TopperH

Moving the blog

2009-10-25T03:35:00.000-07:00

This blog will soon move to http://topperh.blackmamba.kicks-ass.org/

It will not be a blogger's blog anymore and I will use jekyll to generate the pages and disqus for the comments.

It's in very "work in progress" status, anyway, any feedback is welcome.

Managing layman overlays my way

2009-06-10T07:15:00.000-07:00

Layman makes overlay management easy in gentoo, specially if you want to keep track of many different overlays, using different vcs.

On the other hand it may also create some problems, the thing that annoys me the most is when I want to install a single package from an overlay layman forces me to install the whole overlay.
Running ~arch, in the next @world update, it will probably want to update other packages that I don't need to be updated. Portage doesn't provide yet per-overlay mask (I heard paludis does, but haven't tried).

Having many packages in overlays for only a few installed will also considerably slow down some portage operations, may lead to duplicated eclasses, and will make gentoo housekeeping harder. There is also to consider that there are many poorly mantained and/ord bleeding edge and/or experimental overlays that will pull in your tree a considerable amount of crap.

This is my approach: I don't add the "source /usr/local/portage/layman/make.conf" line in my make.conf and I symlink from layman's directories only the directories (directories, not single ebuilds) that I need in a local overlay.

This way portage ignores the existence of the layman tree and sees only a clean and small local overlay. I am able to update layman daily, but only upgrades related to the packages I need will be seen by portage.

Here are the steps to take, assuming you have a working local overlay in /usr/local/portage, to install for example "app-backup/luckybackup" from the sunrise overlay.

emerge -av layman
echo "#source /usr/local/portage/layman/make.conf" >> /etc/make.conf
layman -a sunrise
mkdir /usr/local/portage/app-backup
ln -s /usr/local/portage/layman/sunrise/app-backup/luckybackup /usr/local/portage/app-backup/
emerge -av luckybackup

The reason why I added the line (even if commented) in make.conf is that I might need to uncomment someday. For example when a package has many dependencies on the same overlay. Temoprary removing the comment and doing "emerge -pv package" will give me the exact list of packages who need to be linked in my local overlay.

Get remote irssi notifications without X forwarding

2009-04-03T08:56:00.000-07:00

I was looking for a simple method to have irssi highlight notifications on my local machine while having irssi running on my remote server.

Googoling a bit I found that most methods require X forwarding (and libnotify installed on the server), or screen attached in a terminal in the local machine.

My server has no X, so I'm not going to install libnotify and its dependencies on it, and I don't want to have an irssi terminal open, unless I need it.

Here I found a nice solution:

Server side:

I assume sshd is working on server machine and there is key authentication, so no password is required)

wget http://www.leemhuis.info/files/fnotify/fnotify
cp fnotify ~/.irssi/scripts/fnotify.pl
cd ~/.irssi/scripts
ln -s fnotify.pl autostart/
touch ~/irssi/fnotify

then I reload irssi or I type "/RUN fnotify.pl" inside irssi (I do this step just the first time, then it will be done automatically at irssi startup).

From now, every higlighted message will be logged in this file.

On client side I cd my favourite bin directory (foe me is ~/scripts, but can also be /usr/local/bin) and create a file called irssi-notification.sh:

#!/bin/sh

ssh user@host tail -F ~/.irssi/fnotify | sed -u 's/[<@&]//g' |while read heading message
do notify-send -i gtk-dialog-info -t 300000 -- "${heading}" "${message}"
done

Change the red part with your username and host for the server machine and chmod +x the file.

Make sure x11-libs/libnotify is installed in your sistem (I think that some distro call this package libnotify-bin... don't ask me, debian and ubuntu like to have things complicated).

Now run the file and notifications will appear.

Gentoo releases, my point of view

2009-03-30T07:42:00.000-07:00

This entry follows up this nice article by Jeremy Olexa (darkside) and the relative comments.

I'm not a developer and I don't know much about the technical stuff that my idea involves, it's just a personal and different approach on the question that Jeremy asks.

Reading the article and the comments it looks like PR and advertising are the main issues. I couldn't agree more. When a distro comes out with a new version, popular sites (slashdot, distrowatch...) write an article, popular bloggers try the distro and write their opinion, other bloggers publish screenshots... A lot of buzz is generated, and people are aware that the new distro is out.

Gentoo is always up to date

Gentoo is a rolling release distro, it never needs upgrades, but just updates. The installation in my workstation (made in 2006) is as much up-to-date as the shiny new install on my laptop. That's gentoo magic, you sync, you emerge world, and have every day the latest and greatest.
Most people don't realise that, and this is why all the "gentoo is dead" thing grows.

Installation media

So, when is it that I have a new gentoo release? Maybe when a new installation media is out?
Well, I used just once the minimal gentoo cd to actually do an installation, then I realised that there are better ways to install gentoo. I think gentoo could invest less manpower on installation media releases. What we need is a very minimal cd, with basic tools for networking and partitioning (lvm and raid), that is updated not more than every 12-18 months, and, a very clear and complete chapter in the handbook, explaining how gentoo can be built using livecds like systemrescucd, knoppix, sabayon, or even ubuntu, and how is possible for people coming from other distros to install gentoo in a partition without leaving the environment they are are familiar with.

So, what makes a new release?

If I look other people workstations I can most of times tell at first sight whether they are using Ubuntu, Windows XP, Suse, OSX... The fact is that a lot of people don't care too much about theming their desktops, they just keep the vanilla install as it is.

Let's be honest: I'm sure Ubuntu developers did a lot of background work, but, what comes out from the press for the next release? "A new notification style, and a shining new color theme" Wow... those guys are great in PR stuff.

Gentoo doesn't have a consistent artwork theme, and if I publish a post of my desktop today I will look more or less the same as my desktop 2 years ago.
So, here comes my suggestion: a new gentoo release every time a new artwork theme is ready. I'm not kidding, let's see how it should work...

How it works

The gentoo artwork team provides consistent themes and wallpapers for the most popular DEs, login managers, toolkits, framebuffer and grub. (Sabayon guys are really good in it: from the moment you boot, till the moment you are into the graphical environment the transition looks really smooth). All those themes will be shipped in a package called media-gfx/gentoo-artwork and versioned like gentoo releases (2009.0, 2010.1 ecc.). Those packages will be slotted.

This package will have an USE flag for all the packages we have a theme for. For example "grub framebuffer xdm gdm kdm slim gnome xfce kdm openbox wallpaper", and according with the selected ones the relevant parts will be extracted.

The extracted themes will be named according to version (gtk-theme-gentoo-2009.0, gtk-theme-gentoo-2010.1) and with a symlink (gtk-theme-gentoo-default) that will be managed by an eselect module.

Assuming I have a default installation with no personal customizations, when a new version of gentoo-artwork comes out all I have to do is "eselect gentoo-artwork set n" and tah-dah, my whole gentoo changes shape and I'm ready to publish screenshots of my new gentoo in this blog.

Of course, if this new artwork comes along with a new major version of portage, or a new stabilized gcc, I will have something more to blog about :P

What else?

Gentoo is all about choices, so, if I want to keep the current behaviour all I have to to is add a "-gentoo-artork" USE in my make.conf.

My 2 cents...

Irssi init script (and I don't use quassel anymore)

2009-02-19T09:59:00.000-08:00

I've been a quassel user since the time it joined the gentoo tree.

I really like the idea wich it is built on, and I find it a great replacement for screen + irssi. For those who don't know the quassel core can be built on a server machine (that stays on irc 24/7) and the client part can be built in the machines you use to actually chat on irc.

The main quassel isssue is that, being based on qt, is not really portable. Yes, I know qt is said to work on a lot of platforms (windows, unix, mac...), but is not as easy as having putty.exe on a usb stick. The other problem is that having to install qt on an headless server, just for irc, is quite an overkill.

