DD-WRT Bash script at startup issue - bash

Hey all I have the following BASH script running at startup on my WRT1900ac linksys:
USER="admin"
PASS="passhere"
PROTOCOL="http"
ROUTER_IP="192.168.1.1"
# Port to connect to which will provide the JSON data.
PORT=9898
while [ 1 ]
do
# Grab connected device MAC addresses through router status page.
MACS=$(curl -s --user $USER:$PASS $PROTOCOL://$ROUTER_IP/Status_Wireless.live.asp)
# clear temp JSON file
echo > temp.log
# Get hostname and IP (just in case there is no hostname).
for MAC in $(echo $MACS | grep -oE "wl_mac::[a-z0-9]{2}:[a-z0-9]{2}:[a-z0-9]{2}:[a-z0-9]{2}:[a-z0-9]{2}:[a-z0-9]{2}" | cut -c 9-);
do
grep 0x /proc/net/arp | awk '{print $1 " " $4}' | while IFS= read -r line
do
IP=$(echo $line | cut -d' ' -f1)
MACTEMP=$(echo $line | cut -d' ' -f2)
HOST=$(arp -a | grep $IP | cut -d' ' -f1)
# if no hostname exists, just use IP.
if [ "$HOST" == "" ]
then
HOST=$IP
fi
if [ "$MAC" == "$MACTEMP" ]
then
JSON="{'hostname' : '$HOST', 'mac_address' : '$MAC'}"
echo $JSON >> temp.log
fi
done
done
# Provide the JSON formatted output on $PORT of router.
# This allows one connection before closing the port (connect, receive data, close).
# Port will reopen every 5 minutes with new data as setup in a cron job.
echo -e "HTTP/1.1 200 OK\n\n $(cat temp.log)" | nc -l -p$PORT >/dev/null
# Wait for 10 seconds and do it all over.
sleep 10
done
And for some reason when I reboot the router and then try to visit http://192.168.1.1:9898 it just shows a blank page even though I have my android cell phone connected via wifi to the router and the router shows the MAC address on the status page.
What should be on that page is all the wireless MAC address that are currently connected to the router and displaying them out in JSON form.
Any BASH guru's here that can help spot the problem?


I think it should be
echo -e "HTTP/1.1 200 OK\n\n $(cat temp.log)" | nc -l -p$PORT 0.0.0.0 >/dev/null

Related

insert hosts in check via for loop

following problem:
we have a file called "file.conf"
192.168.30.1|192.168.30.1|os
_gateway|192.168.30.2|Linux 2.6.18 - 2.6.22
...
first is the hostname second is the ipv4
now we have a script where he should automaticaly insert the hosts and ip's via automated user from checkMK
#!/bin/bash
FILE=filename
source $FILE
for i in ${FILE}
do
HOSTNAME=$(cat $i | cut -d '|' -f1)
IP=$(cat $i | cut -d '|' -f2)
curl "http://checkmkadress/check_mk/host&user" -d 'request={"hostname":"'"$HOSTNAME"'","folder":"ansible","attributes":{"ipaddress":"'"$IP"'","site":"sitename","tag_agent":"cmk-agent"}}'
done
but if we do it like that we get the following error cause he try's to put in every host in host and every ip in ip without going through all lines
{"result": "Check_MK exception: Failed to parse JSON request: '{\"hostname\":\"allhostnames":{\"ipaddress\":all_ips\",\"site\":\"sitename\",\"tag_agent\":\"cmk-agent\"}}': Invalid control character at: line 1 column 26 (char 25)", "result_code": 1}
how can we make the curl script go through each line to get host and ip individually

Using what you already done, try in this way.
FILE="filename"
source $FILE
while read line
do
HOSTNAME=$(echo $line | cut -d '|' -f1)
IP=$(echo $line | cut -d '|' -f2)
#your curl command here
done <$FILE
OR, which I prefer
while IFS='|' read -r host ip description
do
#your curl command here
echo "$host : $ip : $description"
done <$FILE

