Daily Newspaper on E-Book Reader via Arch Linux

My friend x23b5 has an Amazon Kindle and wants to read daily newspapers each morning.

Time for a cron job on my Pogoplug to send some newspapers to this Kindle 🙂

There is a pretty cool package called calibre.
It can convert online newspapers to ebook specific files and send it via email to your eBook Reader.

Install calibre

pacman -Sy calibre

Calibre uses recipes to fetch news and convert those to ebook files like .mobi or .epub.
You can list all available recipes with the following command:

ebook-convert --list-recipes

Note: It’s a good idea to create a new email for this task because your password will be saved in plain text. Use a freemailer or create a new user on your own mailserver.
Create the following file in /usr/local/bin/calibre-cron and replace all variables and recipes with yours.
I used another usb stick (cacheStick) to save the temp files because I want to protect my arch linux stick.
To change the output format just change the file extension.

nano /usr/local/bin/calibre-cron
#!/bin/bash

TEMPPATH=/media/cacheStick/calibre/
SERVER_ADDRESS=smtp.mailserver.com
USERNAME=myusername
PASSWORD=mypassword
FROM_EMAIL=ebook@mailserver.com
TO_EMAIL=testuser@kindle.com

# Netzpolitik
ebook-convert "Netzpolitik.recipe" ${TEMPPATH}netzpolitik.mobi
calibre-smtp -a ${TEMPPATH}netzpolitik.mobi -s "Netzpolitik" -r $SERVER_ADDRESS -u $USERNAME -p $PASSWORD $FROM_EMAIL $TO_EMAIL "Netzpolitik"

# Heise
ebook-convert "Heise-online.recipe" ${TEMPPATH}heise.mobi
calibre-smtp -a ${TEMPPATH}heise.mobi -s "Heise" -r $SERVER_ADDRESS -u $USERNAME -p $PASSWORD $FROM_EMAIL $TO_EMAIL "Heise"

# Berliner Zeitung
ebook-convert "Berliner Zeitung.recipe" ${TEMPPATH}bz.mobi
calibre-smtp -a ${TEMPPATH}bz.mobi -s "Berliner Zeitung" -r $SERVER_ADDRESS -u $USERNAME -p $PASSWORD $FROM_EMAIL $TO_EMAIL "Berliner Zeitung"

Make it executable:

chmod u+x /usr/local/bin/calibre-cron

Execute it and check whether it works.

Now we have to create a cron job to execute this script each morning.
Execute this command before you edit your crontab file.

export EDITOR="/usr/bin/nano"

Edit your crontab file:

crontab -e
# daily at 5:00
0 5 * * *       /usr/local/bin/calibre-cron > /dev/null 2>&1

> /dev/null 2>&1” because I don’t need any email or log output.

Now enable cron service:

systemctl enable cronie
systemctl start cronie

Calibre is pretty memory and cpu intensive and takes pretty long! You should run it overnight otherwise your plug pc will be very slow.

How it looks like:

Kindle newspaper

Setup Samba 4 on Arch Linux

Updated: [03.05.2015] Sync some performance settings with my own personal settings, added a link to some more performance tips (see bottom).
Updated: [28.09.2015] Sync again

Note: Check this article if you want to install arch on your pogoplug.

Samba 4 is out now 🙂
So I’ll make this tutorial for Samba 4 because it seems to be slightly faster.
This article will tell you how to install it on your PogoplugV2 or another PlugPC.
Should be pretty much the same for all arch installations.

Samba 3 is going to be removed once Samba 4 is installed.
Your old config will be saved to /etc/samba/smb.conf.pacorig

Installation of Samba 4 is pretty easy.

pacman -Sy samba

Now enable the services.

systemctl enable smbd nmbd

Create Users

If you want to create shares for multiple users you have to create new Unix user and add this one to samba as well.
To make it clean we will create a group called “samba”.

groupadd samba

Now we can add a new user to this group. This user “fabian” is not able to login (-s /sbin/nologin) for security purposes.

useradd -m -g samba -s /sbin/nologin fabian

To use this user in samba shares you have to add it to samba

pdbedit -a -u fabian

Create Shares

We are ready to configure our samba shares.
At the beginning configure
To do so edit /etc/samba/smb.conf

nano /etc/samba/smb.conf

Here is an example configuration.
You have to edit the Share definitions below so it fits your setup.

[global]
   workgroup = WORKGROUP
   server string = POGOPLUG
   netbios name = POGOPLUG
   # hosts allow = 192.168.0.
   printcap name = /dev/null
   load printers = no
   disable spoolss = yes
   printing = bsd
   show add printer wizard = no
   print notify backchannel = no
   log file = /var/log/samba/log.%m
   max log size = 50
   security = user
   dns proxy = no
   # For public share without login
   map to guest = Bad User

   # Android bugix for reading files (samba4 bug see: https://bugzilla.samba.org/show_bug.cgi?id=9706)
   unix extensions = false

   # Fix for file batch copy issues (see: http://archlinuxarm.org/forum/viewtopic.php?f=18&t=4864) - seems to be fixed now (28.09.2015)
   # oplocks = no
   # level2 oplocks = no

