god 0.5.0 → 0.6.0

data/History.txt

@@ -1,3 +1,26 @@
+== 0.6.0 / 
+
+* Minor Enhancement
+  * Move Syslog calls into God::Logger and clean up all calling code
+  * Remove god's pid file on user requested termination
+  * Better handling and cleanup of DRb server's unix domain socket
+  * Allow shorthand for requesting a god log
+  * Add `god check` to make it easier to diagnose event problems
+  * Refactor god binary into class/method structure
+  * Implement `god remove` to remove a Task altogether
+* New Conditions
+  * DiskUsage < PollCondition - trigger if disk usage is above limit on mount [Rudy Desjardins]
+
+== 0.5.2 / 2007-10-10
+
+* Minor Enhancement
+  * Allow extra args to pass through to config file
+  
+== 0.5.1 / 2007-10-08
+
+* Bug Fixes
+  * Rescue connection refused in http response code condition
+
 == 0.5.0 / 2007-10-05
 
 * Major Enhancements

data/Manifest.txt

@@ -5,6 +5,7 @@ Rakefile
 bin/god
 examples/events.god
 examples/gravatar.god
+examples/single.god
 ext/god/extconf.rb
 ext/god/kqueue_handler.c
 ext/god/netlink_handler.c
@@ -13,10 +14,15 @@ lib/god.rb
 lib/god/behavior.rb
 lib/god/behaviors/clean_pid_file.rb
 lib/god/behaviors/notify_when_flapping.rb
+lib/god/cli/command.rb
+lib/god/cli/run.rb
+lib/god/cli/version.rb
 lib/god/condition.rb
 lib/god/conditions/always.rb
+lib/god/conditions/complex.rb
 lib/god/conditions/cpu_usage.rb
 lib/god/conditions/degrading_lambda.rb
+lib/god/conditions/disk_usage.rb
 lib/god/conditions/flapping.rb
 lib/god/conditions/http_response_code.rb
 lib/god/conditions/lambda.rb
@@ -50,6 +56,8 @@ test/configs/child_events/child_events.god
 test/configs/child_events/simple_server.rb
 test/configs/child_polls/child_polls.god
 test/configs/child_polls/simple_server.rb
+test/configs/complex/complex.god
+test/configs/complex/simple_server.rb
 test/configs/contact/contact.god
 test/configs/contact/simple_server.rb
 test/configs/daemon_events/daemon_events.god
@@ -70,6 +78,7 @@ test/helper.rb
 test/suite.rb
 test/test_behavior.rb
 test/test_condition.rb
+test/test_conditions_disk_usage.rb
 test/test_conditions_http_response_code.rb
 test/test_conditions_process_running.rb
 test/test_conditions_tries.rb

data/README.txt

@@ -6,14 +6,21 @@ god
 
 == DESCRIPTION:
   
-God is an easy to configure, easy to extend monitoring framework written in Ruby.
+God is an easy to configure, easy to extend monitoring framework written 
+in Ruby.
 
-Keeping your server processes and tasks running should be a simple part of your deployment process. God aims to be the simplest, most powerful monitoring application available.
+Keeping your server processes and tasks running should be a simple part of 
+your deployment process. God aims to be the simplest, most powerful 
+monitoring application available.
 
 == DOCUMENTATION:
 
 See online documentation at http://god.rubyforge.org
 
+== COMMUNITY:
+
+Sign up for the god mailing list at http://groups.google.com/group/god-rb
+
 == INSTALL:
 
   $ sudo gem install god

data/Rakefile

@@ -1,7 +1,7 @@
 require 'rubygems'
 require 'hoe'
 
