Bump version for 1.6.0

Handle corrupt git clones

Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    kamal (1.5.2)
+    kamal (1.6.0)
       activesupport (>= 7.0)
       base64 (~> 0.2)
       bcrypt_pbkdf (~> 1.0)

README.md

@@ -1,6 +1,6 @@
 # Kamal: Deploy web apps anywhere
 
-From bare metal to cloud VMs, deploy web apps anywhere with zero downtime. Kamal uses parachute for zero-downtime deployments. Works seamlessly across multiple hosts, using SSHKit to execute commands. Originally built for Rails apps, Kamal will work with any type of web app that can be containerized with Docker.
+From bare metal to cloud VMs, deploy web apps anywhere with zero downtime. Kamal has the dynamic reverse-proxy Traefik hold requests while a new app container is started and the old one is stopped. Works seamlessly across multiple hosts, using SSHKit to execute commands. Originally built for Rails apps, Kamal will work with any type of web app that can be containerized with Docker.
 
 ➡️ See [kamal-deploy.org](https://kamal-deploy.org) for documentation on [installation](https://kamal-deploy.org/docs/installation), [configuration](https://kamal-deploy.org/docs/configuration), and [commands](https://kamal-deploy.org/docs/commands).
 

lib/kamal/cli.rb

@@ -1,7 +1,6 @@
 module Kamal::Cli
   class LockError < StandardError; end
   class HookError < StandardError; end
-  class BootError < StandardError; end
 end
 
 # SSHKit uses instance eval, so we need a global const for ergonomics

lib/kamal/cli/app.rb

@@ -9,16 +9,16 @@ class Kamal::Cli::App < Kamal::Cli::Base
         # Assets are prepared in a separate step to ensure they are on all hosts before booting
         on(KAMAL.hosts) do
           KAMAL.roles_on(host).each do |role|
-            PrepareAssets.new(host, role, self).run
+            Kamal::Cli::App::PrepareAssets.new(host, role, self).run
           end
         end
 
         # Primary hosts and roles are returned first, so they can open the barrier
-        barrier = Barrier.new if KAMAL.roles.many?
+        barrier = Kamal::Cli::Healthcheck::Barrier.new if KAMAL.roles.many?
 
         on(KAMAL.hosts, **KAMAL.boot_strategy) do |host|
           KAMAL.roles_on(host).each do |role|
-            Boot.new(host, role, self, version, barrier).run
+            Kamal::Cli::App::Boot.new(host, role, self, version, barrier).run
           end
         end
 
@@ -38,17 +38,8 @@ class Kamal::Cli::App < Kamal::Cli::Base
         roles = KAMAL.roles_on(host)
 
         roles.each do |role|
-          app = KAMAL.app(role: role, host: host)
           execute *KAMAL.auditor.record("Started app version #{KAMAL.config.version}"), verbosity: :debug
-          execute *app.start, raise_on_non_zero_exit: false
+          execute *KAMAL.app(role: role, host: host).start, raise_on_non_zero_exit: false
-
-          if role.running_proxy?
-            version = capture_with_info(*app.current_running_version, raise_on_non_zero_exit: false).strip
-            endpoint = capture_with_info(*app.container_endpoint(version: version)).strip
-            raise Kamal::Cli::BootError, "Failed to get endpoint for #{role} on #{host}, did the container boot?" if endpoint.empty?
-
-            execute *KAMAL.proxy.deploy(role.container_prefix, target: endpoint)
-          end
         end
       end
     end
@@ -61,19 +52,8 @@ class Kamal::Cli::App < Kamal::Cli::Base
         roles = KAMAL.roles_on(host)
 
         roles.each do |role|
-          app = KAMAL.app(role: role, host: host)
-          version = capture_with_info(*app.current_running_version, raise_on_non_zero_exit: false).strip
-
           execute *KAMAL.auditor.record("Stopped app", role: role), verbosity: :debug
-
+          execute *KAMAL.app(role: role, host: host).stop, raise_on_non_zero_exit: false
-          if role.running_proxy?
-            endpoint = capture_with_info(*app.container_endpoint(version: version)).strip
-            if endpoint.present?
-              execute *KAMAL.proxy.remove(role.container_prefix, target: endpoint), raise_on_non_zero_exit: false
-            end
-          end
-
-          execute *app.stop, raise_on_non_zero_exit: false
         end
       end
     end

lib/kamal/cli/app/boot.rb

@@ -1,7 +1,7 @@
 class Kamal::Cli::App::Boot
   attr_reader :host, :role, :version, :barrier, :sshkit
   delegate :execute, :capture_with_info, :capture_with_pretty_json, :info, :error, to: :sshkit
-  delegate :assets?, :running_proxy?, to: :role
+  delegate :uses_cord?, :assets?, :running_traefik?, to: :role
 
   def initialize(host, role, sshkit, version, barrier)
     @host = host
@@ -45,13 +45,11 @@ class Kamal::Cli::App::Boot
 
     def start_new_version
       audit "Booted app version #{version}"
+
+      execute *app.tie_cord(role.cord_host_file) if uses_cord?
       hostname = "#{host.to_s[0...51].gsub(/\.+$/, '')}-#{SecureRandom.hex(6)}"
       execute *app.run(hostname: hostname)
-      if running_proxy?
+      Kamal::Cli::Healthcheck::Poller.wait_for_healthy(pause_after_ready: true) { capture_with_info(*app.status(version: version)) }
-        endpoint = capture_with_info(*app.container_endpoint(version: version)).strip
-        raise Kamal::Cli::BootError, "Failed to get endpoint for #{role} on #{host}, did the container boot?" if endpoint.empty?
-        execute *KAMAL.proxy.deploy(role.container_prefix, target: endpoint)
-      end
     end
 