   # Some Tuning (See Optimize Performance)
   socket options = TCP_NODELAY IPTOS_LOWDELAY
   write cache size = 262144
   # sendfile will interrupt data transfer :/ (but cpu usage is less) - seems to be fixed now (03.05.2015)
   use sendfile = true
   getwd cache = yes
   min receivefile size = 16384
   max xmit = 65536
   # Global security
   public = yes

#============================ Share Definitions ==============================

# Public, read only
[Videos]
        comment = Videos for all
        read only = yes
        # use this only for read only shares!
        fake oplocks = yes
        path = /media/zincobi/Videos

# Public, writeable
[Abrechnungen]
        comment = Abrechnungen
        read only = no
        writeable = yes
        path = /media/zincobi/Abrechnungen

# whole HDD, only for fabian
[zincobi]
        comment = Fabians share
        public = no
        valid users = fabian
        read only = no
        writeable = yes
        path = /media/zincobi

Optimize performance

The stock performance of samba isn’t that great. Especially with NTFS.
But there are some parameters which will increase Samba performance significantly.

Add all these settings to the global section in your smb.conf file.

socket options = TCP_NODELAY IPTOS_LOWDELAY

The main problem for slow file transfer speeds is NTFS, because NTFS needs much CPU on linux.
Nevertheless there are 2 options which will boost the speed:

write cache size

If this integer parameter is set to non-zero value, Samba will create an in-memory cache for each oplocked file (it does not do this for non-oplocked files). All writes that the client does not request to be flushed directly to disk will be stored in this cache if possible. The cache is flushed onto disk when a write comes in whose offset would not fit into the cache or when the file is closed by the client. Reads for the file are also served from this cache if the data is stored within it.

This cache allows Samba to batch client writes into a more efficient write size for RAID disks (i.e. writes may be tuned to be the RAID stripe size) and can improve performance on systems where the disk subsystem is a bottleneck but there is free memory for userspace programs.

The integer parameter specifies the size of this cache (per oplocked file) in bytes.

Default: write cache size = 0

Example: write cache size = 262144 # for a 256k cache size per file

Some example values are:

write cache size = 262144

(262144 = 256KB – you should test some values it’s pretty memory intensive)

Don’t forget to start the samba services or reboot:

systemctl start smbd nmbd

For some additonal performance tips check this blogpost: https://linuxengineering.wordpress.com/2014/08/03/performance-tuning-with-pogoplug-v4/

Installing Arch Linux and setting up everything on Plug Computers like PogoplugV2

Some weeks ago I’ve ordered a so called “Plug Computer”.PogoplugV2
These Computers are very small devices with an ARM SoC.
They are pretty cheap! I’ve got my PogoplugV2 for 10€ 🙂

I was lucky and got the gray version 😀
Some device info:

  • ARMv5te CPU (Marvell Kirkwood 1.2GHz)
  • 256MB RAM
  • 128MB NAND
  • 4 USB 2.0 Ports

Pretty great to use this device as a small homeserver/fileserver. Remember 10€!
I want to use my Pogoplug as a fileserver (for Windows PCs) and DLNA server to stream to my TV.
On top of that maybe some additional stuff in the feature 😉

But the default software sucks. No Samba support, no DLNA, no FTP, …
But this doesn’t matter because there is an Arch Linux for ARM Port.
This is an instruction how to install Arch Linux on a PogoplugV2.
Everything after the Arch Linux installation (Install Webmin, Samba, DLNA, …) can be used for all devices which run Arch Linux for ARM.

Prepare device and install Arch Linux

Arch Linux

Everything we have to do to install Arch Linux are 4 steps.

  1. Enabe SSH Access
  2. Format an USB Stick (min. 2GB size)
  3. Flash a custom Bootloader to boot from your USB Stick
  4. Install Arch Linux

An instruction how to do that is available here: http://archlinuxarm.org/platforms/armv5/pogoplug-v2-pinkgray
There are instuctions for other devices as well.
If you use Windows use PuTTY as a SSH client.

Setup everything

1. Login via SSH

Use PuTTY or another SSH client to access your Pogoplug.

username: root
password: root

 2. Change your root password

First you should change your root password.
Use the following command to do so:

passwd root

 3. Update Arch Linux

Arch Linux has an own package manager called pacman.
With pacman you can install/remove packages and update Arch Linux.

To update Arch Linux use the following command:

pacman -Syu

 4. Change hostname and timezone

Note: For GUI lovers: skip this step and set hostname and timezone in the webmin interface (next step)

The default hostname is “alarm”. Let’s change it!

hostnamectl set-hostname myhostname

To get all available timezones use:

timedatectl list-timezones

And to set your timezone:

timedatectl set-timezone <Zone>/<SubZone>

For example:

timedatectl set-timezone Europe/Berlin

Now reboot:

reboot

Wait a few seconds and reconnect to SSH.

