SQUID3 - using multiple auth_param like basic_ncsa_auth & basic_ldap_auth - bash

i tried to setup squid3 with multiple auth_param. Basically, the first choice should be basic_ldap_auth and if this doesnt return OK it should try basic_ncsa_auth with the same values. As far as i know squid doesnt support it however there is the possibility to use "external" ACL
auth_param basic program /usr/lib/squid3/basic_fake_auth
external_acl_type MultAuth %SRC %LOGIN %{Proxy-Authorization} /etc/squid3/multAuth.pl
acl extAuth external MultAuth
my "multAuth.pl"
use URI::Escape;
use MIME::Base64;
$|=1;
while (<>) {
($ip,$user,$auth) = split();
# Retrieve the password from the authentication header
$auth = uri_unescape($auth);
($type,$authData) = split(/ /, $auth);
$authString = decode_base64($authData);
($username,$password) = split(/:/, $authString);
# do the authentication and pass results back to Squid.
$ldap = `/bin/bash auth/ldap.sh`;
if ($ldap == "OK") {
print "OK";
}
$ncsa = `/bin/bash auth/ncsa.sh`;
if ($ncsa == "OK") {
print "OK";
} else {
print "ERR";
}
}
now i am trying to run with ncsa.sh and ldap.sh the "normal" shell command for these auth methods.
./basic_ldap_auth -R -b "dc=domain,dc=de" -D "CN=Administrator,CN=Users,DC=domain,DC=de" -w "password" -f sAMAccountName=%s -h domain.de
user password
and
./basic_ncsa_auth /etc/squid3/users
user password
Therefor i ran:
auth/ncsa.sh
#!/usr/bin/expect
eval spawn [lrange $argv 0 end]
expect ""
send [lindex $argv 1]
send '\r'
expect {
"OK" {
echo OK
exp_continue
}
"ERR" {
echo ERR
exp_continue
}
interact
with
./ncsa.sh "/usr/lib/squid3/basic_ncsa_auth /etc/squid3/users" "user password"
and i generate the following error:
couldn't execute "/usr/lib/squid3/basic_ncsa_auth /etc/squid3/users": no such file or directory
while executing
"spawn {/usr/lib/squid3/basic_ncsa_auth /etc/squid3/users} {user password}"
("eval" body line 1)
invoked from within
"eval spawn [lrange $argv 0 end]"
(file "./ncsa.sh" line 2)
Besides this error, i am not sure how to pass the variables (username & password) forward and i am also not sure how to answer the shell questions like for example the user & pw input for basic_ldap_auth .
Is there a nice way how to solve that? or any other good plan ?
thanks!

FWIW, the following script helped me transition from passwd based to LDAP based authentication.
Contrary to your requirements, my script acts the other way around: It first checks passwd, then LDAP.
#!/usr/bin/env bash
# multiple Squid basic auth checks
# originally posted here: https://github.com/HackerHarry/mSquidAuth
#
# credits
# https://stackoverflow.com/questions/24147067/verify-user-and-password-against-a-file-created-by-htpasswd/40131483
# https://stackoverflow.com/questions/38710483/how-to-stop-ldapsearch1-from-base64-encoding-userpassword-and-other-attributes
#
# requires ldap-utils, openssl and perl
# tested with Squid 4 using a "auth_param basic program /usr/lib/squid/mSquidAuth.sh" line
# authenticate first against squid password file
# if this fails, try LDAP (Active Directory) and also check group membership
# variables
# sLOGFILE=/var/log/squid/mSquidAuth.log
sPWDFILE="/etc/squid/passwd"
sLDAPHOST="ldaps://dc.domain.local:636"
sBASE="DC=domain,DC=local"
sLDS_OPTIONS="-o ldif-wrap=no -o nettimeout=7 -LLL -P3 -x "
sBINDDN="CN=LDAP-read-user,OU=Users,DC=domain,DC=local"
sBINDPW="read-user-password"
sGROUP="Proxy-Users"
# functions
function _grantAccess {
# echo "access granted - $sUSER" >>$sLOGFILE
echo "OK"
}
function _denyAccess {
# echo "access denied - $sUSER" >>$sLOGFILE
echo "ERR"
}
function _setUserAndPass {
local sAuth="$1"
local sOldIFS=$IFS
IFS=' '
set -- $sAuth
IFS=$sOldIFS
# set it globally
sUSER="$1"
sPASS="$2"
}
# loop
while (true); do
read -r sAUTH
sUSER=""
sPASS=""
sSALT=""
sUSERENTRY=""
sHASHEDPW=""
sUSERDN=""
iDNCOUNT=0
if [ -z "$sAUTH" ]; then
# echo "exiting" >>$sLOGFILE
exit 0
fi
_setUserAndPass "$sAUTH"
sUSERENTRY=$(grep -E "^${sUSER}:" "$sPWDFILE")
if [ -n "$sUSERENTRY" ]; then
sSALT=$(echo "$sUSERENTRY" | cut -d$ -f3)
if [ -n "$sSALT" ]; then
sHASHEDPW=$(openssl passwd -apr1 -salt "$sSALT" "$sPASS")
if [ "$sUSERENTRY" = "${sUSER}:${sHASHEDPW}" ]; then
_grantAccess
continue
fi
fi
fi
# LDAP is next
iDNCOUNT=$(ldapsearch $sLDS_OPTIONS -H "$sLDAPHOST" -D "$sBINDDN" -w "$sBINDPW" -b "$sBASE" "(|(sAMAccountName=${sUSER})(userPrincipalName=${sUSER}))" dn 2>/dev/null | grep -cE 'dn::? ')
if [ $iDNCOUNT != 1 ]; then
# user needs a unique account
_denyAccess
continue
fi
# get user's DN
# we need the extra grep in case we get lines back starting with "# refldap" :/
sUSERDN=$(ldapsearch $sLDS_OPTIONS -H "$sLDAPHOST" -D "$sBINDDN" -w "$sBINDPW" -b "$sBASE" "(|(sAMAccountName=${sUSER})(userPrincipalName=${sUSER}))" dn 2>/dev/null | perl -MMIME::Base64 -n -00 -e 's/\n +//g;s/(?<=:: )(\S+)/decode_base64($1)/eg;print' | grep -E 'dn::? ' | sed -r 's/dn::? //')
# try and bind using that DN to check password validity
# also test if that user is member of a particular group
# backslash in DN needs special treatment
if ldapsearch $sLDS_OPTIONS -H "$sLDAPHOST" -D "$sUSERDN" -w "$sPASS" -b "$sBASE" "name=${sGROUP}" member 2>/dev/null | perl -MMIME::Base64 -n -00 -e 's/\n +//g;s/(?<=:: )(\S+)/decode_base64($1)/eg;print' | grep -q "${sUSERDN/\\/\\\\}"; then
_grantAccess
continue
fi
_denyAccess
done

Related

bash script delete mails from specific address (pop3 account)

I want to clean my mailbox from mails from specific address
I have thousands of messages, I want to do this in bash script, and run it from time to time (a receive SPAM from different addresses, and unfortunately my "spam filters" have only small effect on them)
To interact with a mail server through command line, you could use either telnet or openssl.
