If you don’t look at console output and want to hear that host is reachable then just add -a parameter to ping command (Linux).
$ ping -V
ping utility, iputils-sss20101006
$ ping -a 8.8.8.8
PING 8.8.8.8 (8.8.8.8) 56(84) bytes of data.
64 bytes from 8.8.8.8: icmp_req=1 ttl=44 time=38.9 ms (beep)
64 bytes from 8.8.8.8: icmp_req=2 ttl=44 time=39.3 ms (beep)
64 bytes from 8.8.8.8: icmp_req=3 ttl=44 time=39.0 ms (beep)
^C
--- 8.8.8.8 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2003ms
rtt min/avg/max/mdev = 38.997/39.138/39.346/0.273 ms
For this example I will use such tree:
.
├── dir1/
│ ├── dir6/
│ ├── dir7/
│ ├── dir8/
│ ├── file4
│ └── file5
├── dir2/
│ └── dir9/
│ ├── dir10/
│ │ ├── file9
│ │ └── file10
│ ├── file6
│ ├── file7
│ └── file8
├── dir3/
├── dir4/
├── dir5/
├── file1
├── file2
└── file3
List of all files and directories:
$ find . | sort | grep -v ^.$
./dir1
./dir1/dir6
./dir1/dir7
./dir1/dir8
./dir1/file4
./dir1/file5
./dir2
./dir2/dir9
./dir2/dir9/dir10
./dir2/dir9/dir10/file9
./dir2/dir9/dir10/file10
./dir2/dir9/file6
./dir2/dir9/file7
./dir2/dir9/file8
./dir3
./dir4
./dir5
./file1
./file2
./file3
Number of all files and directories:
$ find . | grep -v ^.$ | wc -l
20
List of all directories:
$ find . -type d | grep -v ^.$
./dir3
./dir5
./dir1
./dir1/dir6
./dir1/dir8
./dir1/dir7
./dir2
./dir2/dir9
./dir2/dir9/dir10
./dir4
Count of all directories:
$ find . -type d | grep -v ^.$ |wc -l
10
List of subdirectories in current directory:
$ find . -maxdepth 1 -type d | sort | grep -v ^.$
./dir1
./dir2
./dir3
./dir4
./dir5
Count of subdirectories in current directory:
$ find . -maxdepth 1 -type d | grep -v ^.$ | wc -l
5
List of all files:
$ find . -type f | sort
./dir1/file4
./dir1/file5
./dir2/dir9/dir10/file9
./dir2/dir9/dir10/file10
./dir2/dir9/file6
./dir2/dir9/file7
./dir2/dir9/file8
./file1
./file2
./file3
Count of all files:
$ find . -type f | wc -l
10
List of files only in current directory:
$ find . -maxdepth 1 -type f | sort
./file1
./file2
./file3
Count of files only in current directory:
$ find . -maxdepth 1 -type f | wc -l
3
This simple script will create web thumbnail using cutycapt console command.
I created it to add thumbnails to my bookmarks page.
#!/bin/sh
if [ "$#" != "1" ]; then
exit
fi
# Temporary image
temp_image="$(tempfile).png"
# Output image
output_image=`echo $1 | sed s/^www\.//`
# First, second and third output_image's letters used for directory creation
first_letter=`echo ${output_image} | cut -c1`
second_letter=`echo ${output_image} | cut -c2`
third_letter=`echo ${output_image} | cut -c3`
# Output directory
output_directory="thumbnails/${first_letter}/${second_letter}/${third_letter}"
cutycapt --url=http://$1 --out=$temp_image
convert -resize 120 -crop 120x90+0+0 $temp_image $temp_image
mkdir -p $output_directory
cp ${temp_image} ${output_directory}/${output_image}.png
unlink $temp_imageMy bookmarks-thumbnail.inc.tpl.php from Semanticscuttle looks like this:
$url=parse_url($address);
$host=preg_replace("/^www\./",'',$url['host']);
$image="/thumbnails/". $host[0] . "/" . $host[1] . "/" . $host[2] . "/" . $host . ".png";
if (file_exists($image)) {
echo '<img class="thumbnail" onclick="window.location.href=\''.$address.'\'" src="/thumbnails/'. $host[0] . '/' . $host[1] . '/' . $host[2] . '/' . $host . '.png" alt="" />
}This solutions have couple of limitations and probably doesn’t need so many sub-directories but it will be tuned over time.
