Bump version for 0.8.2

* main:
  Add option to skip audit broadcasts (useful when testing)

Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    mrsk (0.8.1)
+    mrsk (0.8.2)
       activesupport (>= 7.0)
       dotenv (~> 2.8)
       sshkit (~> 1.21)

README.md

@@ -498,7 +498,7 @@ If you wish to remove the entire application, including Traefik, containers, ima
 
 ## Stage of development
 
-This is alpha software. Lots of stuff is missing. Lots of stuff will keep moving around for a while.
+This is beta software. Commands may still move around. But we're live in production at [37signals](https://37signals.com).
 
 ## License
 

lib/mrsk/cli/accessory.rb

@@ -13,7 +13,7 @@ class Mrsk::Cli::Accessory < Mrsk::Cli::Base
           execute *accessory.run
         end
 
-        audit_broadcast "Booted accessory #{name}"
+        audit_broadcast "Booted accessory #{name}" unless options[:skip_broadcast]
       end
     end
   end

lib/mrsk/cli/app.rb

@@ -5,18 +5,27 @@ class Mrsk::Cli::App < Mrsk::Cli::Base
     using_version(options[:version] || most_recent_version_available) do |version|
       say "Start container with version #{version} (or reboot if already running)...", :magenta
 
+      cli = self
+      
       MRSK.config.roles.each do |role|
         on(role.hosts) do |host|
           execute *MRSK.auditor.record("Booted app version #{version}"), verbosity: :debug
 
           begin
-            execute *MRSK.app.stop, raise_on_non_zero_exit: false
+            old_version = capture_with_info(*MRSK.app.current_running_version).strip
             execute *MRSK.app.run(role: role.name)
+
+            cli.say "Waiting #{MRSK.config.readiness_delay}s for app to boot...", :magenta
+            sleep MRSK.config.readiness_delay
+
+            execute *MRSK.app.stop(version: old_version), raise_on_non_zero_exit: false if old_version.present?
+
           rescue SSHKit::Command::Failed => e
             if e.message =~ /already in use/
-              error "Rebooting container with same version already deployed on #{host}"
+              error "Rebooting container with same version #{version} already deployed on #{host} (may cause gap in zero-downtime promise!)"
               execute *MRSK.auditor.record("Rebooted app version #{version}"), verbosity: :debug
 
+              execute *MRSK.app.stop(version: version)
               execute *MRSK.app.remove_container(version: version)
               execute *MRSK.app.run(role: role.name)
             else

lib/mrsk/cli/base.rb

@@ -20,6 +20,8 @@ module Mrsk::Cli
     class_option :config_file, aliases: "-c", default: "config/deploy.yml", desc: "Path to config file"
     class_option :destination, aliases: "-d", desc: "Specify destination to be used for config file (staging -> deploy.staging.yml)"
 
+    class_option :skip_broadcast, aliases: "-B", type: :boolean, default: false, desc: "Skip audit broadcasts"
+
     def initialize(*)
       super
       load_envs

lib/mrsk/cli/healthcheck.rb

@@ -1,5 +1,7 @@
 class Mrsk::Cli::Healthcheck < Mrsk::Cli::Base
-  MAX_ATTEMPTS = 5
+  MAX_ATTEMPTS = 7
+
+  class HealthcheckError < StandardError; end
 
   default_command :perform
 
@@ -18,7 +20,7 @@ class Mrsk::Cli::Healthcheck < Mrsk::Cli::Base
           if status == "200"
             info "#{target} succeeded with 200 OK!"
           else
-            raise "#{target} failed with status #{status}"
+            raise HealthcheckError, "#{target} failed with status #{status}"
           end
         rescue SSHKit::Command::Failed
           if attempt <= MAX_ATTEMPTS
@@ -31,7 +33,7 @@ class Mrsk::Cli::Healthcheck < Mrsk::Cli::Base
             raise
           end
         end
-      rescue SSHKit::Command::Failed => e
+      rescue SSHKit::Command::Failed, HealthcheckError => e
         error capture_with_info(*MRSK.healthcheck.logs)
 
         if e.message =~ /curl/

lib/mrsk/cli/main.rb

@@ -32,7 +32,7 @@ class Mrsk::Cli::Main < Mrsk::Cli::Base
       invoke "mrsk:cli:prune:all"
     end
 
