:u: OPT
do
case $OPT in
"f" ) FROM=$OPTARG ;;
"t" ) TO="$TO,$OPTARG" ;;
"c" ) CC="$CC,$OPTARG" ;;
"s" ) SUBJ=$OPTARG ;;
"n" ) NOENC="$NOENC $OPTARG" ;;
"b" ) BASE64="$BASE64 $OPTARG" ;;
"u" ) UUE="$UUE $OPTARG" ;;
* ) usage ;;
esac
done
shift `expr $OPTIND - 1`
if [ "$TO" = "" ]
then
usage "An addressee must be specified"
fi
# All remaining parameters are files not requiring encoding ...
# ---------------------------------------------------------
# Build up $FILES as the list of non-encoded files. Use sed to remove
# any leading space from the variable.
FILES=`echo $NOENC $*|sed 's/^ //'`
if [ "$BASE64" = "" -a "$FILES" = "" -a "$UUE" = "" ]
then
usage "At least one file must be specified"
fi
# Remove leading commas from TO, CC ...
# ---------------------------------
TO=`echo $TO | sed 's/^,//'`
CC=`echo $CC | sed 's/^,//'`
# Validate that the files exist ...
# -----------------------------
for F in $FILES $BASE64 $UUE
do
if [ ! -r $F ]
then
echo "Error: File $F does not exist / is not readable."
echo "Exiting. ( Mail not sent )."
exit
fi
done
tty -s && echo "Enter text of main message ( finish with CTRL-D ) ..."
# Now do the work ...
# ---------------
# The generated mail message is output onto standard out, which is then
# piped in to sendmail.
(
cat <<!
From: $FROM
Subject: $SUBJ
To: $TO
!
[ "$CC" ] && echo "Cc: $CC"
cat <<!
Mime-Version: 1.0
Content-Type: multipart/mixed; boundary="DMW.Boundary.605592468"
This is a Mime message, which your mail program may not understand. Parts
of the message will appear as text. If the remainder appears as random
characters in the message body, instead of as attachments, then you'll
have to extract these parts and decode them manually.
--DMW.Boundary.605592468
Content-Type: text/plain; name="message.txt"; charset=US-ASCII
Content-Disposition: inline; filename="message.txt"
Content-Transfer-Encoding: 7bit
!
# Read the standard input as the main text of the message ...
# -------------------------------------------------------
cat -
# Now process the non-encrypted attachments ...
# -----------------------------------------
if [ "$FILES" ]
then
for F in $FILES
do
BASE=`basename $F`
echo --DMW.Boundary.605592468
echo Content-Type: application/octet-stream; name="$BASE"
echo Content-Disposition: attachment; filename="$BASE"
echo Content-Transfer-Encoding: 7bit
echo
cat $F
done
fi
# Now process the base64 encrypted attachments ...
# --------------------------------------------
if [ "$BASE64" ]
then
for F in $BASE64
do
BASE=`basename $F`
echo --DMW.Boundary.605592468
echo Content-Type: application/octet-stream; name="$BASE"
echo Content-Disposition: attachment; filename="$BASE"
echo Content-Transfer-Encoding: base64
echo
perl -e '
use MIME::Base64 qw(encode_base64);
local($/) = undef;
print encode_base64(<STDIN> );' < $F
done
fi
# Now process the uuencode encrypted attachments ...
# ----------------------------------------------
# Sorry, this bit is untested - I haven't got a mail-client which can
# handle uuencoded MIME messages automatically, so can't test if the
# 'Content-Transfer-Encoding: uuencode' line is correct and whether I
# need the uuencode "begin" and "end" lines.
if [ "$UUE" ]
then
for F in $UUE
do
BASE=`basename $F`
echo --DMW.Boundary.605592468
echo Content-Type: application/octet-stream; name="$BASE"
echo Content-Disposition: attachment; filename="$BASE"
echo Content-Transfer-Encoding: uuencode
echo
uuencode < $F xxx
done
fi
# append the final boundary line ...
echo --DMW.Boundary.605592468--
) | /usr/lib/sendmail -t
vielen Dank für dein Script, dass ich jedoch leider nicht ganz kapiere :oops: , da ich in Sachen Linux noch wenig Erfahrungswissen habe. Wenn Du Lust und/oder Zeit hast, würde ich gerne noch ein paar Fragen zum Script loswerden. Meine Fragen:
1. Kann ich das Script einfach rauskopieren, ihm einen Namen geben und dann auf meinen Server kopieren?
2. Welche Anpassungen muß ich vornehmen?
3. Kann ich die erforderliche "Scriptzeile" an mein vorhandenes Backupscript anhängen (wie müßte die Zeile dann aussehen?), oder muß ich das Script separat laufen lassen?
Wie gesagt, wenn Du mir weiter helfen kannst (oder willst) wäre ich dir sehr dankbar. Falls nicht, trotzdem vielen Dank für deine Mühe.
