shieldroute

Bluetooth on Virtual box with Mac OS x 10.9.4 as host

2014-09-09T20:42:00.002-07:00

You want to have bluetooth on your Linux/Window in VirtualBox while using Mac OS X as host. The bug (Ticket #2372) has been filed six (6!!!) years ago, go figure.

For 10.9.4 on MBP 2014 you need to execute following:

$sudo su
$launchctl unload /System/Library/LaunchDaemons/com.apple.blued.plist
$kextunload -b com.apple.iokit.IOBluetoothSerialManager
$kextunload -b com.apple.iokit.BroadcomBluetoothHostControllerUSBTransport

Note that the actual extension may vary depending on your Mac hardware, so you can simple list modules and try to unload each and one of them until you can connect bluetooth on virtual box.

$kextstat | grep -i bluetooth

Cheers

Developing for Nordic Semiconductor nRF51822

2014-08-27T00:07:00.002-07:00

What

This post describes installation and first simple example running dev kit nRF51822 on Mac OS X 10.9.4.

Why

Its either you who will rule Internet of Things, or Internet of Things will rule you. Pick a side and spot wining.
I got my hands on nRF51822 and since I have switched to Mac, the installation instructions for mac are confusing... there was no obvious way to run it from thus this is a mental note to my self how to set up environment. So here you go, after tutorial you will be up and running in no time :) with multiple options.

How

First install needed shit, its all below. The instructions are minimal and assumes you can figure out some simple stuff. If not ask, Google, change carrier.

1. Eclipse

I have many copies of eclipse for different developing purpose containing different language. The reason is that plugins do brake each other and I rather reinstall single plugin then have all my "life"dependent on one eclipse.

Download Eclipse http://www.eclipse.org/downloads/packages/eclipse-ide-cc-developers/lunar (I used eclipse-cpp-luna-R-macosx-cocoa-x86_64.tar.gz)
Extract it
Rename it to eclipseARM
Enter folder and rename eclipse.app to ArmLipse.app
Move folder to /Applications/ or /User/Applications/

1.1 Plugins

Lauch ArmLipse
Create new workspace nRF51 (keep it clean)
Help->Install new software
Add http://embsysregview.sourceforge.net/update
and install embsysregview
Go back, select all available sites, search and install GDB Hardware Debugging
Go back and add site http://gnuarmeclipse.sourceforge.net/updates and install plug in for arm
Go back and add http://nrf51osx.sourceforge.net/updates and install nordic plugin

2. GNU Tools

Get tar for mac from here https://launchpad.net/gcc-arm-embedded/ (I got gcc-arm-none-eabi-4_8-2014q2)
Extract it.

3. Jlink

Get libusb to start with brew install libusb
Get Jlink from http://www.segger.com/jlink-software.html
You will need to enter number from the biggest chips on the board
I got JLink_MacOSX_V490b.pkg, double tap and follow instructions

4. Libraries

You will need firmware either s110, s120, s130 depending on the purpose. You can get then form nordic Semiconductor site.
Create folder development in /User/your_name/ and create subfolder nrf51
Copy, SDK, Firmware and GNU tools to it

Mine looks:

->tree -CL 1 nrf51
nrf51
├── gcc-arm-none-eabi-4_8-2014q2
├── nrf51_sdk_v6_0_0_43681
├── s110_nrf51822_7.0.0
├── s120_nrf51822_1.0.0
└── s130_nrf51822_0.5.0-1.alpha

5 directories, 0 files

5. Configure

5.1 Eclipse - hardway

In case you are insane and have no life play here:

embsysregview: Copy sdks/nrf51/nrf51_sdk_v6_0_0_43681/nrf51822/SVD/nrf51.xml to Applications/eclipseARM/plugins/org.eclipse.cdt.embsysregview.data_0.2.4.r168/data/cortex-m0/Nordic

cd ~
mkdir Applications/eclipseARM/plugins/org.eclipse.cdt.embsysregview.data_0.2.4.r168/data/cortex-m0/Nordic
cp development/sdks/nrf51/nrf51_sdk_v6_0_0_43681/nrf51822/SVD/nrf51.xml Applications/eclipseARM/plugins/org.eclipse.cdt.embsysregview.data_0.2.4.r168/data/cortex-m0/Nordic/

Configure Environment, add this to the .bashrc (!!!change username to your username!!!)

cd ~
export GNU_INSTALL_ROOT=/Users/username/development/sdks/nrf51/gcc-arm-none-eabi-4_8-2014q2
export GNU_VERSION=4.8.4

Now, lets remove Nordics bug from Makefile.posix, just delete these lines (why are they there???)
```
GNU_INSTALL_ROOT := /usr/local/gcc-arm-none-eabi-4_8-2014q1
GNU_VERSION := 4.8.3
```
Now follow the rest of configuration described in n29 v1.0 from 1.2 and make sure you can compile blinky.

Dont rewrite, c-c c-v, into startup window

mon speed 10000
mon endian little
mon flash download = 1 
mon flash device = NRF51822
mon reset 0

5.2 Eclipse easy way

If you want time over for your girlfriend just follow these steps:

Create a project how its is described in http://sourceforge.net/p/nrf51osx/wiki/create%20a%20project/
You done!

6 Make that F*** blink

So all done and you happy, almost. Go get this awesome programmer http://sourceforge.net/projects/rknrfgo/. Install, fire up terminal and run:

cdccontrol disable

Then fire up rknrfgo (fancy name) and program your device :)

6.1 Bug fixing

Everybody is changing everything and fixing bugs all over the places. Which leads that there is HUGE amount of inconsistency in EVERY single post you will read or software you install. Trust no one.

When you created project and if you are using SDK 6+ (and you should) uploading ldscripts/mem.ld will be faulty. To make it work just c-c c-v following (assuming you select softdevice :))

/* 
** Linker script to configure memory regions. 
**
** Device type: xxaa, 256K flash, 16K RAM
** Device type: xxab, 128K flash, 16K RAM
**
** Softdevice:  none, all memory available to device,  Code start: 0x0000 , xxaa:0x40000, xxbb:0x20000, RAM start: 0x20000000 
** Softdevice:  s110, uses 80K code, 8K RAM,   Code start: 0x16000, xxaa:0x2c000, xxbb:0x0c000, RAM start: 0x20002000 
** Softdevice:  s120, uses 96K code, 10K RAM,   Code start: 0x18000, xxaa:0x28000, xxbb:0x08000, RAM start: 0x20002800 
*/ 

GROUP(-lgcc -lc -lnosys)

MEMORY
{
  FLASH (rx) : ORIGIN = 0x00016000, LENGTH = 0x2C000
  RAM (rwx)  : ORIGIN = 0x20002000, LENGTH = 0x2000 
}

/*
** All permutations, copy one into the MEMORY section above, removing the leading '**'s
**
** xxaa, no softdevice
**  FLASH (rx) : ORIGIN = 0x00000000, LENGTH = 0x40000  
**  RAM (rwx)  : ORIGIN = 0x20000000, LENGTH = 0x4000 
**
** xxab, no softdevice
**  FLASH (rx) : ORIGIN = 0x00000000, LENGTH = 0x20000 
**  RAM (rwx)  : ORIGIN = 0x20000000, LENGTH = 0x4000 
**
** xxaa s110
**  FLASH (rx) : ORIGIN = 0x00016000, LENGTH = 0x2C000
**  RAM (rwx)  : ORIGIN = 0x20002000, LENGTH = 0x2000 
**
** xxab s110
**  FLASH (rx) : ORIGIN = 0x00016000, LENGTH = 0x0C000
**  RAM (rwx)  : ORIGIN = 0x20002000, LENGTH = 0x2000 
**
** xxaa s120
**  FLASH (rx) : ORIGIN = 0x00018000, LENGTH = 0x28000
**  RAM (rwx)  : ORIGIN = 0x20002800, LENGTH = 0x1800 
**
** xxab s120
**  FLASH (rx) : ORIGIN = 0x00018000, LENGTH = 0x08000
**  RAM (rwx)  : ORIGIN = 0x20002800, LENGTH = 0x1800 
**
*/

7 What you get?! This beauty! :)

Dat Feeling of the #blink! #nordic #ble #internetofthings #iot #sanfrancisco #arduino #california #oneplus #summershots #SoC #nrf51822

Acknowledgment

The biggest thanks goes to the people who spread they knowledge all over the internet, and did load of coding. Go check them out, donate, send flower or fruits!

Roland King (King indeed)
http://sourceforge.net/projects/rknrfgo/
http://sourceforge.net/projects/nrf51osx/

Other sources of ideas
http://hg.cmason.com/nrf
http://robolabwiki.sdu.dk/mediawiki/index.php/Nordic_Semiconductor_nRF51822_development
http://erlblog.lewin.nu/2013/12/setting-up-eclipse-under-mac-os-x-for.html

Recovering a corrupt Eclipse workspace

2014-04-08T02:58:00.000-07:00

I had a computer crash with a running Eclipse which resulted in a corrupt workspace. I was simply getting message that workspace is in use. While there is many proposed solutions on the internet the simplest one is to open .metadata folder in your workspace, make sure that you can see invisible (hidden) files there and remove .lock file. Works like a charm and preserves all projects. :)

Customizing django-registration

2014-03-19T06:29:00.001-07:00

Whatz up?!

The integration of django-register is awesome! But if you want to customize it with additional models it can be confusing.

Here in example I will show how to create your custom backend, form to handle it (custom validator and model choice field).

Note, the django I use is version v1.5.4 django-registration v1.0. Some line breaks are imposed due to layout of this page.

Thus, bellow you will find everything you need to:

Create custom backend
Add custom form
Create custom validation
Badabung! :D

Below is and example extending

from registration.backends.default.views.RegistrationView

First of all, inherit RegistrationView, then override register() and get_form_class() functions. In the examples to follow it is assumed that the class is saved to regbackend.py.

#regbackend.py
from registration.backends.default.views import RegistrationView

#create custom registration backend
class StudentBackend(RegistrationView): 
    '''
    we just want to override register and fix profile creation
    ''' 
    def register(self,request, **kwargs):
        from usermanager.models import Profile
        user=super(StudentBackend,self).register(request, **kwargs)
        user.first_name = kwargs['first_name']
        user.last_name = kwargs['last_name']
        user.save()
        profile = Profile.objects.create(
                                user=user, course=kwargs['course'])
        profile.save()
        
    def get_form_class(self, request):
        """
        Return the default form class used for user registration.

        """
        from usermanager.form import UserRegistrationForm
        return UserRegistrationForm

Below is the code from urls.py. Nothing fancy, just import your registration backend, and form, and redirect accounts/register/ to your backend.

#urls.py
from regbackend import StudentBackend
from form import UserRegistrationForm

urlpatterns = patterns('',
    url(r'^accounts/register/$',
        StudentBackend.as_view(form_class = UserRegistrationForm),
        name='registration_register'),
    (r'^accounts/', include('registration.backends.default.urls')),
    #more url configs
)

Following code is from form.py:

#form.py
from django import forms
from django.contrib.auth.models import User
from usermanager.models import Course, Profile
from registration.forms import RegistrationForm
from django.utils.translation import ugettext_lazy as _



attrs_dict = { 'class': 'required' }

#this function checks for the allowed domain:
def affiliation_valid_email(value):
    allowed_domain=['domain.se','domain.se']
    #not the prettiest soliution 
    v_error=_("Email does not match allowed domains (%s, %s)"
                           %(allowed_domain[0],allowed_domain[1])
    if value.split('@')[1] not in allowed_domain:
        raise forms.ValidationError(v_error)
    else:
        return value


#the registration form is inherited from django-registration   
class UserRegistrationForm(RegistrationForm):  
    
    class Meta:
        model = User
        fields = ("first_name",
                  "last_name",
                  "username", 
                  "email", 
                  "password1", 
                  "password2")
    #username is enforced to some particular rules 
    u_error = _("User name needs to match domain convention.")   
    username = forms.RegexField(regex=r'^\w',
                                max_length=9,
                                widget=forms.TextInput(attrs=attrs_dict),
                                label=_("Username - same as in domain!"),
                                error_messages={ 
                                'invalid': u_error })
    first_name = forms.CharField(max_length=30)
    last_name = forms.CharField(max_length=30)
    course = forms.ModelChoiceField(queryset=Course.objects,required=True)
    email = forms.EmailField(validators=[affiliation_valid_email],
                             widget=forms.TextInput(
                                     attrs=dict(attrs_dict,maxlength=75)
                                     ),
                                     label=_("domain - Email address"))
    

    
    def clean(self):
        #clean with super
        ue_error = _("Email and user name do not match.")
        super(UserRegistrationForm, self).clean()
        #this is just checking that username is the same as in email
        if self.cleaned_data['email'].split('@')[0] == 
                                        self.cleaned_data.get("username"):
            return self.cleaned_data
        else:
            raise forms.ValidationError(ue_error)
         
    def save(self, commit=True):
        user = super(UserRegistrationForm, self).save(commit=True)    
        user.email = self.cleaned_data["email"]
        user.first_name =  self.cleaned_data["first_name"]
        user.last_name =self.cleaned_data["last_name"]
        profile = Profile(
                   systemuser=user, course = self.cleaned_data["course"])

    if self.commit:
        user.save()
        profile.save()
    return user

Models used in this example:

#models.py

class Course(models.Model):
    code = models.CharField(max_length=6)
    teacher = models.CharField(max_length=30)
    
    class Meta:
        app_label = "usermanager"
        verbose_name = "Course"
        verbose_name_plural = "Courses"
    
    def __unicode__(self):
        return "%s" % self.code
    
class Profile(models.Model):
    user=models.ForeignKey(User, unique=True)
    course=models.ForeignKey('usermanager.Course')
    created = models.DateTimeField(auto_now_add=True)
    updated_local = models.DateTimeField(blank=True, null=True)
    

    def getCourse(self):
        return "%s" %self.course
    
    def save(self,*args,**kwargs):
        self.updated_local = datetime.now()
        super(Profile, self).save(*args,**kwargs)
    
    def __unicode__(self):
        return "Profile: %s" %self.user

Finally, remember to enable registration in installed apps before running

python manage.py syncdb

INSTALLED_APPS = (
    'django.contrib.auth',
    'django.contrib.contenttypes',
    'django.contrib.sessions',
    'django.contrib.messages',
    'django.contrib.staticfiles',
    'django.contrib.admin',
    'django.contrib.admindocs',
    'registration',
)

That's it foks, happy coding! L

Quick and dirty multiply raw file conversion

2013-07-22T15:23:00.000-07:00

Multi-raw2dng-converter

I asume that you know all about installiation and usage of Magic Lanters raw video functionality.

If you like me find your self with many raw files, using raw2dng manually can be pain in a**. Here you have python 3.x script working and tested on Windows 8. Given the argument -d "file directory" it will look for any subfolders there and for each raw file in subfolder will create new folder and will run raw2dng on that file. The resulting dng will end up in folder named "ra_file_name" without RAW. The script utilizes as many new processes as there cores in CPU.

Note that this script is just quick and dirty way to fix my problem. I hope it will be useful for other as well. Feel free to modify and extend.

Usage:

D:\dev\python multiConverter.py -d "F:\Images\2013\0722"

Output:

Successfully processed: F:\Images\2013\0722\1\M17-2243.RAW

Script:

#!/usr/bin/python
#searches for raw files and converts then into a directory with the 
#same name as file
import glob
import sys, getopt
import os, multiprocessing, queue, subprocess
from threading import Thread, Lock

FILE = 0
DIR = 1
RAW2DNG = r"C:path\to\raw2dng.exe"
def optionsGet(argv):
 inputdir = ""
 try:
  opts, args = getopt.getopt(argv,"hd:",["ddir="])
 except getopt.GetoptError:
  print('multiconverter.py -d ')
  sys.exit(2)
 for opt, arg in opts:
  if opt == "-h":
   print("multiconverter.py -d ")
   sys.exit()
  elif opt in ("-d", "--ddirectory"):
   inputdir = arg
 return inputdir

def getCPUs():
 # Windows
    try:
        res = int(os.environ['NUMBER_OF_PROCESSORS'])
        if res > 0:
            return res
    except (KeyError, ValueError):
        pass

def createFolder(name):
 pass
 

def getFileNames(name):
 return glob.glob(name + "\*.RAW")

def getsubdirs(inputdir):
 file_list = []
 dirs=os.listdir(r'%s' %inputdir)
 for d in dirs:
  tmp = r'%s\%s' %(inputdir,d)
  file_list.append(tmp)
 return file_list
  
def createJobs(dir_list, q):
 for d in dir_list:
  files = getFileNames(d)
  for f in files:
   q.put((f,d))
   
  
   
def deployWorkers(file_list, l, num_workers, func):
 jobs = multiprocessing.Queue()
 createJobs(file_list, jobs)
 workers = []
 for i in range(num_workers):
  jobs.put(None)
  tmp = multiprocessing.Process(target=func, args=(jobs,))
  tmp.start()
  workers.append(tmp)
 for worker in workers:
  worker.join()
  
def raw2dngWorker(queue):
 while True:
  job = queue.get()
  if job == None:
   break
  else:
   #enter right dir
   os.chdir(job[DIR])
   #create folder
   folder = job[FILE].split(".")[0]
   os.makedirs(folder)
   #change dir
   os.chdir(folder)
   proc = subprocess.Popen([RAW2DNG, job[FILE]] , 
    stdin=subprocess.PIPE,
    stdout=subprocess.PIPE,
    stderr=subprocess.STDOUT, )
   stdout_value, stderr_value = proc.communicate()
   if not stderr_value:
    print("Successfully processed: %s" %job[FILE])
   else:
    print("ERROR %s " %stderr_value)
   
   
 
   
if __name__ == "__main__":
 inputdir = optionsGet(sys.argv[1:])
 cores = getCPUs()
 l = Lock()
 file_list = getsubdirs(inputdir)

 #start cores = workes
 deployWorkers(file_list, l, cores, raw2dngWorker)

Cheers

Extending NS3 with your module and extra libraries

2012-08-24T00:19:00.001-07:00

I have been working on custom module in ns3 for some time now. Due to many changes in NS3, like bug fixes, extensions, improvements, I decided to move to latest version.

This decision consumed couple of days and several gray hair, due to lack of documentation. The biggest difference for custom module developer is the new waf build system and thus how build works.

Reading http://www.nsnam.org/wiki/index.php/HOWTO_use_ns-3_with_other_libraries it seems very simple to add additional libraries, however the tutorial is not valid for module. Because the suggested build steps will only link the individual files with external libraries and NOT the module. The flags are only passed when compiling indivirual object, the bug (?) is reported.

So here is how I managed to solve it

First, add configuration in your wafscript script

def configure(conf):
     conf.env['xercersc'] = conf.check(mandatory=True, lib='xerces-c', uselib_store='XERCES')
     conf.env['ldl'] = conf.check(mandatory=True, lib='dl', uselib_store='LDL')

make sure you change naming to your libs, note also the name of library, you have to omit "lib", "l" and only go by the name. Having mandatory=True will abort the compilation if waf can not find the library.

Second and last

def build(bld):
    #......
    #...... code omited
    #......
    module.use.append("XERCES")
    module.use.append("LDL")
    #......
    #...... the rest of build code

Delete build directory run

./waf configure && ./waf build

If you still having problems with linking it is very useful to run

./waf --no-task-lines -v

In that case after each build step waf will output runner were the compiler and all passed flags are visible. From there you can continue bug hunt :)

Cheers

TinyOS on Ubuntu 12.04 update error, the fix

2012-08-10T00:50:00.003-07:00

TinyOS released the update to tinysos-tools package. However, if updating from GUI of update manager you will get an error

The action would require the installation of packages from not authenticated sources.

And update will be aborted. To fix it, simply open terminal and run

sudo apt-get update
sudo apt-get upgrade

You will get two warning, proceed by answering Y and you are done!

Install and run NS3 on Amazons EC2

2012-08-07T03:30:00.002-07:00

This is simple step-by step on how to run NS3 on Amazon cloud. If you skip to 3 you can use same script on any other machine using Ubuntu.

This script will install needed packages, download ns3, make standard configuration, build ns-3-dev, and will run test.

Assuming you have Amazon account.

Create Ubuntu 12.04 image
Log in to it
Create file installns3.sh

vi installns3.sh
i

Copy following code into it:

#!/bin/bash
#original timer was published in
#http://www.linuxjournal.com/
#http://tiny.cc/en2niw 
#by Mitch Frazier
function timer()
{
 if [[ $# -eq 0 ]]; then
  echo $(date '+%s')
 else
  local  stime=$1
  etime=$(date '+%s')
         if [[ -z "$stime" ]]; then stime=$etime; fi
         dt=$((etime - stime))
         ds=$((dt % 60))
         dm=$(((dt / 60) % 60))
  dh=$((dt / 3600))
  printf '%d:%02d:%02d' $dh $dm $ds
 fi
}
tmr=$(timer)
INSTALL_DIR="/home/ubuntu/dev/ns3"
#files to install 
APPS="gcc g++ python python-dev mercurial bzr gdb valgrind gsl-bin \
 libgsl0-dev libgsl0ldbl flex bison libfl-dev tcpdump sqlite sqlite3 \
 libsqlite3-dev libxml2 libxml2-dev libgtk2.0-0 libgtk2.0-dev vtun \
 lxc uncrustify doxygen graphviz imagemagick texlive texlive-extra-utils \ 
 texlive-latex-extra python-sphinx dia python-pygraphviz python-kiwi \
 python-pygoocanvas libgoocanvas-dev libboost-signals-dev \ 
 libboost-filesystem-dev openmpi-bin libopenmpi-dev openmpi-doc"
# update repos
RETVAL=$(sudo apt-get update)
if [ $? -ne 0 ]; then
 echo "Unable to update, reason: $?"
 echo $RETVAL
 exit 1
else
 echo "Successfully updated app repostitory"
fi
#function to handle instaliation 
function make_install(){
 if [ $# == 0 ]; then
         echo "must give at least one argument"
  exit 1
 fi
 for app in $*
 do
  #we do not care is packet is installed it is success
  RETVAL=$(sudo apt-get install -y $app 2>&1 >/dev/null)
  if [ $? -ne 0 ]; then
   echo "Failure: can't install $app reson: $?"
          echo $RETVAL
          echo "Skipping package $app" 
  else
   echo "Succesfully installed $app"
  fi

 done
}

#create directory
if [ ! -d "$INSTALL_DIR" ]; then
 echo "Creating $INSTALL_DIR"
 RET=$(mkdir -p $INSTALL_DIR 2>&1 >/dev/null)
 if [ $? -ne 0 ]; then
  echo "Can not create directory $INSTALL_DIR. Reason: $?"
  echo $RET
  exit 1
 else
  echo "Succesfully created directory $INSTALL_DIR"
 fi
fi
#check if port is open

#install apps
make_install $APPS

cd $INSTALL_DIR
echo "Cloning repo"
RET=$(hg clone http://code.nsnam.org/ns-3-allinone)
if [ $? -ne 0 ]; then
 echo "Could not clone ns-3-allinone. REASON: $?"
 echo $RET
 exit 1
else
 echo "Cloning was successfull"
fi
cd "ns-3-allinone"
echo "Downloading ns3-dev"
chmod +x download.py
./download.py -n ns-3-dev
cd "ns-3-dev"
echo "Building standart ns3"
./waf configure --enable-tests
./waf build
echo "Running test"
chmod +x test.py
./test.py
printf 'Elapsed time: %s\n' $(timer $tmr) 
echo "Thank you! Enjoy NS3"

Save file and execute it.

#hit esc
:wq
chmod +x installns3.sh
./installns3.sh
#go grab coffee

It will take a while, I was using medium instance which took: ~34 min.

If you are to lazy to copy paste and bother, find "ubuntu-12.04-ns-3-dev" AMI in community AMI's I have published pre-installed and pre-build version of the image. It will save you time and buck too!

Cheers :)

TOSSIM simulations on Amazon AWS

2012-08-06T09:06:00.000-07:00

There can be many reasons why you want to run your TOSSIM simulations on cloud.

Share same apps and environment with your coworkers.
Give access to students.
Run gi-norm-ous simulation.
etc

Edit: if you for some reason do not want to create your own installation, search publics for ubuntu-12.04-tinyos-tossim, I have published an AMI with preinstalled TinyOS 2.1.1 and svn version ready to use!

What ever reason is here is how-to guide for running both release and svn version of TinyOS simulator TOSSIM, one assumption is that you have AWS account.

Create EBS volume.
Create EC2 instance and start it (it is enough to have ssh port open), type does not matter since you can change it later. The only assumption that you run Ubuntu 12.04 64 bit version.
Log in to the instance
Create and edit source file

sudo vi /etc/apt/sources.list.d/tinyos.list
#in vi paste
deb http://tinyos.stanford.edu/tinyos/dists/ubuntu natty main
#hit esc and 
:wq

Install main release of TinyOS

sudo apt-get update && sudo apt-get install -y --force-yes tinyos-2.1.1
#fix ownership 
sudo chown -R ubuntu:ubuntu /opt/tinyos-2.1.1

Setting up libasound2 (1.0.25-1ubuntu10.1) ...
Setting up libasyncns0 (0.8-4) ...
Setting up libatk1.0-data (2.4.0-0ubuntu1) ...
Setting up libatk1.0-0 (2.4.0-0ubuntu1) ...
Setting up libgtk2.0-common (2.24.10-0ubuntu6) ...
Setting up ttf-dejavu-core (2.33-2ubuntu1) ...
Setting up fontconfig-config (2.8.0-3ubuntu9) ...
Setting up libfontconfig1 (2.8.0-3ubuntu9) ...
Setting up libpixman-1-0 (0.24.4-1) ...
Setting up libxcb-render0 (1.8.1-1) ...
Setting up libxcb-shm0 (1.8.1-1) ...
Setting up libxrender1 (1:0.9.6-2build1) ...
Setting up libcairo2 (1.10.2-6.1ubuntu3) ...
Setting up libavahi-common-data (0.6.30-5ubuntu2) ...
Setting up libavahi-common3 (0.6.30-5ubuntu2) ...
Setting up libavahi-client3 (0.6.30-5ubuntu2) ...
Setting up libcups2 (1.5.3-0ubuntu2) ...
Setting up libjpeg-turbo8 (1.1.90+svn733-0ubuntu4.1) ...
Setting up libjpeg8 (8c-2ubuntu7) ...
Setting up libjasper1 (1.900.1-13) ...
Setting up libtiff4 (3.9.5-2ubuntu1.2) ...
Setting up libgdk-pixbuf2.0-common (2.26.1-1) ...
Setting up libgdk-pixbuf2.0-0 (2.26.1-1) ...
Setting up libthai-data (0.1.16-3) ...
Setting up libdatrie1 (0.2.5-3) ...
Setting up libthai0 (0.1.16-3) ...
Setting up libxft2 (2.2.0-3ubuntu2) ...
Setting up fontconfig (2.8.0-3ubuntu9) ...
Cleaning up old fontconfig caches... done.
Regenerating fonts cache... done.
Setting up libpango1.0-0 (1.30.0-0ubuntu3.1) ...
Setting up libxcomposite1 (1:0.4.3-2build1) ...
Setting up libxfixes3 (1:5.0-4ubuntu4) ...
Setting up libxcursor1 (1:1.1.12-1) ...
Setting up libxdamage1 (1:1.1.3-2build1) ...
Setting up libxi6 (2:1.6.0-0ubuntu2) ...
Setting up libxinerama1 (2:1.1.1-3build1) ...
Setting up libxrandr2 (2:1.3.2-2) ...
Setting up shared-mime-info (1.0-0ubuntu4.1) ...
Setting up libgtk2.0-0 (2.24.10-0ubuntu6) ...
Setting up libnspr4 (4.8.9-1ubuntu2) ...
Setting up libnss3 (3.13.1.with.ckbi.1.88-1ubuntu6) ...
Setting up libnss3-1d (3.13.1.with.ckbi.1.88-1ubuntu6) ...
Setting up tzdata-java (2012b-1) ...
Setting up java-common (0.43ubuntu2) ...
Setting up libgif4 (4.1.6-9ubuntu1) ...
Setting up libjson0 (0.9-1ubuntu1) ...
Setting up libogg0 (1.2.2~dfsg-1ubuntu1) ...
Setting up libflac8 (1.2.1-6) ...
Setting up libvorbis0a (1.3.2-1ubuntu3) ...
Setting up libvorbisenc2 (1.3.2-1ubuntu3) ...
Setting up libsndfile1 (1.0.25-4) ...
Setting up libpulse0 (1:1.1-0ubuntu15.1) ...
Setting up x11-common (1:7.6+12ubuntu1) ...
Setting up libxtst6 (2:1.2.0-4) ...
Setting up libgomp1 (4.6.3-1ubuntu5) ...
Setting up libice6 (2:1.0.7-2build1) ...
Setting up libmpfr4 (3.1.0-3ubuntu2) ...
Setting up libquadmath0 (4.6.3-1ubuntu5) ...
Setting up libsm6 (2:1.2.0-2build1) ...
Setting up libxt6 (1:1.1.1-2build1) ...
Setting up libxmu6 (2:1.1.0-3) ...
Setting up libxpm4 (1:3.5.9-4) ...
Setting up libxaw7 (2:1.0.9-3ubuntu1) ...
Setting up libgd2-noxpm (2.0.36~rc1~dfsg-6ubuntu2) ...
Setting up libmpc2 (0.9-4) ...
Setting up tinyos-base (2.1-20080806) ...
Setting up avr-tinyos-base (2.1-20080806) ...
Setting up avr-binutils-tinyos (2.17-20080812) ...
Setting up avr-gcc-tinyos (4.1.2-20080812) ...
Setting up avr-libc-tinyos (1.4.7-20080812) ...
Setting up avrdude-tinyos (5.4-20080812) ...
Setting up avr-tinyos (2.1-20080806) ...
Setting up avr-optional-tinyos (2.1-20090326) ...
Setting up binutils (2.22-6ubuntu1) ...
Setting up cpp-4.6 (4.6.3-1ubuntu5) ...
Setting up cpp (4:4.6.3-1ubuntu5) ...
Setting up libc-dev-bin (2.15-0ubuntu10) ...
Setting up linux-libc-dev (3.2.0-27.43) ...
Setting up libc6-dev (2.15-0ubuntu10) ...
Setting up nesc (1.3.3-20110821) ...
Setting up tinyos-tools (1.4.0-20100326) ...
Setting up deputy-tinyos (1.1-20080807) ...
Setting up fonts-liberation (1.07.0-2ubuntu0.1) ...
Setting up gcc-4.6 (4.6.3-1ubuntu5) ...
Setting up gcc (4:4.6.3-1ubuntu5) ...
Setting up libcdt4 (2.26.3-10ubuntu1) ...
Setting up libcgraph5 (2.26.3-10ubuntu1) ...
Setting up libgraph4 (2.26.3-10ubuntu1) ...
Setting up libpathplan4 (2.26.3-10ubuntu1) ...
Setting up libgvc5 (2.26.3-10ubuntu1) ...
Setting up libgvpr1 (2.26.3-10ubuntu1) ...
Setting up graphviz (2.26.3-10ubuntu1) ...
Setting up hicolor-icon-theme (0.12-1ubuntu2) ...
Setting up libgtk2.0-bin (2.24.10-0ubuntu6) ...
Setting up xorg-sgml-doctools (1:1.10-1) ...
Setting up x11proto-core-dev (7.0.22-1) ...
Setting up libice-dev (2:1.0.7-2build1) ...
Setting up libpthread-stubs0 (0.3-3) ...
Setting up libpthread-stubs0-dev (0.3-3) ...
Setting up libsm-dev (2:1.2.0-2build1) ...
Setting up libxau-dev (1:1.0.6-4) ...
Setting up libxdmcp-dev (1:1.1.0-4) ...
Setting up x11proto-input-dev (2.1.99.6-1) ...
Setting up x11proto-kb-dev (1.0.5-2) ...
Setting up xtrans-dev (1.2.6-2) ...
Setting up libxcb1-dev (1.8.1-1) ...
Setting up libx11-dev (2:1.4.99.1-0ubuntu2) ...
Setting up libx11-doc (2:1.4.99.1-0ubuntu2) ...
Setting up libxt-dev (1:1.1.1-2build1) ...
Setting up manpages-dev (3.35-0.1ubuntu1) ...
Setting up msp430-binutils-tinyos (2.21.1-20110821) ...
Setting up msp430-gcc-tinyos (4.5.3-20110821) ...
Setting up msp430-libc-tinyos (20110612-20110821) ...
Setting up msp430mcu-tinyos (20110613-20110821) ...
Setting up msp430-tinyos (20110821) ...
Setting up tinyos-required-msp430 (2.1-20090326) ...
Setting up tinyos-required-avr (2.1-20090326) ...
Setting up tinyos-required-all (2.1-20090326) ...
Setting up tinyos-2.1.1 (2.1.1-20100401) ...
Setting up ttf-dejavu-extra (2.33-2ubuntu1) ...
Setting up ttf-liberation (1.07.0-2ubuntu0.1) ...
Setting up icedtea-netx-common (1.2-2ubuntu1.1) ...
Setting up openjdk-6-jre-headless (6b24-1.11.3-1ubuntu0.12.04.1) ...

Install svn release of TinyOS, the directory root is of your chois, I just like keep things organised

sudo apt-get install subversion
mkdir /home/your_user/dev
cd dev
mkdir tiny-os-svn
#checkout the trunk version
svn checkout http://tinyos-main.googlecode.com/svn/trunk/ tinyos-main-read-only

Fix .bashrc

vi .bashrc
i
#paste
TOSROOT_BASE="/opt/tinyos-2.1.1"
TOSROOT_SVN="/home/ubuntu/dev/tiny-os-svn/trunk"
#TOSROOT=$TOSROOT_SVN #swap TOSROOT to change to different version of TinyOS
TOSROOT=$TOSROOT_BASE 
TOSDIR="$TOSROOT/tos"
CLASSPATH=.:$TOSROOT/support/sdk/java:$TOSROOT/support/sdk/java/tinyos.jar:$CLASSPATH
MAKERULES="$TOSROOT/support/make/Makerules"
PYTHONPATH=$PYTHONPATH:$TOSROOT/support/sdk/python
export TOSROOT
export TOSDIR
export CLASSPATH
export MAKERULES
export PYTHONPATH

#reload bash
source /home/ubuntu/.bashrc

Fix JNI

sudo tos-install-jni

Install build tools

sudo apt-get install build-essential python python-dev gdb

Fix python version

vi /opt/tinyos-2.1.1/support/make/sim.extra
i
#change PYTHON_VERSION=2.5 to
PYTHON_VERSION ?= $(shell python --version 2>&1 | sed 's/Python 2\.\([0-9]\)\.[0-9]+\{0,1\}/2.\1/')

Test installiation

cd dev/tiny-os-svn/trunk/apps/Blink
make micaz sim
#should end with 
*** Successfully built micaz TOSSIM library.

Create and run simulation

cd /opt/tinyos-2.1.1/apps/RadioCountToLeds
make micaz sim

import sys
from TOSSIM import *
from RadioCountMsg import *

t = Tossim([])
m = t.mac();
r = t.radio();

t.addChannel("RadioCountToLedsC", sys.stdout)
t.addChannel("LedsC", sys.stdout)

for i in range(0, 2):
  m = t.getNode(i);
  m.bootAtTime((31 + t.ticksPerSecond() / 10) * i + 1)

f = open("topo.txt", "r")
lines = f.readlines()
for line in lines:
  s = line.split()
  if (len(s) > 0):
    if (s[0] == "gain"):
      r.add(int(s[1]), int(s[2]), float(s[3]))

noise = open("/opt/tinyos-2.1.1/tos/lib/tossim/noise/meyer-heavy.txt", "r")
lines = noise.readlines()
for line in lines:
  st = line.strip()
  if (st != ""):
    val = int(st)
    for i in range(0, 2):
      t.getNode(i).addNoiseTraceReading(val)

for i in range(0, 60):
  t.runNextEvent()

msg = RadioCountMsg()
msg.set_counter(7)
pkt = t.newPacket()
pkt.setData(msg.data)
pkt.setType(msg.get_amType())
pkt.setDestination(0)

print "Delivering %s to 0 at %s" %(str(msg),str(t.time() + 3))
pkt.deliver(0, t.time() + 3)


for i in range(0, 20):
  t.runNextEvent()

python packet.py

Lots of pretty output. Done!

For more info on simulations see http://docs.tinyos.net/tinywiki/index.php/TOSSIM.

Remember to stop your instance when you done!

Happy WSN simulations!

Wireshark on AWS EC2

2012-08-06T05:44:00.000-07:00

Updated on 28th Mars 2014, thanks Imran Hayder for suggestions! In some cases you may need to run wireshark on remote machine, especially if you want to in depth to understand what is going on.
Without any further blah blah, amusing you are using Ubuntu here is how to:

Create EBS, make sure to un-check "delete on termination"
Create and start Ubuntu 12.04 EC2
Log in to it
Check disk name

sudo fdisk -l

Format disk

sudo mkfs -t ext4 /device/path

Create mount directory

sudo mkdir /home/data-storage

Mount disk

sudo mount /dev/DEVICE /home/data-storage

Make wireshark folder

sudo mkdir /home/data-storage/wireshark
#important change ownership!
sudo chown root:ubuntu /home/data-storage/wireshark
#allow group to read
sudo chmod -R 774 wireshark

Install wireshark

sudo apt-get install wireshark tshark

Run example

sudo su
cd /home/data-storage/wireshark
#tshark will capture eth0 for 10 seconds and save file to my.pcap
tshark -i eth0 -a duration:10 -w my.pcap

Copy file to local

#from your local machine, note intentional line brake due to layout of the blog.
cd wireshark
scp -i your.pem 
   ubuntu@your_ec2_dns.compute.amazonaws.com:/path-to/my.pcap .

Read the comments bellow for trouble shooting. Note, it is not advisable to copy paste commands since you may bet the html representation, and not what you should type in the terminal.

Ubuntu 12.04 on Lenovo x220

2012-04-14T04:12:00.000-07:00

Ubuntu 12.04

There have been a lot of progress in Ubuntu! I got new Lenovo's Thinkpad X220 and immediately installed Ubuntu 12.04 64 bits. A short summary everything worked out of box, however, I did not tested fingerprint reader nor WiMax or mobile Broadband. The rest is smooth flow even with beta of Ubuntu. Boot time is really awesome, bringing the login screen 7s!

Thanks Canonical for Ubuntu! :)

Export specific sub holder from svn on CentOS

2012-04-04T04:43:00.001-07:00

Some time project grows in your repository and you may want to move it to separate repository or you want to share you work and open up repository from tight security. The reasons can be many but the process is the same.

(assuming root access, LATEST refers to the latest available version, procedure was tested on 0.5.8)

Install needed components

yum install subversion-devel httpd-devel
wget http://prdownloads.sourceforge.net/rsvndump/rsvndump-LATEST.tar.gz
tar xvfz rsvndump-LATEST.tar.gz
cd rsvndump-LATEST
./configure
make
sudo make install

Exporting your project
With rsvndump you don't need to use filters to extract your subfolder within svn, it will do it for you just in one command

rsvndump -v -u username -p psw  https://ulr/svn/repos/to/subfolder > subfolder

Create svn repository

svnadmin create /path/to/svn/project

Allow Apache to read it

chown -R apache:apache symphony

Import it dumped file

svnadmin load --force-uuid /path/to/svn/project < subfolder

Enable web access in Apache

LoadModule dav_svn_module     modules/mod_dav_svn.so
LoadModule authz_svn_module   modules/mod_authz_svn.so
<location svn="">
        SSLRequireSSL
        DAV svn
        SVNParentPath /path/to/svn
        AuthzSVNAccessFile /path/to/svn-acl-conf
        AuthType Basic
        AuthName "Subversion repos"
        AuthUserFile /path/to/svn-auth-conf
        Require valid-user
        SVNIndexXSLT "https://url/svnindex.xsl"
</location>

Enable access in svn-acl-conf

[project:/]
@project_users=rw


[repos:/]
@users=rw

Restart Apache

/etc/init.d/httpd restart

More info
The assumption in example is that all your repositories are in the /path/to/svn/ Then you can use only one conf file for Apache and control access to repositories :) and you are done!

Trac on Centos

2012-03-08T23:36:00.000-08:00

cat httpd/conf.d/trac-project.conf

 #configuration 
#permament redirect for https
Redirect permanent http://path https://path
WSGIScriptAlias /project /var/www/html/project/cgi-bin/trac.wsgi
#with multiple project you need only alias it once

Alias /trac/chrome/common /usr/share/trac/htdocs/common

Alias /trac/chrome/site /usr/share/trac/htdocs/site

<Directory /var/www/html/project/cgi-bin/>
WSGIApplicationGroup %{GLOBAL}
Order deny,allow
Allow from all
</Directory>

<Location '/symphony'>
        SSLRequireSSL
        AuthType Basic
        AuthName "Trac symphony"
        AuthUserFile /usr/share/trac/projects/project/conf/trac.htpasswd
        Require valid-user
</Location>

Back up trac
Since 0.13, trac back up include MySQL db back up, always good to have before upgrading.

 $ trac-admin /path/to/projenv hotcopy /path/to/backupdir


More info: http://trac.edgewall.org/wiki/0.13/TracBackup

Code on blogger

2012-03-07T02:04:00.000-08:00

Quite nice formating for posting code on blogger

</code></pre>
<pre style="font-family: Andale Mono, Lucida Console,

Monaco, fixed, monospace; color: #000000;

background-color: #eee;font-size: 12px;border: 1px dashed #999999;

line-height: 14px;padding: 5px; overflow: auto; width: 100%"><code>

</code></pre>

Boost threads

2011-11-23T06:44:00.000-08:00

Good information on boost thread:

General intro:
http://drdobbs.com/cpp/184401518

Using in class:

http://antonym.org/2009/05/threading-with-boost---part-i-creating-threads.html

Official API
http://www.boost.org/doc/libs/1_48_0/doc/html/thread/thread_management.html