     def stop_new_version
@@ -59,7 +57,16 @@ class Kamal::Cli::App::Boot
     end
 
     def stop_old_version(version)
+      if uses_cord?
+        cord = capture_with_info(*app.cord(version: version), raise_on_non_zero_exit: false).strip
+        if cord.present?
+          execute *app.cut_cord(cord)
+          Kamal::Cli::Healthcheck::Poller.wait_for_unhealthy(pause_after_ready: true) { capture_with_info(*app.status(version: version)) }
+        end
+      end
+
       execute *app.stop(version: version), raise_on_non_zero_exit: false
+
       execute *app.clean_up_assets if assets?
     end
 
@@ -73,7 +80,7 @@ class Kamal::Cli::App::Boot
       info "Waiting for the first healthy #{KAMAL.primary_role} container before booting #{role} on #{host}..."
       barrier.wait
       info "First #{KAMAL.primary_role} container is healthy, booting #{role} on #{host}..."
-    rescue Kamal::Cli::BootError
+    rescue Kamal::Cli::Healthcheck::Error
       info "First #{KAMAL.primary_role} container is unhealthy, not booting #{role} on #{host}"
       raise
     end
@@ -82,6 +89,7 @@ class Kamal::Cli::App::Boot
       if barrier.close
         info "First #{KAMAL.primary_role} container is unhealthy on #{host}, not booting other roles"
         error capture_with_info(*app.logs(version: version))
+        error capture_with_info(*app.container_health_log(version: version))
       end
     end
 

lib/kamal/cli/build.rb

@@ -23,7 +23,9 @@ class Kamal::Cli::Build < Kamal::Cli::Base
         say "Building from a local git clone, so ignoring these uncommitted changes:\n #{uncommitted_changes}", :yellow
       end
 
-      prepare_clone
+      run_locally do
+        Clone.new(self).prepare
+      end
     elsif uncommitted_changes.present?
       say "Building with uncommitted changes:\n #{uncommitted_changes}", :yellow
     end
@@ -126,23 +128,4 @@ class Kamal::Cli::Build < Kamal::Cli::Base
         end
       end
     end
-
-    def prepare_clone
-      run_locally do
-        begin
-          info "Cloning repo into build directory `#{KAMAL.config.builder.build_directory}`..."
-
-          execute *KAMAL.builder.create_clone_directory
-          execute *KAMAL.builder.clone
-        rescue SSHKit::Command::Failed => e
-          if e.message =~ /already exists and is not an empty directory/
-            info "Resetting local clone as `#{KAMAL.config.builder.build_directory}` already exists..."
-
-            KAMAL.builder.clone_reset_steps.each { |step| execute *step }
-          else
-            raise
-          end
-        end
-      end
-    end
 end

lib/kamal/cli/build/clone.rb

@@ -0,0 +1,61 @@
+require "uri"
+
+class Kamal::Cli::Build::Clone
+  attr_reader :sshkit
+  delegate :info, :error, :execute, :capture_with_info, to: :sshkit
+
+  def initialize(sshkit)
+    @sshkit = sshkit
+  end
+
+  def prepare
+    begin
+      clone_repo
+    rescue SSHKit::Command::Failed => e
+      if e.message =~ /already exists and is not an empty directory/
+        reset
+      else
+        raise Kamal::Cli::Build::BuildError, "Failed to clone repo: #{e.message}"
+      end
+    end
+
+    validate!
+  rescue Kamal::Cli::Build::BuildError => e
+    error "Error preparing clone: #{e.message}, deleting and retrying..."
+
+    FileUtils.rm_rf KAMAL.config.builder.clone_directory
+    clone_repo
+    validate!
+  end
+
+  private
+    def clone_repo
+      info "Cloning repo into build directory `#{KAMAL.config.builder.build_directory}`..."
+
+      FileUtils.mkdir_p KAMAL.config.builder.clone_directory
+      execute *KAMAL.builder.clone
+    end
+
+    def reset
+      info "Resetting local clone as `#{KAMAL.config.builder.build_directory}` already exists..."
+
+      KAMAL.builder.clone_reset_steps.each { |step| execute *step }
+    rescue SSHKit::Command::Failed => e
+      raise Kamal::Cli::Build::BuildError, "Failed to clone repo: #{e.message}"
+    end
+
+    def validate!
+      status = capture_with_info(*KAMAL.builder.clone_status).strip
+
+      unless status.empty?
+        raise Kamal::Cli::Build::BuildError, "Clone in #{KAMAL.config.builder.build_directory} is dirty, #{status}"
+      end
+
+      revision = capture_with_info(*KAMAL.builder.clone_revision).strip
+      if revision != Kamal::Git.revision
+        raise Kamal::Cli::Build::BuildError, "Clone in #{KAMAL.config.builder.build_directory} is not on the correct revision, expected `#{Kamal::Git.revision}` but got `#{revision}`"
+      end
+    rescue SSHKit::Command::Failed => e
+      raise Kamal::Cli::Build::BuildError, "Failed to validate clone: #{e.message}"
+    end
+end

lib/kamal/cli/env.rb

@@ -13,6 +13,11 @@ class Kamal::Cli::Env < Kamal::Cli::Base
         end
       end
 