-Hoe.new('god', '0.5.0') do |p|
+Hoe.new('god', '0.6.0') do |p|
   p.rubyforge_name = 'god'
   p.author = 'Tom Preston-Werner'
   p.email = 'tom@rubyisawesome.com'
@@ -25,4 +25,11 @@ end
 desc "Upload site to Rubyforge"
 task :site_edge do
   sh "scp -r site/* mojombo@god.rubyforge.org:/var/www/gforge-projects/god/edge"
+end
+
+desc "Run rcov"
+task :coverage do
+  `rm -fr coverage`
+  `rcov test/test_*.rb`
+  `open coverage/index.html`
 end

data/bin/god

@@ -15,7 +15,7 @@ begin
     opts.banner = <<-EOF
   Usage:
     Starting:
-      god -c <config file> [-p <port> | -b] [-P <file>] [-l <file>] [-D]
+      god [-c <config file>] [-p <port> | -b] [-P <file>] [-l <file>] [-D]
       
     Querying:
       god <command> <argument> [-p <port>]
@@ -29,11 +29,13 @@ begin
       stop <task or group name>          stop task or group
       monitor <task or group name>       monitor task or group
       unmonitor <task or group name>     unmonitor task or group
+      remove <task or group name>        remove task or group from god
       load <file>                        load a config into a running god
       log <task name>                    show realtime log for given task
       status                             show status of each task
       quit                               stop god
       terminate                          stop god and all tasks
+      check                              run self diagnostic
       
     Options:
   EOF
@@ -73,223 +75,18 @@ begin
   
   opts.parse!
   