By the way, I wanted to use quassel anyway, so I installed qt on my headless mini-itx (it took me a while to figure out why emerge was pulling in tons of X related stuff, then I figured it was the qt3support USE flag).

On client side, the useless abilty to show web pages previews, pulled in packages like qt-webkit and qt-phonon. This has already been discussed in Diego's post, and came out to be a non issue, since it is due to bad downstream packaging, and will be solved in next release.

Anyway since a couple of days quasselcore is randomly crashing two, three times a day, and this is really pissing me off. I have no idea why this happens and I don't have the skills to investigate it, so I decided to step back to irssi and to put quassel in standby to see where future development is going.

To make irssi really behave like a daemon I need an init script to make it start. The gentoo wiki provides one, but it doesn't look really elegant since it uses su, it doesn't define the user in a separate file etc.

Here is mine:

/etc/init.d/irssi

#!/sbin/runscript

depend() {
        use net.*
}

start() {
        ebegin "Starting irssi"
        start-stop-daemon --start --chuid $USER \
        --env TERM="xterm" \
        --env HOME="/home/$USER" \
        --exec /usr/bin/screen -- -dmS irssi /usr/bin/irssi
        eend $?
}

stop() {
        ebegin "Stopping irssi"
        start-stop-daemon --stop --signal 2 --name irssi
        eend $?
}

/etc/conf.d/irssi

USER="andrea"

Of course the /etc/init.d/irssi needs to be made executable and added to the default runlevel.

Xorg and hal... Removing my xorg.conf

2009-02-05T04:44:00.000-08:00

As easy as the title says: emerged x11-base/xorg-server-1.5.3-r1 with the "hal" USE flag enabled, removed my /etc/X11/xorg.conf and everything works like a charm!

Direct rendering works, fonts are ok, screen resolution is perfect...

I still have to look into advanced synaptics touchpad settings, like mouse speed, scrolldown etc, by the way the touchpad works out of the box.

I'm really happy that I don't have to bother anymore setting up xorg.conf, so thank you X devlopers for making it possible, and thank you gentoo developers for making it available on my system and for making the transition so painless.

P.S. No, I don't use proprietary video drivers, as I have an Intel card, this is probably the reason all this was so easy for me.

DSDT on my ASUS F3E (mostly) fixed

2009-01-15T12:44:00.000-08:00

Since I have my asus F3E laptop brightness keys (fn + F5 and fn + F6) never worked. I had to workaround this issue manually editing the value in /sys/class/backlight/asus-laptop/brightness (provided by asus-laptop kernel module).

Of course I had a script for doing so, but I cuold not assign it to its own key because fn+F5 and fn+F6 were not producing any acpi event.

Investigating on the problem I had a look on my dsdt

cat /sys/firmware/acpi/tables/DSDT > dsdt.dat
iasl -d dsdt.dat
iasl -tc dsdt.dsl

Here I found some errors:

Intel ACPI Component Architecture
ASL Optimizing Compiler version 20080701 [Jan 14 2009]
Copyright (C) 2000 - 2008 Intel Corporation
Supports ACPI Specification Revision 3.0a

dsdt.dsl.old 4565: Method (OGCD, 0, NotSerialized)
Warning 1086 - ^ Not all control paths return a value (OGCD)

dsdt.dsl.old 4618: Method (OGCA, 0, NotSerialized)
Warning 1086 - ^ Not all control paths return a value (OGCA)

dsdt.dsl.old 6782: Acquire (MUTE, 0x03E8)
Warning 1103 - ^ Possible operator timeout is ignored

dsdt.dsl.old 6796: Acquire (MUTE, 0x03E8)
Warning 1103 - ^ Possible operator timeout is ignored

dsdt.dsl.old 6811: Acquire (MUTE, 0x03E8)
Warning 1103 - ^ Possible operator timeout is ignored

dsdt.dsl.old 6826: Acquire (MUTE, 0x0FFF)
Warning 1103 - ^ Possible operator timeout is ignored

dsdt.dsl.old 6840: Acquire (MUTE, 0x03E8)
Warning 1103 - ^ Possible operator timeout is ignored

dsdt.dsl.old 6855: Acquire (MUTE, 0x03E8)
Warning 1103 - ^ Possible operator timeout is ignored

dsdt.dsl.old 6870: Acquire (MUTE, 0x03E8)
Warning 1103 - ^ Possible operator timeout is ignored

dsdt.dsl.old 9254: And (CTRL, 0x1E)
Warning 1104 - ^ Result is not used, operator has no effect

dsdt.dsl.old 13901: Method (_DSM, 4, NotSerialized)
Warning 1086 - ^ Not all control paths return a value (_DSM)

dsdt.dsl.old 13901: Method (_DSM, 4, NotSerialized)
Warning 1079 - ^ Reserved method must return a value (_DSM)

dsdt.dsl.old 13907: Name (_T_0, Zero)
Error 4081 - Use of reserved word ^ (_T_0)

dsdt.dsl.old 13943: Name (_T_1, Zero)
Error 4081 - Use of reserved word ^ (_T_1)

ASL Input: dsdt.dsl.old - 14099 lines, 404107 bytes, 6700 keywords
Compilation complete. 2 Errors, 12 Warnings, 0 Remarks, 35 Optimizations

I havent investigated the warnings (maybe I'll do that another time) but I tried to solve the errors changing _T_0 and _T_1 with T_0 and T_1.
It looks like a name cannot start with underscore.

The recompiled dsdt works fine and now I have brightness keys working.

Current mpd song's lyric on console

2008-12-12T09:15:00.000-08:00

I always listen to music while working, and I use mpd. I prefer it over other players because it just stays out of the way.

As English is not my native language it sometimes happen that I want to read current song's lyrics. Of course an option would be to fire up ncmpc or sonata but, as I said before, I prefer things that stay out of my way.

REQUIREMENTS: mpd, mpc, python, id3lib

I put in my ~/scripts directory this python script from Max Kellermann (this script is also contained in the ncmpc package). It's called ~/scripts/leoslyrics.py

#!/usr/bin/python
#
# Load lyrics from leoslyrics.com
#
# Author: Max Kellermann
#

from sys import argv, exit
from urllib import urlencode, urlopen
from xml.sax import make_parser, SAXException
from xml.sax.handler import ContentHandler

class SearchContentHandler(ContentHandler):
def __init__(self):
self.code = None
self.hid = None

def startElement(self, name, attrs):
if name == 'response':
self.code = int(attrs['code'])
elif name == 'result':
if self.hid is None or attrs['exactMatch'] == 'true':
self.hid = attrs['hid']

def search(artist, title):
query = urlencode({'auth': 'ncmpc',
'artist': artist,
'songtitle': title})
url = "http://api.leoslyrics.com/api_search.php?" + query
f = urlopen(url)
handler = SearchContentHandler()
parser = make_parser()
parser.setContentHandler(handler)
parser.parse(f)
return handler.hid

class LyricsContentHandler(ContentHandler):
def __init__(self):
self.code = None
self.is_text = False
self.text = None

def startElement(self, name, attrs):
if name == 'text':
self.text = ''
self.is_text = True
else:
self.is_text = False

def characters(self, chars):
if self.is_text:
self.text += chars

def lyrics(hid):
query = urlencode({'auth': 'ncmpc',
'hid': hid})
url = "http://api.leoslyrics.com/api_lyrics.php?" + query
f = urlopen(url)
handler = LyricsContentHandler()
parser = make_parser()
parser.setContentHandler(handler)
parser.parse(f)
return handler.text

hid = search(argv[1], argv[2])
if hid is None:
exit(2)
print lyrics(hid).encode('utf-8')

Then I wrote this simple shell script, called ~/scripts/lyrics.sh

#! /bin/sh
### Change this with your mpd root
MUSICROOT="~/Music"
MPCINFO=`mpc | head -n 1`
MUSICPATH=$MUSICROOT/$MPCINFO
ARTIST=`id3info "$MUSICPATH" | grep TPE1 | cut -d: -f2`
TITLE=`id3info "$MUSICPATH" | grep TIT2 | cut -d: -f2`
echo $TITLE by $ARTIST
echo Loading lyrics...
./leoslyrics.py $ARTIST $TITLE | less

Then I gave the execution rights to the script and... abracadabra...