+      on(KAMAL.traefik_hosts) do
+        execute *KAMAL.traefik.make_env_directory
+        upload! KAMAL.traefik.env.secrets_io, KAMAL.traefik.env.secrets_file, mode: 400
+      end
+
       on(KAMAL.accessory_hosts) do
         KAMAL.accessories_on(host).each do |accessory|
           accessory_config = KAMAL.config.accessory(accessory)
@@ -34,6 +39,10 @@ class Kamal::Cli::Env < Kamal::Cli::Base
         end
       end
 
+      on(KAMAL.traefik_hosts) do
+        execute *KAMAL.traefik.remove_env_file
+      end
+
       on(KAMAL.accessory_hosts) do
         KAMAL.accessories_on(host).each do |accessory|
           accessory_config = KAMAL.config.accessory(accessory)

lib/kamal/cli/healthcheck/barrier.rb

@@ -1,4 +1,4 @@
-class Kamal::Cli::App::Barrier
+class Kamal::Cli::Healthcheck::Barrier
   def initialize
     @ivar = Concurrent::IVar.new
   end
@@ -13,7 +13,7 @@ class Kamal::Cli::App::Barrier
 
   def wait
     unless opened?
-      raise Kamal::Cli::BootError.new("Halted at barrier")
+      raise Kamal::Cli::Healthcheck::Error.new("Halted at barrier")
     end
   end
 

lib/kamal/cli/healthcheck/error.rb

@@ -0,0 +1,2 @@
+class Kamal::Cli::Healthcheck::Error < StandardError
+end

lib/kamal/cli/healthcheck/poller.rb

@@ -0,0 +1,63 @@
+module Kamal::Cli::Healthcheck::Poller
+  extend self
+
+  TRAEFIK_UPDATE_DELAY = 5
+
+
+  def wait_for_healthy(pause_after_ready: false, &block)
+    attempt = 1
+    max_attempts = KAMAL.config.healthcheck["max_attempts"]
+
+    begin
+      case status = block.call
+      when "healthy"
+        sleep TRAEFIK_UPDATE_DELAY if pause_after_ready
+      when "running" # No health check configured
+        sleep KAMAL.config.readiness_delay if pause_after_ready
+      else
+        raise Kamal::Cli::Healthcheck::Error, "container not ready (#{status})"
+      end
+    rescue Kamal::Cli::Healthcheck::Error => e
+      if attempt <= max_attempts
+        info "#{e.message}, retrying in #{attempt}s (attempt #{attempt}/#{max_attempts})..."
+        sleep attempt
+        attempt += 1
+        retry
+      else
+        raise
+      end
+    end
+
+    info "Container is healthy!"
+  end
+
+  def wait_for_unhealthy(pause_after_ready: false, &block)
+    attempt = 1
+    max_attempts = KAMAL.config.healthcheck["max_attempts"]
+
+    begin
+      case status = block.call
+      when "unhealthy"
+        sleep TRAEFIK_UPDATE_DELAY if pause_after_ready
+      else
+        raise Kamal::Cli::Healthcheck::Error, "container not unhealthy (#{status})"
+      end
+    rescue Kamal::Cli::Healthcheck::Error => e
+      if attempt <= max_attempts
+        info "#{e.message}, retrying in #{attempt}s (attempt #{attempt}/#{max_attempts})..."
+        sleep attempt
+        attempt += 1
+        retry
+      else
+        raise
+      end
+    end
+
+    info "Container is unhealthy!"
+  end
+
+  private
+    def info(message)
+      SSHKit.config.output.info(message)
+    end
+end

lib/kamal/cli/main.rb

@@ -38,8 +38,8 @@ class Kamal::Cli::Main < Kamal::Cli::Base
       with_lock do
         run_hook "pre-deploy"
 
-        say "Ensure proxy is running...", :magenta
+        say "Ensure Traefik is running...", :magenta
-        invoke "kamal:cli:proxy:boot", [], invoke_options
+        invoke "kamal:cli:traefik:boot", [], invoke_options
 
         say "Detect stale containers...", :magenta
         invoke "kamal:cli:app:stale_containers", [], invoke_options.merge(stop: true)
@@ -54,7 +54,7 @@ class Kamal::Cli::Main < Kamal::Cli::Base
     run_hook "post-deploy", runtime: runtime.round
   end
 
-  desc "redeploy", "Deploy app to servers without bootstrapping servers, starting proxy, pruning, and registry login"
+  desc "redeploy", "Deploy app to servers without bootstrapping servers, starting Traefik, pruning, and registry login"
   option :skip_push, aliases: "-P", type: :boolean, default: false, desc: "Skip image build and push"
   def redeploy
     runtime = print_runtime do
@@ -107,7 +107,7 @@ class Kamal::Cli::Main < Kamal::Cli::Base
 
   desc "details", "Show details about all containers"
   def details
-    invoke "kamal:cli:proxy:details"
+    invoke "kamal:cli:traefik:details"
     invoke "kamal:cli:app:details"
     invoke "kamal:cli:accessory:details", [ "all" ]
   end
@@ -189,12 +189,12 @@ class Kamal::Cli::Main < Kamal::Cli::Base
     end
   end
 