-    audit_broadcast "Deployed app in #{runtime.to_i} seconds"
+    audit_broadcast "Deployed app in #{runtime.to_i} seconds" unless options[:skip_broadcast]
   end
 
   desc "redeploy", "Deploy app to servers without bootstrapping servers, starting Traefik, pruning, and registry login"
@@ -47,7 +47,7 @@ class Mrsk::Cli::Main < Mrsk::Cli::Base
       invoke "mrsk:cli:app:boot"
     end
 
-    audit_broadcast "Redeployed app in #{runtime.to_i} seconds"
+    audit_broadcast "Redeployed app in #{runtime.to_i} seconds" unless options[:skip_broadcast]
   end
 
   desc "rollback [VERSION]", "Rollback app to VERSION"
@@ -55,14 +55,22 @@ class Mrsk::Cli::Main < Mrsk::Cli::Base
     MRSK.version = version
 
     if container_name_available?(MRSK.config.service_with_version)
-      say "Stop current version, then start version #{version}...", :magenta
+      say "Start version #{version}, then stop the old version...", :magenta
+
+      cli = self
 
       on(MRSK.hosts) do |host|
-        execute *MRSK.app.stop, raise_on_non_zero_exit: false
+        old_version = capture_with_info(*MRSK.app.current_running_version).strip.presence
+
         execute *MRSK.app.start
+
+        cli.say "Waiting #{MRSK.config.readiness_delay}s for app to start...", :magenta
+        sleep MRSK.config.readiness_delay
+
+        execute *MRSK.app.stop(version: old_version), raise_on_non_zero_exit: false
       end
 
-      audit_broadcast "Rolled back app to version #{version}"
+      audit_broadcast "Rolled back app to version #{version}" unless options[:skip_broadcast]
     else
       say "The app version '#{version}' is not available as a container (use 'mrsk app containers' for available versions)", :red
     end

lib/mrsk/commands/app.rb

@@ -18,8 +18,10 @@ class Mrsk::Commands::App < Mrsk::Commands::Base
     docker :start, service_with_version
   end
 
-  def stop
+  def stop(version: nil)
-    pipe current_container_id, xargs(docker(:stop))
+    pipe \
+      version ? container_id_for_version(version) : current_container_id,
+      xargs(docker(:stop))
   end
 
   def info
@@ -132,6 +134,10 @@ class Mrsk::Commands::App < Mrsk::Commands::Base
       end
     end
 
+    def container_id_for_version(version)
+      container_id_for(container_name: service_with_version(version))    
+    end
+
     def service_filter
       [ "--filter", "label=service=#{config.service}" ]
     end

lib/mrsk/configuration.rb

@@ -136,6 +136,9 @@ class Mrsk::Configuration
     { "path" => "/up", "port" => 3000 }.merge(raw_config.healthcheck || {})
   end
 
+  def readiness_delay
+    raw_config.readiness_delay || 7
+  end
 
   def valid?
     ensure_required_keys_present && ensure_env_available

lib/mrsk/configuration/role.rb

@@ -61,7 +61,7 @@ class Mrsk::Configuration::Role
           "traefik.http.routers.#{config.service}.rule" => "PathPrefix(`/`)",
           "traefik.http.services.#{config.service}.loadbalancer.healthcheck.path" => config.healthcheck["path"],
           "traefik.http.services.#{config.service}.loadbalancer.healthcheck.interval" => "1s",
-          "traefik.http.middlewares.#{config.service}.retry.attempts" => "3",
+          "traefik.http.middlewares.#{config.service}.retry.attempts" => "5",
           "traefik.http.middlewares.#{config.service}.retry.initialinterval" => "500ms"
         }
       else

lib/mrsk/version.rb

@@ -1,3 +1,3 @@
 module Mrsk
-  VERSION = "0.8.1"
+  VERSION = "0.8.2"
 end

test/cli/app_test.rb

@@ -2,7 +2,16 @@ require_relative "cli_test_case"
 
 class CliAppTest < CliTestCase
   test "boot" do
-    assert_match /Running docker run --detach --restart unless-stopped/, run_command("boot")
+    # Stub current version fetch
+    SSHKit::Backend::Abstract.any_instance.stubs(:capture)
+      .returns("999") # new version
+      .then
+      .returns("123") # old version
+
+    run_command("boot").tap do |output|
+      assert_match /docker run --detach --restart unless-stopped/, output
+      assert_match /docker container ls --all --filter name=app-123 --quiet | xargs docker stop/, output
+    end
   end
 
   test "boot will reboot if same version is already running" do
@@ -11,12 +20,17 @@ class CliAppTest < CliTestCase
     # Prevent expected failures from outputting to terminal
     Thread.report_on_exception = false
 
