ec2onrails 0.9.6 → 0.9.7

data/History.txt

@@ -1,3 +1,20 @@
+== 0.9.7 2008-03-03
+
+* ami-5c0aef35 (32-bit) and ami-540aef3d (64-bit)
+* memcached installed, and capistrano tasks now support a new role named "memcache". Memcached won't run on hosts that aren't in the "memcache" role.
+* New Capistrano task ec2onrails:get_public_key_from_server to copy the public key file from the server before first use.
+* cap ec2onrails:server:install_gems can now install a specific version of a gem instead of the latest (use "gemname -v version" instead of "gemname" in deploy.rb)
+* Now automatically runs hourly, daily, weekly and monthly scripts in current application's script directory, they run as "app" user, with $RAILS_ENV set.
+* Support for different named rails environments (e.g. "production", "staging1", "staging2"), the current environment name is defined in /etc/mongrel_cluster/app.yml
+* s3.yml now supports multiple sections for different rails environment names
+* cap ec2onrails:server:upgrade_gems now does "gem update --system"
+* Added admin user to adm group to allow read access to syslog without using sudo
+* Added "sudoers" group, if you add named administrator accounts, add them to this group
+* Set PATH and RUBYLIB in /etc/environment, removed and disabled $HOME/.ssh/environment because it's no longer necessary
+* Fixed bug #16423: "hostname defined in deploy.rb shouldn't have to match what the instance thinks it's hostname is"
+* Removed rfacebook gem
+
+
 == 0.9.6 2007-12-21
 
 * ami-e620c58f (32-bit) and ami-f920c590 (64-bit)
@@ -5,12 +22,14 @@
 * Fixed server:install_gems task, issue #16006
 * Fixed db:set_root_password task for Capistrano 2.1, issue #15256
 * Fixed apache config to remove warning on restart, issue #16213
+* Now includes Rails 2.0.1 as well as 1.2.6
 * Removed dependency on tar for server:deploy_files task, now using all-ruby tar lib
 * Rotate logs (including rails production.log) and archive to S3 daily
 * Symlink all log files in /mnt/log from /var/log so that programs (e.g. logwatch) that expect them to be in /var/log still work
 * New script to archive any file to s3
 * New command-line flags on s3 archive & restore scripts
 
+
 == 0.9.5 2007-10-28
 
 * ami-86c025ef (32-bit) and ami-99c025f0 (64-bit)
@@ -18,11 +37,12 @@
 * Fixed rebundling.
 * Generates new ssh keys on first startup
 
+
 == 0.9.4 2007-10-25
 
 * ami-ecc12485 (32-bit) and ami-efc12486 (64-bit)
 * This is another test release, still not intended to be production-ready until it's tested more thoroughly. Look for version 1.0.
-* _The rebundling doesn't seem to be working._ I'll fix that in a new version. Details "here":http://rubyforge.org/tracker/?func=detail&aid=15047&group_id=4552&atid=17558
+* _The rebundling doesn't seem to be working._ I'll fix that in a new version. Details at http://rubyforge.org/tracker/?func=detail&aid=15047&group_id=4552&atid=17558
 * Ubuntu 7.10/Gutsy
 * Experimental multiple instance support.
 ** *Mongrel & MySQL now listen on all ports.* Be sure your firewall settings are correct, and set a MySQL root password.
@@ -35,6 +55,7 @@
 * syslog now logs to /mnt/log.
 * Build now cleans apt cache, image is now 1/2 the size.
 
+
 == 0.9.3 2007-10-06
 
 * ami-a3f91cca
@@ -45,18 +66,21 @@
 * Fixed missing /etc/mongrel_cluster
 * Installed more ruby standard lib packages
 
+
 == 0.9.2 2007-09-13
 
 * ami-0cf61365
 * This is another test release, still not intended to be production-ready until it's tested more thoroughly. Look for version 1.0.
 * Fixed missing kernel modules
 
+
 == 0.9.1 2007-09-11
 
 * ami-99f712f0
 * This is another test release, still not intended to be production-ready until it's tested more thoroughly. Look for version 1.0.
 * Fixed permissions on /var/lib/dhcp3
 
+
 == 0.9.0 2007-09-07
 
 * ami-bdf411d4
@@ -65,6 +89,7 @@
 * Removed "http://packages.ubuntu.com/feisty/libs/sun-java6-jre":Java and it's massive list of dependencies because it was only needed for the EC2 API tools. This saved a huge amount of space.
 * Removed ImageMagick, RMagick and hpricot due to popular demand. There will soon be cap2 tasks to customize which packages and gems are installed.
 