-  desc "remove", "Remove proxy, app, accessories, and registry session from servers"
+  desc "remove", "Remove Traefik, app, accessories, and registry session from servers"
   option :confirmed, aliases: "-y", type: :boolean, default: false, desc: "Proceed without confirmation question"
   def remove
     confirming "This will remove all containers and images. Are you sure?" do
       with_lock do
-        invoke "kamal:cli:proxy:remove", [], options.without(:confirmed)
+        invoke "kamal:cli:traefik:remove", [], options.without(:confirmed)
         invoke "kamal:cli:app:remove", [], options.without(:confirmed)
         invoke "kamal:cli:accessory:remove", [ "all" ], options
         invoke "kamal:cli:registry:logout", [], options.without(:confirmed)
@@ -231,8 +231,8 @@ class Kamal::Cli::Main < Kamal::Cli::Base
   desc "server", "Bootstrap servers with curl and Docker"
   subcommand "server", Kamal::Cli::Server
 
-  desc "proxy", "Manage load balancer proxy"
+  desc "traefik", "Manage Traefik load balancer"
-  subcommand "proxy", Kamal::Cli::Proxy
+  subcommand "traefik", Kamal::Cli::Traefik
 
   private
     def container_available?(version)

lib/kamal/cli/proxy.rb

@@ -1,120 +0,0 @@
-class Kamal::Cli::Proxy < Kamal::Cli::Base
-  desc "boot", "Boot proxy on servers"
-  def boot
-    with_lock do
-      on(KAMAL.proxy_hosts) do
-        execute *KAMAL.registry.login
-        execute *KAMAL.proxy.start_or_run
-      end
-    end
-  end
-
-  desc "reboot", "Reboot proxy on servers (stop container, remove container, start new container)"
-  option :rolling, type: :boolean, default: false, desc: "Reboot proxy on hosts in sequence, rather than in parallel"
-  option :confirmed, aliases: "-y", type: :boolean, default: false, desc: "Proceed without confirmation question"
-  def reboot
-    confirming "This will cause a brief outage on each host. Are you sure?" do
-      with_lock do
-        host_groups = options[:rolling] ? KAMAL.proxy_hosts : [ KAMAL.proxy_hosts ]
-        host_groups.each do |hosts|
-          host_list = Array(hosts).join(",")
-          run_hook "pre-proxy-reboot", hosts: host_list
-          on(hosts) do
-            execute *KAMAL.auditor.record("Rebooted proxy"), verbosity: :debug
-            execute *KAMAL.registry.login
-            execute *KAMAL.proxy.stop, raise_on_non_zero_exit: false
-            execute *KAMAL.proxy.remove_container
-            execute *KAMAL.proxy.run
-          end
-          run_hook "post-proxy-reboot", hosts: host_list
-        end
-      end
-    end
-  end
-
-  desc "start", "Start existing proxy container on servers"
-  def start
-    with_lock do
-      on(KAMAL.proxy_hosts) do
-        execute *KAMAL.auditor.record("Started proxy"), verbosity: :debug
-        execute *KAMAL.proxy.start
-      end
-    end
-  end
-
-  desc "stop", "Stop existing proxy container on servers"
-  def stop
-    with_lock do
-      on(KAMAL.proxy_hosts) do
-        execute *KAMAL.auditor.record("Stopped proxy"), verbosity: :debug
-        execute *KAMAL.proxy.stop, raise_on_non_zero_exit: false
-      end
-    end
-  end
-
-  desc "restart", "Restart existing proxy container on servers"
-  def restart
-    with_lock do
-      stop
-      start
-    end
-  end
-
-  desc "details", "Show details about proxy container from servers"
-  def details
-    on(KAMAL.proxy_hosts) { |host| puts_by_host host, capture_with_info(*KAMAL.proxy.info), type: "Proxy" }
-  end
-
-  desc "logs", "Show log lines from proxy on servers"
-  option :since, aliases: "-s", desc: "Show logs since timestamp (e.g. 2013-01-02T13:23:37Z) or relative (e.g. 42m for 42 minutes)"
-  option :lines, type: :numeric, aliases: "-n", desc: "Number of log lines to pull from each server"
-  option :grep, aliases: "-g", desc: "Show lines with grep match only (use this to fetch specific requests by id)"
-  option :follow, aliases: "-f", desc: "Follow logs on primary server (or specific host set by --hosts)"
-  def logs
-    grep = options[:grep]
-
-    if options[:follow]
-      run_locally do
-        info "Following logs on #{KAMAL.primary_host}..."
-        info KAMAL.proxy.follow_logs(host: KAMAL.primary_host, grep: grep)
-        exec KAMAL.proxy.follow_logs(host: KAMAL.primary_host, grep: grep)
-      end
-    else
-      since = options[:since]
-      lines = options[:lines].presence || ((since || grep) ? nil : 100) # Default to 100 lines if since or grep isn't set
-
-      on(KAMAL.proxy_hosts) do |host|
-        puts_by_host host, capture(*KAMAL.proxy.logs(since: since, lines: lines, grep: grep)), type: "Proxy"
-      end
-    end
-  end
-
-  desc "remove", "Remove proxy container and image from servers"
-  def remove
-    with_lock do
-      stop
-      remove_container
-      remove_image
-    end
-  end
-
-  desc "remove_container", "Remove proxy container from servers", hide: true
-  def remove_container
-    with_lock do
-      on(KAMAL.proxy_hosts) do
-        execute *KAMAL.auditor.record("Removed proxy container"), verbosity: :debug
-        execute *KAMAL.proxy.remove_container
-      end
-    end
-  end
-
-  desc "remove_image", "Remove proxy image from servers", hide: true
-  def remove_image
-    with_lock do
-      on(KAMAL.proxy_hosts) do
-        execute *KAMAL.auditor.record("Removed proxy image"), verbosity: :debug
-        execute *KAMAL.proxy.remove_image
-      end
-    end
-  end
-end

