Lint Roller? I Barely Know Her
RMH
tumblr dot com
No title available
ojovivo
No title available
art blog(derogatory)
Fai_Ryy
Origami Around
★
Today's Document
KIROKAZE
occasionally subtle
taylor price
2025 on Tumblr: Trends That Defined the Year
Color Me Curious
Game Changer & Make Some Noise
No title available
One Nice Bug Per Day
No title available
seen from Norway
seen from United Kingdom
seen from Italy
seen from Iraq
seen from Türkiye
seen from Indonesia
seen from China
seen from Mexico
seen from United States
seen from Brazil
seen from United States
seen from Jordan
seen from United States
seen from Chile
seen from Argentina
seen from Germany
seen from India
seen from Brazil
seen from United States
seen from Iraq
@pedrocarrico
How to install latest node.js in your raspberry pi...
I remember the old days when you were supposed to download the node.js source into your raspberry pi and then compile it, after almost 1 hour you were ready to start using node on your pi.
But now thanks to the node-arm project you can get it up and running in about 5 minutes.
Just:
wget http://node-arm.herokuapp.com/node_latest_armhf.deb sudo dpkg -i node_latest_armhf.deb
Easy...simple...quick...
History in your irb console...
To add a bash like history to your irb console (and also your rails console).
Create/edit your irbrc:
vi ~/.irbrc
And then add this:
require 'irb/ext/save-history' IRB.conf[:SAVE_HISTORY] = 100 IRB.conf[:HISTORY_FILE] = "#{ENV['HOME']}/.irb-history"
After that fire up irb and start browsing your history using the up/down keys.
Vagrant acting up...
If for some reason your getting something like this after you do a "vagrant up":
==> default: Machine booted and ready! ==> default: Checking if machine supports zones: no JSON::ParserError: 757: unexpected token at 'bash: line 2: syntax error near unexpected token `newline'
Odds are that your vagrant plugins json database somehow got corrupted (~/.vagrant.d/plugins.json), just reinstall every vagrant plugin you have and you should be fine.
Happy vms!
Strip down raspbian to a bare minimum...
With the lack of a proper raspbian server edition I usually try to strip my raspbian installation to a minimum, so, on a clean raspbian install and after doing the initial config I usually run this:
apt-get update apt-get purge xserver* -y apt-get purge ^x11 -y apt-get purge ^libx -y apt-get purge ^lx -y apt-get purge samba* -y apt-get purge supercollider* -y apt-get purge netsurf* -y apt-get purge omxplayer -y apt-get purge ^epiphany-browser -y apt-get purge sonic-pi -y apt-get purge wolfram-engine -y apt-get purge scratch -y apt-get purge penguinspuzzle -y apt-get purge desktop-base -y apt-get purge lightdm -y apt-get purge obconf -y apt-get purge openbox -y apt-get purge raspberrypi-artwork -y apt-get autoremove -y apt-get upgrade -y apt-get clean
After this your raspbian installation will be stripped of all the X server's packages and a bunch of others that are not needed when all you want to do is a "server install".
Be sure to comment and suggest more packages you always strip from your raspbian installation
How to setup your wireless in arch linux
To setup the wireless on your raspberry pi on arch linux, just:
cd /etc/netctl #make a new wireless profile install -m640 examples/wireless-wpa wireless-at-home vi wireless-at-home <change config> netctl start wireless-at-home # to start on boot netctl enable wireless-at-home # check your IP and if everything went OK ifconfig
Easy, quick and simple...
Upgrading major versions of Rails sucks. I was there from 0.9 to 1.0, all the way to the famed 2 to 3 release, and they were all painful. I have good news though: Rails 4 is right around the corner, and it’s a much cleaner upgrade. While you are getting your app ready for Rails 4 are you...
Testing email layouts in rails...
While developing I sometimes feel the need to test the layouts of my emails just to get a feel how they are rendered in Gmail or in some other mail client.
For that I usually use the mailcatcher gem but sometimes I really want to send the email and check out the layout and how the various email clients behave with it.