-  if options[:version]
+  if !options[:config] && options[:version]
     require 'god'
-    
-    # print version
-    puts "Version #{God::VERSION}"
-    exit
-  elsif options[:info]
+    God::CLI::Version.version
+  elsif !options[:config] && options[:info]
     require 'god'
-    
-    puts "Version: #{God::VERSION}"
-    puts "Polls: enabled"
-    puts "Events: " + God::EventHandler.event_system
-    
-    exit
-  elsif command = ARGV[0]
+    God::CLI::Version.version_extended
+  elsif !options[:config] && command = ARGV[0]
     require 'god'
-    
-    # a command was specified
-    
-    # connect to drb unix socket
-    DRb.start_service
-    server = DRbObject.new(nil, God::Socket.socket(options[:port]))
-    
-    begin
-      server.ping
-    rescue DRb::DRbConnError
-      puts "The server is not available (or you do not have permissions to access it)"
-      abort
-    end
-    
-    if command == 'load'
-      file = ARGV[1]
-      
-      puts "Sending '#{command}' command"
-      puts
-      
-      unless File.exist?(file)
-        abort "File not found: #{file}"
-      end
-      
-      names, errors = *server.running_load(File.read(file), File.expand_path(file))
-      
-      # output response
-      unless names.empty?
-        puts 'The following tasks were affected:'
-        names.each do |w|
-          puts '  ' + w
-        end
-      end
-      
-      unless errors.empty?
-        puts errors
-        exit(1)
-      end
-    elsif command == 'status'
-      watches = server.status
-      watches.keys.sort.each do |name|
-        state = watches[name][:state]
-        puts "#{name}: #{state}"
-      end
-    elsif command == 'log'
-      begin
-        Signal.trap('INT') { exit }
-        name = ARGV[1]
-        t = Time.at(0)
-        loop do
-          print server.running_log(name, t)
-          t = Time.now
-          sleep 1
-        end
-      rescue God::NoSuchWatchError
-        puts "No such watch"
-      rescue DRb::DRbConnError
-        puts "The server went away"
-      end
-    elsif command == 'quit'
-      begin
-        server.terminate
-        abort 'Could not stop god'
-      rescue DRb::DRbConnError
-        puts 'Stopped god'
-      end
-    elsif command == 'terminate'
-      t = Thread.new { loop { STDOUT.print('.'); STDOUT.flush; sleep(1) } }
-      if server.stop_all
-        t.kill; STDOUT.puts
-        puts 'Stopped all watches'
-      else
-        t.kill; STDOUT.puts
-        puts 'Could not stop all watches within 10 seconds'
-      end
-      
-      begin
-        server.terminate
-        abort 'Could not stop god'
-      rescue DRb::DRbConnError
-        puts 'Stopped god'
-      end
-    else
-      # get the name of the watch/group
-      name = ARGV[1]
-      
-      begin
-        puts "Sending '#{command}' command"
-        
-        t = Thread.new { loop { sleep(1); STDOUT.print('.'); STDOUT.flush; sleep(1) } }
-        
-        # send command
-        watches = server.control(name, command)
-        
-        # output response
-        t.kill; STDOUT.puts
-        unless watches.empty?
-          puts 'The following watches were affected:'
-          watches.each do |w|
-            puts '  ' + w
-          end
-        else
-          puts 'No matching task or group'
-        end
-      rescue God::InvalidCommandError
-        t.kill rescue nil; STDOUT.puts
-        abort "Command '#{command}' is not valid. Run 'god --help' for usage"
-      end
-    end
+    God::CLI::Command.new(command, options, ARGV)
   else
-    # start god
-    $run = true
-    
-    if !options[:daemonize]
-      require 'god'
-      
-      if options[:port]
-        God.port = options[:port]
-      end
-      
-      if options[:config]
-        unless File.exist?(options[:config])
-          abort "File not found: #{options[:config]}"
-        end
-        
-        begin
-          load File.expand_path(options[:config])
-        rescue Exception => e
-          if e.instance_of?(SystemExit)
-            raise
-          else
-            puts e.message
-            puts e.backtrace.join("\n")
-            abort "There was an error in your configuration file (see above)"
-          end
-        end
-      end
-    else
-      # trap and ignore SIGHUP
-      Signal.trap('HUP') {}
-      
-      pid = fork do
-        begin
-          require 'god'
-          
-          log_file = options[:log] || "/dev/null"
-          
-          unless God::EventHandler.loaded?
-            puts
-            puts "***********************************************************************"
-            puts "*"
-            puts "* Event conditions are not available for your installation of god."
-            puts "* You may still use and write custom conditions using the poll system"
-            puts "*"
-            puts "***********************************************************************"
-            puts
-          end
-          
-          # set port if requested
-          if options[:port]
-            God.port = options[:port]
-          end
-          
-          # load config
-          if options[:config]
-            unless File.exist?(options[:config])
-              abort "File not found: #{options[:config]}"
-            end
-            
-            begin
-              load File.expand_path(options[:config])
-            rescue Exception => e
-              if e.instance_of?(SystemExit)
-                raise
-              else
-                puts e.message
-                puts e.backtrace.join("\n")
-                abort "There was an error in your configuration file (see above)"
-              end
-            end
-          end
-          
-          # reset file descriptors
-          STDIN.reopen "/dev/null"
-          STDOUT.reopen(log_file, "a")
-          STDERR.reopen STDOUT
-        rescue => e
-          puts e.message
-          puts e.backtrace.join("\n")
-          abort "There was a fatal system error while starting god (see above)"
-        end
-      end
-      
-      if options[:pid]
-        File.open(options[:pid], 'w') { |f| f.write pid }
-      end
-      
-      ::Process.detach pid
-      
-      exit
-    end
+    require 'god/cli/run'
+    God::CLI::Run.new(options)
   end
 rescue Exception => e
   if e.instance_of?(SystemExit)