lib/kamal/cli/prune.rb

@@ -28,6 +28,7 @@ class Kamal::Cli::Prune < Kamal::Cli::Base
       on(KAMAL.hosts) do
         execute *KAMAL.auditor.record("Pruned containers"), verbosity: :debug
         execute *KAMAL.prune.app_containers(retain: retain)
+        execute *KAMAL.prune.healthcheck_containers
       end
     end
   end

lib/kamal/cli/templates/deploy.yml

@@ -63,6 +63,17 @@ registry:
 #     directories:
 #       - data:/data
 
+# Configure custom arguments for Traefik. Be sure to reboot traefik when you modify it.
+# traefik:
+#   args:
+#     accesslog: true
+#     accesslog.format: json
+
+# Configure a custom healthcheck (default is /up on port 3000)
+# healthcheck:
+#   path: /healthz
+#   port: 4000
+
 # Bridge fingerprinted assets, like JS and CSS, between versions to avoid
 # hitting 404 on in-flight requests. Combines all files from new and old
 # version inside the asset_path.

lib/kamal/cli/templates/sample_hooks/post-proxy-reboot.sample

@@ -1,3 +0,0 @@
-#!/bin/sh
-
-echo "Rebooted proxy on $KAMAL_HOSTS"

lib/kamal/cli/templates/sample_hooks/post-traefik-reboot.sample

@@ -0,0 +1,3 @@
+#!/bin/sh
+
+echo "Rebooted Traefik on $KAMAL_HOSTS"

lib/kamal/cli/templates/sample_hooks/pre-proxy-reboot.sample

@@ -1,3 +0,0 @@
-#!/bin/sh
-
-echo "Rebooting proxy on $KAMAL_HOSTS..."

lib/kamal/cli/templates/sample_hooks/pre-traefik-reboot.sample

@@ -0,0 +1,3 @@
+#!/bin/sh
+
+echo "Rebooting Traefik on $KAMAL_HOSTS..."

lib/kamal/cli/traefik.rb

@@ -0,0 +1,120 @@
+class Kamal::Cli::Traefik < Kamal::Cli::Base
+  desc "boot", "Boot Traefik on servers"
+  def boot
+    with_lock do
+      on(KAMAL.traefik_hosts) do
+        execute *KAMAL.registry.login
+        execute *KAMAL.traefik.start_or_run
+      end
+    end
+  end
+
+  desc "reboot", "Reboot Traefik on servers (stop container, remove container, start new container)"
+  option :rolling, type: :boolean, default: false, desc: "Reboot traefik on hosts in sequence, rather than in parallel"
+  option :confirmed, aliases: "-y", type: :boolean, default: false, desc: "Proceed without confirmation question"
+  def reboot
+    confirming "This will cause a brief outage on each host. Are you sure?" do
+      with_lock do
+        host_groups = options[:rolling] ? KAMAL.traefik_hosts : [ KAMAL.traefik_hosts ]
+        host_groups.each do |hosts|
+          host_list = Array(hosts).join(",")
+          run_hook "pre-traefik-reboot", hosts: host_list
+          on(hosts) do
+            execute *KAMAL.auditor.record("Rebooted traefik"), verbosity: :debug
+            execute *KAMAL.registry.login
+            execute *KAMAL.traefik.stop, raise_on_non_zero_exit: false
+            execute *KAMAL.traefik.remove_container
+            execute *KAMAL.traefik.run
+          end
+          run_hook "post-traefik-reboot", hosts: host_list
+        end
+      end
+    end
+  end
+
+  desc "start", "Start existing Traefik container on servers"
+  def start
+    with_lock do
+      on(KAMAL.traefik_hosts) do
+        execute *KAMAL.auditor.record("Started traefik"), verbosity: :debug
+        execute *KAMAL.traefik.start
+      end
+    end
+  end
+
+  desc "stop", "Stop existing Traefik container on servers"
+  def stop
+    with_lock do
+      on(KAMAL.traefik_hosts) do
+        execute *KAMAL.auditor.record("Stopped traefik"), verbosity: :debug
+        execute *KAMAL.traefik.stop, raise_on_non_zero_exit: false
+      end
+    end
+  end
+
+  desc "restart", "Restart existing Traefik container on servers"
+  def restart
+    with_lock do
+      stop
+      start
+    end
+  end
+
+  desc "details", "Show details about Traefik container from servers"
+  def details
+    on(KAMAL.traefik_hosts) { |host| puts_by_host host, capture_with_info(*KAMAL.traefik.info), type: "Traefik" }
+  end
+
+  desc "logs", "Show log lines from Traefik on servers"
+  option :since, aliases: "-s", desc: "Show logs since timestamp (e.g. 2013-01-02T13:23:37Z) or relative (e.g. 42m for 42 minutes)"
+  option :lines, type: :numeric, aliases: "-n", desc: "Number of log lines to pull from each server"
+  option :grep, aliases: "-g", desc: "Show lines with grep match only (use this to fetch specific requests by id)"
+  option :follow, aliases: "-f", desc: "Follow logs on primary server (or specific host set by --hosts)"
+  def logs
+    grep = options[:grep]
+
+    if options[:follow]
+      run_locally do
+        info "Following logs on #{KAMAL.primary_host}..."
+        info KAMAL.traefik.follow_logs(host: KAMAL.primary_host, grep: grep)
+        exec KAMAL.traefik.follow_logs(host: KAMAL.primary_host, grep: grep)
+      end
+    else
+      since = options[:since]
+      lines = options[:lines].presence || ((since || grep) ? nil : 100) # Default to 100 lines if since or grep isn't set
+
+      on(KAMAL.traefik_hosts) do |host|
+        puts_by_host host, capture(*KAMAL.traefik.logs(since: since, lines: lines, grep: grep)), type: "Traefik"
+      end
+    end
+  end
+
+  desc "remove", "Remove Traefik container and image from servers"
+  def remove
+    with_lock do
+      stop
+      remove_container
+      remove_image
+    end
+  end
+
+  desc "remove_container", "Remove Traefik container from servers", hide: true
+  def remove_container
+    with_lock do
+      on(KAMAL.traefik_hosts) do
+        execute *KAMAL.auditor.record("Removed traefik container"), verbosity: :debug
+        execute *KAMAL.traefik.remove_container
+      end
+    end
+  end
+
+  desc "remove_image", "Remove Traefik image from servers", hide: true
+  def remove_image
+    with_lock do
+      on(KAMAL.traefik_hosts) do
+        execute *KAMAL.auditor.record("Removed traefik image"), verbosity: :debug
+        execute *KAMAL.traefik.remove_image
+      end
+    end
+  end
+end

