Huawei E220 in Gutsy Gibbon (Kubuntu)

shakir    Information Insemination     bash, gutsy, hsdpa, linux, script, ubuntu

I received a package earlier today, and was very happy to find out that it’s the Huawei E220 that I ordered few days ago. This Huawei E220 is a 3G card that supports speed of up to 3.6 Mbps, using HSDPA technology, and it makes me can’t wait to have it work with Gutsy Gibbon. After struggling for quite some time, I found out that the answer to my problem was just these two lines of command:

sudo modprobe -r uhci_hcd
sudo modprobe uhci_hcd

If that short version of my solution doesn’t help, please read ahead…

I’ve been trying to get this modem to work with Gutsy Gibbon, but the online howtos that I found on the net doesnt reliably work for me.

Weird enough for me initially as sometime I can just use wvdial (more on this afterward) and get connected, while most of the time I got various errors about the modem. With these non-reproducible errors and successes, I do some trial and error while looking at the kernel logs, and came up with this one final observation;

shakir@herugrim ~ $ uname -r  # just to show the kernel I'm using, as your mileage may vary
2.6.22-14-generic

At one shell I run this to see kernel messages;

sudo tail -f  /var/log/messages

and got this messages upon plugging in my Huawei E220 to the USB port;

usb 3-1: new full speed USB device using uhci_hcd and address 5
usb 3-1: configuration #1 chosen from 1 choice
usbcore: registered new interface driver libusual
usbcore: registered new interface driver usbserial
/build/buildd/linux-source-2.6.22-2.6.22/drivers/usb/serial/usb-serial.c: USB Serial support registered for generic
usbcore: registered new interface driver usbserial_generic
/build/buildd/linux-source-2.6.22-2.6.22/drivers/usb/serial/usb-serial.c: USB Serial Driver core
/build/buildd/linux-source-2.6.22-2.6.22/drivers/usb/serial/usb-serial.c: USB Serial support registered for GSM modem (1-port)
option 3-1:1.0: GSM modem (1-port) converter detected
usb 3-1: GSM modem (1-port) converter now attached to ttyUSB0

Nice, it has detected my Huawei E220 as GSM modem, and attach it to /dev/ttyUSB0. Let’s create a wvdial configuration to connect to the telco (Celcom in my case). My /etc/wvdial.conf looks like this;

[Dialer Celcom3G]
Phone = *99***1#
Modem = /dev/ttyUSB0
Username = user
Password = pass
ISDN = 0
New PPPD = yes
Baud = 1843200
Init2 = ATZ
Init3 = ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0
Modem Type = Analog Modem

So let’s just connect;

shakir@herugrim ~ $ sudo wvdial Celcom3G
WvDial<*1>: WvDial: Internet dialer version 1.56
WvModem<*1>: Cannot get information for serial port.
WvDial<*1>: Initializing modem.
WvDial<*1>: Sending: ATZ
WvDial<*1>: Sending: ATQ0
WvDial<*1>: Re-Sending: ATZ
WvDial: Modem not responding.

Ouch, an error. The modem has been detected, what could go wrong? After some time, and no success with the available tutorials, I finally got this working, as the problem was related to the USB drivers. What I did was to reload the uhci_hcd module

shakir@herugrim ~ $ sudo modprobe -r uhci_hcd

Lets take a look at what the kernel says about this;

[..snipped..]
Dec  4 22:45:24 herugrim kernel: [ 1343.504000] usb 3-1: USB disconnect, address 6
Dec  4 22:45:24 herugrim kernel: [ 1343.504000] option1 ttyUSB0: GSM modem (1-port) converter now disconnected from ttyUSB0
Dec  4 22:45:24 herugrim kernel: [ 1343.504000] option 3-1:1.0: device disconnected
[..snipped..]

and then lets load the module back again

shakir@herugrim ~ $ sudo modprobe uhci_hcd

and see what happens;

USB Universal Host Controller Interface driver v3.0
ACPI: PCI Interrupt 0000:00:1a.0[A
uhci_hcd 0000:00:1a.0: UHCI Host Controller
[..snipped..]
usb 3-1: new full speed USB device using uhci_hcd and address 2
usb 3-1: configuration #1 chosen from 1 choice
option 3-1:1.0: GSM modem (1-port) converter detected
usb 3-1: GSM modem (1-port) converter now attached to ttyUSB0
option 3-1:1.1: GSM modem (1-port) converter detected
usb 3-1: GSM modem (1-port) converter now attached to ttyUSB1
option 3-1:1.2: GSM modem (1-port) converter detected
usb 3-1: GSM modem (1-port) converter now attached to ttyUSB2
[..snipped..]

