Tapestry

h1. Installing the Tapestry application

The recommended ruby stack for Tapestry is Ruby Enterprise Edition (1.8.7), installed via rvm.

Install rvm and REE:

<pre>

sudo -i

# curl -L https://get.rvm.io | sudo bash -s stable

# exit

sudo -i

# rvm install ree

# exit

</pre>

Start a new login shell.  Make sure RVM is working and @ruby -v@ identifies itself as REE 1.8.7.

 rvm use ree

ruby -v

→ *@ruby 1.8.7 (2012-02-08 MBARI 8/0x6770 on patchlevel 358) [x86_64-linux], MBARI 0x6770, Ruby Enterprise Edition 2012.02@*

Install rails.

<pre>

sudo -i

# rvm use ree

# gem install rails

# gem install passenger

# passenger-install-apache2-module

</pre>

Add three lines (provided by passenger-install-apache2-module) and two more lines (see below) to @/etc/apache2/conf.d/passenger.conf@.

<pre>

LoadModule passenger_module /usr/local/rvm/gems/ree-1.8.7-2012.02/gems/passenger-3.0.17/ext/apache2/mod_passenger.so

PassengerRoot /usr/local/rvm/gems/ree-1.8.7-2012.02/gems/passenger-3.0.17

PassengerRuby /usr/local/rvm/wrappers/ree-1.8.7-2012.02/ruby

PassengerUser YOUR_USERNAME_HERE

RackEnv "development"

</pre>

Install some prerequisites for gems.

 sudo apt-get install libxslt-dev libxml2-dev

Get a copy of the source tree -- e.g., in /var/www/tapestry

<pre>

cd /var/www

sudo chmod a+w,+t .

git clone git://git.clinicalfuture.com/tapestry.git

sudo chmod go-w,-t .

cd tapestry && git submodule init && git submodule update

</pre>

Then execute:

 bundle install

Point a virtualhost to the @public@ dir in your git repo

<pre>

<VirtualHost *:80>

   ServerName tapestry.example.org

   DocumentRoot /var/www/tapestry/public

   <Directory /var/www/tapestry/public>

         AllowOverride all

         Options -MultiViews

   </Directory>

   Alias /warehouse /warehouse

   <Directory /warehouse>

         # This is important; downloads from /warehouse take a long time

         # and we do not want to tie up passenger processes with those.

         PassengerEnabled off

         Options None

         AllowOverride None

         Order allow,deny

         allow from all

   </Directory>

</VirtualHost>

</pre>

Copy @config/database.example@ to @config/database.yml@ and edit to suit. Make sure to select mysql as the database server.  Generate a password using @pwgen 12 1@ or @head -c12345 /dev/urandom|md5sum|head -c12;echo@.  For example:

 development:

  adapter: mysql

  server: localhost

  database: mypg

  username: mypg

  password: c16fbe415d29

  timeout: 5000

Copy @config/environments/development.rb.sample@ to @config/environments/development.rb@ and edit to suit.  In particular the ROOT_URL (you may skip the "http://" prefix):

 ROOT_URL = 'your.vhost.name'

Generate a secret token with something like

* @ruby -e 'puts rand(2**256).to_s(36)'@

* @head -c1234567 /dev/urandom | sha256sum@

* @pwgen 64 1@

Put the secret token in @config/initializers/secret_token.rb@ like this

 Tapestry::Application.configure do

 config.secret_token = "40ih46sqyihsiupgwce5u2oirflnor3atgmvzaqeqng42dwa0o"

end

Set up the database:

 rake db:setup

rake db:schema:load

Set up data directory (replace USERNAME with the user who owns the GIT directory this code is in, and ROOT_URL matching the one above):

 sudo mkdir -p /data/ROOT_URL

sudo chown -R USERNAME:USERNAME /data/ROOT_URL

After you do a git pull, you'll probably want to do

  rake db:migrate

to bring your database up to the latest release.

If the Gemfile has changed, you will also have to run

  bundle install

Set up cron jobs -- for example, @/etc/cron.d/tapestry-get-blog-feed@

 PATH=/usr/local/bin:/usr/bin:/bin

0 * * * * www-data cd /var/www/tapestry/current/script && ./get-blog-feed.rb development http://blog.personalgenomes.org/feed/

To enable large uploads with warehouse storage:

* @sudo apt-get install runit fuse-utils@

* @sudo mkdir -p /etc/service/whmount/log/main@

* Set up supervised whmount service:

<pre>

sudo tee /etc/service/whmount/run <<EOF >/dev/null

#!/bin/sh

modprobe fuse

mkdir -p /warehouse

chown www-data:www-data /warehouse

sudo -u www-data fusermount -u /warehouse

exec sudo -u www-data whmount --no-detach / /warehouse

EOF

sudo tee /etc/service/whmount/log/run <<EOF >/dev/null

#!/bin/sh

exec svlogd -tt main

EOF

</pre>

* Make sure config/environments/development.rb matches your Apache and filesystem mount points:

<pre>

  WAREHOUSE_WEB_ROOT = "/warehouse" # Apache alias pointing to whmount target

  WAREHOUSE_FS_ROOT = "/warehouse" # whmount target

</pre>