lib/kamal/commander.rb

@@ -3,7 +3,7 @@ require "active_support/core_ext/module/delegation"
 
 class Kamal::Commander
   attr_accessor :verbosity, :holding_lock, :connected
-  delegate :hosts, :roles, :primary_host, :primary_role, :roles_on, :proxy_hosts, :accessory_hosts, to: :specifics
+  delegate :hosts, :roles, :primary_host, :primary_role, :roles_on, :traefik_hosts, :accessory_hosts, to: :specifics
 
   def initialize
     self.verbosity = :info
@@ -85,6 +85,10 @@ class Kamal::Commander
     @docker ||= Kamal::Commands::Docker.new(config)
   end
 
+  def healthcheck
+    @healthcheck ||= Kamal::Commands::Healthcheck.new(config)
+  end
+
   def hook
     @hook ||= Kamal::Commands::Hook.new(config)
   end
@@ -105,8 +109,8 @@ class Kamal::Commander
     @server ||= Kamal::Commands::Server.new(config)
   end
 
-  def proxy
+  def traefik
-    @proxy ||= Kamal::Commands::Proxy.new(config)
+    @traefik ||= Kamal::Commands::Traefik.new(config)
   end
 
 

lib/kamal/commander/specifics.rb

@@ -18,8 +18,8 @@ class Kamal::Commander::Specifics
     roles.select { |role| role.hosts.include?(host.to_s) }
   end
 
-  def proxy_hosts
+  def traefik_hosts
-    config.proxy_hosts & specified_hosts
+    config.traefik_hosts & specified_hosts
   end
 
   def accessory_hosts

lib/kamal/commands/app.rb

@@ -1,5 +1,5 @@
 class Kamal::Commands::App < Kamal::Commands::Base
-  include Assets, Containers, Execution, Images, Logging
+  include Assets, Containers, Cord, Execution, Images, Logging
 
   ACTIVE_DOCKER_STATUSES = [ :running, :restarting ]
 
@@ -20,6 +20,7 @@ class Kamal::Commands::App < Kamal::Commands::Base
       "-e", "KAMAL_CONTAINER_NAME=\"#{container_name}\"",
       "-e", "KAMAL_VERSION=\"#{config.version}\"",
       *role.env_args(host),
+      *role.health_check_args,
       *role.logging_args,
       *config.volume_args,
       *role.asset_volume_args,
@@ -56,10 +57,6 @@ class Kamal::Commands::App < Kamal::Commands::Base
     container_id_for(container_name: container_name(version), only_running: only_running)
   end
 
-  def container_name(version = nil)
-    [ role.container_prefix, version || config.version ].compact.join("-")
-  end
-
   def current_running_version
     pipe \
       current_running_container(format: "--format '{{.Names}}'"),

lib/kamal/commands/app/containers.rb

@@ -23,10 +23,9 @@ module Kamal::Commands::App::Containers
     docker :container, :prune, "--force", *filter_args
   end
 