Last time the GSM modem was attached to just /dev/ttyUSB0, but now it’s also attached to /dev/ttyUSB1 and /dev/ttyUSB2. And I guess it’s time to connect using wvdial;

WvDial<*1>: WvDial: Internet dialer version 1.56
WvModem<*1>: Cannot get information for serial port.
WvDial<*1>: Initializing modem.
WvDial<*1>: Sending: ATZ
WvDial Modem<*1>: ATZ
WvDial Modem<*1>: OK
WvDial<*1>: Sending: ATZ
WvDial Modem<*1>: ATZ
WvDial Modem<*1>: OK
WvDial<*1>: Sending: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0
WvDial Modem<*1>: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0
WvDial Modem<*1>: OK
WvDial<*1>: Modem initialized.
WvDial<*1>: Sending: ATDT*99***1#
WvDial<*1>: Waiting for carrier.
WvDial Modem<*1>: ATDT*99***1#
WvDial Modem<*1>: CONNECT
WvDial<*1>: Carrier detected.  Waiting for prompt.
WvDial: Don't know what to do!  Starting pppd and hoping for the best.
WvDial: Starting pppd at Tue Dec  4 22:49:25 2007
WvDial: Pid of pppd: 6433
WvDial<*1>: Using interface ppp0
WvDial<*1>: local  IP address 10.188.15.175
WvDial<*1>: remote IP address 10.64.64.64
WvDial<*1>: primary   DNS address 202.188.0.133
WvDial<*1>: secondary DNS address 202.188.1.5

It works, yeay As usual, lets create a script for this;

#!/bin/bash

sudo modprobe -r uhci_hcd
sudo modprobe uhci_hcd

echo "Waiting kernel to detect modem device"

# Wait till the kernel recognizes our E220
while true; do

        # Read log
        tail -n 20 /var/log/messages  > access.history
        sleep 1
        tail -n 20 /var/log/messages  > access.current

        LOG=`diff access.history access.current | grep ">" | tr -d ">"`

        # Check if it is detected
        if [[ "$LOG" =~ "converter now attached to ttyUSB1" ]]; then
                echo "Modem device detected, dialing..."
                # If it is, get out of the loop immedietly
                break
        fi
done

rm access.history access.current

# Start dialing
sudo wvdial Celcom3G

p/s: The script is now maintained here. Please go there to see the latest updates and discussions

Nikon D40x in Ubuntu Linux

shakir    Information Insemination, Nerd Public Journal?     bash, linux, photography, script

I’ve been using Nikon D40X on my previous trip to Jakarta (and Bandung). For this I need to update my previous script as there are few differences, as the following:

• I set the camera to use RAW format rather than jpeg
• Data transfer is through the laptop’s built in MMC card slot, rather than USB
• I’m now a (K)Ubuntu user

First step, is the software part. As RAW format is not really usable for "presentation", I need to convert it to another format, say JPEG. For this I’m gonna need  ufraw. Let’s install it;

sudo apt-get install ufraw
sudo apt-get install gimp-ufraw #optional

Using ufraw-batch (included in the ufraw package), I came up with this self explanatory, lightly commented bash script;

#!/bin/bash

#Set some variables
MOUNT_DIR=/media/mmc
MOUNT_DEV=/dev/mmcblk0p1
PHOTO_DIR=/home/shakir/Photos/temp/D40X
#Copy files from the MMC card
sudo mkdir $MOUNT_DIR
sudo mount -o uid=shakir,gid=shakir $MOUNT_DEV -t vfat $MOUNT_DIR
mv -v /media/mmc/dcim/100nd40x/* $PHOTO_DIR
sudo umount $MOUNT_DIR
sudo rmdir $MOUNT_DIR
cd $PHOTO_DIR
mkdir jpeg
#Convert RAW files to low quality JPEGs
for i in `ls *.nef`; do
        ufraw-batch –out-type=jpeg –out-path=jpeg/  \
          –compression=70 –size=600,600 –overwrite $i
        mv -v $i nef/
done
mkdir nef
mv *.nef nef/