+
 == 0.8.1 2007-07-29
 
 * ami-b79a7fde
@@ -74,6 +99,7 @@
 * Added empty /etc/apache2/sites-available/app.custom for custom Apache configuration in a separate file (so it can be deployed by rsync more easily).
 * Added ImageMagick, RMagick and hpricot due to popular demand.
 
+
 == 0.8.0 2007-06-20
 
 * ami-4e907527

data/lib/ec2onrails/recipes.rb

@@ -87,6 +87,34 @@ Capistrano::Configuration.instance.load do
       puts "64-bit server image for EC2 on Rails #{ec2onrails_version}: #{image_id_64_bit}"
     end
     
+    desc <<-DESC
+      Copies the public key from the server using the external "ssh"
+      command because Net::SSH, which is used by Capistrano, needs it.
+      This will only work if you have an ssh command in the path.
+      If Capistrano can successfully connect to your EC2 instance you
+      don't need to do this. It will copy from the first server in the
+      :app role, this can be overridden by specifying the HOST 
+      environment variable
+    DESC
+    task :get_public_key_from_server do
+      host = find_servers_for_task(current_task).first.host
+      privkey = ssh_options[:keys][0]
+      pubkey = "#{privkey}.pub"
+      msg = <<-MSG
+      Your first key in ssh_options[:keys] is #{privkey}, presumably that's 
+      your EC2 private key. The public key will be copied from the server 
+      named '#{host}' and saved locally as #{pubkey}. Continue? [y/n]
+      MSG
+      choice = nil
+      while choice != "y" && choice != "n"
+        choice = Capistrano::CLI.ui.ask(msg).downcase
+        msg = "Please enter 'y' or 'n'."
+      end
+      if choice == "y"
+        run_local "scp -i '#{privkey}' app@#{host}:.ssh/authorized_keys #{pubkey}"
+      end
+    end
+        
     desc <<-DESC
       Start a new server instance and prepare it for a cold deploy.
     DESC
@@ -96,6 +124,7 @@ Capistrano::Configuration.instance.load do
       server.install_packages
       server.install_gems
       server.deploy_files
+      server.set_rails_env
       server.restart_services
       deploy.setup
       server.set_roles
@@ -116,7 +145,6 @@ Capistrano::Configuration.instance.load do
     
     namespace :ec2 do      
       desc <<-DESC
-        Start an instance, using the AMI of the correct version to match this gem.
       DESC
       task :start_instance, :roles => [:web, :db, :app] do
         # TODO
@@ -131,26 +159,26 @@ Capistrano::Configuration.instance.load do
     
     namespace :db do
       desc <<-DESC
-        [internal] Load configuration info for the production database from \
+        [internal] Load configuration info for the database from 
         config/database.yml.
       DESC
       task :load_config, :roles => :db do
-        db_config = YAML::load(ERB.new(File.read("config/database.yml")).result)['production']
-        cfg[:production_db_name] = db_config['database']
-        cfg[:production_db_user] = db_config['username']
-        cfg[:production_db_password] = db_config['password']
-        cfg[:production_db_host] = db_config['host']
-        cfg[:production_db_socket] = db_config['socket']
+        db_config = YAML::load(ERB.new(File.read("config/database.yml")).result)[rails_env]
+        cfg[:db_name] = db_config['database']
+        cfg[:db_user] = db_config['username']
+        cfg[:db_password] = db_config['password']
+        cfg[:db_host] = db_config['host']
+        cfg[:db_socket] = db_config['socket']
         
-        if (cfg[:production_db_host].nil? || cfg[:production_db_host].empty?) && 
-          (cfg[:production_db_host] != 'localhost' || cfg[:production_db_host] != '127.0.0.1') && 
-          (cfg[:production_db_socket].nil? || cfg[:production_db_socket].empty?)
-            raise "ERROR: missing database config. Make sure database.yml contains a 'production' section with either 'host: localhost' or 'socket: /var/run/mysqld/mysqld.sock'."
+        if (cfg[:db_host].nil? || cfg[:db_host].empty?) && 
+          (cfg[:db_host] != 'localhost' || cfg[:db_host] != '127.0.0.1') && 
+          (cfg[:db_socket].nil? || cfg[:db_socket].empty?)
+            raise "ERROR: missing database config. Make sure database.yml contains a '#{rails_env}' section with either 'host: localhost' or 'socket: /var/run/mysqld/mysqld.sock'."
         end
         