-  def container_endpoint(version:)
+  def container_health_log(version:)
     pipe \
       container_id_for(container_name: container_name(version)),
-      xargs(docker(:inspect, "--format", "'{{.NetworkSettings.IPAddress}}{{range $k, $v := .NetworkSettings.Ports}}{{printf \":%s\" $k}}{{break}}{{end}}'")),
+      xargs(docker(:inspect, "--format", DOCKER_HEALTH_LOG_FORMAT))
-      [ :sed, "-e", "'s/\\/tcp$//'" ]
   end
 end

lib/kamal/commands/app/cord.rb

@@ -0,0 +1,22 @@
+module Kamal::Commands::App::Cord
+  def cord(version:)
+    pipe \
+      docker(:inspect, "-f '{{ range .Mounts }}{{printf \"%s %s\\n\" .Source .Destination}}{{ end }}'", container_name(version)),
+      [ :awk, "'$2 == \"#{role.cord_volume.container_path}\" {print $1}'" ]
+  end
+
+  def tie_cord(cord)
+    create_empty_file(cord)
+  end
+
+  def cut_cord(cord)
+    remove_directory(cord)
+  end
+
+  private
+    def create_empty_file(file)
+      chain \
+        make_directory_for(file),
+        [ :touch, file ]
+    end
+end

lib/kamal/commands/builder.rb

@@ -2,7 +2,8 @@ require "active_support/core_ext/string/filters"
 
 class Kamal::Commands::Builder < Kamal::Commands::Base
   delegate :create, :remove, :push, :clean, :pull, :info, :validate_image, to: :target
-  delegate :clone_directory, :build_directory, to: :"config.builder"
+
+  include Clone
 
   def name
     target.class.to_s.remove("Kamal::Commands::Builder::").underscore.inquiry
@@ -54,23 +55,6 @@ class Kamal::Commands::Builder < Kamal::Commands::Base
     end
   end
 
-  def create_clone_directory
-    make_directory clone_directory
-  end
-
-  def clone
-    git :clone, Kamal::Git.root, path: clone_directory
-  end
-
-  def clone_reset_steps
-    [
-      git(:remote, "set-url", :origin, Kamal::Git.root, path: build_directory),
-      git(:fetch, :origin, path: build_directory),
-      git(:reset, "--hard", Kamal::Git.revision, path: build_directory),
-      git(:clean, "-fdx", path: build_directory)
-    ]
-  end
-
   private
     def ensure_local_docker_installed
       docker "--version"

lib/kamal/commands/builder/clone.rb

@@ -0,0 +1,28 @@
+module Kamal::Commands::Builder::Clone
+  extend ActiveSupport::Concern
+
+  included do
+    delegate :clone_directory, :build_directory, to: :"config.builder"
+  end
+
+  def clone
+    git :clone, Kamal::Git.root, path: clone_directory
+  end
+
+  def clone_reset_steps
+    [
+      git(:remote, "set-url", :origin, Kamal::Git.root, path: build_directory),
+      git(:fetch, :origin, path: build_directory),
+      git(:reset, "--hard", Kamal::Git.revision, path: build_directory),
+      git(:clean, "-fdx", path: build_directory)
+    ]
+  end
+
+  def clone_status
+    git :status, "--porcelain", path: build_directory
+  end
+
+  def clone_revision
+    git :"rev-parse", :HEAD, path: build_directory
+  end
+end

lib/kamal/commands/proxy.rb

@@ -1,79 +0,0 @@
-class Kamal::Commands::Proxy < Kamal::Commands::Base
-  delegate :argumentize, :optionize, to: Kamal::Utils
-  delegate :container_name, to: :proxy_config
-
-  attr_reader :proxy_config
-
-  def initialize(config)
-    super
-    @proxy_config = config.proxy
-  end
-
-  def run
-    docker :run,
-      "--name", container_name,
-      "--detach",
-      "--restart", "unless-stopped",
-      *proxy_config.publish_args,
-      "--volume", "/var/run/docker.sock:/var/run/docker.sock",
-      "--volume", "#{container_name}:/root/.config/parachute",
-      *config.logging_args,
-      *proxy_config.docker_options_args,
-      proxy_config.image
-  end
-
-  def start
-    docker :container, :start, container_name
-  end
-
-  def stop
-    docker :container, :stop, container_name
-  end
-
-  def start_or_run
-    combine start, run, by: "||"
-  end
-
-  def deploy(service, target:)
-    optionize({ target: target })
-    docker :exec, container_name, :parachute, :deploy, service, *optionize({ target: target }), *proxy_config.deploy_command_args
-  end
-
-  def remove(service, target:)
-    docker :exec, container_name, :parachute, :remove, service, *optionize({ target: target })
-  end
-
-  def info
-    docker :ps, "--filter", "name=^#{container_name}$"
-  end
-
-  def logs(since: nil, lines: nil, grep: nil)
-    pipe \
-      docker(:logs, container_name, (" --since #{since}" if since), (" --tail #{lines}" if lines), "--timestamps", "2>&1"),
-      ("grep '#{grep}'" if grep)
-  end
-
-  def follow_logs(host:, grep: nil)
-    run_over_ssh pipe(
-      docker(:logs, container_name, "--timestamps", "--tail", "10", "--follow", "2>&1"),
-      (%(grep "#{grep}") if grep)
-    ).join(" "), host: host
-  end
-
-  def remove_container
-    docker :container, :prune, "--force", "--filter", container_filter
-  end
-
-  def remove_image
-    docker :image, :prune, "--all", "--force", "--filter", image_filter
-  end
-
-  private
-    def container_filter
-      "label=org.opencontainers.image.title=parachute"
-    end
-
-    def image_filter
-      "label=org.opencontainers.image.title=parachute"
-    end
-end

lib/kamal/commands/prune.rb

@@ -20,6 +20,10 @@ class Kamal::Commands::Prune < Kamal::Commands::Base
       "while read container_id; do docker rm $container_id; done"
   end
 