So first I configure my actionmailer to send emails through gmail by adding this to my development.rb:
config.action_mailer.deconfig.action_mailer.delivery_method = :smtp config.action_mailer.smtp_settings = { :address => "smtp.gmail.com", :port => 587, :domain => 'gmail' :user_name => '[email protected]', :password => 'password', :authentication => 'plain', :enable_starttls_auto => true }
And afterwards I usually I set up an action mailer interceptor like this:
class OverrideMailRecipient def self.delivering_email(mail) mail.to = "[email protected]" end end ActionMailer::Base.register_interceptor(OverrideMailRecipient)
This ensures that every email sent from my application will have my email address in the to address thus making me receive every email message.
With this I can preview my emails in every email client and eliminate some potential layout bugs that might show up.
Another trick I've used recently is using an interceptor to divert some email through another email server, for example, if you want to send some bulk email through the rails console and send it through another email server:
class OverrideMailSettings def self.delivering_email(mail) mail.delivery_method.settings.merge!({ :user_name => "username", :password => "password", :domain => "host", :address => "smtp.server.net", :port => 587, :authentication => :plain, :enable_starttls_auto => true }) end end ActionMailer::Base.register_interceptor(OverrideMailSettings)
Upon doing this you can just call your mailers like:
UserMailer.welcome_email(@user).deliver
And it will send the email through the other email server.
Why should we cook a raspberry pi and how...
Lately I've been having some SD card corruption on some of my raspberry pis and so I thought I could automate the process of installing everything onto it using chef.
I'm really not certain of what causes the SD card corruption on the raspberry pi, some say that overclocking might cause it (see here) while others say that it's most probable to occur if it's not shutdown properly (see here).
Thing is that I need a way to get my little raspberry pi up and running as soon as possible and here is where chef comes to the rescue.
Soon I started to use chef and knife solo to prepare my pis but soon preparing a new raspbian and installing/preparing chef became too cumbersome due to the time it took so I had the idea of preparing a minimal raspian image with chef pre-installed and afterwards I just had to do all the cooking.
So how do I install chef onto my raspberry pis:
Flash a clean raspbian onto an SD card: sudo dd bs=1m if=./2012-10-28-wheezy-raspbian.img of=/dev/<raw device> #(if you're unsure what raw device you should check this out)
Boot it into the raspberry pi but don't do "sudo raspi-config" or anything else.
Get your ssh public keys onto the root user authorized_keys
On another computer: knife solo prepare root@raspberrypi
After the chef installation shutdown raspbian, and insert the SD card onto another computer.
Make an image of the raspbian with chef installed: sudo dd bs=1m if=/dev/<raw device> of=./2012-10-28-wheezy-raspbian-chef.img
Now you have an clean image of raspbian with chef install that can be flashed onto an SD card with: sudo dd bs=1m if=./2012-10-28-wheezy-raspbian-chef.img of=/dev/<raw device>
Flash it onto an SD card, boot the raspberry pi and cook to your heart's content :-)
Lately I've been working on a set of chef recipes that I keep on github. I've uploaded my whole "kitchen" of recipes but in the future as the recipes will start to evolve and getting bigger and bigger I'll try and make individual repositories for each of them and/or incorporate them in the opscode cookbooks.
Download my raspbian with chef image here (raspbian version 2012-10-28 and chef 11) and there are some instructions here for the installation procedure.
Configuring and installing the Ralink RT5370 on raspbian...
Installing the Ralink RT5370 on raspbian is pretty straightforward:
Install the required packages:
sudo apt-get install wpasupplicant usbutils wireless-tools
Edit /etc/wpa_supplicant/wpa_supplicant.conf
sudo vi /etc/wpa_supplicant/wpa_supplicant.conf
Copy and alter the following template:
ctrl_interface=/var/run/wpa_supplicant ctrl_interface_group=0 update_config=1 network={ ssid="YOUR SSID here" scan_ssid=1 mode=0 proto=WPA2 auth_alg=OPEN pairwise=CCMP group=CCMP key_mgmt=WPA-PSK psk="YOUR PASSWORD HERE" id_str="YOUR COMPUTER ID HERE" priority=1 }
Edit /etc/network/interfaces:
sudo vi /etc/network/interfaces
And append the following lines:
auto wlan0 iface wlan0 inet dhcp wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf
Bring the interface up (or restart it if it was up but not fully configured):
sudo ifdown wlan0 # ensure it's down sudo ifup wlan0
PROFIT!
If you're having trouble be sure to run "dmesg" or "iwlist wlan0 scan" to check the status of your adapter or network.
PS: if you're getting "wpa_supplicant: /sbin/wpa_supplicant daemon failed to start" when you try to bring the interface up then be sure to check your wpa_supplicant configuration file format and options, if you get one option incorrect or mess up the format the daemon will fail to start. Check the manual page for wpa_supplicant for a list of options and settings.
Nginx and websockets madness...
Nginx does not yet support the websockets protocol (it is still planned on for version 1.3.x on their roadmap) but there's a solution with Weibin Yao's third party module (https://github.com/yaoweibin/nginx_tcp_proxy_module).
Installing it is as easy as (following the instructions on the github repo, and tweak them a bit to download latest nginx):
$ wget 'http://nginx.org/download/nginx-1.3.7.tar.gz' $ tar -xzvf nginx-1.3.7.tar.gz $ git clone https://github.com/yaoweibin/nginx_tcp_proxy_module.git $ cd nginx-1.3.7/ $ patch -p1 < ../nginx_tcp_proxy_module/tcp.patch $ ./configure --add-module=/path/to/nginx_tcp_proxy_module --prefix=/your_path_to/nginx $ make $ make install
But configuring it might be a little bit tricky as I've found many misleading configuration examples and I had some websocket timeouts after 75s or so (that's the default keep alive for nginx).
After tinkering a little bit and browsing some of the project's issues on github I managed to get a working configuration for my application:
tcp { upstream websockets { server 127.0.0.1:4567; check interval=3000 rise=2 fall=5 timeout=1000 type=http; check_http_expect_alive http_2xx http_3xx; } server { server_name your.server.name; listen 80; timeout 1d; websocket_connect_timeout 1d; websocket_read_timeout 1d; websocket_send_timeout 1d; so_keepalive on; tcp_nodelay on; websocket_buffer 1k; websocket_pass websockets; } }
However there are some limitations to the nginx_tcp_proxy_module as you can’t use the same listening port with the HTTP module and so you can't use some of the configuration options of the http module (such as locations).
Thoughts questions? Please comment :-)
Introducing Pimon...
Pimon is a pet project of mine that was born from my need to lower my electricity bill.
Instead of having my pimped up Intel Core I7 supercharged and overclocked for gaming supercomputer that spends close to 400W I chose the raspberry pi to handle all my torrent downloads and home NAS needs for around 2W (plus around 20W for the ocasional external disk).
With Pimon I can monitor the raspberry pi statistics (cpu. memory, swap and disk) in a web page and check it's health from anywhere.
Pimon is coded in Ruby and uses Sinatra and redis. I tried it to be as lightweight as possible and so far it's performing rather well. There's still a lot to be done and it's in a very alpha stage.
You can see it in action running on my raspberrypi at pec.no-ip.biz and you can see the source and contribute at https://github.com/pedrocarrico/pimon , pull requests and code reviews are more than welcome.
Compiling and installing ruby on the raspberry pi using rbenv...
Installing ruby on the raspberry pi might be tricky due to the memory constraints and other some quirks.
So here's how I did it on the latest raspbian "wheezy":
First increase your swapfile: sudo dd if=/dev/zero of=swapfile bs=1M count=256 sudo swapoff /var/swap sudo mkswap swapfile sudo swapon swapfile
Change your memory split to CPU 224/32 VRAM using: sudo raspi-config
Update your system so you don't have the dreaded bundler error: sudo apt-get install zlib1g-dev openssl libopenssl-ruby1.9.1 libssl-dev libruby1.9.1 libreadline-dev git
After that increase your /tmp size (raspbian comes with 100MB default /tmp size and with tmpfs): sudo vi /etc/default/tmpfs And change TMP_SIZE to 1G (or whatever you want, be sure to add at least 200MB to compile the ruby binaries).
Install rbenv/ruby-build using the instructions in here and in here.
And then: rbenv install 1.9.3-p194
After about 1h45m you can enjoy the latest ruby on your raspberry pi. Enjoy!
PS: If you're getting the following error: compiling load.c gcc: internal compiler error: Killed (program cc1) Please submit a full bug report, with preprocessed source if appropriate. See <file:///usr/share/doc/gcc-4.6/README.Bugs> for instructions. ... Then you should increase your swapfile and be sure to change the memory split on your raspberry pi to (CPU 224/32 VRAM).
Setting up HAProxy in your development environment...
For some years now I've been wanting to experiment with HAProxy and never got the chance, so after having to debug a strange behaviour in one application I started out by setting up my local (development) environment with a similar configuration.
Setting up a HAProxy in your development machine might come in handy if you want to test your application behaviour with a load-balancer. It might also be good to test different balancer options and check out if your application is working properly.
First of all install HAProxy in your development environment (in my case a Macbook running Lion), for this you'll have to: 1 - Get latest (stable) HAProxy: $ curl -O http://haproxy.1wt.eu/download/1.4/src/haproxy-1.4.21.tar.gz $ tar xzpvf haproxy-1.4.21.tar.gz $ cd haproxy-1.4.21 2 - Compile: $ make TARGET=generic ARCH=x86_64 USE_PCRE=1 3 - Install: $ make install DESTDIR='~/opt/haproxy' PREFIX=''
After installing HAProxy get the following sample configuration and save it to a file named haproxy.cfg .
Configure your application to have two instances that run on port 8081 and 8082. If you're running a rails application open up two terminals and run: $ rails server -p 8081 $ rails server -p 8082
Run HAProxy with: $ ~/opt/haproxy/sbin/haproxy -f haproxy.conf
Point your browser to http://localhost:8080 and check out the logs on each of your application instances, you'll notice you're being balanced between each instance.
A good technique to try out different balancer options is to have more than one listen block on your local HAProxy configuration for example: listen roundrobin_balancer localhost:8080 mode http option httplog balance roundrobin server app1 localhost:8081 server app2 localhost:8082 listen source_balancer localhost:8090 mode http option httplog balance source server app1 localhost:8081 server app2 localhost:8082 If you set it this way you may test your application balanced on different ports using different balancer strategies and check out if it behaves properly. Testing the balancer strategies with a coworker on another computer interacting with your application will make the test more accurate.
PS: If compiling HAProxy gives you an error about missing the PCRE lib then you may install it via brew with: $ brew install pcre
Rbenv quirks in Linux Mint 13...
If you're getting: $ gem install bundler ERROR: Loading command: install (LoadError) cannot load such file -- zlib ERROR: While executing gem ... (NameError)
After a clean installation of Linux Mint 13 "Maya" and after the rbenv ruby installation, then be sure to: $ rm -rf ~/.rbenv/versions/YOUR_RUBY_VERSION_HERE/ $ sudo apt-get install zlib1g-dev openssl libopenssl-ruby1.9.1 libssl-dev libruby1.9.1 libreadline-dev $ rbenv install YOUR_RUBY_VERSION_HERE $ rbenv rehash $ gem install bundler After this gem works flawlessly and you can go back to coding.
PS: I read reports that this may also occur in the latest Ubuntu distributions.
Sinatra file upload progress demo...
As promised here's a sample project using sinatra that uses the configuration on my last post.
Some quirks of putting it all together:
Jquery upload progress will only work if the progress resource answers in JSON so you'll have to use the "upload_progress_json_output" directive on nginx to make it work (as described in here).
In the nginx documentation it says that the track_uploads directive must be the last directive of the location and that it must be in a proxy_pass location. The passenger_enabled directive for nginx works as a proxy_pass so you may declare it before to make the track_uploads directive work.
The load paths in the rackup file for the Sinatra application are b0rked when running it with passenger so I had to put a "require './sinatra-upload-progress'" to make it work.
If you have any problems putting it all together be sure to comment here.
Github pull requests are more than welcome :-)
Installing nginx with passenger and HttpUploadProgressModule...
Lately I've been tinkering with nginx, passenger and file upload progress and since some of my fellow programmers said they were having problems putting it all together I'll try and put here some of the steps I've taken to get it all working from scratch.
Get latest nginx (version 1.2.0):
$ curl -C - -O http://nginx.org/download/nginx-1.2.0.tar.gz $ tar xzpvf nginx-1.2.0.tar.gz
Get Nginx HttpUploadProgressModule (version 0.9.0):
curl -L -o nginx-upload-progress.tar.gz https://github.com/masterzen/nginx-upload-progress-module/tarball/v0.9.0 tar xzpvf nginx-upload-progress.tar.gz
Install passenger:
$ #If you're using rbenv don't forget to "rbenv rehash" afterwards $ gem install passenger
Compile Nginx:
$ cd nginx-1.2.0 $ export PREFIX= $ ./configure --add-module=../masterzen-nginx-upload-progress-module-82b35fc --add-module=`passenger-config --root`/ext/nginx --prefix=$PREFIX $ make $ make install
Check if everything's OK with:
$ sudo /sbin/nginx
Now go to your browser and go to localhost:80 and you'll shall see "Welcome to nginx!"
Here's a sample project for you to test this configuration.
Check the following sites for more on this:
http://wiki.nginx.org/HttpUploadProgressModule
https://github.com/masterzen/nginx-upload-progress-module
https://github.com/drogus/jquery-upload-progress
PS: If you get this error while running ./configure:
"./configure: error: the HTTP rewrite module requires the PCRE library.You can either disable the module by using --without-http_rewrite_module option, or install the PCRE library into the system, or build the PCRE library statically from the source with nginx by using --with-pcre= option."
Then you just have to get libpcre and if you're using brew on Mac OSX then:
$ brew install pcre
PS2: If you're getting something like:
... objs/ngx_modules.o \ /Users/pecarrico/.rbenv/versions/1.9.3-p125/lib/ruby/gems/1.9.1/gems/passenger-3.0.12/ext/nginx/../common/libpassenger_common.a /Users/pecarrico/.rbenv/versions/1.9.3-p125/lib/ruby/gems/1.9.1/gems/passenger-3.0.12/ext/nginx/../common/libboost_oxt.a -lstdc++ -lpthread -lm -lpcre -lcrypto -lcrypto -lz Undefined symbols for architecture x86_64: "_pcre_free_study", referenced from: _ngx_pcre_free_studies in ngx_regex.o ld: symbol(s) not found for architecture x86_64 collect2: ld returned 1 exit status make[1]: *** [objs/nginx] Error 1 make: *** [build] Error 2
Add --with-ld-opt="-L /usr/local/lib" to the ./configure options (check http://trac.nginx.org/nginx/ticket/94 for more information).
Stuff about BDD...
I recently read "The RSpec book" and I must say that it's one of the best technical books I've read in a long time.
It starts out with light introduction to BDD and then it keeps the reader engaged by presenting clear examples on how to use RSpec and Cucumber.
Throughout the examples they explain the philosophy of BDD and great concepts like "Red/Green/Refactor" that in my opinion should be engraved in every programmer's brain.
The chapters on Rspec Expectations and Mocks are so good that I've read them twice and the examples they provide really show why we should use mocks and stubs to isolate certain components of our code.
I really recommend this book to every programmer and even if you're not programming in Ruby you'll get a really nice overview of every aspect of BDD and the concepts behind code testing.
Clear, precise...recommended!
On my way to "The Cucumber Book"!