You can connect to your pop server using the following command (I've taken gmail as an example. You'll have to look for your email host pop3 address and socket.) :
openssl s_client -connect pop.gmail.com:995 -quiet
As this command is interactive, it will ask for a username, a password and a serie of commands.
expect is a tool that can automate interaction with interactive commands. The basic syntax is as follow : expect "somestring" action -> If the program we monitor displays "somestring", we execute the action.
Here is a script that would delete all the messages present on your email address :
#!/usr/bin/expect
#you can modify the timeout if the script fails
set timeout 1
#our connection variables
set ip "pop.gmail.com"
set socket "995"
set user "user.name"
set pass "password"
#we set the address we want to remove mails from here. Escape special regex characters such as dots.
set target_address "mail\.example#gmail\.com"
#we launch the subprocess we want to interact with
spawn openssl s_client -connect $ip:$socket -quiet
#if connection went all right, we try to login
expect -re ".OK.*" {send "user $user\r"}
expect -re ".OK.*" {send "pass $pass\r"}
#if login went alright, we try to count the messages on the server
#you will get the following output :
#+OK NB_MSG TOTAL_SIZE
expect -re ".OK.*" {send "stat\r"}
#if the stat command went alright ...
expect -re ".OK.*" {
#we extract the number of mail from the output of the stat command
set mail_count [lindex [split [lindex [split $expect_out(buffer) \n] 1] " "] 1]
#we iterate through every email...
for {set i 1} {$i <= $mail_count} {incr i 1} {
#we retrieve the header of the email
send "top $i 0\r"
#if the header contains "To: $target_address" (or "To: <$target_address>" or "To: Contact Name <$target_address>" ...)
#to filter according to the sender, change the regex to "\nFrom: ..."
expect -re "\nTo: \[^\n\]*$target_address" {
#we delete the email
send "dele $i\r"
}
}
}
expect default
You might need to alter your email account settings to allow the use of external programs
Here I have written a similar script in PHP, by use of the IMAP function of PHP. I did not have Linux, so I could not use the bash script. However. I thought that sharing the PHP-script would provide a solution for those running WINDOWS and having access
<?php
echo "My showmail.php script";
// This is the format for opening the email connection:
//
// domain prt foldr[.<.subfolder........>] my#blp.com xxxxxxxx
// $connection = imap_open('{imap.one.com:143}Inbox.<write subfolder here>','email-name','password');
// Execution: (write this in the browsers address bar)
http://digitalageinstitute.com/showmail.php?fromdate=01%20January%202009&todate=28%20August%202018&search=#weheartit.com&email=david.svarrer#digitalageinstitute.com&password=xxxxxxx&domain=imap.one.com&folder=Inbox
//
// This line above, entered into a browsers address bar will execute this script (named showmail.php),
// from the servers root (www.digitalageinstitute.com), and
//
// will delete emails from the mail folder "Inbox" (case sensitive) for the user david.svarrer#digitalageinstitute.com, by use of
// the IMAP domain (this is where you access emails from outside) imap.one.com and at port 143 (see this, further down)
//
$emailaddress = $_GET["email"];
$emailpassword= $_GET["password"];
$fromdate = $_GET["fromdate"];
$todate = $_GET["todate"];
$search = $_GET["search"];
$domain = $_GET["domain"];
$folder = $_GET["folder"];
echo $emailaddress." ".$emailpassword." ".$fromdate." ".$todate." ".$search." ".$domain." ".$folder;
$connection = imap_open('{'.$domain.':143}'.$folder,$emailaddress,$emailpassword);
$ccount = imap_num_msg($connection);
$maxtoexpunge=20000;
echo "Parameters = ".$fromdate.",".$todate.",".$search.":".'ALL FROM "'.$search.'" SINCE "'.$fromdate.'" BEFORE "'.$todate.'"';
// $some = imap_search($connection, 'ALL FROM "Twitter"');
$some = imap_search($connection, 'ALL FROM "'.$search.'" SINCE "'.$fromdate.'" BEFORE "'.$todate.'"');
print_r ($some);
echo "<br/>Elements: ".count($some)."<br/>";
$expunge = 0;
for($msgno = 0; $msgno < count($some) ; $msgno++) {
// echo "Fetching element ".$some[$msgno]."<br/>";
$headers = imap_headerinfo($connection, $some[$msgno]);
//$position = True;
$position = strpos(" ".strtolower($headers->fromaddress), strtolower($search));
$position2 = strpos(" ".strtolower($headers->subject), strtolower($search));
if ($position || $position2) {
if ($expunge<$maxtoexpunge) {
imap_delete ($connection, $some[$msgno]);
echo "<br/>Deleting:".$expunge." sequence [".$msgno."]=".$some[$msgno]." ".$headers->fromaddress.", subject:".$headers->subject;
$expunge++;
} else {
echo "<br/>Skipping:"." sequence [".$msgno."]=".$some[$msgno]." ".$headers->fromaddress.", subject:".$headers->subject;
}
}
}
// The expunge command deletes after all action has been taken. DON'T call it until very end, as it will otherwise mess up message numbers!!
imap_expunge ($connection);
echo "<br/>Expunged!!<br/>";
/* Here are the keywords to be used for the imap search order.
toaddress - full to: line, up to 1024 characters
to - an array of objects from the To: line, with the following properties: personal, adl, mailbox, and host
fromaddress - full from: line, up to 1024 characters
from - an array of objects from the From: line, with the following properties: personal, adl, mailbox, and host
ccaddress - full cc: line, up to 1024 characters
cc - an array of objects from the Cc: line, with the following properties: personal, adl, mailbox, and host
bccaddress - full bcc: line, up to 1024 characters
bcc - an array of objects from the Bcc: line, with the following properties: personal, adl, mailbox, and host
reply_toaddress - full Reply-To: line, up to 1024 characters
reply_to - an array of objects from the Reply-To: line, with the following properties: personal, adl, mailbox, and host
senderaddress - full sender: line, up to 1024 characters
sender - an array of objects from the Sender: line, with the following properties: personal, adl, mailbox, and host
return_pathaddress - full Return-Path: line, up to 1024 characters
return_path - an array of objects from the Return-Path: line, with the following properties: personal, adl, mailbox, and host
remail -
date - The message date as found in its headers
Date - Same as date
subject - The message subject
Subject - Same as subject
in_reply_to -
message_id -
newsgroups -
followup_to -
references -
Recent - R if recent and seen, N if recent and not seen, ' ' if not recent.
Unseen - U if not seen AND not recent, ' ' if seen OR not seen and recent
Flagged - F if flagged, ' ' if not flagged
Answered - A if answered, ' ' if unanswered
Deleted - D if deleted, ' ' if not deleted
Draft - X if draft, ' ' if not draft
Msgno - The message number
MailDate -
Size - The message size
udate - mail message date in Unix time
fetchfrom - from line formatted to fit fromlength characters
fetchsubject - subject line formatted to fit subjectlength characters
*/
?>
I was looking for a bash script and as I did not found one I made my own, only external programm needed is socat for SSL/TLS connect to server:
#!/bin/bash -
#===============================================================================
#
# FILE: delete-mail.sh
#
USAGE="./delete-mail.sh user pass [server] [port] [max_delete]"
#
# PARAMETER: user pass - login credentials, mandatory
# server - mail server, default: imap.1und1.de
# port - port on mail server, default: pop3s
# max_delete - maximum # of messages to delete, default: 1000
#
DESCRIPTION="delete max_delete messages in given pop3 mailbox"
#
# AUTHOR: KayM (), kay#rrr.de
# CREATED: 09.06.2021 18:03:58
# REVISION: ---
#===============================================================================
# check args
if [[ "$1" == "-h"* ]]; then
printf "usage: %s\ndescr: %s\n" "$USAGE" "$DESCRIPTION" 1>&2
exit
elif [ "$#" -lt 2 ]; then
printf "Missing args, usage: %s\n" "$USAGE"
exit 1
fi
# assign values
USER="$1"
PASS="$2"
SERVER="${3:-imap.1und1.de}"
PORT="${4:-pop3s}"
MAX_MESSAGE="${5:-100}"
# define actions
popserver() { socat - "OPENSSL:$SERVER:$PORT" 2>/dev/null; }
poplogin() {
printf "USER %s\n" "$USER"
sleep 0.5
printf "PASS %s\n" "$PASS"
sleep 0.5
}
deletemsg() {
for (( j = 1 ; j <= MAX_MESSAGE; j++ ))
do
printf "DELE %s\n" "$j"
sleep 0.5
done
}
popstat() { echo "STAT"; }
popquit() { echo "QUIT"; }
checkresult() {
local line
while read -r line
do
[[ "$line" == "-ERR "* ]] && MAX_MESSAGE=0
printf "%s\n" "$line"
done
}
# get message count, remove newline
RES="$({ poplogin; popstat; } | popserver | tail -n 1 )"
RES="${RES//[$'\r\n']}"
if [[ "$RES" == "+OK 0 "* ]]; then
printf "No messages to delete (%s)\n" "$RES"
exit
elif [[ "$RES" == "+OK "* ]]; then
: "${RES#+OK }"; NUM="${_%% *}"
printf "%s Messages found (%s)\n" "$NUM" "$RES"
[ "$NUM" -lt "$MAX_MESSAGE" ] && MAX_MESSAGE="$NUM"
else
printf "Can't get number of messages: %s\n" "$RES"
exit 1
fi
if [ "$MAX_MESSAGE" = "0" ]; then
printf "Keep all messages\n"
exit
fi
# delete messages
printf "Delete %s messages ...\n" "$MAX_MESSAGE"
{ poplogin; deletemsg; popstat; popquit;} | popserver | checkresult

I want to use parallel-ssh to run a bash script on multiple servers, but it simple prints the echo statements

I have a bash script called sr_run_batch.sh which does super resolution of images. Now I want to do testing on different servers in parallel at the same time. ie. 1 Virtual machine at one given point of time. then 2 virtual machines at one point of time , 3 and then 4.
I tried writing into it the commands
for host in $(cat hosts.txt); do ssh "$host" "$command" >"output.$host"; done
ssh-keygen && for host in $(cat hosts.txt); do ssh-copy-id $host; done
where the file hosts.txt contains the list of servers: username#ip(format) but when I run this, it gives me substitution error
Hence, I tried pssh (parallel-ssh)
pssh -h hosts-file -l username -P $command
command being ./sr_run_batch.sh
but it didn't run, so I modified this to
pssh -h hosts-file -l ben -P -I<./sr_run_batch.sh
But, for some unknown reason, it just prints the echo statements in the code.
here is the code :
NList=(5)
VList=(1)
FList=("input/flower1.jpg" "input/flower2.jpg" "input/flower3.jpg" "input/flower4.jpg")
IList=("320X240" "640X480" "1280X960" "1920X1200")
SList=(2 3)
for VM in ${VList[#]}; do
for ((index=0; index < ${#FList};)) do
file=$FList[$index]
image_size=$IList[$index]
width=`echo $image_size|cut -d "X" -f1`
height=`echo $image_size|cut -d "X" -f2`
for scale_factor in ${SList[#]}; do
for users in ${NList[#]}; do
echo "V: $VM, " "F: $file, " "S: $scale_factor, " "I: $width $height , " "N: $users"
for i in `seq 1 $users` ; do
./sr_run_once.sh $file $width $height $scale_factor &
done
wait
done # for users
done # for scale_factor
done # for index
done # for VM
exit 0
Have you also tried to use pssh with a simple bash-script so see if the communication is set up ok?
$ pssh -h hosts.txt -A -l ben -P -I<./uptime.sh
Warning: do not enter your password if anyone else has superuser
privileges or access to your account.