Telnet to one computer from multiple computer

I have scenario where I have list of 100s of server . Want to check whether those all server can reach to specified destination server or not by telneting from all server to that destination server.
I have written a code as below,
#!/bin/bash
#bash to check telnet status.
#set -x;
#
#clear
SetParam() {
export URLFILE="Host_PortFile.txt"
export TIME=`date +%d-%m-%Y_%H.%M.%S`
export port=80
export STATUS_UP=`echo -e "\E[32m[ RUNNING ]\E[0m"`
export STATUS_DOWN=`echo -e "\E[31m[ DOWN ]\E[0m"`
export MAIL_TO="admin(at)techpaste(dot)com"
export SHELL_LOG="`basename $0`.log"
}
Telnet_Status() {
SetParam
cat $URLFILE | while read next
do
server=`echo $next | cut -d : -f1`
port=`echo $next | awk -F":" '{print $2}'`
TELNETCOUNT=`sleep 5 | telnet $server $port | grep -v "Connection refused" | grep "Connected to" | grep -v grep | wc -l`
if [ $TELNETCOUNT -eq 1 ] ; then
echo -e "$TIME : Port $port of URL http://$server:$port/ is \E[32m[ OPEN ]\E[0m";
else
echo -e "$TIME : Port $port of URL http://$server:$port/ is \E[31m[ NOT OPEN ]\E[0m";
echo -e "$TIME : Port $port of URL http://$server:$port/ is NOT OPEN" | mailx -s "Port $port of URL $server:$port/ is DOWN!!!" $MAIL_TO;
fi
done;
}
Main() {
Telnet_Status
}
SetParam
Main | tee -a $SHELL_LOG
My Host_PortFile.txt file looks like,
gmail.com:443
But here, i need to go to individual server and has to run this which consumes more time. Is there any modification I can do so that I can run the script from one machine to read all source server name from text file or any and can check that server can reach the destination server or not? Can anyone suggest on this please?

Email Alerts when service or server automatically comes up

I am working on a bash script that helps to ping and get the network interface level status of the host machines and services.
This script will send a email alerts in case of failure.
#!/bin/bash
HOSTS="192.168.8.200"
COUNT=4
for myHost in $HOSTS
do
count=$(ping -c $COUNT $myHost | grep 'received' | awk -F',' '{ print $2 }' | awk '{ print $1 }')
if [ $count -eq 0 ]; then
# 100% failed
echo -e "HOST:$myHost is down (ping failed) at $(date)" | mailx -A gmail -s “Mail subject” anymail#anydomain.com
fi
done
This works fine.
But need help to get a one single email alert when host automatically comes up (ping success).

You need to save the state of the host (up/down) during the calls of your script.
if the host is "up" and the former state was "down" then you need to send an email.
You can just write the result of the "check command" to a file in /tmp/
if the check returns that the server is up you read the content of the file. if the state is "down" in the file, then send an email an write "up" to the file.
on the next check if the server is up, there will be no additional email sent, because the server was also "up" before.
#!/bin/bash
HOSTS="192.168.8.200 192.168.8.201 192.168.122.1"
COUNT=4
STATE="/tmp/ping_state.txt"
for myHost in $HOSTS
do
count=$(ping -c $COUNT $myHost | grep 'received' | awk -F',' '{ print $2 }' | awk '{ print $1 }')
if [ $count -eq 0 ]; then
# 100% failed
#echo -e "HOST:$myHost is down (ping failed) at $(date)" | mailx -A gmail -s “Mail subject” anymail#anydomain.com
echo "host $myHost down"
#delete all previous entries of that ip
sed -i "/$myHost/d" $STATE
#mark host as down
echo "$myHost - down" >> $STATE
else
CHECK=`grep "$myHost" $STATE | grep -o "down"`
if [ "$CHECK" = "down" ]; then
echo "host $myHost up again"
#insert email for host up here
fi
#delete all previous entries of that ip
sed -i "/$myHost/d" $STATE
echo "$myHost - up" >> $STATE
fi
done
for simple testing I just used an echo statement instead of sending an email.