-    MRSK.app.stubs(:run).raises(SSHKit::Command::Failed.new("already in use")).then.returns([ :docker, :run ])
+    MRSK.app.stubs(:run)
+      .raises(SSHKit::Command::Failed.new("already in use"))
+      .then
+      .raises(SSHKit::Command::Failed.new("already in use"))
+      .then
+      .returns([ :docker, :run ])
 
     run_command("boot").tap do |output|
-      assert_match /Rebooting container with same version already deployed/, output # Can't start what's already running
+      assert_match /Rebooting container with same version 999 already deployed/, output # Can't start what's already running
-      assert_match /docker ps --quiet --filter label=service=app \| xargs docker stop/, output # Stop what's running
+      assert_match /docker container ls --all --filter name=app-999 --quiet | xargs docker container rm/, output # Stop old running
-      assert_match /docker container ls --all --filter name=app-999 --quiet \| xargs docker container rm/, output # Remove old container
+      assert_match /docker container ls --all --filter name=app-999 --quiet | xargs docker container rm/, output # Remove old container
       assert_match /docker run/, output # Start new container
     end
   ensure

test/cli/main_test.rb

@@ -19,7 +19,7 @@ class CliMainTest < CliTestCase
     Mrsk::Cli::Main.any_instance.stubs(:container_name_available?).returns(true)
 
     run_command("rollback", "123").tap do |output|
-      assert_match /Stop current version, then start version 123/, output
+      assert_match /Start version 123, then stop the old version/, output
       assert_match /docker ps -q --filter label=service=app | xargs docker stop/, output
       assert_match /docker start app-123/, output
     end

test/commands/app_test.rb

@@ -14,7 +14,7 @@ class CommandsAppTest < ActiveSupport::TestCase
 
   test "run" do
     assert_equal \
-      "docker run --detach --restart unless-stopped --log-opt max-size=10m --name app-999 -e RAILS_MASTER_KEY=\"456\" --label service=\"app\" --label role=\"web\" --label traefik.http.routers.app.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.services.app.loadbalancer.healthcheck.path=\"/up\" --label traefik.http.services.app.loadbalancer.healthcheck.interval=\"1s\" --label traefik.http.middlewares.app.retry.attempts=\"3\" --label traefik.http.middlewares.app.retry.initialinterval=\"500ms\" dhh/app:999",
+      "docker run --detach --restart unless-stopped --log-opt max-size=10m --name app-999 -e RAILS_MASTER_KEY=\"456\" --label service=\"app\" --label role=\"web\" --label traefik.http.routers.app.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.services.app.loadbalancer.healthcheck.path=\"/up\" --label traefik.http.services.app.loadbalancer.healthcheck.interval=\"1s\" --label traefik.http.middlewares.app.retry.attempts=\"5\" --label traefik.http.middlewares.app.retry.initialinterval=\"500ms\" dhh/app:999",
       @app.run.join(" ")
   end
 
@@ -22,7 +22,7 @@ class CommandsAppTest < ActiveSupport::TestCase
     @config[:volumes] = ["/local/path:/container/path" ]
 
     assert_equal \
-      "docker run --detach --restart unless-stopped --log-opt max-size=10m --name app-999 -e RAILS_MASTER_KEY=\"456\" --volume /local/path:/container/path --label service=\"app\" --label role=\"web\" --label traefik.http.routers.app.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.services.app.loadbalancer.healthcheck.path=\"/up\" --label traefik.http.services.app.loadbalancer.healthcheck.interval=\"1s\" --label traefik.http.middlewares.app.retry.attempts=\"3\" --label traefik.http.middlewares.app.retry.initialinterval=\"500ms\" dhh/app:999",
+      "docker run --detach --restart unless-stopped --log-opt max-size=10m --name app-999 -e RAILS_MASTER_KEY=\"456\" --volume /local/path:/container/path --label service=\"app\" --label role=\"web\" --label traefik.http.routers.app.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.services.app.loadbalancer.healthcheck.path=\"/up\" --label traefik.http.services.app.loadbalancer.healthcheck.interval=\"1s\" --label traefik.http.middlewares.app.retry.attempts=\"5\" --label traefik.http.middlewares.app.retry.initialinterval=\"500ms\" dhh/app:999",
       @app.run.join(" ")
   end
 