Lets say that we have one big file and want to split it into 10M parts.
$ ls -lh
razem 44M
-rw-r----- 1 milosz milosz 44M 2011-02-27 16:28 file.png
To divide file into smaller parts we will use split command.
$ split -d -b10M file.png file1.part
$ ls -lh
razem 87M
-rw-r--r-- 1 milosz milosz 10M 2011-02-27 16:39 file1.part00
-rw-r--r-- 1 milosz milosz 10M 2011-02-27 16:39 file1.part01
-rw-r--r-- 1 milosz milosz 10M 2011-02-27 16:39 file1.part02
-rw-r--r-- 1 milosz milosz 10M 2011-02-27 16:39 file1.part03
-rw-r--r-- 1 milosz milosz 3,3M 2011-02-27 16:39 file1.part04
-rw-r----- 1 milosz milosz 44M 2011-02-27 16:28 file.png
You can merge them in couple different ways.
$ cat file1.part0{0,1,2,3,4} > file_m.png
$ diff file.png file_m.png
$$ ls -1 file1.* | sort | xargs cat > file_m2.png
$ diff file.png file_m2.png
$
$ cat file1.part0* > file_m1.png
$ diff file.png file_m1.png
$
I don’t remember source but it is very easy to remember.
$ cat ip_file
10.0.0.2
10.3.0.1
192.168.0.3
192.168.9.11
192.168.9.10
10.2.2.1
10.2.2.21
10.2.2.17
192.168.2.1
10.0.0.1
10.0.0.100
$ sort -t. +0n -1n +1n -2n +2n -3n +3n ip_file
10.0.0.1
10.0.0.2
10.0.0.100
10.2.2.1
10.2.2.17
10.2.2.21
10.3.0.1
192.168.0.3
192.168.2.1
192.168.9.10
192.168.9.11
Just extracted it from some very old script. Worth to remember.
Managing SPS switches could be real fun if you use expect to automate your tasks.
Script inserted below will backup configuration of specified switches. It has couple of shortcomings but works quite well. By modifying expect script you can read logs, copy running config to startup config and of course modify configuration.
#!/bin/sh
# Script designed to backup configuration of SPS switches
#
# Example usage:
# $ sh sps_backup.sh 10.2.2.3 10.2.2.4
#
# Commands
expect=`which expect`
sed=`which sed`
# Archive name
ar_date=`date +"%d.%m.%y_%H%M%S"`
ar_name="sps_${ar_date}.tgz"
# SSH
ssh=`which ssh`
ssh_port="22"
# get_credentials - Get username and password
get_credentials(){
echo -n "Username: "
read user
oldmodes=`stty -g`
stty -echo
echo -n "Password: "
read pass
stty $oldmodes
echo
export pass
export user
}
# get_config - create expect script
get_config() {
cat << EOF
#!${expect} -f
log_user 0
exp_internal 0
sleep 3
spawn ${ssh} ${host} -p ${ssh_port}
expect {
"The authenticity of host" {
send "yes\r"
}
}
expect "User Name:" {
send "$user"
send "\r"
}
expect "Password:" {
send "$pass"
send "\r"
}
expect {
"User Name:" {
exit 1
}
"SPS*#" {
send "show startup-config\r"
while (1) {
expect {
timeout break
"<return>" {
puts "\$expect_out(buffer)"
send " "
}
"SPS*#" {
puts "\$expect_out(buffer)"
break
}
}
}
}
}
send "exit\r"
exit 0
EOF
}
# Get login details
get_credentials
# Create temporary directory
temp_dir=`mktemp -d` || exit 1
for host in ${*}; do
temp_file="${temp_dir}/${host}"
# Get config
get_config ${host} | ${expect} -f - | while read line
do
echo $line >> $temp_file
done
# Remove unnecessary stuff
if [ -f ${temp_file} ]; then
sed -ci -e "/^User.*/d" $temp_file
sed -ci -e "/^Pass.*/d" $temp_file
sed -ci -e "/More.*return*/d" $temp_file
sed -ci -e "/show startup-config/d" $temp_file
sed -ci -e "/^SPS.*#$/d" $temp_file
sed -ci -e "s/^M//g" $temp_file
sed -ci -e "/^ $/d" $temp_file
sed -ci -e "s/^\ //g" $temp_file
sed -ci -e "/^$/d" $temp_file
fi
# Check if we got config
# It will contain "password ..." line
file_check=`cat ${temp_file} | grep "^password"}`
if [ -n "${file_check}"]; then
echo -e "${host}\tOK"
else
echo -e "${host}\tError"
rm ${temp_file}
fi
done
# Create archive
directory=`pwd`
cd ${temp_dir} && tar cfz ${directory}/${ar_name} .