data/examples/single.god

@@ -0,0 +1,66 @@
+RAILS_ROOT = "/Users/tom/dev/gravatar2"
+
+God.watch do |w|
+  w.name = "local-3000"
+  w.interval = 5.seconds # default
+  w.start = "mongrel_rails start -c #{RAILS_ROOT} -P #{RAILS_ROOT}/log/mongrel.pid -p 3000 -d"
+  w.stop = "mongrel_rails stop -P #{RAILS_ROOT}/log/mongrel.pid"
+  w.restart = "mongrel_rails restart -P #{RAILS_ROOT}/log/mongrel.pid"
+  w.pid_file = File.join(RAILS_ROOT, "log/mongrel.pid")
+  
+  # clean pid files before start if necessary
+  w.behavior(:clean_pid_file)
+  
+  # determine the state on startup
+  w.transition(:init, { true => :up, false => :start }) do |on|
+    on.condition(:process_running) do |c|
+      c.running = true
+    end
+  end
+  
+  # determine when process has finished starting
+  w.transition([:start, :restart], :up) do |on|
+    on.condition(:process_running) do |c|
+      c.running = true
+    end
+    
+    # failsafe
+    on.condition(:tries) do |c|
+      c.times = 5
+      c.transition = :start
+    end
+  end
+
+  # start if process is not running
+  w.transition(:up, :start) do |on|
+    on.condition(:process_exits)
+  end
+  
+  # restart if memory or cpu is too high
+  w.transition(:up, :restart) do |on|
+    on.condition(:memory_usage) do |c|
+      c.interval = 20
+      c.above = 50.megabytes
+      c.times = [3, 5]
+    end
+    
+    on.condition(:cpu_usage) do |c|
+      c.interval = 10
+      c.above = 10.percent
+      c.times = [3, 5]
+    end
+  end
+  
+  # lifecycle
+  w.lifecycle do |on|
+    on.condition(:flapping) do |c|
+      c.to_state = [:start, :restart]
+      c.times = 5
+      c.within = 5.minute
+      c.transition = :unmonitored
+      c.retry_in = 10.minutes
+      c.retry_times = 5
+      c.retry_within = 2.hours
+    end
+  end
+end

data/ext/god/netlink_handler.c

@@ -28,6 +28,8 @@ nlh_handle_events()
   struct nlmsghdr *hdr;
   struct proc_event *event;
   
+  VALUE extra_data;
+  
   fd_set fds;
   
   FD_ZERO(&fds);
@@ -61,16 +63,27 @@ nlh_handle_events()
           return INT2FIX(0);
         }
       
-        rb_funcall(cEventHandler, m_call, 2, INT2FIX(event->event_data.exit.process_pid), ID2SYM(proc_exit));
+        extra_data = rb_hash_new();
+        rb_hash_aset(extra_data, ID2SYM(rb_intern("pid")), INT2FIX(event->event_data.exit.process_pid));
+        rb_hash_aset(extra_data, ID2SYM(rb_intern("exit_code")), INT2FIX(event->event_data.exit.exit_code));
+        rb_hash_aset(extra_data, ID2SYM(rb_intern("exit_signal")), INT2FIX(event->event_data.exit.exit_signal));
+        rb_hash_aset(extra_data, ID2SYM(rb_intern("thread_group_id")), INT2FIX(event->event_data.exit.process_tgid));
+
+        rb_funcall(cEventHandler, m_call, 3, INT2FIX(event->event_data.exit.process_pid), ID2SYM(proc_exit), extra_data);
         return INT2FIX(1);
       
-      /* TODO: On fork, call and pass pid of child */
       case PROC_EVENT_FORK:
         if (Qnil == rb_funcall(cEventHandler, m_watching_pid, 1, INT2FIX(event->event_data.fork.parent_pid))) {
           return INT2FIX(0);
         }
+        
+        extra_data = rb_hash_new();
+        rb_hash_aset(extra_data, rb_intern("parent_pid"), INT2FIX(event->event_data.fork.parent_pid));
+        rb_hash_aset(extra_data, rb_intern("parent_thread_group_id"), INT2FIX(event->event_data.fork.parent_tgid));
+        rb_hash_aset(extra_data, rb_intern("child_pid"), INT2FIX(event->event_data.fork.child_pid));
+        rb_hash_aset(extra_data, rb_intern("child_thread_group_id"), INT2FIX(event->event_data.fork.child_tgid));
       
-        rb_funcall(cEventHandler, m_call, 2, INT2FIX(event->event_data.fork.parent_pid), ID2SYM(proc_fork));
+        rb_funcall(cEventHandler, m_call, 3, INT2FIX(event->event_data.fork.parent_pid), ID2SYM(proc_fork), extra_data);
         return INT2FIX(1);
 
       case PROC_EVENT_NONE: