Gnuton's Blog

Compiling a custom Ubuntu Kernel step by step

gnuton — Sun, 27 Dec 2009 11:30:59 +0000

I want to list here brefly all steps to compile a custom Ubuntu kernel

sudo -s
apt-get install kernel-source kernel-package libncurses5-dev fakeroot
cd /usr/src
tar xvjf tar xjvf linux-source-`uname -r`.tar.bz2
ln -s /usr/src/linux-source-2.6.31 linux
cd /usr/src/linux
cp /boot/config-`uname -r` /usr/src/linux/.config
make menuconfig
Now you can modify the kernel configuration…
make-kpkg clean
fakeroot make-kpkg –initrd –append-to-version=-custom kernel_image kernel_headers

After some time you will have debian packages like: linux-image-xxx.deb linux-headers-xxx.deb in /usr/src

Use dpkg -i *.deb to install it, reboot your machine and select them in grub/lilo.

Translate

Text Adventures or Interactive Fiction

gnuton — Tue, 15 Dec 2009 13:22:06 +0000

Today I found an Interactive fiction interpreter running in several platforms… and of course Linux is one of them…
After trying it in my linux box I tried to compile it for my N900. Still there are some issues compiling JAM (a replacement for make) for X86 target…

The interpreter is called Gargoyle and it claims to be able read all kind of IF file types.

Ubuntu packages are available here: https://launchpad.net/~radix/+archive/ppa
A lot of Quests are available… use google to find them…

Acquistare un account megavideo premium a 30% in meno dall Europa

gnuton — Mon, 30 Nov 2009 09:53:08 +0000

Grazie a questo video apparso su youtube, son venuto a conoscenza di un trucchetto carino per salvare qualche euro durante l’acquisto di un account megavideo premium.
In pratica tale “procedura” ti consente di acquistare un account megavideo/megaupload premium pagando in dollari l’importo che avresti pagato in euro.
Per esempio invece di pagare 79,99 Euro un account di 2 anni, lo paghi 79,99 dollari (=53 euro circa).

Sfortunatamente i proxy americani sono difficili da reperire. Ce ne sono di gratuiti, ma non permettono la connessione a Paypal o son lentissimi.

Io ho usato questo: planetlab03.cs.washington.edu:3128 e questa e’ la procedura da seguire.
1. mi registro a megavideo.com
2. attivo il proxy americano
3. essendo ancora loggato nel sito, vado sul tab “premium” e clicco sull icona di paypal per acuistare l opzione premium desiderata
4. la connessione si interrompe e’ viene mostrata un pagina di errore, questo accade perche il proxy non mi permette l acquisto. (NO PROBLEM!)
5. disattivo il proxy
6. aggiorno la pagina che mostra l errore di connessione, cliccando su SI quando la finestra di dialogo chiede se si vuole
inviare i dati nuovamente.
7. Posso ora accedere a paypal. L importo da pagare viene mostrato nella schermata subito o dopo il login.

8. Pochi minuti dopo il pagamento una mail di notifica ci conferma l’attivazione del nostro account premium.

Sharing di un Account Megavideo tra utenti:

L account megavideo dovrebbe permettere la connessione contemporanea di max 5 utenti. L’account viene disattivato per 6 ore nel caso in cui + di 5 IP diversi sono connessi con lo stesso account premium.
Dopo le 6 ore, la password dell account viene resettata ed inviata alla mail usata al momento della registrazione dell account.

My first post written with ScribeFire

gnuton — Sun, 22 Nov 2009 08:53:34 +0000

This is a test post written with a Firefox add-on..

DD-WRT: Remove SSH login banner

gnuton — Mon, 16 Nov 2009 22:15:49 +0000

Despite DD-WRT is a cool firmware there are some default things that looks very crap from a security point of view. One of these is the ssh login banner used by sshd.

To remove it use Web interface > Administration > Commands > paste

echo "Unauthorized access is prohibited." > /tmp/loginprompt killall dropbear dropbear -b /tmp/loginprompt -r /tmp/root/.ssh/ssh_host_rsa_key -d /tmp

and press "Save Startup" button.
Power light will blink for 30 seconds. After that sshd will show new banner.

gnuton@iron:~$ ssh root@192.168.1.1
Unauthorized access is prohibited.
root@192.168.1.1's password:

Edit webpages with in your web browser

gnuton — Wed, 24 Jun 2009 19:08:57 +0000

There are some javascript tricks able to enable the hidden design mode.
To do that you just need to put this: “javascript:document.body.contentEditable%20=%20′true’;%20document.designMode=’on’;%20void%200″ in your address bar and you can switch it off with “javascript:document.body.contentEditable%20=%20′false’;%20document.designMode=’off’;%20void%200″

The latest command doesn’t work in Firefox3.
These commands should instead should work in Firefox3 too:
“javascript:document.body.contentEditable=”true”;void0″ and “javascript:document.body.contentEditable=”false”;void0″

Resume scp transfers

gnuton — Mon, 22 Jun 2009 20:31:15 +0000

If you are coping a file (uploading/downloading) with ssh and the connection is interrupted. Then if you need to continue to downloading/uploading the latest processed file, then you can define scpresume:
alias scpresume=”rsync –partial –progress –rsh=ssh”
and then you can use it with the same syntax used by scp.

resolv.conf must be a symlink

gnuton — Mon, 22 Jun 2009 20:26:03 +0000

resolvconf: Error: /etc/resolv.conf must be a symlink
fix:
/etc$ sudo ln -s /etc/resolvconf/run/resolv.conf

Mirrorin a website + cookies

gnuton — Sat, 30 May 2009 14:50:14 +0000

If you want to mirror a phpbb or other CMS that ask for an authentication, you can get the coockis from the firefox dir.
In my case is:
cp /home/gnuton/.mozilla/firefox/xxxxxx.default/cookies.sqlite .
Then convert the coockies file from sqlite to plain paper with this script

gnuton@iron:/tmp/SS$ cat SS
#!/bin/bash
#

sqlite3 $1 < .mode tabs
.header off
select host as domain,
case substr(host,1,1)=’.’ when 0 then ‘FALSE’ else ‘TRUE’ end as flag,
path,
case isSecure when 0 then ‘FALSE’ else ‘TRUE’ end as secure,
expiry as expiration, name, value from moz_cookies;

EOF

The you have your plaintext cookies file and you can run now wget to do the job.
$ wget –load-cookies cookies.txt -m -np http://mywebsite.com

Mirroring a website n Linux

admin — Fri, 01 May 2009 11:17:09 +0000

Well, in linux the best tool to take a copy of a entire website is WGet.
Scenario: To get all the pages, images and the other contents that are in a hypothetic book/ directory of a hypothetic website (eg: example.com), we can run:
wget -m -np -p http://www.example.com/book/

where “m” stands for “mirror”, “np” is “no parent” (we want only the pages in the book/ directory) and the “p” flags tells to wget taking all the resources (like images) needed to display correctly the web pages mirrored in offline mode.