The script would generate a low quality / resolution JPEGs, based on the un-processed RAWs. The JPEGs are useful to actually help me deciding which photos are meant for deletion, and which to keep. Later I would manually go to the selected RAW files folder, and;

ufraw *.nef

to edit the RAW files. This final step would give me the files that I mean to keep;

ufraw-batch --out-type=jpeg --out-path=jpeg/ --compression=100 --overwrite *.nef

I bring back no souvenirs but these photos;

Recover deleted files with photorec

shakir    Information Insemination     bash, linux, script

I was just watching a movie while suddenly all the files except the hidden ones in my home directory were gone, just as someone issued the command

sudo rm -rf /home/shakir/*

To avoid further data loss, I quickly power off and boot into my Ubuntu live CD to create an image of my /home partition. Here’s how to see in which partition is my /home directory:

shakir@herugrim ~ $ cat /proc/partitions
major minor  #blocks  name
8     0  156290904 sda
8     1   21494938 sda1
8     2    8795587 sda2
8     3    2939895 sda3
8     4          1 sda4
8     5    9775521 sda5
8     6  113282316 sda6

From the partition sizes I can tell my /home partition was in the /dev/sda6, and the next step is to really create an image of the partition to safely doing the recovery process without risking of losing more data. /media/usbdisk is where my external usb harddisk is mounted

sudo dd if=/dev/sda6 of=/media/usbdisk/herugrim-sda6-20071010.img

Photorec is part of testdisk package, and this is how to install it in ubuntu;

sudo apt-get update
sudo apt-get install testdisk

It’s time to actually do run the program

sudo photorec /home/shakir/temp/herugrim-sda6-20071010.img

After going through some options, photorec starts doing it’s job.

Photorec stores recovered files in recup_dir.<sequence>/<sequence>.<file extension> in the specified target directory, which is not very useful. Here’s a script I wrote to find all the recovered JPEG files and move/rename it accordingly. Almost the same technique can be used for other file formats.

#!/bin/bash
PHOTODIR=/home/shakir/temp/photorec
cd $PHOTODIR
mkdir JPEG
for i in `ls | grep recup`;do
	for j in `ls $i/*.jpg`; do
		if FILE=`exiv2 $j 2>/dev/null | grep timestamp | awk ‘{ print $4"-"$5 }’ | tr -d ‘:’ | grep 200`; then
			cp -v $j "JPEG/"$FILE".jpg"
		fi
	done
done

Automatic backup using rsync

shakir    Information Insemination     bash, linux, script

I’ve suddenly lost all the files in my home folders, and I quickly do these to not let the impact to be as bad, if it ever recurs.

Say 192.168.1.10 is my backup server (anduril), herugrim is my laptop’s name, these are the steps I did to let my laptop automatically rsync to the backup server.

Configure Password-less Login

Generate RSA key for use with SSH. Make sure you dont use any passphrase.

shakir@herugrim ~ $ ssh-keygen -t rsa -C shakir@192.168.1.10
Generating public/private rsa key pair.
Enter file in which to save the key (/home/shakir/.ssh/id_rsa): /home/shakir/.ssh/id_rsa.192.168.1.10
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /home/shakir/.ssh/id_rsa.192.168.1.10.
Your public key has been saved in /home/shakir/.ssh/id_rsa.192.168.1.10.pub.
The key fingerprint is: 11:d9:23:2e:68:05:59:d9:ac:5a:00:69:17:3a:b4:24  

Copy the public key to the server

shakir@herugrim ~ $ ssh-copy-id -i .ssh/id_rsa.192.168.1.10.pub shakir@192.168.1.10

Test if passwordless login successful

shakir@herugrim ~ $ ssh shakir@192.168.1.10 -i .ssh/id_rsa.192.168.1.10
shakir@anduril:~$

Configure Target Directory

shakir@anduril:~$ mkdir backup/herugrim -p

Manual Rsync (test power)

shakir@herugrim ~ $ rsync --verbose --progress --compress --rsh=ssh --recursive --times --perms --links --delete /home/shakir anduril:/home/shakir/backup/herugrim -e "ssh -i /home/shakir/.ssh/id_rsa.192.168.1.10"

Configure cron

At your machine To edit your crontab entry, issue this command

crontab -e

Add this entry to your crontab

30      9       *       *       1-5     rsync --compress --rsh=ssh --recursive --times --perms --links --delete /home/shakir anduril:/home/shakir/backup/herugrim -e "ssh -i /home/shakir/.ssh/id_rsa.192.168.1.10"

Run cron

sudo /etc/init.d/cron start     #depending on your linux distro

And I guess that’s all..

backup guys, backup…