How to get the primary IP address of the local machine on Linux and OS X? [closed]

Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
This question does not appear to be about a specific programming problem, a software algorithm, or software tools primarily used by programmers. If you believe the question would be on-topic on another Stack Exchange site, you can leave a comment to explain where the question may be able to be answered.
Closed 4 years ago.
Improve this question
I am looking for a command line solution that would return me the primary (first) IP address of the localhost, other than 127.0.0.1
The solution should work at least for Linux (Debian and RedHat) and OS X 10.7+
I am aware that ifconfig is available on both but its output is not so consistent between these platforms.

Use grep to filter IP address from ifconfig:
ifconfig | grep -Eo 'inet (addr:)?([0-9]*\.){3}[0-9]*' | grep -Eo '([0-9]*\.){3}[0-9]*' | grep -v '127.0.0.1'
Or with sed:
ifconfig | sed -En 's/127.0.0.1//;s/.*inet (addr:)?(([0-9]*\.){3}[0-9]*).*/\2/p'
If you are only interested in certain interfaces, wlan0, eth0, etc. then:
ifconfig wlan0 | ...
You can alias the command in your .bashrc to create your own command called myip for instance.
alias myip="ifconfig | sed -En 's/127.0.0.1//;s/.*inet (addr:)?(([0-9]*\.){3}[0-9]*).*/\2/p'"
A much simpler way is hostname -I (hostname -i for older versions of hostname but see comments). However, this is on Linux only.

The following will work on Linux but not OSX.
This doesn't rely on DNS at all, and it works even if /etc/hosts is not set correctly (1 is shorthand for 1.0.0.0):
ip route get 1 | awk '{print $NF;exit}'
or avoiding awk and using Google's public DNS at 8.8.8.8 for obviousness:
ip route get 8.8.8.8 | head -1 | cut -d' ' -f8
A less reliable way: (see comment below)
hostname -I | cut -d' ' -f1

For linux machines (not OS X) :
hostname --ip-address

Solution
$ ip -o route get to 8.8.8.8 | sed -n 's/.*src \([0-9.]\+\).*/\1/p'
192.168.8.16
Explanation
The correct way to query network information is using ip:
-o one-line output
route get to get the actual kernel route to a destination
8.8.8.8 Google IP, but can use the real IP you want to reach
e.g. ip output:
8.8.8.8 via 192.168.8.254 dev enp0s25 src 192.168.8.16 uid 1000 \ cache
To extract the src ip, sed is the ligthest and most compatible with regex support:
-n no output by default
's/pattern/replacement/p' match pattern and print replacement only
.*src \([0-9.]\+\).* match the src IP used by the kernel, to reach 8.8.8.8
e.g. final output:
192.168.8.16
Other answers
I think none of the preceding answer are good enough for me, as they don't work in a recent machine (Gentoo 2018).
Issues I found with preceding answers:
use of positional column in command output;
use of ifconfig which is deprecated and -- for example -- don't list multple IPs;
use of awk for a simple task which sed can handle better;
ip route get 1 is unclear, and is actually an alias for ip route get to 1.0.0.0
use of hostname command, which don't have -I option in all appliance and which return 127.0.0.1 in my case.

on Linux
hostname -I
on macOS
ipconfig getifaddr en0
hostname -I can return multiple addresses in an unreliable order (see the hostname manpage), but for me it just returns 192.168.1.X, which is what you wanted.