# Clean up
rm ${temp_dir}/*
rmdir ${temp_dir}
echo "Archive contents:"
tar tfz ${directory}/${ar_name}For example to see logs on SPS switch you could use such code:
send "show logging\r"
while (1) {
expect {
timeout break
"<return>" {
puts "$expect_out(buffer)"
send " "
}
"SPS*#" {
puts "$expect_out(buffer)"
break
}
}
}
send "exit\r"If have couple of SRW switches then there is a little problem. If you try connecting using ssh you will see error:
ssh_rsa_verify: RSA modulus too small: 512 < minimum 768 bits
key_verify failed for server_host_key
To solve it you need to download ssh source, change SSH_RSA_MINIMUM_MODULUS_SIZE in ssh.h and compile it by hand.
After that you can access lcli (it’s a little bit tricky but easily scriptable) :
expect "Password:"
send "$env(user)"
send "\t"
send "$env(pass)"
send "\r"
expect "logout"
send \x1a
expect "TERMINATED"
send \r
expect ">"
send "lcli\r"
expect "User Name:"
send "$env(user)"
send "\r"
expect "Password:"
send "$env(pass)"
send "\r"
expect "#"
Just one note: There are couple of assumptions used, one of them is switch hostname=switch model (like hostname SPS224G4).
If you want to get rid of spaces in file/directory names then just look at this simple script:
#!/bin/bash
ls | grep "\ " | while read item
do
new=`echo ${item} | tr [:blank:] _`
old=${item}
mv "${old}" ${new}
doneScript works in current directory for files and directories.
Sometimes I need to downscale video (like music) for portable device.
After some research I ended up using mencoder within such script:
#!/bin/sh
OUTPUT_DIR="/home/milosz/converted"
mencoder ${1} -ovc xvid -xvidencopts bitrate=300 -vf scale=320:-3 -ofps 25 -oac lavc -lavcopts acodec=mp3:abitrate=64 -o ${OUTPUT_DIR}/${1##*/}
To check if your Nginx server has properly configured compression just execute command:
$ curl -s -I --compressed http://my.web.site | grep Content-Encoding
Empty result means that it doesn’t support compression.
Proper results should look like this:
Using KDE notification area within shell script gives a lot of useful possibilities. You can easily get feedback from scripts running in background and checking logs or running services.
To create above notification as example I used quite simple script that uses kdialog command. and basic HTML tags.
#!/bin/sh
message(){
cat << EOF
Say <strong>hello</strong> using your <font color=red>nofication</font> area.
Date: <em>`date`</em>
Used space: `df -h | grep sda1 | awk '{ print $5 }'`
Load average: <u>`uptime | awk '{ print $10 $11 $12 }' | sed s/\,/\ /g`</u>
EOF
}
timeout=20
kdialog=`which kdialog`
${kdialog} --passivepopup "`message`" ${timeout}