SWAT's blog

Recently I received my new phone, which is a Samsung Galaxy S GT-I9000 (Europe version, running Android 2.2), and I was looking for a hypothetical way to get root access on it. After reading a lot of howto’s and informational posts/pages (see references at the bottom), which sometimes seemed overly complicated or superfluous, I decided to find an easy way to do it better (and by solely using Linux). This solution “works for me” and will probably void your warranty, so proceed at your own risk.

Root your Samsung Galaxy in 4 easy steps:

1. Make sure the “3 button combo” is working. Boot your phone once holding “volume button up + home button + power button” and one holding “volume button down + home button + power button” and check if you get into “Recovery mode” and “Download mode”.
2. Install “CF-root” by using Heimdall, which is a cross-platform open-source tool suite used to flash firmware (aka ROMs) onto Samsung Galaxy S devices. The CF-root archive contains a zImage which can easily be flashed onto your phone (make sure your phone is in “Download mode” – “volume button down + home button + power button”). I had to use the JPM version of CF-root. Note: Heimdall is cross-platform whereas Odin is Windows-only.
   heimdall flash –kernel zImage
3. Install the ClockWorkMod by putting the update.zip on the phone (from your PC) and applying it using the “Recovery mode”. You do not need a 3rd party tool like “ROM manager” to do this. Note: ROM manager can send out anonymous statistics etc.
4. Install Darky’s ROM by putting the ZIP file on the phone (from your PC) and applying it using “Recovery mode” (which now is ClockWorkMod) using the “Install ZIP from SD card” option. You will see a strange blue reboot text and the phone will reboot, but do not be alarmed, have patience!

Note: thanks go out to the developers and how to / post writers for providing the necessary required information to make this possible

References:

• CF-root (on XDA-developers)
• Flash/Root/ADM (on XDA-developers)
• Darky’s ROM (on XDA-developers)

Eventually I also got convinced to use GNOME-Do (Launchpad)  and I do not regret it. I use the keyboard even more and use the mouse less, what is there not to love? You can even enable searching your bookmarks, contacts etc. which means that you can open up a conversation window or an e-mail draft right away (yes, it is that easy). Unfortunately all the really cool stuff is still upstream, so if you want to latest/greatest version you have 3 options available:

1. Download it from the website (don’t do it, you want the packaged/maintained version)
2. Use the PPA of the GNOME Do Core Team to get the latest upstream version which is packaged (I used this version which is for those of us who like to live on the edge)
3. Upgrade to Jaunty (the release is only 2 days away!)

If you choose option 2), I can give you this small how-to:

1. Add the PPA’s key to your system:
   

   sudo apt-key adv –recv-keys –keyserver keyserver.ubuntu.com a5d19fdcaa6abb440cd3464628a8205077558dd0

   Source

2. Add the repository to your /etc/apt/sources.list
   

   deb http://ppa.launchpad.net/do-core/ppa/ubuntu intrepid main
   deb-src http://ppa.launchpad.net/do-core/ppa/ubuntu intrepid main

3. Update your cache and install
   

   sudo apt-get update && sudo apt-get install gnome-do gnome-do-plugins

Note for 64-bit users! There is a bug in Launchpad’s ppa that causes 64-bit mono package builds to fail, but David was in the marines for 12 years and only learned one thing- Leave no man behind. This message doesn’t apply to plugins, however. Plugins are architecture independent and build fine in the PPA. You should use plugins from the PPA, and for core we’ve got a package for you at http://launchpad.net/do/+download (source)

The title of this blog post, pretty much says it all. Since reading “long” documents or manuals is an extremely hard and time-consuming occupation for most people, I just made this ‘how to’ for all the really lazy people .

Note: Only register your nickname and get a cloak if you tend to use IRC on a very regular basis
Why should you care about getting a cloak? A cloak hides your hostname. A hostname can be used to the IP address which belongs to it. So if you do not want everyone to know your IP address, get a cloak. Furthermore, a cloak enables IRC channel operators to give a someone an ‘invite’ based on it, instead of a hostname invite. This is extremely usefull if some users have dynamic IP addresses and the operator does not want to change the invite list on a regular basis.

1. Choose and switch to your (not already registered or occupied) nickname.

   /nick your_nickname

2. Register your nickname.

   /msg nickserv register your_password your@email.address

3. Hide your e-mail.

   /msg nickserv set hidemail on

4. Change to your alternate nickname. I suggest you use, your normal nickname appended with an underscore.

   /nick alternate_nickname

5. Link your alternate nickname to your nickname.

   /msg nickserv group

6. Switch back to your normal nickname.

   /nick original_nickname

7. Check who is currently on-duty.

   /stats p

   If nobody is on duty, you can get a list of the staffers who could help you out.

   /who freenode/staff/*

   If nobody is on duty, go to #freenode, ask politely for an unaffiliated cloak and wait. Do not request such a cloak repetitively if you do not get an answer immediately, but wait at least 3 hours between requests.

8. Ask the staffer politely for an unaffiliated cloak.
9. Set up your IRC client to automatically identify (log in) yourself to nickserv. You can also do this manually:

   /msg nickserv identify

Source: http://freenode.net/faq.shtml

Some months ago, I searched for a new music player yet again. A little history: When I started with linux, I was started using XMMS as my default music player. After a while I heard that Beep-media-player was the improved version of XMMS and even had Winamp skins support. This just had to be the player I was searching for… and it was.
After my iPod nano purchase, I made a habit of using Banshee for all my iPod related transfers. Although I do not like .Net and therefore Mono, I give this application the benefit of the doubt, since it gets the work done. I still was not completely satisfied with my regular music player though.
A friend of mine suggested using MPD and Sonata and it is a total bliss. Imagine hearing your music play, even when your X.org is down! MPD is the music daemon, to whom the Sonata client connects. If you do not like Sonata, you can use another MPD client, there are several available in the repository.
You can even easily extend MPD to include Icecast, so you can stream your music. This is quite cool though. If you are on a remote location and connect to your MPD server with an MPD client and connect to your icecast stream, you can control the playlist and listen to your music everywhere. That is, if you have sufficient bandwidth of course.

Note: You will need one MPD client to control the remote playlist and a music player to listen to the stream

1. Install the needed software

   sudo apt-get install mpd icecast2 sonata

2. Setup icecast2

   Edit /etc/default/icecast2 and change this to ‘true’:

   ENABLED=true

   Change at least the <authentification> and <hostname> section of /etc/icecast2/icecast.xml :

   <authentification>
   <source-password>foobar21</source-password>
   <relay-password>foobar54</relay-password>
   <admin-user>admin</admin-user>
   <admin-password>foobar532</admin-password>
   </authentification>
   
   <hostname>foobar5126</hostname>

3. Setup mpd

   Edit your /etc/mpd.conf and edit the ‘audio_output’ section:

   audio_output {
   type            "shout"
   name            "Icecast2 stream"
   host            "foobar5126"
   port            "8000"
   mount           "/mpd_stream.ogg"
   password        "foobar21"
   bitrate         "192"
   format          "44100:16:1"
   # Optional Paramters
   user        "source"
   description "Music"
   genre       "misc"
   }

That is it. Enjoy your music stream either on the server or on any client that listens to your icecast stream!

It took me a while to get Red Alert and the expansion Counterstrike working correctly on Ubuntu. I wanted to use a nocd patch, but unfortunately I could not find a working one that also works with Counterstrike. I tried Cedega and Wine and they both did not work completely. Sometimes the CD could not be found or Counterstrike could not be installed etc.. The solution was: dosbox!