-        [cfg[:production_db_name], cfg[:production_db_user], cfg[:production_db_password]].each do |s|
+        [cfg[:db_name], cfg[:db_user], cfg[:db_password]].each do |s|
           if s.nil? || s.empty?
-            raise "ERROR: missing database config. Make sure database.yml contains a 'production' section with a database name, user, and password."
+            raise "ERROR: missing database config. Make sure database.yml contains a '#{rails_env}' section with a database name, user, and password."
           elsif s.match(/['"]/)
             raise "ERROR: database config string '#{s}' contains quotes."
           end
@@ -158,25 +186,25 @@ Capistrano::Configuration.instance.load do
       end
       
       desc <<-DESC
-        Create the MySQL production database. Assumes there is no MySQL root \
+        Create the MySQL database. Assumes there is no MySQL root \
         password. To create a MySQL root password create a task that's run \
         after this task using an after hook.
       DESC
       task :create, :roles => :db do
         on_rollback { drop }
         load_config
-        run "echo 'create database #{cfg[:production_db_name]};' | mysql -u root"
-        run "echo \"grant all on #{cfg[:production_db_name]}.* to '#{cfg[:production_db_user]}'@'%' identified by '#{cfg[:production_db_password]}';\" | mysql -u root"
+        run "echo 'create database if not exists #{cfg[:db_name]};' | mysql -u root"
+        run "echo \"grant all on #{cfg[:db_name]}.* to '#{cfg[:db_user]}'@'%' identified by '#{cfg[:db_password]}';\" | mysql -u root"
       end
       
       desc <<-DESC
-        Drop the MySQL production database. Assumes there is no MySQL root \
+        Drop the MySQL database. Assumes there is no MySQL root \
         password. If there is a MySQL root password, create a task that removes \
         it and run that task before this one using a before hook.
       DESC
       task :drop, :roles => :db do
         load_config
-        run "echo 'drop database if exists #{cfg[:production_db_name]};' | mysql -u root"
+        run "echo 'drop database if exists #{cfg[:db_name]};' | mysql -u root"
       end
       
       desc <<-DESC
@@ -224,7 +252,18 @@ Capistrano::Configuration.instance.load do
       DESC
       task :set_roles, :roles => [:web_admin, :db_admin, :app_admin] do
         # TODO generate this based on the roles that actually exist so arbitrary new ones can be added
-        sudo "/usr/local/ec2onrails/bin/set_roles.rb web=#{hostnames_for_role(:web)} app=#{hostnames_for_role(:app)} db_primary=#{hostnames_for_role(:db, :primary => true)}"
+        sudo "/usr/local/ec2onrails/bin/set_roles.rb web=#{hostnames_for_role(:web)} app=#{hostnames_for_role(:app)} db_primary=#{hostnames_for_role(:db, :primary => true)} memcache=#{hostnames_for_role(:memcache)}"
+      end
+
+      desc <<-DESC
+        Change the default value of RAILS_ENV on the server. Technically
+        this changes the server's mongrel config to use a different value
+        for "environment". The value is specified in :rails_env
+      DESC
+      task :set_rails_env, :roles => [:web_admin, :db_admin, :app_admin] do
+        rails_env = fetch(:rails_env, "production")
+        sudo "/usr/local/ec2onrails/bin/set_rails_env #{rails_env}"
+        deploy.restart
       end
       
       desc <<-DESC
@@ -239,7 +278,19 @@ Capistrano::Configuration.instance.load do
         Upgrade to the newest versions of all rubygems.
       DESC
       task :upgrade_gems, :roles => [:web_admin, :db_admin, :app_admin] do
-        sudo "gem update -y --no-rdoc --no-ri"
+        sudo "gem update --system --no-rdoc --no-ri"
+        sudo "gem update -y --no-rdoc --no-ri" do |ch, str, data|
+          ch[:data] ||= ""
+          ch[:data] << data
+          if data =~ />\s*$/
+            puts data
+            choice = Capistrano::CLI.ui.ask("The gem command is asking for a number:")
+            ch.send_data("#{choice}\n")
+          else
+            puts data
+          end
+        end
+        
       end
       
       desc <<-DESC
@@ -248,7 +299,7 @@ Capistrano::Configuration.instance.load do
         NOTE: the package installation will be non-interactive, if the packages \
         require configuration either log in as 'admin' and run \
         'dpkg-reconfigure packagename' or replace the package's config files \
-        using the 'ec2onrails:deploy_config_files' task.
+        using the 'ec2onrails:server:deploy_files' task.
       DESC
       task :install_packages, :roles => [:web_admin, :db_admin, :app_admin] do
         if cfg[:packages] && cfg[:packages].any?
@@ -261,17 +312,18 @@ Capistrano::Configuration.instance.load do
         be with an array of strings.
       DESC
       task :install_gems, :roles => [:web_admin, :db_admin, :app_admin] do
-        if cfg[:rubygems] && cfg[:rubygems].any?
-          sudo "gem install #{cfg[:rubygems].join(' ')} -y --no-rdoc --no-ri" do |ch, str, data|
-            ch[:data] ||= ""
-            ch[:data] << data
-            if data =~ />\s*$/
-              puts data
-              puts "The gem command is asking for a number:"
-              choice = STDIN.gets
-              ch.send_data(choice)
-            else
-              puts data
+        if cfg[:rubygems]
+          cfg[:rubygems].each do |gem|
+            sudo "gem install #{gem} -y --no-rdoc --no-ri" do |ch, str, data|
+              ch[:data] ||= ""
+              ch[:data] << data
+              if data =~ />\s*$/
+                puts data
+                choice = Capistrano::CLI.ui.ask("The gem command is asking for a number:")
+                ch.send_data("#{choice}\n")
+              else
+                puts data
+              end
             end
           end
         end
@@ -356,12 +408,6 @@ Capistrano::Configuration.instance.load do
       task :run_script, :roles => [:web_admin, :db_admin, :app_admin] do
         # TODO
       end
-      
-      desc <<-DESC
-      DESC
-      task :archive_logs, :roles => [:web_admin, :db_admin, :app_admin] do
-        # TODO
-      end
     end
     
   end

data/lib/ec2onrails/version.rb

@@ -20,10 +20,10 @@ module Ec2onrails #:nodoc:
   module VERSION #:nodoc:
     MAJOR = 0
     MINOR = 9
-    TINY  = 6
+    TINY  = 7
     STRING = [MAJOR, MINOR, TINY].join('.')
     
-    AMI_ID_32_BIT = 'ami-e620c58f'
-    AMI_ID_64_BIT = 'ami-f920c590'
+    AMI_ID_32_BIT = 'ami-5c0aef35'
+    AMI_ID_64_BIT = 'ami-540aef3d'
   end
 end

data/website/index.html

@@ -33,7 +33,7 @@
     <h1>EC2 on Rails</h1>
     <div id="version" class="clickable" onclick='document.location = "http://rubyforge.org/projects/ec2onrails"; return false'>
       <p>Get Version</p>
-      <a href="http://rubyforge.org/projects/ec2onrails" class="numbers">0.9.6</a>
+      <a href="http://rubyforge.org/projects/ec2onrails" class="numbers">0.9.7</a>
     </div>
     <h2>Deploy a Ruby on Rails app on <span class="caps">EC2</span> in five minutes</h2>
 
@@ -67,6 +67,7 @@ your Rails app.</p>
 		<li>Amazon <span class="caps">AMI</span> tools installed</li>
 		<li>A script to rebundle a customized version of the image in one step if required</li>
 		<li>MySQL, Apache, and syslog configured to use /mnt for data and logging so you don&#8217;t fill up <span class="caps">EC2</span>&#8217;s small root filesystem</li>
+		<li>Automatically archives Rails and Apache logs to S3 nightly.</li>
 		<li>32-bit and 64-bit images available (supports all instance types, small to extra large).</li>
 		<li>Created using a build file, full source is available.</li>
 		<li>Multiple instance support.</li>
@@ -144,16 +145,10 @@ If you&#8217;re using multiple instances, be sure to allow them network access t
 But for some reason Capistrano requires that you have both the public key and the private key files together on the client.</p>
 
 
-	<p>So wherever you stored your private key file, you need to put the public key file beside it, with the extension &#8221;.pub&#8221;. 
-You can copy it from the server using scp like this:</p>
+	<p>From within the root of your rails app do:</p>
 
 
-<pre>scp -i /home/you/.ssh/your-private-key admin@ec2-xxx....amazonaws.com:.ssh/authorized_keys /home/you/.ssh/your-private-key.pub</pre>
-
-	<p>This uses scp to connect to the server using your private key. It copies the file &#8220;authorized_keys&#8221; from /home/admin/.ssh/ and 
-saves it in your .ssh dir. So after this you should have two files, one named &#8220;yourkey&#8221; (this is your private key) and one 
-named (yourkey.pub), this is your public key.</p>
-
+<pre>cap ec2onrails:get_public_key_from_server</pre>
 