Password:
10.0.0.67: 11:06:50 up 28 min, 2 users, load average: 0.00, 0.00, 0.00
[1] 11:06:50 [SUCCESS] 10.0.0.67
10.0.0.218: 11:06:50 up 24 min, 2 users, load average: 0.00, 0.05, 0.20
[2] 11:06:50 [SUCCESS] 10.0.0.218

Expect fails but I don't see why

I have a bash script that gets info from Heroku so that I can pull a copy of my database. That script works fine in cygwin. But to run it in cron it halts because the shell that it uses halts as Heroku's authentication through Heroku Toolbelt.
Here is my crontab:
SHELL=/usr/bin/bash
5 8-18 * * 1-5 /cygdrive/c/Users/sam/work/push_db.sh >>/cygdrive/c/Users/sam/work/output.txt
I have read the Googles and the man page within cygwin to come up with this addition:
#!/usr/bin/bash
. /home/sam.walton/.profile
echo $SHELL
curl -H "Accept: application/vnd.heroku+json; version=3" -n https://api.heroku.com/
#. $HOME/.bash_profile
echo `heroku.bat pgbackups:capture --expire`
#spawn heroku.bat pgbackups:capture --expire
expect {
"Email:" { send -- "$($HEROKU_LOGIN)\r"}
"Password (typing will be hidden):" { send -- "$HEROKU_PW\r" }
timeout { echo "timed out during login"; exit 1 }
}
sleep 2
echo "first"
curl -o latest.dump -L "$(heroku.bat pgbackups:url | dos2unix)"
Here's the output from the output.txt
/usr/bin/bash
{
"links":[
{
"rel":"schema",
"href":"https://api.heroku.com/schema"
}
]
}
Enter your Heroku credentials. Email: Password (typing will be hidden): Authentication failed. Enter your Heroku credentials. Email: Password (typing will be hidden): Authentication failed. Enter your Heroku credentials. Email: Password (typing will be hidden): Authentication failed.
As you can see it appears that the output is not getting the result of the send command as it appears it's waiting. I've done so many experiments with the credentials and the expect statements. All stop here. I've seen few examples and attempted to try those out but I'm getting fuzzy eyed which is why I'm posting here. What am I not understanding?
Thanks to comments, I'm reminded to explicitly place my env variables in .bashrc:
[[ -s $USERPROFILE/.pik/.pikrc ]] && source "$USERPROFILE/.pik/.pikrc"
export HEROKU_LOGIN=myEmailHere
export HEROKU_PW=myPWhere
My revised script per #Dinesh's excellent example is below:
. /home/sam.walton/.bashrc echo $SHELL echo $HEROKU_LOGIN curl -H "Accept: application/vnd.heroku+json; version=3" -n https://api.heroku.com/
expect -d -c " spawn heroku.bat pgbackups:capture --expire --app gw-inspector expect {
"Email:" { send -- "myEmailHere\r"; exp_continue}
"Password (typing will be hidden):" { send -- "myPWhere\r" }
timeout { puts "timed out during login"; exit 1 } } " sleep 2 echo "first"
This should work but while the echo of the variable fails, giving me a clue that the variable is not being called, I am testing hardcoding the variables directly to eliminate that as a variable. But as you can see by my output not only is the echo yielding nothing, there is no clue that any diagnostics are being passed which makes me wonder if the script is even being called to run from expect, as well as the result of the spawn command. To restate, the heroku.bat command works outside the expect closure but the results are above. The result of the command directly above is:
/usr/bin/bash
{
"links":[
{
"rel":"schema",
"href":"https://api.heroku.com/schema"
}
]
}
What am I doing wrong that will show me diagnostic notes?
If you are going to use the expect code inside your bash script, instead of calling it separately, then you should have use the -c flag option.
From your code, I assume that you have the environmental variables HEROKU_LOGIN and HEROKU_PW declared in the bashrc file.
#!/usr/bin/bash
#Your code here
expect -c "
spawn <your-executable-process-here>
expect {
# HEROKU_LOGIN & HEROKU_PW will be replaced with variable values.
"Email:" { send -- "$HEROKU_LOGIN\r";exp_continue}
"Password (typing will be hidden):" { send "$HEROKU_PW\r" }
timeout { puts"timed out during login"; exit 1 }
}
"
#Your further bash code here
You should not use echo command inside expect code. Use puts instead. The option of spawning the process inside expect code will be more robust than spawning it outside.
Notice the use of double quotes with the expect -c flag. If you use single quotes, then bash script won't do any form of substitution. So, if you need bash variable substitution, you should use double quotes for the expect with -c flag.
To know about usage of -c flag, have a look at here
If you still have any issue, you can debug by appending -d with the following way.
expect -d -c "
our code here
"

Hide password using expect script

I used the script to log into the power broker and connected
ssh through the network. As per my knowledge every single more is
tracked. I am ok with that.
My question is how to hide the password, below sample I tried:
#!/usr/bin/expect -f
set timeout 2
spawn pbrun /bin/su - big_admin
expect {
"Password for casper#BIG_BANK.COM:" {send "foobar_pass!\r" ; exp_continue }
"*$*" ;
}
interact
I tried storing my password in a /home/casper/.password and then doing a chmod 400 on it and then
putting it in the {send "/home/casper/.password\r" exp continue}, but it did not work.
Any ideas?
Source : http://wiki.tcl.tk/3594
Procedure to encrypt device passwords
Takes pd (password list to be encrypted) and filename (name of file for resulting encrypted password list) as input
Returns nothing
Format of the list of passwords (current and older generations by device type?) depends on how calling program needs them - suggest separated by \n character for readability
proc utility_encrypt {pd filename} {
global key HOME
#catch [exec echo "$pd" | des -e -k $key -b > $HOME/pwdir/$filename]
catch [exec des -e -k $key -b > [file join $HOME pwdir $filename] << $pd]
return
}
Procedure to decrypt device passwords
Takes filename (name of password file to decrypt)
Returns dpd (list of passwords in plaintext)
proc utility_decrypt filename {
global key HOME
catch {exec cat $HOME/pwdir/$filename | des -d -b -k $key} dpd
# Some people write the previous command as
# catch {exec des -d -b -k $key < $HOME/pwdir/$filename} dpd
return $dpd
}

Minimal web server using netcat

I'm trying to set up a minimal web server using netcat (nc). When the browser calls up localhost:1500, for instance, it should show the result of a function (date in the example below, but eventually it'll be a python or c program that yields some data).
My little netcat web server needs to be a while true loop in bash, possibly as simple as this:
while true ; do echo -e "HTTP/1.1 200 OK\n\n $(date)" | nc -l -p 1500 ; done
When I try this the browser shows the currently available data during the moment when nc starts. I want the browser displays the data during the moment the browser requests it, though. How can I achieve this?
Try this:
while true ; do nc -l -p 1500 -c 'echo -e "HTTP/1.1 200 OK\n\n $(date)"'; done
The -cmakes netcat execute the given command in a shell, so you can use echo. If you don't need echo, use -e. For further information on this, try man nc. Note, that when using echo there is no way for your program (the date-replacement) to get the browser request. So you probably finally want to do something like this:
while true ; do nc -l -p 1500 -e /path/to/yourprogram ; done
Where yourprogram must do the protocol stuff like handling GET, sending HTTP 200 etc.
I had the problem where I wanted to return the result of executing a bash command:
$ while true; do { echo -e 'HTTP/1.1 200 OK\r\n'; sh test; } | nc -l 8080; done
NOTE:
This command was taken from: http://www.razvantudorica.com/08/web-server-in-one-line-of-bash
This executes a bash script and returns the result to a browser client connecting to the server running this command on port 8080.
My script does this:
$ nano test
#!/bin/bash
echo "************PRINT SOME TEXT***************\n"
echo "Hello World!!!"
echo "\n"
echo "Resources:"
vmstat -S M
echo "\n"
echo "Addresses:"
echo "$(ifconfig)"
echo "\n"
echo "$(gpio readall)"
and my web browser is showing
************PRINT SOME TEXT***************
Hello World!!!
Resources:
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
r b swpd free buff cache si so bi bo in cs us sy id wa
0 0 0 314 18 78 0 0 2 1 306 31 0 0 100 0
Addresses:
eth0 Link encap:Ethernet HWaddr b8:27:eb:86:e8:c5
inet addr:192.168.1.83 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:27734 errors:0 dropped:0 overruns:0 frame:0
TX packets:26393 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:1924720 (1.8 MiB) TX bytes:3841998 (3.6 MiB)
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:65536 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
GPIOs:
+----------+-Rev2-+------+--------+------+-------+
| wiringPi | GPIO | Phys | Name | Mode | Value |
+----------+------+------+--------+------+-------+
| 0 | 17 | 11 | GPIO 0 | IN | Low |
| 1 | 18 | 12 | GPIO 1 | IN | Low |
| 2 | 27 | 13 | GPIO 2 | IN | Low |
| 3 | 22 | 15 | GPIO 3 | IN | Low |
| 4 | 23 | 16 | GPIO 4 | IN | Low |
| 5 | 24 | 18 | GPIO 5 | IN | Low |
| 6 | 25 | 22 | GPIO 6 | IN | Low |
| 7 | 4 | 7 | GPIO 7 | IN | Low |
| 8 | 2 | 3 | SDA | IN | High |
| 9 | 3 | 5 | SCL | IN | High |
| 10 | 8 | 24 | CE0 | IN | Low |
| 11 | 7 | 26 | CE1 | IN | Low |
| 12 | 10 | 19 | MOSI | IN | Low |
| 13 | 9 | 21 | MISO | IN | Low |
| 14 | 11 | 23 | SCLK | IN | Low |
| 15 | 14 | 8 | TxD | ALT0 | High |
| 16 | 15 | 10 | RxD | ALT0 | High |
| 17 | 28 | 3 | GPIO 8 | ALT2 | Low |
| 18 | 29 | 4 | GPIO 9 | ALT2 | Low |
| 19 | 30 | 5 | GPIO10 | ALT2 | Low |
| 20 | 31 | 6 | GPIO11 | ALT2 | Low |
+----------+------+------+--------+------+-------+
Add -q 1 to the netcat command line:
while true; do
echo -e "HTTP/1.1 200 OK\n\n $(date)" | nc -l -p 1500 -q 1
done
The problem you are facing is that nc does not know when the web client is done with its request so it can respond to the request.
A web session should go something like this.
TCP session is established.
Browser Request Header: GET / HTTP/1.1
Browser Request Header: Host: www.google.com
Browser Request Header: \n #Note: Browser is telling Webserver that the request header is complete.
Server Response Header: HTTP/1.1 200 OK
Server Response Header: Content-Type: text/html
Server Response Header: Content-Length: 24
Server Response Header: \n #Note: Webserver is telling browser that response header is complete
Server Message Body: <html>sample html</html>
Server Message Body: \n #Note: Webserver is telling the browser that the requested resource is finished.
The server closes the TCP session.
Lines that begin with "\n" are simply empty lines without even a space and contain nothing more than a new line character.
I have my bash httpd launched by xinetd, xinetd tutorial. It also logs date, time, browser IP address, and the entire browser request to a log file, and calculates Content-Length for the Server header response.
user#machine:/usr/local/bin# cat ./bash_httpd
#!/bin/bash
x=0;
Log=$( echo -n "["$(date "+%F %T %Z")"] $REMOTE_HOST ")$(
while read I[$x] && [ ${#I[$x]} -gt 1 ];do
echo -n '"'${I[$x]} | sed -e's,.$,",'; let "x = $x + 1";
done ;
); echo $Log >> /var/log/bash_httpd
Message_Body=$(echo -en '<html>Sample html</html>')
echo -en "HTTP/1.0 200 OK\nContent-Type: text/html\nContent-Length: ${#Message_Body}\n\n$Message_Body"
To add more functionality, you could incorporate.
METHOD=$(echo ${I[0]} |cut -d" " -f1)
REQUEST=$(echo ${I[0]} |cut -d" " -f2)
HTTP_VERSION=$(echo ${I[0]} |cut -d" " -f3)
If METHOD = "GET" ]; then
case "$REQUEST" in
"/") Message_Body="HTML formatted home page stuff"
;;
/who) Message_Body="HTML formatted results of who"
;;
/ps) Message_Body="HTML formatted results of ps"
;;
*) Message_Body= "Error Page not found header and content"
;;
esac
fi
Happy bashing!
Another way to do this
while true; do (echo -e 'HTTP/1.1 200 OK\r\n'; echo -e "\n\tMy website has date function" ; echo -e "\t$(date)\n") | nc -lp 8080; done
Let's test it with 2 HTTP request using curl
In this example, 172.16.2.6 is the server IP Address.
Server Side
admin#server:~$ while true; do (echo -e 'HTTP/1.1 200 OK\r\n'; echo -e "\n\tMy website has date function" ; echo -e "\t$(date)\n") | nc -lp 8080; done
GET / HTTP/1.1 Host: 172.16.2.6:8080 User-Agent: curl/7.48.0 Accept:
*/*
GET / HTTP/1.1 Host: 172.16.2.6:8080 User-Agent: curl/7.48.0 Accept:
*/*
Client Side
user#client:~$ curl 172.16.2.6:8080
My website has date function
Tue Jun 13 18:00:19 UTC 2017
user#client:~$ curl 172.16.2.6:8080
My website has date function
Tue Jun 13 18:00:24 UTC 2017
user#client:~$
If you want to execute another command, feel free to replace $(date).
I had the same need/problem but nothing here worked for me (or I didn't understand everything), so this is my solution.
I post my minimal_http_server.sh (working with my /bin/bash (4.3.11) but not /bin/sh because of the redirection):
rm -f out
mkfifo out
trap "rm -f out" EXIT
while true
do
cat out | nc -l 1500 > >( # parse the netcat output, to build the answer redirected to the pipe "out".
export REQUEST=
while read -r line
do
line=$(echo "$line" | tr -d '\r\n')
if echo "$line" | grep -qE '^GET /' # if line starts with "GET /"
then
REQUEST=$(echo "$line" | cut -d ' ' -f2) # extract the request
elif [ -z "$line" ] # empty line / end of request
then
# call a script here
# Note: REQUEST is exported, so the script can parse it (to answer 200/403/404 status code + content)
./a_script.sh > out
fi
done
)
done
And my a_script.sh (with your need):
#!/bin/bash
echo -e "HTTP/1.1 200 OK\r"
echo "Content-type: text/html"
echo
date
mkfifo pipe;
while true ;
do
#use read line from pipe to make it blocks before request comes in,
#this is the key.
{ read line<pipe;echo -e "HTTP/1.1 200 OK\r\n";echo $(date);
} | nc -l -q 0 -p 8080 > pipe;
done
Here is a beauty of a little bash webserver, I found it online and forked a copy and spruced it up a bit - it uses socat or netcat I have tested it with socat -- it is self-contained in one-script and generates its own configuration file and favicon.
By default it will start up as a web enabled file browser yet is easily configured by the configuration file for any logic. For files it streams images and music (mp3's), video (mp4's, avi, etc) -- I have tested streaming various file types to Linux,Windows and Android devices including a smartwatch!
I think it streams better than VLC actually. I have found it useful for transferring files to remote clients who have no access beyond a web browser e.g. Android smartwatch without needing to worry about physically connecting to a USB port.
If you want to try it out just copy and paste it to a file named bashttpd, then start it up on the host with $> bashttpd -s
Then you can go to any other computer (presuming the firewall is not blocking inbound tcp connections to port 8080 -- the default port, you can change the port to whatever you want using the global variables at the top of the script). http://bashttpd_server_ip:8080
#!/usr/bin/env bash
#############################################################################
###########################################################################
### bashttpd v 1.12
###
### Original author: Avleen Vig, 2012
### Reworked by: Josh Cartwright, 2012
### Modified by: A.M.Danischewski, 2015
### Issues: If you find any issues leave me a comment at
### http://scriptsandoneliners.blogspot.com/2015/04/bashttpd-self-contained-bash-webserver.html
###
### This is a simple Bash based webserver. By default it will browse files and allows for
### retrieving binary files.
###
### It has been tested successfully to view and stream files including images, mp3s,
### mp4s and downloading files of any type including binary and compressed files via
### any web browser.
###
### Successfully tested on various browsers on Windows, Linux and Android devices (including the
### Android Smartwatch ZGPAX S8).
###
### It handles favicon requests by hardcoded favicon image -- by default a marathon
### runner; change it to whatever you want! By base64 encoding your favorit favicon
### and changing the global variable below this header.
###
### Make sure if you have a firewall it allows connections to the port you plan to
### listen on (8080 by default).
###
### By default this program will allow for the browsing of files from the
### computer where it is run.
###
### Make sure you are allowed connections to the port you plan to listen on
### (8080 by default). Then just drop it on a host machine (that has bash)
### and start it up like this:
###
### $192.168.1.101> bashttpd -s
###
### On the remote machine you should be able to browse and download files from the host
### server via any web browser by visiting:
###
### http://192.168.1.101:8080
###
#### This program requires (to work to full capacity) by default:
### socat or netcat (w/ '-e' option - on Ubuntu netcat-traditional)
### tree - useful for pretty directory listings
### If you are using socat, you can type: bashttpd -s
###
### to start listening on the LISTEN_PORT (default is 8080), you can change
### the port below.
### E.g. nc -lp 8080 -e ./bashttpd ## <-- If your nc has the -e option.
### E.g. nc.traditional -lp 8080 -e ./bashttpd
### E.g. bashttpd -s -or- socat TCP4-LISTEN:8080,fork EXEC:bashttpd
###
### Copyright (C) 2012, Avleen Vig <avleen#gmail.com>
###
### Permission is hereby granted, free of charge, to any person obtaining a copy of
### this software and associated documentation files (the "Software"), to deal in
### the Software without restriction, including without limitation the rights to
### use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
### the Software, and to permit persons to whom the Software is furnished to do so,
### subject to the following conditions:
###
### The above copyright notice and this permission notice shall be included in all
### copies or substantial portions of the Software.
###
### THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
### IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
### FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
### COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
### IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
### CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
###
###########################################################################
#############################################################################
### CHANGE THIS TO WHERE YOU WANT THE CONFIGURATION FILE TO RESIDE
declare -r BASHTTPD_CONF="/tmp/bashttpd.conf"
### CHANGE THIS IF YOU WOULD LIKE TO LISTEN ON A DIFFERENT PORT
declare -i LISTEN_PORT=8080
## If you are on AIX, IRIX, Solaris, or a hardened system redirecting
## to /dev/random will probably break, you can change it to /dev/null.
declare -a DUMP_DEV="/dev/random"
## Just base64 encode your favorite favicon and change this to whatever you want.
declare -r FAVICON="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"
declare -i DEBUG=1
declare -i VERBOSE=0
declare -a REQUEST_HEADERS
declare REQUEST_URI=""
declare -a HTTP_RESPONSE=(
[200]="OK"
[400]="Bad Request"
[403]="Forbidden"
[404]="Not Found"
[405]="Method Not Allowed"
[500]="Internal Server Error")
declare DATE=$(date +"%a, %d %b %Y %H:%M:%S %Z")
declare -a RESPONSE_HEADERS=(
"Date: $DATE"
"Expires: $DATE"
"Server: Slash Bin Slash Bash"
)
function warn() { ((${VERBOSE})) && echo "WARNING: $#" >&2; }
function chk_conf_file() {
[ -r "${BASHTTPD_CONF}" ] || {
cat >"${BASHTTPD_CONF}" <<'EOF'
#
# bashttpd.conf - configuration for bashttpd
#
# The behavior of bashttpd is dictated by the evaluation
# of rules specified in this configuration file. Each rule
# is evaluated until one is matched. If no rule is matched,
# bashttpd will serve a 500 Internal Server Error.
#
# The format of the rules are:
# on_uri_match REGEX command [args]
# unconditionally command [args]
#
# on_uri_match:
# On an incoming request, the URI is checked against the specified
# (bash-supported extended) regular expression, and if encounters a match the
# specified command is executed with the specified arguments.
#
# For additional flexibility, on_uri_match will also pass the results of the
# regular expression match, ${BASH_REMATCH[#]} as additional arguments to the
# command.
#
# unconditionally:
# Always serve via the specified command. Useful for catchall rules.
#
# The following commands are available for use:
#
# serve_file FILE
# Statically serves a single file.
#
# serve_dir_with_tree DIRECTORY
# Statically serves the specified directory using 'tree'. It must be
# installed and in the PATH.
#
# serve_dir_with_ls DIRECTORY
# Statically serves the specified directory using 'ls -al'.
#
# serve_dir DIRECTORY
# Statically serves a single directory listing. Will use 'tree' if it is
# installed and in the PATH, otherwise, 'ls -al'
#
# serve_dir_or_file_from DIRECTORY
# Serves either a directory listing (using serve_dir) or a file (using
# serve_file). Constructs local path by appending the specified root
# directory, and the URI portion of the client request.
#
# serve_static_string STRING
# Serves the specified static string with Content-Type text/plain.
#
# Examples of rules:
#
# on_uri_match '^/issue$' serve_file "/etc/issue"
#
# When a client's requested URI matches the string '/issue', serve them the
# contents of /etc/issue
#
# on_uri_match 'root' serve_dir /
#
# When a client's requested URI has the word 'root' in it, serve up
# a directory listing of /
#
# DOCROOT=/var/www/html
# on_uri_match '/(.*)' serve_dir_or_file_from "$DOCROOT"
# When any URI request is made, attempt to serve a directory listing
# or file content based on the request URI, by mapping URI's to local
# paths relative to the specified "$DOCROOT"
#
#unconditionally serve_static_string 'Hello, world! You can configure bashttpd by modifying bashttpd.conf.'
DOCROOT=/
on_uri_match '/(.*)' serve_dir_or_file_from
# More about commands:
#
# It is possible to somewhat easily write your own commands. An example
# may help. The following example will serve "Hello, $x!" whenever
# a client sends a request with the URI /say_hello_to/$x:
#
# serve_hello() {
# add_response_header "Content-Type" "text/plain"
# send_response_ok_exit <<< "Hello, $2!"
# }
# on_uri_match '^/say_hello_to/(.*)$' serve_hello
#
# Like mentioned before, the contents of ${BASH_REMATCH[#]} are passed
# to your command, so its possible to use regular expression groups
# to pull out info.
#
# With this example, when the requested URI is /say_hello_to/Josh, serve_hello
# is invoked with the arguments '/say_hello_to/Josh' 'Josh',
# (${BASH_REMATCH[0]} is always the full match)
EOF
warn "Created bashttpd.conf using defaults. Please review and configure bashttpd.conf before running bashttpd again."
# exit 1
}
}
function recv() { ((${VERBOSE})) && echo "< $#" >&2; }
function send() { ((${VERBOSE})) && echo "> $#" >&2; echo "$*"; }
function add_response_header() { RESPONSE_HEADERS+=("$1: $2"); }
function send_response_binary() {
local code="$1"
local file="${2}"
local transfer_stats=""
local tmp_stat_file="/tmp/_send_response_$$_"
send "HTTP/1.0 $1 ${HTTP_RESPONSE[$1]}"
for i in "${RESPONSE_HEADERS[#]}"; do
send "$i"
done
send
if ((${VERBOSE})); then
## Use dd since it handles null bytes
dd 2>"${tmp_stat_file}" < "${file}"
transfer_stats=$(<"${tmp_stat_file}")
echo -en ">> Transferred: ${file}\n>> $(awk '/copied/{print}' <<< "${transfer_stats}")\n" >&2
rm "${tmp_stat_file}"
else
## Use dd since it handles null bytes
dd 2>"${DUMP_DEV}" < "${file}"
fi
}
function send_response() {
local code="$1"
send "HTTP/1.0 $1 ${HTTP_RESPONSE[$1]}"
for i in "${RESPONSE_HEADERS[#]}"; do
send "$i"
done
send
while IFS= read -r line; do
send "${line}"
done
}
function send_response_ok_exit() { send_response 200; exit 0; }
function send_response_ok_exit_binary() { send_response_binary 200 "${1}"; exit 0; }
function fail_with() { send_response "$1" <<< "$1 ${HTTP_RESPONSE[$1]}"; exit 1; }
function serve_file() {
local file="$1"
local CONTENT_TYPE=""
case "${file}" in
*\.css)
CONTENT_TYPE="text/css"
;;
*\.js)
CONTENT_TYPE="text/javascript"
;;
*)
CONTENT_TYPE=$(file -b --mime-type "${file}")
;;
esac
add_response_header "Content-Type" "${CONTENT_TYPE}"
CONTENT_LENGTH=$(stat -c'%s' "${file}")
add_response_header "Content-Length" "${CONTENT_LENGTH}"
## Use binary safe transfer method since text doesn't break.
send_response_ok_exit_binary "${file}"
}
function serve_dir_with_tree() {
local dir="$1" tree_vers tree_opts basehref x
## HTML 5 compatible way to avoid tree html from generating favicon
## requests in certain browsers, such as browsers in android smartwatches. =)
local no_favicon=" <link href=\"data:image/x-icon;base64,${FAVICON}\" rel=\"icon\" type=\"image/x-icon\" />"
local tree_page=""
local base_server_path="/${2%/}"
[ "$base_server_path" = "/" ] && base_server_path=".."