 5. Install Webmin

Webmin is a pretty cool web-based interface for system administration.
It’s easy to use  and will help you to configure stuff faster.
To install it use:

pacman -Sy webmin perl-net-ssleay

Now we have to allow access from more IP addresses.
You have to edit the configuration file.

nano /etc/webmin/miniserv.conf

Find the following line: allow=127.0.0.1
Now add a new allow line with your local network broadcast ip or a specific ip and save the file.
For example:

allow=127.0.0.1 ## Allow local access
allow=192.168.0.0 ## Allows all users from 192.168.0.1 - 192.168.0.255

Now we’ll enable & start the webmin service so it autostarts.

systemctl enable webmin
systemctl start webmin

To access the Webmin interface open a browser and go to:
https://deviceip:10000

To lower memory usage go to:
Webmin>Webmin Configuration>Advanced Configuration and disable Pre-load Webmin functions library?”

 6. USB Auto Mount

If you want to remove your USB HDD and use it anywhere else it would be cool to have automounting like Windows.
There are a lot of auto mounting mechanisms for Arch Linux but a lot of them are outdated.
This one is using udevil to auto mount all USB HDDs on attach as /media/PARTITION_LABEL. So make sure all partitions have a label!

pacman -Sy udevil

I want to access my HDDs via Samba to use them in Windows so I have to use either FAT32 or NTFS.
This sucks because FAT32 isn’t able to handle files >4GB so it’s useless.
And NTFS is fucking slow on Linux.

But with some special mount options we are able to increase the speed dramatically!
I’ve did some benchmarks with hdparm and dd before and after the optimization.
The read speed was pretty good already (about 29MB/s –> USB 2.0 limit).
But the write speed was really bad!

Before: ~6,5 MB/s write speed
After: ~28,5 MB/s write speed

As you can see the write speed on NTFS was really slow before. But there is a way to fix it 🙂
We have to edit the mount options and add a special ntfs-3g option to our udevil automount settings.
Open the udevil config file and edit it:

nano /etc/udevil/udevil.conf

Search for default_options_ntfs= and allowed_options=

Now we add the option “big_writes” to both lines so it looks like:

..., noatime, big_writes, uid=...

Save the file.

Create the /media directory

mkdir /media

Add the udevil service to autostart.

systemctl enable devmon@root

Let’s reboot to see whether it works

reboot

Wait a few seconds and reconnect to SSH.
Now you’ll have a new folder: /media/yourHDDname where your HDD is mounted.
And you should see something like /dev/sdX1 on /media/… if you type:

mount

Note: All USB HDDs should go to standby automatically.
You can check with:

hdparm -C /dev/sda

7. Blink LED to HDD activity

The PogoplugV2 has a green and orange LED.
So I thought it would be a cool idea change the color to HDD activity.
I’ve created an systemd service.

———————————

Only If you have a PogoplugV2!

You have to correct your arcNumber and matchid because there is a bug in the current uboot and your Pogoplug isn’t detected as a PogoplugV2.

pacman -Sy uboot-env
fw_setenv arcNumber 3542
fw_setenv machid dd6
reboot

———————————

We need the iostat binary to create our own deamon.
Iostat will check hdd activity.

pacman -Sy sysstat

Create a new file called /usr/lib/systemd/system/blinkled.service with the following content:

[Unit]
Description=Blink LED if hard drive is active daemon

[Service]
Type=oneshot
ExecStart=/usr/local/bin/blinkled
ExecStop=/usr/bin/killall iostat
RemainAfterExit=true

[Install]
WantedBy=multi-user.target

Create another file called /usr/local/bin/blinkled
This command is pretty long 😉 It will execute iostat every 3 seconds. If there is HDD activity the Pogoplug will start to blink orange.
You can change heartbeat to default-on or timer as well.

#!/bin/bash
/usr/bin/iostat -z -d 3 | /usr/bin/sed -u -e '1d' -e 's/Device.*/none/' -e 's/sd.*/heartbeat/' -e '/^$/d' > /sys/class/leds/status:orange:fault/trigger &

And make it executable

chmod u+x /usr/local/bin/blinkled

To start our service use:

systemctl start blinkled

To autostart use:

systemctl enable blinkled

 8. Update Kernel to >3.2

Warning: This specific instruction works for PogoplugV2 only!
You can brick your device!

The current kernel for PogoplugV2 is 3.1.x but 3.8.x or newer is available.
You have to update manually because newer kernels need a new uboot.

You need the newest uboot to boot Kernels >3.2 so you have to install it before:
If this is a fresh installation you probably have the newest uboot already and you don’t have to do that!

cd /tmp
wget http://jeff.doozan.com/debian/uboot/install_uboot_mtd0.sh
chmod +x install_uboot_mtd0.sh
./install_uboot_mtd0.sh

Then you have to correct your arcNumber and matchid if you didn’t follow “Blink LED to HDD activity”.

pacman -Sy uboot-env
fw_setenv arcNumber 3542
fw_setenv machid dd6
reboot

Now you can install the new Kernel files with:

pacman -Sy linux-kirkwood linux-headers-kirkwood
reboot

 9. Install Samba

See this post

10. Install MiniDLNA

Install MiniDLNA and enable the service.

pacman -Sy minidlna
systemctl enable minidlna

Now edit the config file

nano /etc/minidlna.conf

It’s pretty easy to understand the config file and the documentation is well.
You have to set at least one media_dir
All USB devices are mounted at /media/DEVICE_LABEL
For example:

media_dir=V,/media/zincobi/Videos

Here are some additional options you could set:

friendly_name=POGOPLUG DLNA Server
db_dir=/var/cache/minidlna
log_dir=/var/log
presentation_url=http://YOUR_DEVICE_IP_HERE:8200/