 	<h4>6. Deploy the app with Capistrano</h4>
 
@@ -165,10 +160,10 @@ cap deploy:migrations
 
 	<p>Yes, it&#8217;s that easy! The setup task will set the server&#8217;s timezone, install any 
 gems and Ubuntu packages that you specified in the config file, and 
-create your production database.</p>
+create your database.</p>
 
 
-	<p>That&#8217;s it, you&#8217;re done.</p>
+	<p>That&#8217;s it, your app is now running on <span class="caps">EC2</span>!!</p>
 
 
 	<h2>Capistrano tasks</h2>
@@ -234,6 +229,9 @@ bucket and deploy the app, useful to set up a staging server.</li>
 	</ul>
 
 
+	<p>To get a full list of the Capistrano tasks at any time type <code>cap -T</code> from with your rails app root.</p>
+
+
 	<h2>Saving your own customized version of the image.</h2>
 
 
@@ -312,7 +310,7 @@ or send me an <a href="http://pauldowman.com/contact/">email</a>.</p>
 
 	<p>This code is free to use under the terms of the <span class="caps">GPL</span> v2.</p>
     <p class="coda">
-      <a href="http://pauldowman.com/contact/">Paul Dowman</a>, 21st December 2007<br>
+      <a href="http://pauldowman.com/contact/">Paul Dowman</a>, 2nd March 2008<br>
       Theme extended from <a href="http://rb2js.rubyforge.org/">Paul Battley</a>
     </p>
 </div>

data/website/index.txt

@@ -27,6 +27,7 @@ Features of the EC2 image:
 * Amazon AMI tools installed
 * A script to rebundle a customized version of the image in one step if required
 * MySQL, Apache, and syslog configured to use /mnt for data and logging so you don't fill up EC2's small root filesystem
+* Automatically archives Rails and Apache logs to S3 nightly.
 * 32-bit and 64-bit images available (supports all instance types, small to extra large).
 * Created using a build file, full source is available.
 * Multiple instance support.
@@ -89,14 +90,9 @@ h4. 5. Copy your public key from the server to keep Capistrano happy
 Technically all you should need to connect to the server is the private key file, the public key is on the server. 
 But for some reason Capistrano requires that you have both the public key and the private key files together on the client.
 
-So wherever you stored your private key file, you need to put the public key file beside it, with the extension ".pub". 
-You can copy it from the server using scp like this:
+From within the root of your rails app do:
 
-<pre>scp -i /home/you/.ssh/your-private-key admin@ec2-xxx....amazonaws.com:.ssh/authorized_keys /home/you/.ssh/your-private-key.pub</pre>
-
-This uses scp to connect to the server using your private key. It copies the file "authorized_keys" from /home/admin/.ssh/ and 
-saves it in your .ssh dir. So after this you should have two files, one named "yourkey" (this is your private key) and one 
-named (yourkey.pub), this is your public key.
+<pre>cap ec2onrails:get_public_key_from_server</pre>
 
 
 h4. 6. Deploy the app with Capistrano
@@ -108,9 +104,9 @@ cap deploy:migrations
 
 Yes, it's that easy! The setup task will set the server's timezone, install any 
 gems and Ubuntu packages that you specified in the config file, and 
-create your production database.
+create your database.
 
-That's it, you're done.
+That's it, your app is now running on EC2!!
 
 h2. Capistrano tasks
 
@@ -158,6 +154,8 @@ step, everything except deploying the app.
 - Another convenience task to recreate the db, restore data to it from an s3
 bucket and deploy the app, useful to set up a staging server.
 
+To get a full list of the Capistrano tasks at any time type <code>cap -T</code> from with your rails app root.
+
 
 h2. Saving your own customized version of the image.
 

metadata

@@ -1,15 +1,15 @@
 --- !ruby/object:Gem::Specification 
 name: ec2onrails
 version: !ruby/object:Gem::Version 
-  version: 0.9.6
-platform: ""
+  version: 0.9.7
+platform: ruby
 authors: 
 - Paul Dowman
 autorequire: 
 bindir: bin
 cert_chain: []
 
-date: 2007-12-21 00:00:00 -05:00
+date: 2008-03-03 00:00:00 -05:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -101,7 +101,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
 requirements: []
 
 rubyforge_project: ec2onrails
-rubygems_version: 0.9.5
+rubygems_version: 1.0.1
 signing_key: 
 specification_version: 2
 summary: Client-side libraries (Capistrano tasks) for managing and deploying to EC2 on Rails servers.