local tree_opts="--du -h -a --dirsfirst"
add_response_header "Content-Type" "text/html"
# The --du option was added in 1.6.0. "/${2%/*}"
read _ tree_vers x < <(tree --version)
tree_page=$(tree -H "$base_server_path" -L 1 "${tree_opts}" -D "${dir}")
tree_page=$(sed "5 i ${no_favicon}" <<< "${tree_page}")
[[ "${tree_vers}" == v1.6* ]]
send_response_ok_exit <<< "${tree_page}"
}
function serve_dir_with_ls() {
local dir="$1"
add_response_header "Content-Type" "text/plain"
send_response_ok_exit < \
<(ls -la "${dir}")
}
function serve_dir() {
local dir="$1"
# If `tree` is installed, use that for pretty output.
which tree &>"${DUMP_DEV}" && \
serve_dir_with_tree "$#"
serve_dir_with_ls "$#"
fail_with 500
}
function urldecode() { [ "${1%/}" = "" ] && echo "/" || echo -e "$(sed 's/%\([[:xdigit:]]\{2\}\)/\\\x\1/g' <<< "${1%/}")"; }
function serve_dir_or_file_from() {
local URL_PATH="${1}/${3}"
shift
URL_PATH=$(urldecode "${URL_PATH}")
[[ $URL_PATH == *..* ]] && fail_with 400
# Serve index file if exists in requested directory
[[ -d "${URL_PATH}" && -f "${URL_PATH}/index.html" && -r "${URL_PATH}/index.html" ]] && \
URL_PATH="${URL_PATH}/index.html"
if [[ -f "${URL_PATH}" ]]; then
[[ -r "${URL_PATH}" ]] && \
serve_file "${URL_PATH}" "$#" || fail_with 403
elif [[ -d "${URL_PATH}" ]]; then
[[ -x "${URL_PATH}" ]] && \
serve_dir "${URL_PATH}" "$#" || fail_with 403
fi
fail_with 404
}
function serve_static_string() {
add_response_header "Content-Type" "text/plain"
send_response_ok_exit <<< "$1"
}
function on_uri_match() {
local regex="$1"
shift
[[ "${REQUEST_URI}" =~ $regex ]] && \
"$#" "${BASH_REMATCH[#]}"
}
function unconditionally() { "$#" "$REQUEST_URI"; }
function main() {
local recv=""
local line=""
local REQUEST_METHOD=""
local REQUEST_HTTP_VERSION=""
chk_conf_file
[[ ${UID} = 0 ]] && warn "It is not recommended to run bashttpd as root."
# Request-Line HTTP RFC 2616 $5.1
read -r line || fail_with 400
line=${line%%$'\r'}
recv "${line}"
read -r REQUEST_METHOD REQUEST_URI REQUEST_HTTP_VERSION <<< "${line}"
[ -n "${REQUEST_METHOD}" ] && [ -n "${REQUEST_URI}" ] && \
[ -n "${REQUEST_HTTP_VERSION}" ] || fail_with 400
# Only GET is supported at this time
[ "${REQUEST_METHOD}" = "GET" ] || fail_with 405
while IFS= read -r line; do
line=${line%%$'\r'}
recv "${line}"
# If we've reached the end of the headers, break.
[ -z "${line}" ] && break
REQUEST_HEADERS+=("${line}")
done
}
if [[ ! -z "{$1}" ]] && [ "${1}" = "-s" ]; then
socat TCP4-LISTEN:${LISTEN_PORT},fork EXEC:"${0}"
else
main
source "${BASHTTPD_CONF}"
fail_with 500
fi
LOL, a super lame hack, but at least curl and firefox accepts it:
while true ; do (dd if=/dev/zero count=10000;echo -e "HTTP/1.1\n\n $(date)") | nc -l 1500 ; done
You better replace it soon with something proper!
Ah yes, my nc were not exactly the same as yours, it did not like the -p option.
If you're using Apline Linux, the BusyBox netcat is slightly different:
while true; do nc -l -p 8080 -e sh -c 'echo -e "HTTP/1.1 200 OK\n\n$(date)"'; done
And another way using printf:
while true; do nc -l -p 8080 -e sh -c "printf 'HTTP/1.1 200 OK\n\n%s' \"$(date)\""; done
while true; do (echo -e 'HTTP/1.1 200 OK\r\nConnection: close\r\n';) | timeout 1 nc -lp 8080 ; done
Closes connection after 1 sec, so curl doesn't hang on it.
Type in nc -h and see if You have -e option available. If yes, You can create a script, for example:
script.sh
echo -e "HTTP/1.1 200 OK\n\n $(date)"
and run it like this:
while true ; do nc -l -p 1500 -e script.sh; done
Note that -e option needs to be enabled at compilation to be available.
I think the problem that all the solution listed doesn't work, is intrinsic in the nature of http service, the every request established is with a different client and the response need to be processed in a different context, every request must fork a new instance of response...
The current solution I think is the -e of netcat but I don't know why doesn't work... maybe is my nc version that I test on openwrt...
with socat it works....
I try this https://github.com/avleen/bashttpd
and it works, but I must run the shell script with this command.
socat tcp-l:80,reuseaddr,fork EXEC:bashttpd &
The socat and netcat samples on github doesn't works for me, but the socat that I used works.
Actually, the best way to close gracefully the connection is to send the Content-Length header like following. Client (like curl will close the connection after receiving the data.
DATA="Date: $(date)";
LENGTH=$(echo $DATA | wc -c);
echo -e "HTTP/1.1 200 OK\nContent-Length: ${LENGTH}\n\n${DATA}" | nc -l -p 8000;
On OSX you can use :
while true; do echo -e "HTTP/1.1 200 OK\n\n $(date)" | nc -l localhost 1500 ; done

Resources