Note: Red alert has a DOS and a Windows95 version. I installed the DOS version, which has a great installer (ow sweet, fond memories)
Some general dosbox tips:

• Mount a CD: “mount D D:\ -t cdrom”
• Mount a CD (if it is not recognized by Red Alert correctly): “mount D D:\ -t cdrom -usecd 0″
• Capture/Release the mouse (if you need to use the mouse within DOSBox): CTRL + F10
• Umount: “mount -u d:”
• Switch to full-screen (and back): ALT + ENTER

How do you get Red Alert working with dosbox on your computer?

1. Make images of the original CD’s since nocd patches will get you nowhere. (only if you do not want to use the original CD’s)

cdrdao read-cd --read-raw --datafile game_name.bin --device /dev/scd0 --driver generic-mmc-raw game_name.toc

2. Create a custom dosbox directory

mkdir dosbox_redalert

3. Start dosbox and mount your dosbox directory and CD image

mount c: /home/swat/dosbox_redalert/
imgmount d: /home/swat/redalert.bin -t iso

4. Install Red Alert and any expansions. I installed Counterstrike after installing Red Alert.

d:
setup

5. If you want, you can patch the game to any version (I installed the 1.08 patch). The patch will first ‘unarchive’ itself and then you will have to copy the PATCH.* files to the Red Alert directory and run the patching wizard.
Enjoy the good old feeling of Red Alert!

Before I had this blog, I used drupal as blogging software. After I was using it for a while, I found it bloated and it used PHP so it became less appealing to me. Apparently I ‘lost’ one precious, use full blog entry, which I had to retrieve using an old database backup. Here it is:
My HD crashed (due to power-failure) and I fixed it

Maybe this story has a strange title but it’s true. Yesterday the entire region here had no power for about 2 hours. I was just working and suddenly my PC’s and monitors turned off. I thought I did something (like consuming too much power), because I just connected another computer to the power-outlet. As I went downstairs to check the fuse-box, I saw that everything was OK. Then I went outside, talked to the neighbours and they had the same problem. Damn, total power-outage.

TV, radio, PC, internet (router), fridge, phone (digital) were out-of-order. Even my cellphone had bad reception, because the cell-stations had no power. When the power went on again, I saw the damage that was done to one of my HD’s. The horror, the corruption, the evil…

swat@voyager:~$ mount /mnt/hd200/
mount: wrong fs type, bad option, bad superblock on /dev/hdb1,
missing codepage or other error
In some cases useful info is found in syslog - try
dmesg | tail  or so

swat@voyager:~$ dmesg | grep -i hdb
[4294730.583000] hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error }
[4294730.583000] hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=183, high=0, low=183, sector=168
[4294730.583000] end_request: I/O error, dev hdb, sector 168
[4294731.923000] hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error }
[4294731.923000] hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=183, high=0, low=183, sector=176
[4294731.923000] end_request: I/O error, dev hdb, sector 176
[4294731.923000] Buffer I/O error on device hdb, logical block 22
[4294733.338000] hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error }
[4294733.338000] hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=184, high=0, low=184, sector=184
[4294733.338000] end_request: I/O error, dev hdb, sector 184
[4294734.686000] hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error }
[4294734.686000] hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=183, high=0, low=183, sector=183
[4294734.686000] end_request: I/O error, dev hdb, sector 183
[4294734.686000] EXT3-fs error (device hdb1): ext3_get_inode_loc: unable to read inode block - inode=8, block=15
[4294827.909000] hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error }
[4294827.909000] hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=183, high=0, low=183, sector=183
[4294827.909000] end_request: I/O error, dev hdb, sector 183
[4294827.910000] EXT3-fs error (device hdb1): ext3_get_inode_loc: unable to read inode block - inode=8, block=15

swat@voyager:~$ fsck /dev/hdb1
fsck.ext3: Attempt to read block from filesystem resulted in short read while trying to open /dev/hdb1
Could this be a zero-length partition?

It looks bad, doesn’t it? Luckily it could be fixed by something this simple:

mke2fs -n /dev/hdb1
e2fsck -b 32768 /dev/hdb1

Time to party, like it’s 1999