Now I can see current song's lyrics wherever I have an open terminal, or even inside a vim session.

Automatic TV series download --- Part 2 --- An alternative way

2008-12-02T05:50:00.000-08:00

I posted before about automatic TV series download.
Since my old home server passed away, and I'm building a new one from scratch, I was looking for a better way to do this.

This morning on #gentoo-it (freenode) earcar pointed me out that transmission has a built in web interface (no webserver required), can run on low resources, and has very few dependencies.

Unfortunatly it lacks the "folder watch" feature, that made me choose rtorrent for the previous howto. BTW, it's pretty easy to implement it with a script and a cron job.

INSTALLING TRANSMISSION

As today (december 2nd 2008) the portage version of transmission is a little bit out of date, it might work for the scope of this howto too, but I'm not sure.
I installed version 1.40 on my own local overlay.
If you are not familiar with local overlays read here, personally I don't use layman for this scope, just added PORTDIR_OVERLAY in my /etc/make.conf.

The ebuild can be found in gentoo's bugzilla.

CREATING THE INIT SCRIPT

The package doesn't come with an init script but fortunatly someone added it in gentoo's bugzilla.

/etc/init.d/transmission

#!/sbin/runscript
#

# DO NOT EDIT!
# ------------
# All configurable options are set in /etc/conf.d/transmission

NAME=transmission-daemon
declare -a OPTIONS
OPTIONS+=" -a ${TR_ACL:=+127.0.0.1}"
if [ -z "$TR_BLOCK" -o "$TR_BLOCK" = "no" ]; then
OPTIONS+=" -B"
else
OPTIONS+=" -b"
fi
OPTIONS+=" -g ${TR_HOME:-/var/transmission/config}"
OPTIONS+=" -p ${CTL_PORT:-9091}"
if [ -z "$REMOTE_USER" -o -z "$REMOTE_PASS" ]; then
OPTIONS+=" -T"
else
OPTIONS+=" -t"
OPTIONS+=" -u $REMOTE_USER"
OPTIONS+=" -v $REMOTE_PASS"
fi
OPTIONS+=" -w ${TR_DOWNLOAD:-/var/transmission/downloads}"
declare -a EXTRA_OPT
if [ -z "$TR_UP_SPEED" -o "$TR_UP_SPEED" = "unlimited" ]; then
EXTRA_OPT+=" -U"
else
EXTRA_OPT+=" -u $TR_UP_SPEED"
fi
if [ -z "$TR_DN_SPEED" -o "$TR_DN_SPEED" = "unlimited" ]; then
EXTRA_OPT+=" -D"
else
EXTRA_OPT+=" -d $TR_DN_SPEED"
fi
if [ -z "$TR_PX" -o "$TR_PX" = "no" ]; then
EXTRA_OPT+=" -X"
else
EXTRA_OPT+=" -x"
fi
if [ -z "$TR_PMAP" -o "$TR_PMAP" = "no" ]; then
EXTRA_OPT+=" -M"
else
EXTRA_OPT+=" -m"
fi
if [ -z "$TR_ENCRYPT" -o "$TR_ENCRYPT" = "tolerated" ]; then
EXTRA_OPT+=" -et"
elif [ "$TR_ENCRYPT" = "prefered" ]; then
EXTRA_OPT+=" -ep"
elif [ "$TR_ENCRYPT" = "required" ]; then
EXTRA_OPT+=" -er"
else
EXTRA_OPT+=" -et"
fi

depend() {
need net
}

start() {
ebegin "Starting transmission daemon"
start-stop-daemon --start --quiet \
--chuid ${TR_USERNAME:-nobody} \
--exec /usr/bin/transmission-daemon -- ${OPTIONS[@]}
sleep 2
transmission-remote ${CTL_PORT:-9091} \
${REMOTE_PASS:+-n $REMOTE_USER:$REMOTE_PASS} \
--port ${TR_PORT:-54318} ${EXTRA_OPT[@]} > /dev/null
eend $?
}

stop() {
ebegin "Stopping transmission daemon"
start-stop-daemon --stop --quiet --retry=TERM/30/KILL/5 \
--name $NAME
eend $?
}

# vim: set ft=gentoo-init-d ts=3 sw=3 et:

/etc/conf.d/transmission

# For information on options, see the transmission-daemon(1) man page.
#
# The commented variables in this file are the defaults that are used
# in the init-script. You don't need to uncomment them except to
# customize them to different values.

#
# ----- MAIN CONFIGURATION -----
#

# The folder where Transmission stores its config and .torrent files
TR_HOME=/home/andrea/transmission/config ### EDIT THIS

# The folder where Transmission stores downloads
TR_DOWNLOAD=/home/andrea/transmission/downloads ### EDIT THIS

# The port Transmission uses to connect to other peers
# TR_PORT=54318

# The name of the user that runs transmission-daemon and owns the files
TR_USERNAME=andrea ### EDIT THIS

# Login credentials for the Web- and the RPC-interface
# - default: leaveve commented/empty to disable access authentication
REMOTE_USER="andrea" ### EDIT THIS
REMOTE_PASS="very secure password" ### EDIT THIS
# - alternative: set a user name and password
#REMOTE_USER=admin
#REMOTE_PASS=password

#
# ----- ADVANCED OPTIONS -----
#

# Acces control lists
# (machines/networks allowed/denied to control transmission -- comma separated)
# TR_ACL="+127.0.0.1"

# Enable use of block lists
# (read the Wiki about what else needs to be done)
# TR_BLOCK=no

# Control port used
# only privileged users can use ports < ctl_port="9091" tr_up_speed="20" style="color: rgb(255, 0, 0);">### YOU MIGHT WANT TO EDIT THIS
TR_DN_SPEED=200 ### YOU MIGHT WANT TO EDIT THIS

# Peer exchange
# TR_PX=no

# Port mapping using NAT/UPnP
# TR_PMAP=no

# Encryption
# Any of :
# TR_ENCRYPT=tolerated
TR_ENCRYPT=prefered ### YOU MIGHT WANT TO EDIT THIS
# TR_ENCRYPT=required

Make sure to create the needed directories, and that the user you choose has write permissions on them.

TEST TRANSMISSION

Run the init script, "/etc/init.d/transmission start" and point a browser to http://my.host/:9091 (that will be localhost or 127.0.0.1 if you run the browser from the same machine).

You will be asked for username and password (the one you provided in the config file) and will be able to upload torrent files.

INSTALLING PYTVSHOWS

[EDIT] Thanks to Jeremy Olexa aka Darkside, net-misc/pytvshows is now in portage tree, so you can safely skip this part and just emerge it. [/EDIT]

There are many scripts to automate the download of tv series, btw I like pytvshows, because it just does what it's intented to do: read a config file, grab the new episodes from tvrss, and downloads the torrents in a given folder, that is, in our case, the folder our script will watch.

$ mkdir tmp
$ mkdir scripts
$ cd tmp
$ wget http://ovh.dl.sourceforge.net/sourceforge/pytvshows/pytvshows-0.2.tar.gz
$ tar xvzf pytvshows-0.2.tar.gz
$ cd pytvshows-0.2
$ python setup.py build
$ cp build/scripts-2.5/pytvshows ~/scripts/

Now we create and edit ~/.pytvshows.cfg with the exact names (like in tvrss.net) of our favourite series:

[Lost]

[Heroes]

[24]

[The+Sarah+Connor+Cronichles]

[Fringe]

Don't forget the "+" instead of spaces.
Now we launch manually the script (only for the first time),

$ mkdir ~/transmission/watch
$ ~/scripts/pytvshows -o ~/transmission/watch

and see what happened in our config file...
The script assumes that you have alredy seen all the episodes broadcasted until today and just wanna download the ones coming next. To change this behavior just change the number of the episode to the first one you want to download -1, for example if I want to download from the third episode of Fringe till today I change its config like this:

[Fringe]
episode = 2
season = 1
show_type = seasonepisode
human_name = Fringe
time = 2008-10-15 05:20:06

If I want all the episodes I change the "episode" value to 0.
Launching the script again will now fulfill our ~/transmission/watch/ folder with Fringe's latest episodes torrents.

CREATING THE SCRIPT

We want to tell transmission that we have now new torrents and we want to download them, here is a little script that I called ~/scripts/autodownoad.sh

#!/bin/bash

for file in ~/transmission/watch/*.torrent

do
if [ "$file" = "~/transmission/watch/*.torrent" ]; then
echo "No torrent in watchdir."
else
echo "$file" added to queue.
/usr/bin/transmission-remote -a "$file" -n user:password ### EDIT THIS
rm "$file"
fi
done

exit 0

Let's test it:

~/scripts/autodownload.sh

Now fire up again your browser in your transmission page... You should see the new episodes downloading.

AUTOMATING IT ALL

You will need a cron daemon that support per-user cron tabs, such as vixie cron. As regular user run "crontab -e" and add the following lines:

05,35 * * * * ~/scripts/pytvshows -o ~/transmission/watch/
08,58 * * * * ~/scripts/autodownload.sh

Every 30 minutes, pytvshows will search for new episodes of your favourite tv series, and right after, transmission will download them.

Surrender

2008-11-18T05:03:00.000-08:00

Yes, I surrendered to social networking...

I never had Myspace, Facebook, Flickr...

But twitter looks so cool, so: http://twitter.com/TopperH

Dear lazyweb...

2008-11-14T03:06:00.000-08:00

The laptop I used as my home server passed away yesterday and needs to be replaced.

I bought this on ebay and it is going to arrive me on tuesday/wednesday.

Of course I'm going to install gentoo.

I don't have a monitor to attach on it (it's gonna be an headless server), so I asked the seller to setup the bios for cd boot.

I will do all the installation via ssh but I know for sure that both the gentoo livecd and systemrescuecd need user interaction to have ssh up and running (you have to setup a root password).

Probably there will be no problem to it blind:

insert the cd
power the machine on
wait for a while
type "passwd" "******" "******"
type "/etc/init.d/sshd restart"

But I guess if there is a live distro that has a fixed root password and starts ssh and dhcp by itself.
I've been told on irc that xbox's distros can do that, but I don't think I will be able to chroot in an x86_64 enviroment from that...

Does anybody have a better idea?

A better way to use mutt + gMail's IMAP

2008-10-24T09:31:00.000-07:00

Imap is cool and gMail's imap is really cool. It allows users to have their mail clients syncronyzed and always up to date. The only issue with imap is speed.
On slow connections messages take some time to load, and attachments too. Evolution and kmail can use "offline imap" fetching locally all the messages and giving you a fast response.

Unfortunatly I don't feel confortable with neither of those clients, the only one I find really usable (I'm a mailing lists addicted) is mutt.

Mutt has imap support and can cache the headers of the messages, and the alredy read messages in local files, but that's not enough. Fetching messages still slow.

The idea is to reproduce the behavior of kmail or evoultion "offline imap" in mutt.
Offlineimap can do what I'm looking for:

# emerge -av offlineimap

Now we need to configure offlineimap for gmail, the file to edit is ~/.offlineimaprc:

[general]
accounts = ANameOfChoice

# Gives you a nice blinky output on the console so you know what's happening.
# ui = Curses.Blinkenlights

# If uncommented, this would show nothing at all. Great for cronjobs or background-processes
ui = Noninteractive.Quiet

[Account ANameOfChoice]
localrepository = mylocal
# Profile-Name for the remote Mails for a given Account
remoterepository = Gmail

[Repository mylocal]
# Way of storing Mails locally. Only Maildir is currently supported
type = Maildir

# Place where the synced Mails should be
localfolders = ~/Mail

# fetches your mails every 5 Minutes
autorefresh = 5

[Repository Gmail]

type = Gmail

# Specify the Gmail user name. This is the only mandatory parameter.

remoteuser = myusername@gmail.com
remotepass = mypassword

# Deleting a message from a Gmail folder via the IMAP interface will
# just remove that folder's label from the message: the message will
# continue to exist in the '[Gmail]/All Mail' folder. If `realdelete`
# is set to `True`, then deleted messages will really be deleted
# during `offlineimap` sync, by moving them to the '[Gmail]/Trash'
# folder. BEWARE: this will deleted a messages from *all folders* it
# belongs to!
#
# See http://mail.google.com/support/bin/answer.py?answer=77657&topic=12815
# See http://mail.google.com/support/bin/answer.py?answer=77657&topic=12815

realdelete = no

# The trash folder name may be different from [Gmail]/Trash
# for example on german googlemail, this setting should be
#
# trashfolder = [Google Mail]/Papierkorb
#
# The same is valid for the spam folder
#
# spamfolder = [Google Mail]/Spam

We don't want other users to know our password, so:

$ chmod 600 ~/.offlineimaprc
$ mkdir ~/Mail

For mail sending I use msmtp:

#emerge -av msmtmp

You can still use ssmtp (gentoo's default) or postfix or whatever, but you can't have two sendmail programs installed at the same time.

My ~/.msmtprc looks like this:

account default
host smtp.gmail.com
port 587
protocol smtp
auth on
from username@gmail.com
user username@gmail.com
password mypassword
tls on
tls_starttls on

Again I don't like people to read my password:

$ chmod 600 ~/.msmtprc

The first time offlineimap is launched can take a lot of time, specially if you keep ALL your mail since 1999 stored in your gmail account, so we will start it manually

$ offlineimap

Once done we can create a cron job to sync our folder every 10 minutes:

$ crontab -e

*/10 * * * * /usr/bin/offlineimap

You will need a cron daemon that supports users crontabs, such as vixie-cron.

And here is part of my ~/.mutt/muttrc

## ~/.mutt/muttrc
set mbox_type=Maildir
set folder=$HOME/Mail
set spoolfile=+inbox
set header_cache=~/.hcache
set sendmail="/usr/bin/msmtp"
set edit_headers=yes
set record=+SENT
set postponed=+DRAFTS
mailboxes +INBOX
set pager_context=1
set pager_index_lines=6 #show a mini-index in pager
set menu_scroll
set pgp_verify_sig=yes #dont show pgp in pager
set status_on_top #put status line at top
set sort=threads #sort by message threads in index
set from=username@gmail.com
set realname="My Name"
set reverse_name=yes
set reverse_realname=no
set certificate_file=~/.mutt/certificates
set move=no
set delete=yes
set include
set reply_to
set abort_nosubject=no
set hdr_format="%4C %Z %{%m/%d} %-15.15F (%4c) %s" # format of the index
hdr_order From: Date: User-Agent: X-Mailer To: Cc: Reply-To: Subject:
ignore *
unignore From: Date: User-Agent: X-Mailer To: Cc: Reply-To: Subject:
my_hdr Reply-To:

set editor="vim"
set alias_file=~/.mutt/mail_aliases
#source ~/.mutt/mail_aliases
source ~/.mutt/gpg.rc

###
send-hook . 'set signature=~/.signature'

That should make the trick :)

Automatic TV series download, using rtorrent and a nice web interface.

2008-10-21T03:41:00.000-07:00

This howto helps to set up a linux box that automatically downloads your favourite tv series episodes as soon as they come out.

EDIT: another version of this howto can be found here.

The box can be controlled remotely or locally via a nice web interface.

What do we need:

A working unix box with a webserver up and running (for this howto I assume gentoo with lighttpd)
Xmlrpc-c
Rtorrent
Wtorrent
Pytvshows
Dtach
A cron daemon that supports users crontab (for example vixie-cron)
A web browser (only on the client)

CONFIGURING THE WEB SERVER
Skip this part if you are a real geek and don't need a web ui for rtorrent

Php must be installed with the most common use flags (and some unusual like pdo, cgi, xml, sqlite and curl), here are mine:

Installed versions: 5.2.6-r7(5)(09:55:18 PM 10/20/2008)(berkdb bzip2 cgi cjk cli curl crypt ctype exif gdbm iconv imap mysql mysqli ncurses nls pcre pdo posix postgres readline reflection session spell spl sqlite ssl truetype unicode xml zlib -adabas -apache2 -bcmath -birdstep -calendar -cdb -concurrentmodphp -curlwrappers -db2 -dbase -dbmaker -debug -discard-path -doc -empress -empress-bcs -esoob -fastbuild -fdftk -filter -firebird -flatfile -force-cgi-redirect -frontbase -ftp -gd -gd-external -gmp -hash -inifile -interbase -iodbc -ipv6 -java-external -json -kerberos -kolab -ldap -ldap-sasl -libedit -mcve -mhash -msql -mssql -oci8 -oci8-instant-client -odbc -pcntl -pic -qdbm -recode -sapdb -sharedext -sharedmem -simplexml -snmp -soap -sockets -solid -suhosin -sybase -sybase-ct -sysvipc -threads -tidy -tokenizer -wddx -xmlreader -xmlrpc -xmlwriter -xpm -xsl -yaz -zip -zip-external)

And for lighty:

Installed versions: 1.4.20(11:14:49 PM 10/17/2008)(bzip2 fam fastcgi gdbm mysql pcre php ssl -doc -ipv6 -ldap -lua -memcache -minimal -rrdtool -test -webdav -xattr)

The minimal use flag MUST be turned off, because it triggers scgi, that we require.

Time to put the hands in /etc/lighttpd/lighttpd.conf ... In the modules section we add this line:

"mod_scgi",

then we put this at the end of the file:

scgi.server = (
"/RPC2" => # RT_DIR
( "127.0.0.1" =>
(
"host" => "127.0.0.1", # Ip where rtorrent is listening
"port" => 5000, # Port specified in .rtorrent.rc
"check-local" => "disable"
)
)
)

Here there are some hints if you want to set up auth for accesing the /RPC2 location and for configuring apache instead of lighttpd.

INSTALLING RTORRENT

Now it's time to install rtorrent with xmlrpc support, in gentoo it's quite straightforward:

#echo net-p2p/rtorrent xmlrpc >> /etc/portage/package.use
#emerge -av rtorrent

(xmlrpc support is only requred if you plan to use the web interface, so, real geeks can skip this useflag too)

I like to have all my stuff in a certain order so I created a dedicated rtorrent user with its own home folder, but you can do it as your own everyday user too

# useradd -m -G users,cron torrent
# su torrent
$ cd
$ mkdir Downloads
$ mkdir TorrentsToWatch
$ bzcat /usr/share/doc/rtorrent-0.8.2-r5/rtorrent.rc.bz2 > .rtorrent.rc

Now we edit the file we just created (/home/torrent/.rtorrent.rc) following the hints in the comments. Here are the mandatory parts:

# Default directory to save the downloaded torrents.
directory = /home/torrent/Downloads

# Watch a directory for new torrents, and stop those that have been
# deleted.
schedule = watch_directory,5,5,load_start=/home/torrent/TorrentsToWatch/*.torrent

# Scgi port
scgi_port = 127.0.0.1:5000

(scgi port is only needed for web interface users)

CREATING AN INIT SCRIPT

Now we log out and log back in as root to create a rtorrent init script (we also need dtach, so we can follow the rtorrent process in a console or via ssh in case something goes wrong with the web interface).

# emerge -av app-misc/dtach

[EDIT] There is a bug here with up-to-date rtorrent init scripts [EDIT]
And create the /etc/init.d/rtorrentd file with this content:

depend() {
use net.*
}

start() {
ebegin "Starting rtorrent"
start-stop-daemon --start --chuid $USER \
--env TERM="xterm" \
--env HOME="/home/$USER" \
--exec /usr/bin/dtach -- -n $SOCKET /usr/bin/rtorrent
eend $?
}

stop() {
ebegin "Stopping rtorrent"
start-stop-daemon --stop --signal 2 --name rtorrent
eend $?
}

And its configuration file, /etc/conf.d/rtorrentd

USER="torrent"
SOCKET="/tmp/rtorrent.dtach"

Now we launch the init script and we check if rtorrent works correctly:

#/etc/init.d/rtorrentd start
# dtach -a /tmp/rtorrent.dtach

You need to press "ctrl+\" to detach the attached screen.

INSTALLING PYTVSHOWS

[EDIT] Thanks to Jeremy Olexa aka Darkside, net-misc/pytvshows is now in portage tree, so you can safely skip this part and just emerge it. [/EDIT]

There are many scripts to automate the download of tv series, btw I like pytvshows, because it just does what it's intented to do: read a config file, grab the new episodes from tvrss, and downloads the torrents in a given folder, that is, in our case, the folder rtorrent is watching.

# su torrent
$ mkdir tmp
$ cd tmp
$ wget http://ovh.dl.sourceforge.net/sourceforge/pytvshows/pytvshows-0.2.tar.gz
$ tar xvzf pytvshows-0.2.tar.gz
$ cd pytvshows-0.2
$ python setup.py build
$ cp build/scripts-2.5/pytvshows ~/

Now we create and edit /home/torrent/.pytvshows.cfg with the exact names (like in tvrss.net) of our favourite series:

[Lost]

[Heroes]

[24]

[The+Sarah+Connor+Cronichles]

[Fringe]

Don't forget the "+" instead of spaces.
Now we launch manually the script (only for the first time),

$ ~/pytvshows -o ~/TorrentsToWatch/

[Fringe]
episode = 2
season = 1
show_type = seasonepisode
human_name = Fringe
time = 2008-10-15 05:20:06

If I want all the episodes I change the "episode" value to 0.
Launching the script again will now fulfill our ~/TorrentsToWatch/ folder with Fringe's latest episodes torrents.
If our rtorrent init script is still running rtorrent will be automagicaly alredy downloading Fringe :)
Check it with this command:

$ dtach -a /tmp/rtorrent.dtach

AUTOMATING THINGS

The scope of this howto is to download episodes as soon as they are out with no need of interaction... This is where a cron job comes useful:

#su torrent
$crontab -e

and we add this line to the cron jobs:

19,49 * * * * ~/pytvshows -o ~/TorrentsToWatch/

Now the script will check every 30 minutes for new episodes and add the torrents in the rtorrent watched folder.

INSTALLING A WEB UI

There are many rtorrent grapihcal clients listed in this page, the reasons why I've chosen wtorrent are the following:

I alredy have a web server installed
It's php, so it's not consuming resources while I'm not using it
It's simple and clean
It works

Get the latest wtorrent form the project page and extract it in your web root:

#cd /var/www/localhost/htdocs
#tar -zxvf /path/to/wtorrent-alpha*.tar.gz

Now we create the file wtorrent/conf/user.conf.php

And point a web browser to http://my.host/wtorrent/install.php to test the configuration, then create a user with a strong password and logout.

Now it's time to remove the install.php file and point our web browser to http://my.host/wtorrent/

Have fun :D

Keeping a hostname even when not on lan

2008-09-22T09:14:00.000-07:00

I use to connect to my home server from my laptop, from my LAN when I'm at home and from the internet when I'm not.

My server has a static ip address in my lan (192.168.1.5) and a dyndns name on the internet.

The server's hostname is "fandango" and the dyndns name is something like "fandango.foo.bar".

I had this line in my /etc/hosts:

192.168.1.5 fandango

This configuration was a pain in the ass because from home I had to "ssh TopperH@fandango", while from the outside I had to "ssh TopperH@fandango.foo.bar". I also had double passwords saved in my web browser, double quassel configuration ecc.

The idea is to always relate to my server as "fandango", wether at home or not, so I made two scripts and created a postup in my /etc/conf.d/net.

/root/scripts/hosts.world

#!/bin/bash
MYFILE='/etc/hosts.backup'
OLDHOST=`grep fandango $MYFILE | awk '{ print $1 }'`
NEWHOST=`host fandangofoo.bar | gawk '{print $4}'`
sed s/$OLDHOST/$NEWHOST/ $MYFILE > /etc/hosts

/root/scripts/hosts.home

#!/bin/bash
MYFILE='/etc/hosts.backup'
OLDHOST=`grep fandango $MYFILE | awk '{ print $1 }'`
NEWHOST=192.168.1.5
sed s/$OLDHOST/$NEWHOST/ $MYFILE > /etc/hosts

/etc/conf.d/net

[snip]
postup() {
if [[ ${IFACE} == "ppp1" ]] ; then
/root/scripts/hosts.home
elif [[ ${IFACE} == "ppp2" ]] ; then
/root/scripts/hosts.world
fi
return 0
}

I'm sure there are more elegant ways to achieve the same results, and comments are welcome... By the way it just works :)

Forwarding local mail to Gmail using postfix

2008-09-19T21:46:00.000-07:00

On my workstation I have postfix set up to delivery local mail to a maildir in my $HOME, so that I can read it using my mail client of choice.

I also have a server and I often forget to ssh in it and open mutt to read the emails that the system (mostly cron) sends me.

I know there are simple ways to be notified every time I open a console, for example this:

echo "MAILCHECK=30" >> ~/.bashrc
echo 'MAILPATH=~/.maildir/new?"You have a new mail. Read it with 'mutt'."' >> ~/.bashrc

But as long as the server works fine I don't need to login that often.

So, why not sending all the local mail to my gmail account, so that I can read it wherever I am, even on my BlackBerry? Here I found a nice howto.

First of all I need postfix set up:

# emerge -C ssmtp
# echo mail-mta/postfix mbox pam sasl ssl >> /etc/portage/package.use
# emerge postfix

Once is emerged I edit /etc/postfix/main.cf being careful to change XXX with something meaningful:

inet_interfaces = 127.0.0.1
relayhost = [smtp.gmail.com]:587
smtp_use_tls = yes
smtp_tls_CAfile = /etc/postfix/cacert.pem
smtp_tls_cert_file = /etc/postfix/XXX-cert.pem
smtp_tls_key_file = /etc/postfix/XXX-key.pem
smtp_tls_session_cache_database = btree:/var/run/smtp_tls_session_cache smtp_sasl_auth_enable = yes
smtp_sasl_password_maps = hash:/etc/postfix/saslpass
smtpd_sasl_local_domain = $myhostname
smtp_sasl_security_options = noanonymous

Then, according to this tutorial I create the tls certificate:

# /etc/ssl/misc/CA.pl -newca
# openssl req -new -nodes -subj '/CN=domain.com/O=Name/C=US/ST=State/L=Location/emailAddress=user@gmail.com' -keyout XXX-key.pem -out XXX-req.pem -days 3650

Domain, name, country, state, location and email address must be substituted and remembered, to be used in next step (once again XXX must be filled as above):

# openssl ca -out XXX-cert.pem -infiles XXX-req.pem
# cp demoCA/cacert.pem XXX-key.pem XXX-cert.pem /etc/postfix
# chmod 644 /etc/postfix/XXX-cert.pem /etc/postfix/cacert.pem
# chmod 400 /etc/postfix/XXX-key.pem

Now I edit the /etx/postfix/saslpass using my gmail username and password:

[smtp.gmail.com]:587 user@gmail.com:password

and I create the associated hash file:

# cd /etc/postfix
# postmap saslpass
# chmod 600 saslpass
# chmod 644 saslpass.db

Now, as regular user, specify the local forward:

$ echo 'user@gmail.com' > ~/.forward

I also set up local aliases in /etc/mail/aliases:

root: username
operator: username

Postfix needs a few commands before being started:

# postfix upgrade-configuration
# postfix check
# newaliases
# /etc/init.d/postfix start

Now all my local emails should be sent to my gmail account, let's see if thigs are working:

# emerge -av mail-client/mailx
$ andrea@fandango ~ $ mail root
Subject: postfix works?
Yes it does!!!
Cc:

This is the output of /var/log/messages

Sep 20 13:58:40 fandango postfix/pickup[23235]: 3F61AF066C: uid=1000 from=
Sep 20 13:58:40 fandango postfix/cleanup[23243]: 3F61AF066C: message-id=<20080920115840.3f61af066c@localhost>
Sep 20 13:58:40 fandango postfix/qmgr[23239]: 3F61AF066C: from=, size=339, nrcpt=1 (queue active)
Sep 20 13:58:40 fandango postfix/cleanup[23243]: 41AAEF066B: message-id=<20080920115840.3f61af066c@localhost>
Sep 20 13:58:40 fandango postfix/qmgr[23239]: 41AAEF066B: from=, size=471, nrcpt=1 (queue active)
Sep 20 13:58:40 fandango postfix/qmgr[23239]: 3F61AF066C: removed
Sep 20 13:58:40 fandango postfix/local[23245]: 3F61AF066C: to=, orig_to=, relay=local, delay=0.01, delays=0.01/0/0/0, dsn=2.0.0, status=sent (forwarded as 41AAEF066B)
Sep 20 13:58:43 fandango postfix/qmgr[23239]: 41AAEF066B: removed
Sep 20 13:58:43 fandango postfix/smtp[23246]: 41AAEF066B: to=, orig_to=, relay=smtp.gmail.com[72.14.221.109]:587, delay=3.3, delays=0/0/1.4/1.9, dsn=2.0.0, status=sent (250 2.0.0 OK 1221912634 12sm2163798fgg.0)

Forwarding VirtualBox guest apps on host's X server

2008-08-26T09:11:00.000-07:00

SCENARIO:

My host machine runs a stable amd64 Gentoo and Virtualbox. My guest operating system is again gentoo (this time a bleeding edge unstable ~x86 with evil overlays), but every X11 capable guest should work.

The guest machine network uses NAT, so there's nothing to configure, just dhcp and internet works.

GOAL:

Have guest apps to look native on my host DE, using systray, being able to copy/paste, while still running on a protected enviroment.

No need to run session managers (xdm) in the guest machine, neither to configure Xorg or to install VirtualBox Guest Additions.

HOWTO:

NAT doesn't allow host to ssh in guest, but there's a workaround:

$ VBoxManage setextradata "VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/HostPort" 2222
$ VBoxManage setextradata "VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/GuestPort" 22
$ VBoxManage setextradata "VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/Protocol" TCP

Read here to find how to undo those steps.

The guest now must be shut down (a simple reboot will not work).

Now the host machine can ssh into guest using this command:

$ ssh -p 2222 "user"@localhost

On the host machine we uncomment/edit those lines in /etc/ssh/ssh_config

ForwardAgent yes
ForwardX11 yes

On the guest machine we uncomment/edit this line in /etc/ssh/sshd_config

X11Forwarding yes

Remaining in the guest machine we reload the ssh revice:

# /etc/init.d/sshd reload

Now we run (from the host) a little test to see if everything works fine:

$ ssh -p 2222 "user"@localhost -YC xclock

If it complains about wrong autenthication remove from the guest ~/.Xauthority and try again.

Now we make things nicer:

$ echo "alias runguest='ssh -p 2222 "user"@localhost -YC '" >> .bashrc

Now we can open a terminal and:

$ runguest firefox

Cool, isn't it?

DEAR LAZYWEB:

One of the nicest features of VirtualBox is the ability to "hibernate" the guest "saving the machine state". Can anybody find a way to ctrl-z a forwarded app, suspend the Virtual Machine and then foreground it once resumed the VM?

Xorg-server with hal support and keybord layout

2008-07-31T01:48:00.000-07:00

Recent version of xorg-server (1.4.2) can be compiled with the "hal" use flag, since I have this flag globally enabled, I got my package compiled this way.

Flameeyes told me on irc that the hal support is useless in most desktop systems and may cause some problems, by the way I encountered only one major issue: the keyboard layout setting in /etc/X11/xorg.conf is ignored.

This happens because now it is up to hal to take care of it, so I had to create a file called /etc/hal/fdi/policy/10-x11-input.fdi

http://rafb.net/p/pkcBG846.html

(I had to uplad the content of the file to nopaste since blogger's wysiwyg editor seems to have some problems handling xml tags)

After reloading hal and restarting X, I had my Italian keyboard layout back.

/etc revision control with git

2008-07-17T09:37:00.000-07:00

After reading this article I tough that having all of my config files under revision control would be a nice idea.

Since geeks seem to love git, I googled a bit and found some interesting howtos:
Bart's Blog
Cyclopedia Square
OpenSuse

First thing to do is to install git, in gentoo this step comes pretty simple:

# emerge -av dev-util/git

Then tell git something about me (this step is optional):

$ git config --global user.name "Your Name Comes Here"
$ git config --global user.email you@yourdomain.example.com

Create the new repo (a .git folder)

# cd /etc
# git init .

Now, create a new (and useful) file inside the git controlled folder:

# vim /etc/cron.hourly/hourly_etc_git

#!/bin/bash
MYDATE=`date`
echo "etc_git - $MYDATE" | tee -a /var/log/etc_git.log
pushd /etc
git add . | tee -a /var/log/etc_git.log
git commit -m"$MYDATE" | tee -a /var/log/etc_git.log
popd

And register the changes:

# cd /etc
# git add .
#git commit -m"Initial /etc commit"

There are some files I don't want to be tracked by git:

# vim /etc/.gitignore

mtab
ld.so.cache
asound.state
resolv*
adjtime

Assuming cron is up and running the /etc folder will be checked every hour.

To see the last 3 changes:

# git whatchanged -3

To see the complete log with all the diffs:

# git-log -p

And finally the one I like best, it gives an overview of the changes made in each step:

# git-log --stat --summary

Encrypting home partition

2008-07-03T22:02:00.000-07:00

Following this howto I encrypted my /home using AES with a 256-bit key.

Edit: there was also this very good howto, but the link seems to be down.

First of all I needed to recompile my kernel to have appropriate support:

CONFIG_DM_CRYPT=y
CONFIG_CRYPTO_CBC=y
CONFIG_CRYPTO_SHA256=y
CONFIG_CRYPTO_AES=y
CONFIG_CRYPTO_AES_X86_64=y

I also needed a free partition for my home (in my case /dev/sda4), and emerged some stuff:

#emerge -av sys-fs/cryptsetup

Let's begin with encryption:

#cryptsetup luksFormat -c aes-cbc-essiv:sha256 /dev/sda4

I usually have my home folder and all sensitive data in an ext3 partition, so:

#mke2fs -j /dev/mapper/secure-home

Make a temporary mount point and mount:

#mkdir /mnt/secure-home
#mount /dev/mapper/secure-home /mnt/secure-home

Now I could have synced all my data from the old home with rsync in this way:

#rsync -tarv /home/* /mnt/secure-home

But I had a lot of cruft in my home dir, so I decided to selectively copy by hand what I need and start up with a brand new home, backing up the old one inside a backup dir in the encrypted partition:

#mkdir /mnt/secure-home/backup
#rsync -tarv /home/* /mnt/secure-home/backup

The /etc/fstab needs some mods:

/dev/mapper/secure-home /home ext3 noatime

And so does the /etc/conf.d/dmcrypt:

## /home with passphrase
target=secure-home
source='/dev/sda4'

Now it's time to remove the old home (mine was in the / partition so I only needed to rm it)

#rm -rf /home/*

At reboot I've been asked for my password and everything works fine, with no visible slowdowns.

A smarter way to hibernate

2008-07-02T11:02:00.000-07:00

While I like the way linux handles memory caches, it can become a pain in the ass when you need to hibernate (suspend to disk) your laptop.

The memory cache size grows the more you work and tends to fulfill all your free ram. While this is good because it speeds up your system while doing repetitive jobs it can consistently increase the hibernate time.

Along with that it writes and reads lots of data every time you hibernate/resume stressing your disks.

For example my new laptop has 3gb of ram and sometimes the hibernate process has to write on the swap partition more than one gb.

I wrote a little script that (hopefully) safely wipes the memory cache and hibernates my machine.

I called it /usr/local/sbin/freeze.

#! /bin/sh
sync
echo 1 > /proc/sys/vm/drop_caches
sync
hibernate

I'm using this instead of hibernate since a couple of days and haven't encountered any issues yet.

Ebuild for SMILE

2008-06-30T21:46:00.000-07:00

I needed a slideshow application to make a dvd of my daughter's pictures for my grandparents. I always knew, even if never tried that there is an app called "manslide" to do it.

After some google searches I found out that this app is broken and unmanteined, and the kde-apps page has been removed.

Smile (Slideshow Maker In linux Enviroment) seems to be a nice qt4 rewrite of manslide. I haven't been able to find an ebuild for it, neither in bugzilla or in some overlays, so I wrote one by myself. I've never been good in writing ebuilds, btw this one just works:

# Copyright 1999-2008 Gentoo Foundation
# Distributed under the terms of the GNU General Public License v2
# $Header: $

inherit eutils qt4
DESCRIPTION="Slideshow Maker In Linux Environement"
HOMEPAGE="http://www.kde-apps.org/content/show.php/SMILE?content=83276"
SRC_URI="http://www.mandrivalinux-online.eu/manslide/${P}.tar.gz"
SLOT="0"

LICENSE="GPL-2"
KEYWORDS="~amd64"
IUSE=""

DEPEND=">=media-sound/sox-14.0.1
media-video/mplayer
( $(qt4_min_version 4.2.0) )
"

RDEPEND="${DEPEND}"

src_compile() {
cd smile && eqmake4 smile.pro
emake || die
}

src_install() {
dodir /usr/bin
exeinto /usr/bin
doexe smile/smile

make_desktop_entry Smile Smile smile "Qt;AUdioVideo;Video"
}

Howto: AuthenTec AES1610 Fingerprint reader and gentoo

2008-06-20T09:22:00.000-07:00

My laptop comes with a fingerprint reader that can be used for authentication:

Bus 006 Device 007: ID 08ff:1600 AuthenTec, Inc.

There is a project called fprint that allows users to use it under linux. There is not yet an ebuilt for it in the portage tree, but I found it in an overlay.

The main issue I found is that libfprint does not compile with gcc-4.1.2 (which is currently stable for gentoo amd64) because of an "-fgnu89-inline" error.
This option is not yet implemented in 4.1.2 so I decided to switch to gcc-4.3.1.

I assume layman is installed and working, otherwise take a look here.

# layman -a wschlich-testing
# echo =app-misc/fprint_demo-0.4 >> /etc/portage/package.keywords/general
# echo =media-libs/libfprint-0.0.6 >> /etc/portage/package.keywords/general
# echo =sys-auth/pam_fprint-0.2 >> /etc/portage/package.keywords/general
# emerge -av fprint_demo pam_fprint

and now, as a regular user under X:

$ fprint_demo

In the enroll tab enroll the right index finger moving it smoothly . Than test it in the verify tab. It will take dozens of tries before getting a correct match, this is why we are going to use the usual password method if the match fails.

Now we have to edit the /etc/pam.d/system-auth file inserting this line:

auth sufficient pam_fprint.so

before this one:

auth sufficient pam_unix.so try_first_pass likeauth nullok

Howto synaptics touchpad and usb mouse work together in gentoo

2008-06-14T06:59:00.000-07:00

First of all event interface must be enabled in kernel:

Device Drivers --->
Input device support --->
<*> Event interface

Then xorg-server must be compiled with the "evdev" USE:

#echo 'INPUT_DEVICES="evdev keyboard mouse synaptics"' >> /etc/make.conf
#emerge -av1 xorg-server

Here is the relevant part of the /etc/X11/xorg.conf file:

Section "ServerLayout"
[...]
InputDevice "Mouse0" "AlwaysCore"
InputDevice "TouchPad" "SendCoreEvents"
EndSection

Section "InputDevice"
Identifier "Mouse0"
Driver "mouse"
Option "Protocol" "auto"
Option "ZAxisMapping" "4 5 6 7"
EndSection

Section "InputDevice"
Driver "synaptics"
Identifier "TouchPad"
Option "SendCoreEvents"
Option "Protocol" "auto-dev"
Option "SHMConfig" "on"
EndSection

What made the trick was to put the USB mouse in "AlwaysCore" and the touchpad in "SendCoreEvents"

Chicony Electronics UVC embedded webcam on Asus PRO60E + gentoo

2008-05-29T04:14:00.000-07:00

Setting up the embedded 1.3 megapixel webcam on my notebook was pretty easy:

#lsusb
Bus 002 Device 002: ID 04f2:b029 Chicony Electronics Co., Ltd

I found out this to be an UVC (USB Video Class device) and at the time I'm writing there's not a kernel module for it.
BTW there is a project to handle support for such devices: linux-ucv
The gentoo tree provides an ebuild (that is currently ~amd64 keyworded):

[D] media-video/linux-uvc
Available versions: ~*0.1.0_pre100 ~*0.1.0_pre126 ~*0.1.0_pre157 ~*0.1.0_pre173 {isight kernel_linux}
Installed versions: 0.1.0_pre173(18:42:13 05/28/08)(kernel_linux)
Homepage: http://linux-uvc.berlios.de/
Description: Linux driver and user-space tools for USB Video Class devices.

I just needed to emerge it to have the uvcvideo module compiled and ready to be modprobed.
Of course v4l support must be compiled in kernel (or as a module):

revolver ~ # lsmod | grep uvcv
uvcvideo 44932 0
compat_ioctl32 8064 1 uvcvideo
videodev 31424 2 uvcvideo,compat_ioctl32
v4l1_compat 11524 2 uvcvideo,videodev

I tested the cam with kopete and works fine!

Asus PRO60Eseries + Gentoo = LOVE

2008-05-23T05:55:00.000-07:00

OVERVIEW:

CPU : Duo T7250
DISPLAY: 13.3" WXGA
WIRELESS: 802.11abgn + Bluetooth
HD: 250GB
RAM: 3GB

revolver ~ # lspci
00:00.0 Host bridge: Intel Corporation Mobile Memory Controller Hub (rev 03)
00:02.0 VGA compatible controller: Intel Corporation Mobile Integrated Graphics Controller (rev 03)
00:02.1 Display controller: Intel Corporation Mobile Integrated Graphics Controller (rev 03)
00:1a.0 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Contoller #4 (rev 03)
00:1a.1 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #5 (rev 03)
00:1a.7 USB Controller: Intel Corporation 82801H (ICH8 Family) USB2 EHCI Controller #2 (rev 03)
00:1b.0 Audio device: Intel Corporation 82801H (ICH8 Family) HD Audio Controller (rev 03)
00:1c.0 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 1 (rev 03)
00:1c.1 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 2 (rev 03)
00:1c.2 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 3 (rev 03)
00:1c.3 PCI bridge: Intel Corporation 82801H (ICH8 Family) PCI Express Port 4 (rev 03)
00:1d.0 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #1 (rev 03)
00:1d.1 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #2 (rev 03)
00:1d.2 USB Controller: Intel Corporation 82801H (ICH8 Family) USB UHCI Controller #3 (rev 03)
00:1d.7 USB Controller: Intel Corporation 82801H (ICH8 Family) USB2 EHCI Controller #1 (rev 03)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev f3)
00:1f.0 ISA bridge: Intel Corporation Mobile LPC Interface Controller (rev 03)
00:1f.1 IDE interface: Intel Corporation Mobile IDE Controller (rev 03)
00:1f.2 SATA controller: Intel Corporation Mobile SATA AHCI Controller (rev 03)
01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 01)
02:00.0 Network controller: Intel Corporation Unknown device 4229 (rev 61)
05:00.0 SATA controller: JMicron Technologies, Inc. JMicron 20360/20363 AHCI Controller (rev 02)

What does not work:

The fingerprint reader (Bus 006 Device 002: ID 08ff:1600 AuthenTec, Inc.) [EDIT now it works]
The LCD backlight controls (CONFIG_ASUS_LAPTOP=Y doesn't help and acpi4asus doesn't compile)

What needs workarounds:

ALSA

00:1b.0 Audio device: Intel Corporation 82801H (ICH8 Family) HD Audio Controller (rev 03)
I needed to add "options snd-hda-intel model=lenovo" to /etc/modules.d/alsa and update modules.

What I have not tried yet:

Integrated 1.3 megapixels webcam (I don't really need a webcam)
[edit] now it works, see here

revolver# uname -snrv
Linux revolver 2.6.25-tuxonice-r4 #4 SMP PREEMPT Thu May 22 16:04:52

revolver ~ # eselect profile show
Current make.profile symlink
/usr/portage/profiles/default/linux/amd64/2008.0/desktop

Kernel config
make.conf

Gentoo init script for GPRS dialout connections

2008-05-22T22:08:00.000-07:00

So I decided to get rid of wvdial and to use the powerful gentoo init scripts to estabilish a dialup connection with my Nokia cell phone via bluetooth.
Of course the bluetooth connection is already up 'n running and works fine on /dev/rfcomm0 (I might blog about obtaining it another time).
Here's the /etc/conf.d/net relevant part:

config_ppp2=( "ppp" )
link_ppp2="/dev/rfcomm0"
username_ppp2='wind'
password_ppp2='wind'
pppd_ppp2=(
"noauth"
"defaultroute"
"usepeerdns"
)

phone_number_ppp2=( "*99***1#" )

chat_ppp2=('ABORT' 'BUSY'
'ABORT' 'ERROR'
'ABORT' 'NO ANSWER'
'ABORT' 'NO CARRIER'
'ABORT' 'NO DIALTONE'
'ABORT' 'Invalid Login'
'ABORT' 'Login incorrect'
'TIMEOUT' '5'
'' 'ATZ'
'OK' 'AT+CGDCONT=1,"ip","internet.wind","",0,0'
'OK' 'ATDT\T'
'TIMEOUT' '60'
'CONNECT' ''
'TIMEOUT' '5'
'~--' ''
)

This works for the Italian operator wind, btw, changhing the IP, phone number, username and password should work for all.

Let's make the symlink:

#cd /etc/init.d
#ln -s net.lo net.ppp2

And see if it works:

#/etc/init.d/net.ppp2 start