@@ -30,7 +30,7 @@ class CommandsAppTest < ActiveSupport::TestCase
     @config[:healthcheck] = { "path" => "/healthz" }
 
     assert_equal \
-      "docker run --detach --restart unless-stopped --log-opt max-size=10m --name app-999 -e RAILS_MASTER_KEY=\"456\" --label service=\"app\" --label role=\"web\" --label traefik.http.routers.app.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.services.app.loadbalancer.healthcheck.path=\"/healthz\" --label traefik.http.services.app.loadbalancer.healthcheck.interval=\"1s\" --label traefik.http.middlewares.app.retry.attempts=\"3\" --label traefik.http.middlewares.app.retry.initialinterval=\"500ms\" dhh/app:999",
+      "docker run --detach --restart unless-stopped --log-opt max-size=10m --name app-999 -e RAILS_MASTER_KEY=\"456\" --label service=\"app\" --label role=\"web\" --label traefik.http.routers.app.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.services.app.loadbalancer.healthcheck.path=\"/healthz\" --label traefik.http.services.app.loadbalancer.healthcheck.interval=\"1s\" --label traefik.http.middlewares.app.retry.attempts=\"5\" --label traefik.http.middlewares.app.retry.initialinterval=\"500ms\" dhh/app:999",
       @app.run.join(" ")
   end
 

test/configuration/role_test.rb

@@ -42,7 +42,7 @@ class ConfigurationRoleTest < ActiveSupport::TestCase
   end
 
   test "special label args for web" do
-    assert_equal [ "--label", "service=\"app\"", "--label", "role=\"web\"", "--label", "traefik.http.routers.app.rule=\"PathPrefix(\\`/\\`)\"", "--label", "traefik.http.services.app.loadbalancer.healthcheck.path=\"/up\"", "--label", "traefik.http.services.app.loadbalancer.healthcheck.interval=\"1s\"", "--label", "traefik.http.middlewares.app.retry.attempts=\"3\"", "--label", "traefik.http.middlewares.app.retry.initialinterval=\"500ms\""], @config.role(:web).label_args
+    assert_equal [ "--label", "service=\"app\"", "--label", "role=\"web\"", "--label", "traefik.http.routers.app.rule=\"PathPrefix(\\`/\\`)\"", "--label", "traefik.http.services.app.loadbalancer.healthcheck.path=\"/up\"", "--label", "traefik.http.services.app.loadbalancer.healthcheck.interval=\"1s\"", "--label", "traefik.http.middlewares.app.retry.attempts=\"5\"", "--label", "traefik.http.middlewares.app.retry.initialinterval=\"500ms\""], @config.role(:web).label_args
   end
 
   test "custom labels" do
@@ -66,7 +66,7 @@ class ConfigurationRoleTest < ActiveSupport::TestCase
       c[:servers]["beta"] = { "traefik" => "true", "hosts" => [ "1.1.1.5" ] }
     })
 
-    assert_equal [ "--label", "service=\"app\"", "--label", "role=\"beta\"", "--label", "traefik.http.routers.app.rule=\"PathPrefix(\\`/\\`)\"", "--label", "traefik.http.services.app.loadbalancer.healthcheck.path=\"/up\"", "--label", "traefik.http.services.app.loadbalancer.healthcheck.interval=\"1s\"", "--label", "traefik.http.middlewares.app.retry.attempts=\"3\"", "--label", "traefik.http.middlewares.app.retry.initialinterval=\"500ms\"" ], config.role(:beta).label_args
+    assert_equal [ "--label", "service=\"app\"", "--label", "role=\"beta\"", "--label", "traefik.http.routers.app.rule=\"PathPrefix(\\`/\\`)\"", "--label", "traefik.http.services.app.loadbalancer.healthcheck.path=\"/up\"", "--label", "traefik.http.services.app.loadbalancer.healthcheck.interval=\"1s\"", "--label", "traefik.http.middlewares.app.retry.attempts=\"5\"", "--label", "traefik.http.middlewares.app.retry.initialinterval=\"500ms\"" ], config.role(:beta).label_args
   end
 
   test "env overwritten by role" do

test/fixtures/deploy_with_accessories.yml

@@ -27,3 +27,5 @@ accessories:
     port: 6379
     directories:
       - data:/data
+
+readiness_delay: 0