MiniDLNA will update the DB automatically.
If you have a lot of files (>8192) you should increase your max inotify value for a single user so MiniDLNA can watch all files.
Create /etc/sysctl.d/90-inotify.conf and insert the following:

# Increase inotify max watchs per user for local minidlna
fs.inotify.max_user_watches = 100000

MiniDLNA is able to use thumbnails or cover images.
I’ve created a small bash script which will generate thumbnails for videos recursively.
You need ffmpegthumbnailer to create these thumbnails.

pacman -Sy ffmpegthumbnailer

Let’s create the script:

nano /usr/local/bin/generateThumbs.sh

This script will generate a thumbnails for every video file recursively.
You can add even more extensions to the TYPES array.

#!/bin/bash

DIR=$1

if [ -z "$DIR" ]; then
	printf "%sn" "This script will generate thumbnails for all video files inside the given directory"
	printf "%sn" "moviexy.avi --> moviexy.jpg"
	printf "%sn" "ffmpegthumbnailer is required"
	printf "%sn" "Usage: generateThumbs "
	exit 1
fi

TYPES=( mov mp4 avi mkv m4v xvid divx wmv mpg mpeg )

# Create a regex of the extensions for the find command
TYPES_RE="\("${TYPES[1]}
for t in "${TYPES[@]:1:${#TYPES[*]}}"; do
    TYPES_RE="${TYPES_RE}\|${t}"
done
TYPES_RE="${TYPES_RE}\)"

find "$DIR" -regex ".*.${TYPES_RE}" -type f | while read -r FILEPATH
do
	printf "%sn" "--- Start Thumbnailcreation for ---"
	printf "%sn" "$FILEPATH"
	THUMBFILE="${FILEPATH%.*}.jpg" # remove video ext. and add .jpg
	if [ -f "$THUMBFILE" ]; then
		printf "e[1;33m%se[0mn" "Thumbnail exists"
	else
		printf "e[1;32m%se[0mn" "Generating thumbnail..."
		ffmpegthumbnailer -i "$FILEPATH" -o "$THUMBFILE" -s 160 -q 10
	fi
	printf "%sn" "--- End Thumbnailcreation ---"
done
exit 0

See: https://github.com/Obihoernchen/generateThumbs

Make it executable:

chmod u+x /usr/local/bin/generateThumbs.sh

Execute it with something like:

generateThumbs.sh /media/YOURDEVICELABEL/Videos

Now start MiniDLNA

systemctl start minidlna

12. Configure static IP

Be careful you can mess up your whole network connection!

I don’t like dynamic IP addresses in my LAN. Especially for servers.
They should have a static IP like every server. Of course there are hostnames but not all devices are able to resolve them.

Arch uses netcfg to configure your network adapters. You can create multiple profile.
To create a new static ip profile create a new profile in /etc/network.d/ named wired-static-eth0

nano /etc/network.d/wired-static-eth0

And add the following content, replace everything with your values and save the file.
Note: I use custom DNS settings because my router doesn’t allow me to edit DNS settings.

CONNECTION='ethernet'
DESCRIPTION='A basic static ethernet connection using iproute'
INTERFACE='eth0'
IP='static'

#IP of your pogoplug
ADDR='192.168.0.2'

#ROUTES=('192.168.0.0/24 via 192.168.1.2')

#Gateway (For example your router)
GATEWAY='192.168.0.1'

#DNS set to ('Gateway IP') if you want to use the DNS your router provides
DNS=('8.8.8.8' '8.8.4.4')

## For IPv6 autoconfiguration
#IP6=stateless
## For IPv6 static address configuration
#IP6='static'
#ADDR6=('1234:5678:9abc:def::1/64' '1234:3456::123/96')
#ROUTES6=('abcd::1234')
#GATEWAY6='1234:0:123::abcd'

Now you need to change the default network profile to the new one.

nano /etc/conf.d/netcfg

Find NETWORKS= and edit it:

NETWORKS=(wired-static-eth0)

You have to be sure that everything is correct otherwise you are going to loose network connection and have to edit all files from your USB stick with another PC.
Just restart the netcfg service or restart to apply the new settings

systemctl restart netcfg

 13. Enable color in Bash prompt

The current terminal doesn’t look that nice. Everything is white and boring…
But we can change this so your terminal will look like this:

Terminal with color

Just replace /etc/bash.bashrc with the following content:

# /etc/bash.bashrc
#
# https://wiki.archlinux.org/index.php/Color_Bash_Prompt
#
# This file is sourced by all *interactive* bash shells on startup,
# including some apparently interactive shells such as scp and rcp
# that can't tolerate any output. So make sure this doesn't display
# anything or bad things will happen !

# Test for an interactive shell. There is no need to set anything
# past this point for scp and rcp, and it's important to refrain from
# outputting anything in those cases.

# If not running interactively, don't do anything!
[[ $- != *i* ]] && return

# Bash won't get SIGWINCH if another process is in the foreground.
# Enable checkwinsize so that bash will check the terminal size when
# it regains control.
#  (E11)
shopt -s checkwinsize

# Enable history appending instead of overwriting.
shopt -s histappend

case ${TERM} in
        xterm*|rxvt*|Eterm|aterm|kterm|gnome*)
                PROMPT_COMMAND=${PROMPT_COMMAND:+$PROMPT_COMMAND; }'printf "3]0;%s@%s:%s
# /etc/bash.bashrc
#
# https://wiki.archlinux.org/index.php/Color_Bash_Prompt
#
# This file is sourced by all *interactive* bash shells on startup,
# including some apparently interactive shells such as scp and rcp
# that can't tolerate any output. So make sure this doesn't display
# anything or bad things will happen !
# Test for an interactive shell. There is no need to set anything
# past this point for scp and rcp, and it's important to refrain from
# outputting anything in those cases.
# If not running interactively, don't do anything!
[[ $- != *i* ]] && return
# Bash won't get SIGWINCH if another process is in the foreground.
# Enable checkwinsize so that bash will check the terminal size when
# it regains control.
#  (E11)
shopt -s checkwinsize
# Enable history appending instead of overwriting.
shopt -s histappend
case ${TERM} in
xterm*|rxvt*|Eterm|aterm|kterm|gnome*)
PROMPT_COMMAND=${PROMPT_COMMAND:+$PROMPT_COMMAND; }'printf "\033]0;%s@%s:%s\007" "${USER}" "${HOSTNAME%%.*}" "${PWD/#$HOME/~}"'
;;
screen)
PROMPT_COMMAND=${PROMPT_COMMAND:+$PROMPT_COMMAND; }'printf "\033_%s@%s:%s\033\\" "${USER}" "${HOSTNAME%%.*}" "${PWD/#$HOME/~}"'
;;
esac
# fortune is a simple program that displays a pseudorandom message
# from a database of quotations at logon and/or logout.
# Type: "pacman -S fortune-mod" to install it, then uncomment the
# following line:
# [[ "$PS1" ]] && /usr/bin/fortune
# Welcome Message
# Arch Linux Logo Welcome Message
c1="$(tput sgr0)$(tput setaf 4)"
c2="$(tput bold)$(tput setaf 4)"
c3="$(tput bold)$(tput setaf 0)"
echo
echo "        $c2,$c1                       _     _ _"
echo "       $c2/$c1#$c2\\$c1        __ _ _ __ ___| |__ | (_)_ __  _   ___  __"
echo "      $c2/$c1###$c2\\$c1      / _\` | '__/ __| '_ \\| | | '_ \\| | | \\ \\/ /"
echo "     $c2/$c1#####$c2\\$c1    | (_| | | | (__| | | | | | | | | |_| |>  <"
echo "    $c2/$c1##,-,##$c2\\$c1    \\__,_|_|  \\___|_| |_|_|_|_| |_|\\__,_/_/\\_\\"
echo "   $c2/$c1##(   )##$c2\\$c1"
echo "  $c2/$c1#.--   --.#$c2\\  $c3 A simple, elegant GNU/Linux distribution."
echo " $c2/$c1\`           \`$c2\\$(tput sgr0)"$'\n'
# Show date
DATE="$(tput setaf 1)[$(tput setaf 6)$(date)"
DATE="${DATE}$(tput setaf 1)]"
echo $DATE
# Show kernel info
KERNEL="$(tput setaf 1)[$(tput setaf 6)$(uname -srmn)"
KERNEL="${KERNEL}$(tput setaf 1)]"
echo "$KERNEL"$'\n'
# Reset colors
tput sgr0
# Set colorful PS1 only on colorful terminals.
# dircolors --print-database uses its own built-in database
# instead of using /etc/DIR_COLORS. Try to use the external file
# first to take advantage of user additions. Use internal bash
# globbing instead of external grep binary.
# sanitize TERM:
safe_term=${TERM//[^[:alnum:]]/?}
match_lhs=""
[[ -f ~/.dir_colors ]] && match_lhs="${match_lhs}$(<~/.dir_colors)"
[[ -f /etc/DIR_COLORS ]] && match_lhs="${match_lhs}$(</etc/DIR_COLORS)"
[[ -z ${match_lhs} ]] \
&& type -P dircolors >/dev/null \
&& match_lhs=$(dircolors --print-database)
if [[ $'\n'${match_lhs} == *$'\n'"TERM "${safe_term}* ]] ; then
# we have colors <img src="http://obihoernchen.net/wordpress/wp-includes/images/smilies/icon_smile.gif" alt=":-)" class="wp-smiley">
# Enable colors for ls, etc. Prefer ~/.dir_colors
if type -P dircolors >/dev/null ; then
if [[ -f ~/.dir_colors ]] ; then
eval $(dircolors -b ~/.dir_colors)
elif [[ -f /etc/DIR_COLORS ]] ; then
eval $(dircolors -b /etc/DIR_COLORS)
fi
fi
# -- CUSTOM PS1 String START --
PS1="\$? \$(if [[ \$? == 0 ]]; then echo \"\[\033[01;32m\]\342\234\223\"; else echo \"\[\033[01;31m\]\342\234\227\"; fi) $(if [[ ${EUID} == 0 ]]; then echo '\[\033[01;31m\]\h'; else echo '\[\033[01;32m\]\u@\h'; fi)\[\033[01;34m\] \w \$\[\033[00m\] "
#PS1="\[\033[0;37m\]\342\224\214\342\224\200\$([[ \$? != 0 ]] && echo \"[\[\033[0;31m\]\342\234\227\[\033[0;37m\]]\342\224\200\")[$(if [[ ${EUID} == 0 ]]; then echo '\[\033[0;31m\]\u\[\033[0;37m\]@\[\033[0;96m\]\h'; else echo '\[\033[0;33m\]\u\[\033[0;37m\]@\[\033[0;96m\]\h'; fi)\[\033[0;37m\]]\342\224\200[\[\033[0;32m\]\w\[\033[0;37m\]]\n\[\033[0;37m\]\342\224\224\342\224\200\342\224\200\076 \[\033[0m\]"
# --- Custom PS1 String END ---
# Use this other PS1 string if you want \W for root and \w for all other users:
# PS1="$(if [[ ${EUID} == 0 ]]; then echo '\[\033[01;31m\]\h\[\033[01;34m\] \W'; else echo '\[\033[01;32m\]\u@\h\[\033[01;34m\] \w'; fi) \$([[ \$? != 0 ]] && echo \"\[\033[01;31m\]:(\[\033[01;34m\] \")\$\[\033[00m\] "
alias ls="ls --color=auto"
alias dir="dir --color=auto"
alias grep="grep --colour=auto"
alias dmesg='dmesg --color'
man() {
env LESS_TERMCAP_mb=$'\E[01;31m' \
LESS_TERMCAP_md=$'\E[01;38;5;74m' \
LESS_TERMCAP_me=$'\E[0m' \
LESS_TERMCAP_se=$'\E[0m' \
LESS_TERMCAP_so=$'\E[38;5;246m' \
LESS_TERMCAP_ue=$'\E[0m' \
LESS_TERMCAP_us=$'\E[04;38;5;146m' \
man "$@"
}
# Uncomment the "Color" line in /etc/pacman.conf instead of uncommenting the following line...!
# alias pacman="pacman --color=auto"
else
# show root@ when we do not have colors
PS1="\u@\h \w \$([[ \$? != 0 ]] && echo \":( \")\$ "
# Use this other PS1 string if you want \W for root and \w for all other users:
# PS1="\u@\h $(if [[ ${EUID} == 0 ]]; then echo '\W'; else echo '\w'; fi) \$([[ \$? != 0 ]] && echo \":( \")\$ "
fi
PS2="> "
PS3="> "
PS4="+ "
# Try to keep environment pollution down, EPA loves us.
unset safe_term match_lhs
# Try to enable the auto-completion (type: "pacman -S bash-completion" to install it).
[ -r /usr/share/bash-completion/bash_completion ] && . /usr/share/bash-completion/bash_completion
# Try to enable the "Command not found" hook ("pacman -S pkgfile" to install it).
# See also: https://wiki.archlinux.org/index.php/Bash#The_.22command_not_found.22_hook
[ -r /usr/share/doc/pkgfile/command-not-found.bash ] && . /usr/share/doc/pkgfile/command-not-found.bash
7" "${USER}" "${HOSTNAME%%.*}" "${PWD/#$HOME/~}"' ;; screen) PROMPT_COMMAND=${PROMPT_COMMAND:+$PROMPT_COMMAND; }'printf "3_%s@%s:%s3\\" "${USER}" "${HOSTNAME%%.*}" "${PWD/#$HOME/~}"' ;; esac # fortune is a simple program that displays a pseudorandom message # from a database of quotations at logon and/or logout. # Type: "pacman -S fortune-mod" to install it, then uncomment the # following line: # [[ "$PS1" ]] && /usr/bin/fortune # Welcome Message # Arch Linux Logo Welcome Message c1="$(tput sgr0)$(tput setaf 4)" c2="$(tput bold)$(tput setaf 4)" c3="$(tput bold)$(tput setaf 0)" echo echo " $c2,$c1 _ _ _" echo " $c2/$c1#$c2\$c1 __ _ _ __ ___| |__ | (_)_ __ _ ___ __" echo " $c2/$c1###$c2\$c1 / _\` | '__/ __| '_ \\| | | '_ \\| | | \\ \\/ /" echo " $c2/$c1#####$c2\$c1 | (_| | | | (__| | | | | | | | | |_| |> <" echo " $c2/$c1##,-,##$c2\$c1 \\__,_|_| \\___|_| |_|_|_|_| |_|\\__,_/_/\\_\\" echo " $c2/$c1##( )##$c2\$c1" echo " $c2/$c1#.-- --.#$c2\\ $c3 A simple, elegant GNU/Linux distribution." echo " $c2/$c1\` \`$c2\$(tput sgr0)"$'\n' # Show date DATE="$(tput setaf 1)[$(tput setaf 6)$(date)" DATE="${DATE}$(tput setaf 1)]" echo $DATE # Show kernel info KERNEL="$(tput setaf 1)[$(tput setaf 6)$(uname -srmn)" KERNEL="${KERNEL}$(tput setaf 1)]" echo "$KERNEL"$'\n' # Reset colors tput sgr0 # Set colorful PS1 only on colorful terminals. # dircolors --print-database uses its own built-in database # instead of using /etc/DIR_COLORS. Try to use the external file # first to take advantage of user additions. Use internal bash # globbing instead of external grep binary. # sanitize TERM: safe_term=${TERM//[^[:alnum:]]/?} match_lhs="" [[ -f ~/.dir_colors ]] && match_lhs="${match_lhs}$(<~/.dir_colors)" [[ -f /etc/DIR_COLORS ]] && match_lhs="${match_lhs}$(</etc/DIR_COLORS)" [[ -z ${match_lhs} ]] \ && type -P dircolors >/dev/null \ && match_lhs=$(dircolors --print-database) if [[ $'\n'${match_lhs} == *$'\n'"TERM "${safe_term}* ]] ; then # we have colors <img src="http://obihoernchen.net/wordpress/wp-includes/images/smilies/icon_smile.gif" alt=":-)" class="wp-smiley"> # Enable colors for ls, etc. Prefer ~/.dir_colors if type -P dircolors >/dev/null ; then if [[ -f ~/.dir_colors ]] ; then eval $(dircolors -b ~/.dir_colors) elif [[ -f /etc/DIR_COLORS ]] ; then eval $(dircolors -b /etc/DIR_COLORS) fi fi # -- CUSTOM PS1 String START -- PS1="$? $(if [[ $? == 0 ]]; then echo \"\[3[01;32m\]243\"; else echo \"\[3[01;31m\]247\"; fi) $(if [[ ${EUID} == 0 ]]; then echo '\[3[01;31m\]\h'; else echo '\[3[01;32m\]\u@\h'; fi)\[3[01;34m\] \w $\[3[00m\] " #PS1="\[3[0;37m\]244240$([[ $? != 0 ]] && echo \"[\[3[0;31m\]247\[3[0;37m\]]240\")[$(if [[ ${EUID} == 0 ]]; then echo '\[3[0;31m\]\u\[3[0;37m\]@\[3[0;96m\]\h'; else echo '\[3[0;33m\]\u\[3[0;37m\]@\[3[0;96m\]\h'; fi)\[3[0;37m\]]240[\[3[0;32m\]\w\[3[0;37m\]]\n\[3[0;37m\]2442402406 \[3[0m\]" # --- Custom PS1 String END --- # Use this other PS1 string if you want \W for root and \w for all other users: # PS1="$(if [[ ${EUID} == 0 ]]; then echo '\[3[01;31m\]\h\[3[01;34m\] \W'; else echo '\[3[01;32m\]\u@\h\[3[01;34m\] \w'; fi) $([[ $? != 0 ]] && echo \"\[3[01;31m\]:(\[3[01;34m\] \")$\[3[00m\] " alias ls="ls --color=auto" alias dir="dir --color=auto" alias grep="grep --colour=auto" alias dmesg='dmesg --color' man() { env LESS_TERMCAP_mb=$'\E[01;31m' \ LESS_TERMCAP_md=$'\E[01;38;5;74m' \ LESS_TERMCAP_me=$'\E[0m' \ LESS_TERMCAP_se=$'\E[0m' \ LESS_TERMCAP_so=$'\E[38;5;246m' \ LESS_TERMCAP_ue=$'\E[0m' \ LESS_TERMCAP_us=$'\E[04;38;5;146m' \ man "$@" } # Uncomment the "Color" line in /etc/pacman.conf instead of uncommenting the following line...! # alias pacman="pacman --color=auto" else # show root@ when we do not have colors PS1="\u@\h \w $([[ $? != 0 ]] && echo \":( \")$ " # Use this other PS1 string if you want \W for root and \w for all other users: # PS1="\u@\h $(if [[ ${EUID} == 0 ]]; then echo '\W'; else echo '\w'; fi) $([[ $? != 0 ]] && echo \":( \")$ " fi PS2="> " PS3="> " PS4="+ " # Try to keep environment pollution down, EPA loves us. unset safe_term match_lhs # Try to enable the auto-completion (type: "pacman -S bash-completion" to install it). [ -r /usr/share/bash-completion/bash_completion ] && . /usr/share/bash-completion/bash_completion # Try to enable the "Command not found" hook ("pacman -S pkgfile" to install it). # See also: https://wiki.archlinux.org/index.php/Bash#The_.22command_not_found.22_hook [ -r /usr/share/doc/pkgfile/command-not-found.bash ] && . /usr/share/doc/pkgfile/command-not-found.bash

More information and other examples: https://wiki.archlinux.org/index.php/Color_Bash_Prompt

To search for available packages in repos if your command doesn’t exist on your current system configuration install pkgfile

pacman -Sy pkgfile
pkgfile --update

14. Disable some logging to extend USB stick lifetime

Flash drives have limited number of write cycles and default linux logging will write pretty much so your USB will be destroyed pretty fast.
That’s why a swap file on your USB stick is a really bad idea!
But you can disable some logging to extend the liefetime of your USB stick.

Just edit syslog-ng.conf and disable some of the “log { source(src); filter…” lines at the end of the file.
You can comment out (add a “#” in front of the line) all lines if you want to disable everything.

  nano /etc/syslog-ng/syslog-ng.conf

I’ve disabled some of them. That’s my example:

# Disable some logging
#log { source(src); filter(f_acpid); destination(d_acpid); };
log { source(src); filter(f_authpriv); destination(d_authlog); };
#log { source(src); filter(f_syslog); destination(d_syslog); };
#log { source(src); filter(f_cron); destination(d_cron); };
#log { source(src); filter(f_daemon); destination(d_daemon); };
#log { source(src); filter(f_kernel); destination(d_kernel); };
#log { source(src); filter(f_lpr); destination(d_lpr); };
#log { source(src); filter(f_mail); destination(d_mail); };
#log { source(src); filter(f_news); destination(d_news); };
#log { source(src); filter(f_ppp); destination(d_ppp); };
log { source(src); filter(f_user); destination(d_user); };
#log { source(src); filter(f_uucp); destination(d_uucp); };
#log { source(src); filter(f_debug); destination(d_debug); };
#log { source(src); filter(f_messages); destination(d_messages); };
log { source(src); filter(f_err); destination(d_errors); };
log { source(src); filter(f_emergency); destination(d_console); };
#log { source(src); filter(f_everything); destination(d_everything); };
#log { source(src); filter(f_iptables); destination(d_iptables); };

On top of that you could disable logging of some servers like samba, minidlna and so on.
I don’t do that because I want to have these logs but to do so set the log path in the config files to /dev/null

Das Mysterium – ZRAM

Da ich immer wieder so viele falsche Informationen über ZRAM sehe habe ich beschlossen mal etwas Klarheit in die ganze Materie zu bringen
Oft wird über einen Performance Vorteil gesprochen, da ZRAM schneller als Swap ist. Das ganze mag für normale PCs stimmen aber wir sind hier mit Android unterwegs und da sieht das ganze etwas anders aus!

Erst einmal eine kurze Feststellung für alle, die noch nicht so lange in der Android Szene unterwegs sind:
ZRAM = ramzswap = Compcache

Um ZRAM genauer zu erklären müssen erstmal andere Begriffe genauer geklärt werden:
Swap kann man mit der Auslagerungsdatei unter Windows vergleichen werden. Wird der Arbeitsspeicher (RAM) zu voll kann der PC die Daten, die gerade nicht aktiv gebraucht werden (z.B. Hintergrundanwendungen) auslagern um so wieder RAM frei zu räumen. Dazu werden diese Daten auf eine Festplatte geschrieben. Bei Bedarf können diese Daten dann einfach wieder von dort aus gelesen werden. Selbst die schnellste SSD ist aber langsamer als der Arbeitsspeicher. Unter Android gibt es Swap nicht!

Bei ZRAM werden nicht benötigte Speicherressourcen komprimiert und dann in einen fest reservierten Bereich im RAM verschoben (ZRAM). Also eine Art Swap Partition im Arbeitsspeicher.
Dadurch ist mehr Ram frei, da die Daten dann nur noch ca. 1/4 des ehemaligen Speicherbedarfs haben. Allerdings muss die CPU dadurch mehr arbeiten, da sie die Daten komprimieren muss (oder auch wieder entpacken wenn sie wieder gebraucht werden). Der Vorteil hierbei liegt ganz klar in der Geschwindigkeit. Da die Swap Partition sich im RAM befindet ist diese viel schneller als eine Swap Partition auf einer Festplatte.

An und für sich eine ganz tolle Sache. Aber Android besitzt keine Swap Partition und daher bringt ZRAM unter Android keinen Performance Gewinn wie es bei einem normalen PC der Fall wäre.

Beim normalen PC würde das folgendermaßen aussehen:
Swap = Auslagerungsdatei (auf Festplatte) –> Langsam
ZRAM (Swap im RAM) –> Schneller als Swap
RAM –> Schnell

Bei Android gibt es keine Swap Partition und daher bringt ZRAM auch keinen Performance Boost.
Das einzige was ZRAM bringt ist “mehr” RAM. Durch das komprimieren “vergrößert” sich sozusagen der verfügbare Arbeitsspeicher. Das ist auf Geräten mit sehr wenig RAM (<256MB) auch ziemlich nützlich. Das S2 hat aber 1GB und die reichen mehr als aus. Da muss nicht künstlich auf 1,5GB hochgepusht werden.

Denn aktiviert man ZRAM hat das auch 2 Nachteile. Das komprimieren und dekomprimieren verbraucht CPU Time, welche wiederum höheren Stromverbrauch zur Folge hat.

Grob kann man also sagen (Für Geräte mit mehr als 512MB RAM):
Ohne ZRAM: +CPU Performance | +Battery | -RAM
Mit ZRAM: -CPU Performance | -Battery | +RAM

Für Geräte mit zu wenig RAM also durchaus sinnvoll. Aber wer ballert beim S2 schon sein kompletten RAM voll und braucht dann immer noch mehr?

Überprüfen ob ZRAM läuft kann man im Terminal mit
free oder cat /proc/meminfo

ZRAM aus:

ZRAM an:

Weiterführende Infos:
Swap and Compcache – CyanogenMod Wiki