In a shell script I need to know the value of a variable further down in the code, without running through it first.
This pings $IP which is extracted from $VAR below the while loop.
The $VAR is unknown at the time this is extracted (IP=$(echo $VAR | awk '{print $1}'))
Is it possible to read VAR in before the while loop runs?
The code:
#!/bin/sh
TIMEOUT=10
IP=$(echo $VAR | awk '{print $1}')
while [ $TIMEOUT -ne 0 ];do
ping -c 1 -W 1 "$IP" >/dev/null
rc=$?
if [ $rc -eq 0 ];then
TIMEOUT=0
else
TIMEOUT=$(($TIMEOUT - 1))
echo $TIMEOUT
sleep 1
fi
done
# rest of code to run after while loop
VAR="192.168.0.1 t,r 20,e"
Related
I am working on an existing shell script code which has eval. I feel like that eval is unnecessary here and wanted to remove to avoid Injection.
Could you please check the code and advise why there is an eval in the code.
FILE_PATH=`echo $1 | awk '{ print $10 }' | cut -f2 -d'"'
FILE_PATH=`(eval "echo ${FILE_PATH}")`
if $1 is something like that ---"~/tttttttt.txt.
FILE_PATH will be ~/tttttttt.txt without eval.
but with eval;
FILE_PATH will be /home/user/tttttttt.txt
#!/bin/bash
path='-----"~/tttttttt.txt'
FILE_PATH=`echo $path | awk '{ print $1 }' | cut -f2 -d'"'`
echo "${FILE_PATH}"
ls -lart ${FILE_PATH}
FILE_PATH=`(eval "echo ${FILE_PATH}")`
echo $FILE_PATH
ls -lart ${FILE_PATH}
if run above script, output:
~/tttttttt.txt
ls: cannot access ~/tttttttt.txt: No such file or directory
/home/user/tttttttt.txt
-rw-rw-r-- 1 user user 0 Aug 26 15:54 /home/user/tttttttt.txt
I'm trying to configure tomcat init.d start script to work with multiple instances (at this time 2 instances)
I'm following below sample script to create init.d script
#!/bin/bash
#
# tomcat This shell script takes care of starting and stopping Tomcat
#
# chkconfig: - 80 20
#
### BEGIN INIT INFO
# Provides: tomcat
# Required-Start: $network $syslog
# Required-Stop: $network $syslog
# Default-Start:
# Default-Stop:
# Short-Description: start and stop tomcat
### END INIT INFO
TOMCAT_USER=root
TOMCAT_HOME="/opt/tomcat7/node1"
SHUTDOWN_WAIT=45
tomcat_pid() {
echo `ps aux | grep org.apache.catalina.startup.Bootstrap | grep -v grep | awk '{ print $2 }'`
}
start() {
pid=$(tomcat_pid)
if [ -n "$pid" ]
then
echo "Tomcat is already running (pid: $pid)"
else
# Start tomcat
echo "Starting tomcat service"
/bin/su - -c "cd $TOMCAT_HOME/bin && $TOMCAT_HOME/bin/startup.sh" $TOMCAT_USER
fi
return 0
}
stop() {
pid=$(tomcat_pid)
if [ -n "$pid" ]
then
echo "Stoping Tomcat"
/bin/su - -c "cd $TOMCAT_HOME/bin && $TOMCAT_HOME/bin/shutdown.sh" $TOMCAT_USER
let kwait=$SHUTDOWN_WAIT
count=0
count_by=5
until [ `ps -p $pid | grep -c $pid` = '0' ] || [ $count -gt $kwait ]
do
echo "Waiting for processes to exit. Timeout before we kill the pid: ${count}/${kwait}"
sleep $count_by
let count=$count+$count_by;
done
if [ $count -gt $kwait ]; then
echo "Killing processes which didn't stop after $SHUTDOWN_WAIT seconds"
kill -9 $pid
fi
else
echo "Tomcat is not running"
fi
return 0
}
case $1 in
start)
start
;;
stop)
stop
;;
restart)
stop
start
;;
status)
pid=$(tomcat_pid)
if [ -n "$pid" ]
then
echo "Tomcat is running with pid: $pid"
else
echo "Tomcat is not running"
fi
;;
esac
exit 0
problem is tomcat_pid() method returns process ids of all tomcat instances, because of that, the second instance cannot be started. Is there a better method to handle this?
found a workaround, but expecting better solution
using netstat we can find process id via running port number
echo `netstat -tlnp | awk '/:80 */ {split($NF,a,"/"); print a[1]}'`
So i modified the function tomcat_pid() as below
tomcat_pid() {
echo `netstat -tlnp | awk '/:<port> */ {split($NF,a,"/"); print a[1]}'`
}
I've a short ksh script that is often generated automatically. However, it's missing some stuff and I want to be able to "fix it" like I want by running a simple 1 liner perl to sed it.
Let say that the ksh contains the following function:
foo()
{
/some/command param1
if [[ $? -eq 0 ]]
then
/some/other/command stop param1
fi
/some/command param2
if [[ $? -eq 0 ]]
then
/some/other/command stop param2
fi
#...
}
and what I really want looks like this
foo_force()
{
/some/command param1
if [[ $? -eq 0 ]]
then
/some/other/command stop -f param1
fi
/some/command param2
if [[ $? -eq 0 ]]
then
/some/other/command stop -f param2
fi
#...
}
foo()
{
/some/command param1
if [[ $? -eq 0 ]]
then
/some/other/command stop param1
fi
/some/command param2
if [[ $? -eq 0 ]]
then
/some/other/command stop param2
fi
#...
foo_force
}
So far I've been able to get something "close", ie 2 perl commands, which would be ok with me. However, the second command only replace the last stop by a -f stop... which is not what I'm looking for.
/usr/bin/perl -i -pe "BEGIN{undef $/;}; s/^(foo)(\(\)\n{.*\n)}/\1_force\2}\n\n\1\2\t\1_force\n}/gms" /tmp/foo.ksh
/usr/bin/perl -i -pe "BEGIN{undef $/;}; s/^(foo\(\)\n{.* stop)( .*\n})/\1 -f\2/gms" /tmp/foo.ksh
This might work for you (GNU sed):
sed -i '/^foo()/{:a;$!N;/\n}/!ba;h;s/^foo/&_force/;s/stop /&-f /g;s/$/\n/p;g;s/\n}/\n\n foo_force&/}' /file
Using perl
perl -0777 -i -pe '
s{(^foo\(\).*?\n\})}{
my $foo = my $force = $1;
$foo =~ s/\}$/ foo_force()\n\}/;
$force =~ s/^foo\K/_force/;
$force =~ s/ stop \K/-f /g;
"$foo\n$force"
}mse;
' /tmp/foo.ksh
How can I resolve the name by the given inode in the current directory in the following script that prints all filenames of symlinks pointing to a specified file that is passed as an argument to the script. The list should be sorted by ctime.
#!/usr/bin/ksh
IFS="`printf '\n\t'`"
USAGE="usage: symlink.sh <file>"
get_ctime() {
perl -se 'use File::stat; $file=lstat($filename); print $file->ctime' -- -filename="$1"
}
stat_inode() {
perl -se 'use File::stat; $file=stat($filename); if (defined $file) { print $file->ino; }' -- -filename="$1"
}
lstat_inode() {
perl -se 'use File::stat; $file=lstat($filename); if (defined $file) { print $file->ino; }' -- -filename="$1"
}
if [ $# -eq 0 ]; then
echo "$USAGE"
exit 1
fi
FILE_NAME="$1"
FILE_INODE=$(stat_inode "$FILE_NAME")
if [ ! -e "$FILE_NAME" ]; then
echo "no such file \"$FILE_NAME\""
exit 1
fi
for LINK in ./* ./.[!.]* ;do
if [ -L "$LINK" ]; then
TARGET_INODE=$(stat_inode "$LINK")
if [ ! -z "$TARGET_INODE" ]; then
if [ "$FILE_INODE" -eq "$TARGET_INODE" ]; then
echo $(get_ctime "$LINK") $(lstat_inode "$LINK");
fi
fi
fi
done | sort -nk1 | awk '{print $2}'
Basically, I'd like to pipe awk to some kind of lookup function like this: | awk ' ' | lookup
I'd really appreciate if someone suggested a more elegant way to accomplish the task.
OS: SunOS 5.10
Shell: KSH
Something like this?
$ find . -maxdepth 1 -inum 2883399
./.jshintrc
$
or:
$ echo 2883399 | xargs -IX find . -maxdepth 1 -inum X
./.jshintrc
$
scripts in /etc/NetworkManager/dispatcher.d will got exec and parameters will be passed to the scripts by NetworkManager.
One of my laptop BIOS is malfunctioning, I have to manually sync the time, and do system upgrade BTW. I am working with a script to automate this task.
Here's the script:
#!/bin/sh
IF=$1
STATUS=$2
if [ "$STATUS"x != 'up'x -o "$(date +%Y)" -gt "2012" ] ;then
exit
fi
logger "==$0=="
wait_for_process(){
PNAME=$1
PID=`pgrep $PNAME`
while [ -z "$PID" ];do
logger "waiting $1 running for another 3 sec.."
sleep 3;
PID=`pgrep $PNAME`
done
logger "$1 is running!"
}
wait_for_process nm-applet
wait_for_process lxpanel
export DISPLAY=$(echo $DISPLAY | cut -c -2)
if [ -z $DISPLAY ];then
export DISPLAY=:0
fi
#below cmd will yield null string for $user
user=$(who | grep "$DISPLAY" | awk '{print $1}' | tail -n1)
#so I have to hardcode the user name:(
user=xxx
export XAUTHORITY="/home/$user/.Xauthority"
logger "Display $DISPLAY user $user"
su $user -c "xterm -e 'sudo /usr/bin/ntpd -qdg && sudo yaourt -Syua' &" || logger "cannot run xterm"
(the script is invoked before x window, run as root)
user=$(who | grep "$DISPLAY" | awk '{print $1}' | tail -n1) cannot find the login user name. But it works in xterm.
Can someone help?
I am using archlinux i686 + openbox + lxpanel
edit:
I want to find the real login user name, while the script is run by root.
Are you looking for the name of the user running the script? How about:
user=$( id -un )