Edited (2014-06-01 2018-01-09 2021-07-25)
From some time ago, I use now newer ip tool. But under bash, I will do simply:
read -r _{,} gateway _ iface _ ip _ < <(ip r g 1.0.0.0)
Then
printf '%-12s %s\n' gateway $gateway iface $iface ip $ip
gateway 192.168.1.1
iface eth0
ip 192.168.1.37
From there, the mask is:
while IFS=$' /\t\r\n' read lne lip lmask _;do
[ "$lne" = "inet" ] && [ "$lip" = "$ip" ] && mask=$lmask
done < <(ip a s dev $iface)
echo Mask is $mask bits.
Mask is 24 bits.
Then if you want to see your mask as an IP:
printf -v msk '%*s' $mask ''
printf -v msk %-32s ${msk// /1}
echo $((msk=2#${msk// /0},msk>>24)).$((msk>>16&255)).$((msk>>8&255)).$((msk&255))
255.255.255.0
Edited (2014-06-01 2018-01-09)
For stronger config, with many interfaces and many IP configured on each interfaces, I wrote a pure bash script (not based on 127.0.0.1) for finding correct interface and ip, based on default route. I post this script at very bottom of this answer.
Intro
As both Os have bash installed by default, there is a bash tip for both Mac and Linux:
The locale issue is prevented by the use of LANG=C:
myip=
while IFS=$': \t' read -a line ;do
[ -z "${line%inet}" ] && ip=${line[${#line[1]}>4?1:2]} &&
[ "${ip#127.0.0.1}" ] && myip=$ip
done< <(LANG=C /sbin/ifconfig)
echo $myip
Putting this into a function:
Minimal:
getMyIP() {
local _ip _line
while IFS=$': \t' read -a _line ;do
[ -z "${_line%inet}" ] &&
_ip=${_line[${#_line[1]}>4?1:2]} &&
[ "${_ip#127.0.0.1}" ] && echo $_ip && return 0
done< <(LANG=C /sbin/ifconfig)
}
Simple use:
getMyIP
192.168.1.37
Fancy tidy:
getMyIP() {
local _ip _myip _line _nl=$'\n'
while IFS=$': \t' read -a _line ;do
[ -z "${_line%inet}" ] &&
_ip=${_line[${#_line[1]}>4?1:2]} &&
[ "${_ip#127.0.0.1}" ] && _myip=$_ip
done< <(LANG=C /sbin/ifconfig)
printf ${1+-v} $1 "%s${_nl:0:$[${#1}>0?0:1]}" $_myip
}
Usage:
getMyIP
192.168.1.37
or, running same function, but with an argument:
getMyIP varHostIP
echo $varHostIP
192.168.1.37
set | grep ^varHostIP
varHostIP=192.168.1.37
Nota: Without argument, this function output on STDOUT, the IP and a newline, with an argument, nothing is printed, but a variable named as argument is created and contain IP without newline.
Nota2: This was tested on Debian, LaCie hacked nas and MaxOs. If this won't work under your environ, I will be very interested by feed-backs!
Older version of this answer
( Not deleted because based on sed, not bash. )
Warn: There is an issue about locales!
Quick and small:
myIP=$(ip a s|sed -ne '/127.0.0.1/!{s/^[ \t]*inet[ \t]*\([0-9.]\+\)\/.*$/\1/p}')
Exploded (work too;)
myIP=$(
ip a s |
sed -ne '
/127.0.0.1/!{
s/^[ \t]*inet[ \t]*\([0-9.]\+\)\/.*$/\1/p
}
'
)
Edit:
How! This seem not work on Mac OS...
Ok, this seem work quite same on Mac OS as on my Linux:
myIP=$(LANG=C /sbin/ifconfig | sed -ne $'/127.0.0.1/ ! { s/^[ \t]*inet[ \t]\\{1,99\\}\\(addr:\\)\\{0,1\\}\\([0-9.]*\\)[ \t\/].*$/\\2/p; }')
splitted:
myIP=$(
LANG=C /sbin/ifconfig |
sed -ne $'/127.0.0.1/ ! {
s/^[ \t]*inet[ \t]\\{1,99\\}\\(addr:\\)\\{0,1\\}\\([0-9.]*\\)[ \t\/].*$/\\2/p;
}')
My script (jan 2018):
This script will first find your default route and interface used for, then search for local ip matching network of gateway and populate variables. The last two lines just print, something like:
Interface : en0
Local Ip : 10.2.5.3
Gateway : 10.2.4.204
Net mask : 255.255.252.0
Run on mac : true
or
Interface : eth2
Local Ip : 192.168.1.31
Gateway : 192.168.1.1
Net mask : 255.255.255.0
Run on mac : false
Well, there it is:
#!/bin/bash
runOnMac=false
int2ip() { printf ${2+-v} $2 "%d.%d.%d.%d" \
$(($1>>24)) $(($1>>16&255)) $(($1>>8&255)) $(($1&255)) ;}
ip2int() { local _a=(${1//./ }) ; printf ${2+-v} $2 "%u" $(( _a<<24 |
${_a[1]} << 16 | ${_a[2]} << 8 | ${_a[3]} )) ;}
while IFS=$' :\t\r\n' read a b c d; do
[ "$a" = "usage" ] && [ "$b" = "route" ] && runOnMac=true
if $runOnMac ;then
case $a in
gateway ) gWay=$b ;;
interface ) iFace=$b ;;
esac
else
[ "$a" = "0.0.0.0" ] && [ "$c" = "$a" ] && iFace=${d##* } gWay=$b
fi
done < <(/sbin/route -n 2>&1 || /sbin/route -n get 0.0.0.0/0)
ip2int $gWay gw
while read lhs rhs; do
[ "$lhs" ] && {
[ -z "${lhs#*:}" ] && iface=${lhs%:}
[ "$lhs" = "inet" ] && [ "$iface" = "$iFace" ] && {
mask=${rhs#*netmask }
mask=${mask%% *}
[ "$mask" ] && [ -z "${mask%0x*}" ] &&
printf -v mask %u $mask ||
ip2int $mask mask
ip2int ${rhs%% *} ip
(( ( ip & mask ) == ( gw & mask ) )) &&
int2ip $ip myIp && int2ip $mask netMask
}
}
done < <(/sbin/ifconfig)
printf "%-12s: %s\n" Interface $iFace Local\ Ip $myIp \
Gateway $gWay Net\ mask $netMask Run\ on\ mac $runOnMac

Specific to only certain builds of Ubuntu. Though it may just tell you 127.0.0.1:
hostname -i
or
hostname -I

You can also get IP version 4 address of eth0 by using this command in linux
/sbin/ip -4 -o addr show dev eth0| awk '{split($4,a,"/");print a[1]}'
Output will be like this
[root#localhost Sathish]# /sbin/ip -4 -o addr show dev eth0| awk '{split($4,a,"/");print a[1]}'
192.168.1.22

This works on Linux and OSX
This will get the interface associated to the default route
NET_IF=`netstat -rn | awk '/^0.0.0.0/ {thif=substr($0,74,10); print thif;} /^default.*UG/ {thif=substr($0,65,10); print thif;}'`
Using the interface discovered above, get the ip address.
NET_IP=`ifconfig ${NET_IF} | grep -Eo 'inet (addr:)?([0-9]*\.){3}[0-9]*' | grep -Eo '([0-9]*\.){3}[0-9]*' | grep -v '127.0.0.1'`
OSX
uname -a
Darwin laptop 14.4.0 Darwin Kernel Version 14.4.0: Thu May 28 11:35:04 PDT 2015; root:xnu-2782.30.5~1/RELEASE_X86_64 x86_64
echo $NET_IF
en5
echo $NET_IP
192.168.0.130
CentOS Linux
uname -a
Linux dev-cil.medfx.local 2.6.18-164.el5xen 1 SMP Thu Sep 3 04:03:03 EDT 2009 x86_64 x86_64 x86_64 GNU/Linux
echo $NET_IF
eth0
echo $NET_IP
192.168.46.10

Using some of the other methods You may enter a conflict where multiple IP adresses is defined on the system.
This line always gets the IP address by default used.
ip route get 8.8.8.8 | head -1 | awk '{print $7}'

ifconfig | grep 'inet ' | grep -v '127.0.0.1' | awk '{print $2}'

Im extracting my comment to this answer:
ip route get 1 | sed -n 's/^.*src \([0-9.]*\) .*$/\1/p'
It bases on #CollinAnderson answer which didn't work in my case.

Assuming you need your primary public IP as it seen from the rest of the world, try any of those:
wget http://ipecho.net/plain -O - -q
curl http://icanhazip.com
curl http://ifconfig.me/ip

Finds an IP address of this computer in a network which is a default gateway (for example excludes all virtual networks, docker bridges) eg. internet gateway, wifi gateway, ethernet
ip route| grep $(ip route |grep default | awk '{ print $5 }') | grep -v "default" | awk '/scope/ { print $9 }'
Works on Linux.
Test:
➜ ~ ip route| grep $(ip route |grep default | awk '{ print $5 }') | grep -v "default" | awk '/scope/ { print $9 }'
192.168.0.114
➜ reverse-networking git:(feature/type-local) ✗ ifconfig wlp2s0
wlp2s0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 192.168.0.114 netmask 255.255.255.0 broadcast 192.168.0.255
inet6 fe80::d3b9:8e6e:caee:444 prefixlen 64 scopeid 0x20<link>
ether ac:x:y:z txqueuelen 1000 (Ethernet)
RX packets 25883684 bytes 27620415278 (25.7 GiB)
RX errors 0 dropped 27 overruns 0 frame 0
TX packets 7511319 bytes 1077539831 (1.0 GiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

The shortest way to get your local ipv4-address on your linux system:
hostname -I | awk '{print $1}'

I have to add to Collin Andersons answer that this method also takes into account if you have two interfaces and they're both showing as up.
ip route get 1 | awk '{print $NF;exit}'
I have been working on an application with Raspberry Pi's and needed the IP address that was actually being used not just whether it was up or not. Most of the other answers will return both IP address which isn't necessarily helpful - for my scenario anyway.

Primary network interface IP
ifconfig `ip route | grep default | head -1 | sed 's/\(.*dev \)\([a-z0-9]*\)\(.*\)/\2/g'` | grep -oE "\b([0-9]{1,3}\.){3}[0-9]{1,3}\b" | head -1

ip addr show | grep -E '^\s*inet' | grep -m1 global | awk '{ print $2 }' | sed 's|/.*||'

Another ifconfig variantion that works both on Linux and OSX:
ifconfig | grep "inet " | cut -f2 -d' '

Not sure if this works in all os, try it out.
ifconfig | awk -F"[ :]+" '/inet addr/ && !/127.0/ {print $4}'

I went through a lot of links (StackExchange, AskUbuntu, StackOverflow etc) and came to the decision to combine all the best solutions into one shell script.
In my opinion these two QAs are the best of seen:
How can I get my external IP address in a shell script?
https://unix.stackexchange.com/q/22615
How do I find my internal ip address?
https://askubuntu.com/a/604691
Here is my solution based on some ideas by rsp shared in his repository (https://github.com/rsp/scripts/).
Some of you could say that this script is extremely huge for so simple task but I'd like to make it easy and flexible in usage as much as possible. It supports simple configuration file allowing redefine the default values.
It was successfully tested under Cygwin, MINGW and Linux (Red Hat).
Show internal IP address
myip -i
Show external IP address
myip -e
Source code, also available by the link: https://github.com/ildar-shaimordanov/tea-set/blob/master/home/bin/myip. Example of configuration file is there, next to the main script.
#!/bin/bash
# =========================================================================
#
# Getting both internal and external IP addresses used for outgoing
# Internet connections.
#
# Internal IP address is the IP address of your computer network interface
# that would be used to connect to Internet.
#
# External IP address is the IP address that is visible by external
# servers that you connect to over Internet.
#
# Copyright (C) 2016 Ildar Shaimordanov
#
# =========================================================================
# Details of the actual implementation are based on the following QA:
#
# How can I get my external IP address in a shell script?
# https://unix.stackexchange.com/q/22615
#
# How do I find my internal ip address?
# https://askubuntu.com/a/604691
# =========================================================================
for f in \
"$( dirname "$0" )/myip.conf" \
~/.myip.conf \
/etc/myip.conf
do
[ -f "$f" ] && {
. "$f"
break
}
done
# =========================================================================
show_usage() {
cat - <<HELP
USAGE
$( basename "$0" ) [OPTIONS]
DESCRIPTION
Display the internal and external IP addresses
OPTIONS
-i Display the internal IP address
-e Display the external IP address
-v Turn on verbosity
-h Print this help and exit
HELP
exit
}
die() {
echo "$( basename "$0" ): $#" >&2
exit 2
}
# =========================================================================
show_internal=""
show_external=""
show_verbose=""
while getopts ":ievh" opt
do
case "$opt" in
i )
show_internal=1
;;
e )
show_external=1
;;
v )
show_verbose=1
;;
h )
show_usage
;;
\? )
die "Illegal option: $OPTARG"
;;
esac
done
if [ -z "$show_internal" -a -z "$show_external" ]
then
show_internal=1
show_external=1
fi
# =========================================================================
# Use Google's public DNS to resolve the internal IP address
[ -n "$TARGETADDR" ] || TARGETADDR="8.8.8.8"
# Query the specific URL to resolve the external IP address
[ -n "$IPURL" ] || IPURL="ipecho.net/plain"
# Define explicitly $IPCMD to gather $IPURL using another tool
[ -n "$IPCMD" ] || {
if which curl >/dev/null 2>&1
then
IPCMD="curl -s"
elif which wget >/dev/null 2>&1
then
IPCMD="wget -qO -"
else
die "Neither curl nor wget installed"
fi
}
# =========================================================================
resolveip() {
{
gethostip -d "$1" && return
getent ahostsv4 "$1" \
| grep RAW \
| awk '{ print $1; exit }'
} 2>/dev/null
}
internalip() {
[ -n "$show_verbose" ] && printf "Internal: "
case "$( uname | tr '[:upper:]' '[:lower:]' )" in
cygwin* | mingw* | msys* )
netstat -rn \
| grep -w '0.0.0.0' \
| awk '{ print $4 }'
return
;;
esac
local t="$( resolveip "$TARGETADDR" )"
[ -n "$t" ] || die "Cannot resolve $TARGETADDR"
ip route get "$t" \
| awk '{ print $NF; exit }'
}
externalip() {
[ -n "$show_verbose" ] && printf "External: "
eval $IPCMD "$IPURL" $IPOPEN
}
# =========================================================================
[ -n "$show_internal" ] && internalip
[ -n "$show_external" ] && externalip
# =========================================================================
# EOF

I just utilize Network Interface Names, my custom command is
[[ $(ip addr | grep enp0s25) != '' ]] && ip addr show dev enp0s25 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p' || ip addr show dev eth0 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p'
in my own notebook
[flying#lempstacker ~]$ cat /etc/redhat-release
CentOS Linux release 7.2.1511 (Core)
[flying#lempstacker ~]$ [[ $(ip addr | grep enp0s25) != '' ]] && ip addr show dev enp0s25 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p' || ip addr show dev eth0 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p'
192.168.2.221
[flying#lempstacker ~]$
but if the network interface owns at least one ip, then it will show all ip belong to it
for example
Ubuntu 16.10
root#yakkety:~# sed -r -n 's#"##g;s#^VERSION=(.*)#\1#p' /etc/os-release
16.04.1 LTS (Xenial Xerus)
root#yakkety:~# [[ $(ip addr | grep enp0s25) != '' ]] && ip addr show dev enp0s25 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p' || ip addr show dev eth0 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p'
178.62.236.250
root#yakkety:~#
Debian Jessie
root#jessie:~# sed -r -n 's#"##g;s#^PRETTY_NAME=(.*)#\1#p' /etc/os-release
Debian GNU/Linux 8 (jessie)
root#jessie:~# [[ $(ip addr | grep enp0s25) != '' ]] && ip addr show dev enp0s25 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p' || ip addr show dev eth0 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p'
192.81.222.54
root#jessie:~#
CentOS 6.8
[root#centos68 ~]# cat /etc/redhat-release
CentOS release 6.8 (Final)
[root#centos68 ~]# [[ $(ip addr | grep enp0s25) != '' ]] && ip addr show dev enp0s25 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p' || ip addr show dev eth0 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p'
162.243.17.224
10.13.0.5
[root#centos68 ~]# ip route get 1 | awk '{print $NF;exit}'
162.243.17.224
[root#centos68 ~]#
Fedora 24
[root#fedora24 ~]# cat /etc/redhat-release
Fedora release 24 (Twenty Four)
[root#fedora24 ~]# [[ $(ip addr | grep enp0s25) != '' ]] && ip addr show dev enp0s25 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p' || ip addr show dev eth0 | sed -n -r 's#.*inet (.*)/.*brd.*#\1#p'
104.131.54.185
10.17.0.5
[root#fedora24 ~]# ip route get 1 | awk '{print $NF;exit}'
104.131.54.185
[root#fedora24 ~]#
It seems like that command ip route get 1 | awk '{print $NF;exit}' provided by link is more accurate, what's more, it more shorter.

