@echo off
SETLOCAL
set TARGET_PATH=%~dp0
set IN_FILENAME="%TARGET_PATH%Version.txt"
set OUT_FILENAME="%TARGET_PATH%Version.cs"
set /p BUILD= < %IN_FILENAME%
set BUILD=%BUILD:~22%
if /I "%BUILD%" == "" set BUILD=0
set /a BUILD=%BUILD%+1
echo #define VERSION_BUILD %BUILD% > %IN_FILENAME%
echo using System; > %OUT_FILENAME%
echo namespace Angler { >> %OUT_FILENAME%
echo 	class Version { >> %OUT_FILENAME%
echo 		public const int BuildCount 	= %BUILD%; >> %OUT_FILENAME%
echo 		public const string BuildDate 	= "%DATE%"; >> %OUT_FILENAME%
echo 		public const string BuildTime 	= "%TIME%"; >> %OUT_FILENAME%
echo 	} >> %OUT_FILENAME%
echo } >> %OUT_FILENAME%
ENDLOCAL

I went the easy route with this by simply dropping the build count into an additional file.
I wanted to avoid parsing the value out of the .cs file. But you’re welcome to improve on this :P

When you first run it, it will create the VersionInfo.txt file which will contain the initial build count (for you to adjust).
I place the script in my project directory as “Increase build count.cmd” and add this as a pre-build event: “$(ProjectDir)Increase build count.cmd” (INCLUDING the quotes!)
Now you simply add Version.cs to your project and use its members in your program.

First, you wanna fire up your old friend Wireshark. Now set up the device you want to emulate to send the data to the IP of your machine. Now you can start a Live Capture in Wireshark (Ctrl+E). Now you command your device to send the message by whatever means needed (like pressing the correct button on the device). Now stop the Live Capture in Wireshark (Ctrl+E) and add a filter to get the packet you need.

Let’s assume the devices IP address is 10.0.0.1 and your IP address is 10.0.0.2. Now the correct filter would be

ip.src == 10.0.0.1 && ip.dst == 10.0.0.2

Now, hopefully you’ll see a single packet. In the center panel of Wireshark there should be a Data segment. This is what we want. Right-click it and select Copy -> Bytes (Hex Stream). You’ll get something like:

4e656574732c4b2c0100

(In case you’re interested, this is the payload when you press the first button on a Neets Control EU Standard device.)

Now let’s put this into code. The below example already has 2 payloads that would signal the target device to be turned on or off. For this 2 packets have been captured and can now be selected via the command line. The implementation is straight-forward.

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Net.Sockets;

namespace Neets_Control {
  class Program {

    private static byte[] WORKLOAD_ON   = new byte[] { 0x4e, 0x65, 0x65, 0x74, 0x73, 0x2c, 0x4b, 0x2c, 0x01, 0x00 };
    private static byte[] WORKLOAD_OFF  = new byte[] { 0x4e, 0x65, 0x65, 0x74, 0x73, 0x2c, 0x4b, 0x2c, 0x02, 0x00 };

    private static string TARGET_HOST   = "10.11.110.11";
    private static int    TARGET_PORT   = 5009;

    static void Main( string[] args ) {

      TcpClient sender = new TcpClient( TARGET_HOST, TARGET_PORT );
      NetworkStream stream = sender.GetStream();
      if( args[ 0 ] == "on" ) {
        Console.WriteLine( "Switching on" );
        stream.Write( WORKLOAD_ON, 0, WORKLOAD_ON.Length );

      } else if( args[ 0 ] == "off" ) {
        Console.WriteLine( "Switching off" );
        stream.Write( WORKLOAD_OFF, 0, WORKLOAD_OFF.Length );

      } else {
        Console.WriteLine( "Missing parameter (on/off)." );
      }

      stream.Close();
      sender.Close();

    }
  }
}

So, yeah, I can now just refer back to my introduction. Possibly this wasn’t event worth mentioning ;)

In this project we have a Windows domain of about 100 machines which are running our software are mainly used to display “stuff” (let’s not go into details).
When I arrived on site where we deployed our product I found that some technicians where using .bat files to deploy files to machines in the domain. After deploying their files they would restart the machine remotely through Windows remote desktop feature. They would repeat that process for every machine in the domain manually until the desired group was up-to-date.
Needless to say, for a programmer, this seemed unnecessarily cumbersome. This was when I wrote the first helpful tool.

The tool will read a .xml file which contains target IP addresses (or machine names if DNS is available) and will use them as a parameter to a supplied batch file. Alternatively it would use a list of targets supplied via the command line. Thus enabling you to run a .bat file for a large set of targets.
Let’s have a look at the source:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading;
using System.IO;
using System.Xml;
using System.Diagnostics;

namespace BatchProcessor {
  class Program {

    private const   String      APPLICATION_NAME    = "Batch Processor";
    private const   String      APPLICATION_VERSION = "0.1";

    private static ConsoleColor defaultColor        = ConsoleColor.Gray;

    static void Main( string[] args ) {

      Console.WindowWidth = 160;
      Console.Title = String.Format( "{0} {1}", APPLICATION_NAME, APPLICATION_VERSION );

      Console.ForegroundColor = ConsoleColor.Cyan;
      Console.WriteLine( String.Format( "{0} {1}", APPLICATION_NAME, APPLICATION_VERSION ) );
      Console.ForegroundColor = ConsoleColor.DarkCyan;
      Console.WriteLine( "CPP Studios Event GmbH 2009" );
      Console.ForegroundColor = defaultColor;

      string        batchFile   = String.Empty;
      string        targetsFile = String.Empty;
      List<string>  targets     = new List<string>();

      if( args.Length < 2 ) {
        errorExit( "Incorrect number of arguments" );
        return;
      }

      batchFile   = args[ 0 ];
      targetsFile = args[ 1 ];

      // Check if targetsFile is actually a file
      FileInfo targetsFileInfo = new FileInfo( targetsFile );
      if( !targetsFileInfo.Exists ) {
        for( int targetIndex = 1; targetIndex < args.Length; ++targetIndex ) {
          targets.Add( args[ targetIndex ] );
        }

      } else {
        XmlDocument targetFile = new XmlDocument();
        try {
          Console.WriteLine( String.Format( "Reading targets from {0}...", targetsFile ) );
          targetFile.Load( targetsFile );

          XmlNodeList targetNodes = targetFile.SelectNodes( "/targets/target" );
          foreach( XmlNode targetNode in targetNodes ) {
            targets.Add( targetNode.Attributes[ "ip" ].InnerText );
          }

        } catch( Exception ex ) {
          Console.ForegroundColor = ConsoleColor.Red;
          Console.WriteLine( "Error while reading targets: " + ex.Message );
          Console.ForegroundColor = defaultColor;
        }
      }
      Console.WriteLine( String.Format( "Got {0} targets", targets.Count ) );

      foreach( String target in targets ) {
        string commandLine = String.Format( "{0} {1}", batchFile, target );
        Console.ForegroundColor = ConsoleColor.DarkGreen;
        Console.WriteLine( String.Format( "Executing '{0}'...", commandLine ) );
        Console.ForegroundColor = defaultColor;

        ProcessStartInfo  p     = new ProcessStartInfo( batchFile );
        Process           proc  = new Process();

        p.Arguments               = target;
        p.RedirectStandardOutput  = true;
        p.UseShellExecute         = false;
        proc.StartInfo            = p;

        proc.Start();
        StreamReader outputReader = proc.StandardOutput;

        proc.WaitForExit();
        Console.Write( outputReader.ReadToEnd() );

        Console.ForegroundColor = ConsoleColor.DarkGreen;
        Console.WriteLine( String.Format( "Finished processing '{0}'.", commandLine ) );
        Console.ForegroundColor = defaultColor;
      }

      Console.ForegroundColor = ConsoleColor.Magenta;
      Console.WriteLine( "Operation completed." );
      Console.ForegroundColor = defaultColor;

    }

    private static void errorExit( String errorMessage ) {
      Console.ForegroundColor = ConsoleColor.Red;
      Console.WriteLine( errorMessage );
      Console.ForegroundColor = defaultColor;
      Thread.Sleep( 5000 );
    }

  }
}

Pretty much a straight forward implementation of what I explained above.
What amazed me most about the tool was the simplicity and how much can be achieved by the approach. Everyone who is capable of writing a .bat file could plug it right into the system and apply the command set to our pre-defined groups. I like to think that it turned out to be a very simple, yet versatile tool.

Another thing to notice is the console redirection used here. It’s pretty much what you will find if you google for “c# redirect console output” in a few seconds. Soon I noticed that it’s a very shitty way to redirect console output and leaves a lot to wish for. I will cover that topic in more detail in an upcoming post.

I start with a clean Ubuntu 8.04 install. So, first things first:

sudo apt-get update
sudo apt-get dist-upgrade
sudo shutdown -r now

Now that we’re up-to-date, let’s begin:

sudo apt-get build-dep php5
sudo apt-get install libmcrypt-dev libc-client-dev checkinstall

Now we have all the crap to build PHP with. Let’s get PHP itself:

cd /tmp/
wget http://us3.php.net/get/php-5.3.0.tar.gz/from/this/mirror
tar xvzf php-5.3.0.tar.gz
cd php-5.3.0/

Now all that is left is to build PHP…

./configure --prefix=/usr --with-apxs2=/usr/bin/apxs2 --with-config-file-path=/etc/php5/apache2 --with-config-file-scan-dir=/etc/php5/apache2/conf.d --mandir=/usr/share/man --disable-debug --with-regex=php --disable-rpath --disable-static --with-pic --with-layout=GNU --with-pear=/usr/share/php --enable-calendar --enable-sysvsem --enable-sysvshm --enable-sysvmsg --enable-bcmath --with-bz2 --enable-ctype --with-db4 --without-gdbm --with-iconv --enable-exif --enable-ftp --with-gettext --enable-mbstring --with-pcre-regex=/usr --enable-shmop --enable-sockets --enable-wddx --with-libxml-dir=/usr --with-zlib --with-kerberos=/usr --with-openssl=/usr --enable-soap --enable-zip --with-exec-dir=/usr/lib/php5/libexec --without-mm --with-curl=shared,/usr --with-zlib-dir=/usr --with-gd=shared,/usr --enable-gd-native-ttf --with-gmp=shared,/usr --with-jpeg-dir=shared,/usr --with-xpm-dir=shared,/usr/X11R6 --with-png-dir=shared,/usr --with-freetype-dir=shared,/usr --with-t1lib=shared,/usr --with-ldap=shared,/usr --with-ldap-sasl=/usr --with-mhash=shared,/usr --with-mysql=shared,/usr --with-mysqli=/usr/bin/mysql_config --with-pspell=shared,/usr --with-unixODBC=shared,/usr --with-xsl=shared,/usr --with-snmp=shared,/usr --with-sqlite=shared,/usr --with-mssql=shared,/usr --with-tidy=shared,/usr --with-xmlrpc=shared --with-pgsql=shared,/usr --enable-gd-native-ttf --enable-dba=shared --with-openssl-dir=shared,/usr --enable-gd-jis-conv --enable-json --with-mcrypt=shared,/usr --enable-pcntl --with-pdo-mysql --with-pdo-odbc=unixODBC,/usr --with-pdo-pgsql=shared,/usr --with-pdo-sqlite --enable-xmlreader --with-tsrm-pthreads --with-imap --with-imap-ssl

I’m sure this doesn’t enable ALL features in PHP, but it works for me so far. If I’ll ever find I was missing something, I’ll be sure to update this post.

When building make sure to pass the correct number of CPUs to make with the -j parameter to safe some time.

make -j2

And now we wait…

Now we need to add these lines to our httpd.conf (which is empty in this case)

#LoadModule directive to aid module installations
#LoadModule dummy_module /usr/lib/apache2/modules/mod_dummy.so

Now we can build our .deb for PHP.

sudo checkinstall -D --install=no --fstrans=no --maintainer=your@email --reset-uids=yes --nodoc --pkgname=php5 --pkgversion=5.3 --pkgrelease=200908060830 --arch=amd64

And that’s it. Now you have a .deb which you can install on a different system or (as checkinstall told you) just put on this system with dpkg -i.

Have fun

The first thing we’ll need is the screen package.

sudo apt-get install screen

Now let’s create a user for steam and switch to it.

sudo adduser --disabled-login steam
sudo su - steam

Let’s download gSteam, my frontend, and set it executable for only steam.

wget http://www.dirty-motherfucker.org/steam/gSteam.sh
chmod 700 gSteam.sh

Now we can install the steam client.

./gSteam.sh install steam

As noted in my post regarding the L4D server, if the installation does not seem to properly start or
hang, you can try to create a Steam.cfg file (in the homedir) with preferred content server ids.

http://www.dirty-motherfucker.org/blog/2008/12/26/running-a-source-dedicated-server-for-left-4-dead/

After you get the magic line “Steam Linux Client updated, please retry the command” you are ready to
install some games. So let’s install Team Fortress 2.

./gSteam.sh install tf

The identifiers for the games can be obtained with:

./steam -command list

The “No installation record found at…” messages can be safely ignored.
Now you will have to wait until the game is fully downloaded. This can take from “forever” to
“until the fucking end of time” in some cases. The old Steam.cfg can help here as well, but
I personally found that patience is the best tool.

Once the installation is completed a profile template for tf will be created. gSteam will try its
best to determine some default parameters. You should edit it nevertheless to at least adjust
the name of your server (and possibly the profile).

If everything worked fine you can now simply

./gSteam start tf2

and the thing should be running.

Installing other games is as simple as:

./gSteam install l4d_full

Let me just add that the script is far from fail-proof. Especially error-cases aren’t properly handled. But it saved me a lot of time so I thought I’d share it.
Hope it helps someone else ;)

@echo off
SETLOCAL
set TARGET_PATH=%~dp0
set FILENAME="%TARGET_PATH%build.h"
set /p BUILD= < %FILENAME%
set BUILD=%BUILD:~22%
if /I "%BUILD%" == "" set BUILD=0
set /a BUILD=%BUILD%+1
echo #define VERSION_BUILD %BUILD% > %FILENAME%
echo #define VERSION_DATE L"%DATE%" >> %FILENAME%
echo #define VERSION_TIME L"%TIME%" >> %FILENAME%
ENDLOCAL

In case it’s not obvious what this batch file does. First we construct our target filename so that it is still valid even if VS calls the batch from another directory. Then we read in the first line of our build.h (the VERSION_BUILD #define), then we cut off the part before the build number, add 1 and write it back. I also add 2 #defines for date and time, cause, why not? I made date and time already wide strings but left the build count an integer. This makes it slightly easier to parse for the batch file ;)

So just throw the two files in a folder in your project and run the batch once so the build.h is created. Then open up your project properties and add the batch as a new Pre-Build Event. It makes most sense just to do it for Release builds. Now whenever you wanna reference the build count, just #include the build.h. Done.

I guess it would be desirable to solve this whole thing without macros. But at this time I lack the energy to make it more complex. Maybe in a few years…

A while ago I have ditched PsyBNC in favor of ZNC. And let me tell you, I am not missing that POS that is PsyBNC. So, so much for the software bashing. Back to topic.

When using an IRC bouncer I prefer to have the recorded buffer played back to me when I re-connect. ZNC has several options for that. First of all there are the text buffers it will play back to you. Which is pretty standard. ZNC will play back the message as if the user said it and can either append or prepend the time when it happened. Sadly this does not look like the usual user text and is one of the problems this script targets.
Then there is the buffextras module which will additionally play back generic IRC events as text. These will appear as if a user (*buffextras) sends text to the channel like with the generic buffer playback. The same problems as above here apply.
You can go a little further though. You have the savebuff module which will play back IRC events in a certain format that can be parsed by a client-side script. This is the second problem this script targets.

The script will take all the buffer playback output from ZNC and format it as if it actually happened in the live session. You can grab the script from the ZNC wiki.

So, what to take from this post? If you’re a PsyBNC user, switch to ZNC right now! If you’re a mIRC user and already use ZNC, great, grab the script. If you don’t use mIRC or don’t use IRC at all, well, I have just wasted your time :D

BOOL consoleAttached = AttachConsole( ATTACH_PARENT_PROCESS );

boost::program_options::options_description desc( "Allowed options" );
desc.add_options()
	( "help", "Show this message" )
	( "width", "Backbuffer width" )
	( "height", "Backbuffer height" )
	;

wstring commandLine = GetCommandLine();
int     argc = 0;
LPWSTR* argv = 0;
argv = CommandLineToArgvW( commandLine.c_str(), &argc );

boost::program_options::variables_map vm;
boost::program_options::store( boost::program_options::parse_command_line( argc, argv, desc ), vm );
boost::program_options::notify( vm );

LocalFree( argv );

if( vm.count( "help" ) ) {
	if( consoleAttached ) {
		HANDLE hStdOut = GetStdHandle( STD_OUTPUT_HANDLE );
		if( INVALID_HANDLE_VALUE == hStdOut ) {
			return 1;
		}
		DWORD dwCharsWritten;
		std::stringstream desc_strings;
		desc_strings << desc;
		std::string desc_string = desc_strings.str();

		WriteConsoleA( hStdOut, desc_string.c_str(), desc_string.length(), &dwCharsWritten, NULL );
		FreeConsole();
	}
	return 1;
}

Sadly after the application finishes the prompt doesn’t instantly return. Maybe I’ll look into that some day…

As all of my administration articles (so far) this one is for Ubuntu Server 8.04.1.

Let’s start by adding a new user that will run our servers.

sudo adduser --disabled-login steam

Now to log in with the newly created user to install the servers. We need to switch the user as root as a normal user can not log into this account.

sudo su - steam

No we can install the HLDSUpdateTool.

wget http://www.steampowered.com/download/hldsupdatetool.bin
chmod +x hldsupdatetool.bin
./hldsupdatetool.bin
./steam

In case it will hang at “Checking bootstrapper version…” you can try to set a set of preferred content servers. You do that by creating a file called “steam.cfg” and adding the following line to it:

PreferredContentServerIDs = "id id id"

You can find the ids on http://store.steampowered.com/stats/content/ under “Individual Server Statisistics”. Look for Valve #NN servers and place the number of the servers you want to use inside the quotes. Then retry running steam.

Now we’ll install a game. Let’s start out with Left 4 Dead. If you want to see what games are available, run:

./steam -command list

This will also show you the correct game identifier for Left 4 Dead, which is l4d_full (left4dead was for the demo).

./steam -command update -game l4d_full -dir /home/steam

In case the installation process hangs you can also try setting the preferred content servers as mentioned above. Now, theoretically you’re good to go and you can start your server. But we’re gonna go a step further. I use a start script based on the article mentioned at the start of this article. Now let’s grab it.
But wait, in case you realize that you really don’t want to use any of my shitty scripts and that you just want to get a clean server running, in that case it can stilll be helpful to get the script. You see when you use my script and you start a profile it will print out the command line it used. That way you can build up your desired configuration according to this guide, then run the script and get the correct command line to run your server. Thus I still highly recommend getting the script.

wget http://www.dirty-motherfucker.org/blog/wp-content/uploads/2008/12/steamLauncher.sh
chmod +x steamLauncher.sh

The script allows to easily start and stop Source servers in the background. It also allows for several profiles for several games. So you can start all games through the same interface. I like it, if you don’t then maybe have a look at the original script.

So let’s create a profile for Left 4 Dead.

mkdir profiles
editor profiles/l4d.conf

Now, this is how a profile should look like:

longName                = Left 4 Dead - 1 instance      ; The name of this profile/game
gameName                = left4dead                     ; The name given by steam to the game (equals the foldername ex. cstrike|tf2|left4dead)
daemon                  = /home/steam/l4d/srcds_run     ; Usually srcds_run
updater                 = /home/steam/steam             ; Where is the updater located? (steam binary)
basePath                = /home/steam/l4d               ; The base path of the steam installation (this folder contains tf2/cstrike/left4dead folders)
userName                = steam                         ; What user the server should run as

serverName              = My L4D Server                 ; The name that will appear in the server browser
serverIp                = 11.222.33.111                 ; The IP address to bind the server socket to
serverPort              = 27010                         ; The first port to use for listening (will be treated as base port with forked l4d servers)
serverMap               = l4d_hospital01_apartment      ; The first map to load
serverMaxPlayers        = 4                             ; Maximum number of players allowed on the server
serverPriority          = 0                             ; Renice to set for the server process (-20 is high, 0 is normal, +19 is low)
forkCount               = 0                             ; How many instances to fork
additionalParams        = -nohltv -steamport 27960+## +clientport 25030+## +exec server##.cfg +sv_lan 0 ; Additional params to run the server with

Obviously in that profile you need to adjust at least the serverIp parameter. Now you can run that profile by typing.

./steamLauncher start l4d

Keep in mind though, this is not a state that we want to be comfortable with just yet. Nevertheless start it up and check if there is any trouble. If it comes up, attach to the screen session as the script tells you to.
In case you get an error saying “Cannot open terminal ‘/dev/pts/0′”, you can run “script /dev/null” and then attach to the session. There are other ways to resolve it. If you wanna find out about those they’re easy to find in Google ;)

Now, in the profile of Left 4 Dead in the additionParameters we also tell the server to execute a certain config file. Let’s create that one now.

editor l4d/left4dead/cfg/server01.cfg

In here we can now override the default map, difficulty and whatnot. For example:

hostname        My L4D Server - No Mercy - Normal Difficulty
map             l4d_hospital01_apartment
m_difficulty    normal

Setting the difficulty and map is almost pointless though as they can be changed by vote anyway. But I’m sure you can think of some other server variables that you might want to set. And this is the place to do it.

Now might be a good time to talk about forking with Left 4 Dead servers. Given that there are only 4 or 8 players on a Left 4 Dead server it’s almost a waste only running a single instance. That’s why you can fork multiple instances of servers. Let’s have a look at a Left 4 Dead profile modified to run 4 servers.

longName                = Left 4 Dead - 4 instances     ; The name of this profile/game
gameName                = left4dead                     ; The name given by steam to the game (equals the foldername ex. cstrike|tf2|left4dead)
daemon                  = /home/steam/l4d/srcds_run     ; Usually srcds_run
updater                 = /home/steam/steam             ; Where is the updater located? (steam binary)
basePath                = /home/steam/l4d               ; The base path of the steam installation (this folder contains tf2/cstrike/left4dead folders)
userName                = steam                         ; What user the server should run as

serverName              = My L4D Server                 ; The name that will appear in the server browser
serverIp                = 11.222.33.111                 ; The IP address to bind the server socket to
serverPort              = 27010                         ; The first port to use for listening (will be treated as base port with forked l4d servers)
serverMap               = l4d_hospital01_apartment      ; The first map to load
serverMaxPlayers        = 4                             ; Maximum number of players allowed on the server
serverPriority          = 0                             ; Renice to set for the server process (-20 is high, 0 is normal, +19 is low)
forkCount               = 4                             ; How many instances to fork
additionalParams        = -nohltv -steamport 27960+## +clientport 25030+## +exec server##.cfg +sv_lan 0 ; Additional params to run the server with

You can now also create server02.cfg, server03.cfg and server04.cfg in the l4d folder. This way you can start up each instance with a different map, difficulty, key, …
It remains a mystery to me how to properly send commands to the console with forked instances though. Although from what I read there are ways, but I just never had any luck with those. I also never had any luck with rcon. So basically when I play on my own servers I can’t do shit when someone is getting on my nerves. And we sure don’t want that!
So what do we do? We install SourceMod. It gives us a nice and easy way to kick players when they get on our balls. And we can be happy little server admins. During the installation of SourceMod it may be helpful to run only a single instance profile of Left 4 Dead so you can easily attach to the screen session and send of some commands to test your installation.
So let’s get to it!

First of all we require MetaMod. A very straight-forward guide can be found on the MetaMod Wiki. For completeness sake I’ll replicate the required information here though.

1. Download MetaMod at http://www.metamodsource.net/
2. Put it into the left4dead folder. (cd l4d/left4dead)
3. tar xvzf mmsource-1.7.0.tar.gz (Filename might differ)
4. In l4d/left4dead/addons create a file metamod.vdf and put the following into it:

   "Plugin"
   {
           "file"  "../left4dead/addons/metamod/bin/server_i486.so"
   }
   

5. Go back into your home directory and restart your server (./steamLauncher restart l4d)
6. Attach to the screen session and type “meta version”. Everything other than “Unknown command” is a good sign.

If the installation of MetaMod fails, refer to the MetaMod Wiki or bug me about it.

Now we’re gonna install SourceMod. Again a proper guide can be found at http://wiki.alliedmods.net/Installing_SourceMod.

1. Grab one of the latest snapshots of SourceMod from http://www.sourcemod.net/snapshots-1.2.php
2. Just like we did with MetaMod, put it into the l4d/left4dead folder.
3. Extract it via tar xvzf sourcemod-1.2.0-hg2469.tar.gz (Filename may differ again)
4. Go back into your home directory and restart your server (./steamLauncher restart l4d)
5. Attach to the screen session and type “meta list”. You should see that SourceMod is loaded.

Now last but not least we need to set ourself up as an admin in SourceMod. We do that by editing

l4d/left4dead/addons/sourcemod/configs/admins_simple.ini

Right at the bottom we will now put the following line:

"STEAM_YOURID" "99:z"

If you’re like me then you’ll have no clue what your Steam ID is. So open up steam and connect to any server in any game. Then open up the console and type “status”. In the list that appears there should be your nick somewhere. And close to it, you’ll find your Steam ID.

Now you should be an admin on your Left 4 Dead servers. You can join your own servers easily by invoking “openserverbrowser” in the Left 4 Dead console. To enjoy being an admin it helps to bind “sm_admin” to some key. Then you can open the admin menu and kick those pesky players who just can’t stop running ahead to get themselves killed.

In case this isn’t clear. You’ll have to forward some ports for your servers to work. Which can easily be accomplished by adding two lines to your /etc/ufw/after.rules file. Similar to these two:

# Steam TF2 Server
-A ufw-after-input -d 11.222.33.111 -m udp -p udp --dport 26900:27040 -m conntrack --ctstate NEW,RELATED,ESTABLISHED -j ACCEPT
-A ufw-after-input -d 11.222.33.111 -m tcp -p tcp --dport 27000:27020 -m conntrack --ctstate NEW,RELATED,ESTABLISHED -j ACCEPT

I hope I haven’t left any open questions. In case I did, please let me know ;)

But, oh well, I didn’t make the finals :( And I don’t plan on entering it again. Although after seeing the source of the finalists I assume I would even be able to add more features. Which would be using techniques that I assumed were frowned upon ;)

So, because of the length of some lines, no code in this post, only a link to the source.
And this is the resulting swf file.

I hope someone gets something out of this, cause I sure didn’t :D