There's a node package for everything. It's cross-platform and easy to use.
$ npm install --global internal-ip-cli
$ internal-ip
fe80::1
$ internal-ip --ipv4
192.168.0.3
This is a controversial approach, but using npm for tooling is becoming more popular, like it or not.

If you know the network interface (eth0, wlan, tun0 etc):
ifconfig eth0 | grep addr: | awk '{ print $2 }' | cut -d: -f2

ifconfig | grep "inet addr:" | grep -v "127.0.0.1" | grep -Eo '[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}' | head -1

ifconfig $(netstat -rn | grep -E "^default|^0.0.0.0" | head -1 | awk '{print $NF}') | grep 'inet ' | awk '{print $2}' | grep -Eo '([0-9]*\.){3}[0-9]*'

Works on Mac, Linux and inside Docker Containers:
$ hostname --ip-address 2> /dev/null || (ifconfig | sed -En 's/127.0.0.1//;s/.*inet (addr:)?(([0-9]*\.){3}[0-9]*).*/\2/p' | awk '{print$1; exit}')
Also works on Makefile as:
LOCAL_HOST := ${shell hostname --ip-address 2> /dev/null || (ifconfig | sed -En 's/127.0.0.1//;s/.*inet (addr:)?(([0-9]*\.){3}[0-9]*).*/\2/p' | awk '{print $1; exit}')}

For linux, what you need is this command:
ifconfig $1|sed -n 2p|awk '{ print $2 }'|awk -F : '{ print $2 }'
type this in your shell and you will simply know your ip.

This is easier to read:
ifconfig | grep 'inet addr:' |/usr/bin/awk '{print $2}' | tr -d addr:

If you have npm and node installed : npm install -g ip && node -e "const ip = require('ip'); console.log(ip.address())"

Network status change notification in mac os x

In Mac Os X how my application get informed when the status of network connection changes? I tried using SCNetworkConnectionGetStatus from SCNetworkConnection Reference. But it has to be called continuously. I need an API which will will inform me as soon as network status chages.

This is what I ended up using. Once I had this script I just put it into a basic while loop and voila -- network connectivity change monitoring.
#!/bin/bash
set -o pipefail
configured_ip_addresses="$((ifconfig | \
grep -iEo '[0-9]+\.[0-9]+\.[0-9]+\.[0-9]+' | \
grep -vi '127.0.0.1' | tr '\n' ' ') || echo NONE_CONFIGURED)"
externally_visible_ip_address="$(curl -m 1 ipinfo.io/ip 2>/dev/null || echo NO_CONNECTIVITY)"
computed_state="Actual: $externally_visible_ip_address, Configured: $configured_ip_addresses"
statefile="/tmp/net-watcher.state"
if [ -f $statefile ]; then
echo "$computed_state" > "${statefile}-new"
new_chksum="$(md5 "${statefile}-new" | awk '{print $NF}')"
existing_chksum="$(md5 "${statefile}" | awk '{print $NF}')"
if [[ "${new_chksum}" != "${existing_chksum}" ]]; then
mv "${statefile}-new" "${statefile}"
osascript -e "display notification \"$(cat $statefile)\" with title \"ALERT: Network Changed\""
else
rm "${statefile}-new"
fi
else
echo "$computed_state" > $statefile
fi

Resources