+  def healthcheck_containers
+    docker :container, :prune, "--force", *healthcheck_service_filter
+  end
+
   private
     def stopped_containers_filters
       [ "created", "exited", "dead" ].flat_map { |status| [ "--filter", "status=#{status}" ] }
@@ -35,4 +39,8 @@ class Kamal::Commands::Prune < Kamal::Commands::Base
     def service_filter
       [ "--filter", "label=service=#{config.service}" ]
     end
+
+    def healthcheck_service_filter
+      [ "--filter", "label=service=#{config.healthcheck_service}" ]
+    end
 end

lib/kamal/commands/traefik.rb

@@ -0,0 +1,124 @@
+class Kamal::Commands::Traefik < Kamal::Commands::Base
+  delegate :argumentize, :optionize, to: Kamal::Utils
+
+  DEFAULT_IMAGE = "traefik:v2.10"
+  CONTAINER_PORT = 80
+  DEFAULT_ARGS = {
+    "log.level" => "DEBUG"
+  }
+  DEFAULT_LABELS = {
+    # These ensure we serve a 502 rather than a 404 if no containers are available
+    "traefik.http.routers.catchall.entryPoints" => "http",
+    "traefik.http.routers.catchall.rule" => "PathPrefix(`/`)",
+    "traefik.http.routers.catchall.service" => "unavailable",
+    "traefik.http.routers.catchall.priority" => 1,
+    "traefik.http.services.unavailable.loadbalancer.server.port" => "0"
+  }
+
+  def run
+    docker :run, "--name traefik",
+      "--detach",
+      "--restart", "unless-stopped",
+      *publish_args,
+      "--volume", "/var/run/docker.sock:/var/run/docker.sock",
+      *env_args,
+      *config.logging_args,
+      *label_args,
+      *docker_options_args,
+      image,
+      "--providers.docker",
+      *cmd_option_args
+  end
+
+  def start
+    docker :container, :start, "traefik"
+  end
+
+  def stop
+    docker :container, :stop, "traefik"
+  end
+
+  def start_or_run
+    any start, run
+  end
+
+  def info
+    docker :ps, "--filter", "name=^traefik$"
+  end
+
+  def logs(since: nil, lines: nil, grep: nil)
+    pipe \
+      docker(:logs, "traefik", (" --since #{since}" if since), (" --tail #{lines}" if lines), "--timestamps", "2>&1"),
+      ("grep '#{grep}'" if grep)
+  end
+
+  def follow_logs(host:, grep: nil)
+    run_over_ssh pipe(
+      docker(:logs, "traefik", "--timestamps", "--tail", "10", "--follow", "2>&1"),
+      (%(grep "#{grep}") if grep)
+    ).join(" "), host: host
+  end
+
+  def remove_container
+    docker :container, :prune, "--force", "--filter", "label=org.opencontainers.image.title=Traefik"
+  end
+
+  def remove_image
+    docker :image, :prune, "--all", "--force", "--filter", "label=org.opencontainers.image.title=Traefik"
+  end
+
+  def port
+    "#{host_port}:#{CONTAINER_PORT}"
+  end
+
+  def env
+    Kamal::Configuration::Env.from_config \
+      config: config.traefik.fetch("env", {}),
+      secrets_file: File.join(config.host_env_directory, "traefik", "traefik.env")
+  end
+
+  def make_env_directory
+    make_directory(env.secrets_directory)
+  end
+
+  def remove_env_file
+    [ :rm, "-f", env.secrets_file ]
+  end
+
+  private
+    def publish_args
+      argumentize "--publish", port unless config.traefik["publish"] == false
+    end
+
+    def label_args
+      argumentize "--label", labels
+    end
+
+    def env_args
+      env.args
+    end
+
+    def labels
+      DEFAULT_LABELS.merge(config.traefik["labels"] || {})
+    end
+
+    def image
+      config.traefik.fetch("image") { DEFAULT_IMAGE }
+    end
+
+    def docker_options_args
+      optionize(config.traefik["options"] || {})
+    end
+
+    def cmd_option_args
+      if args = config.traefik["args"]
+        optionize DEFAULT_ARGS.merge(args), with: "="
+      else
+        optionize DEFAULT_ARGS, with: "="
+      end
+    end
+
+    def host_port
+      config.traefik["host_port"] || CONTAINER_PORT
+    end
+end

lib/kamal/configuration.rb

@@ -6,7 +6,7 @@ require "erb"
 require "net/ssh/proxy/jump"
 
 class Kamal::Configuration
-  delegate :service, :image, :port, :servers, :labels, :registry, :stop_wait_time, :hooks_path, :logging, to: :raw_config, allow_nil: true
+  delegate :service, :image, :servers, :labels, :registry, :stop_wait_time, :hooks_path, :logging, to: :raw_config, allow_nil: true
   delegate :argumentize, :optionize, to: Kamal::Utils
 
   attr_reader :destination, :raw_config
@@ -107,16 +107,16 @@ class Kamal::Configuration
     raw_config.allow_empty_roles
   end
 
-  def proxy_roles
+  def traefik_roles
-    roles.select(&:running_proxy?)
+    roles.select(&:running_traefik?)
   end
 
-  def proxy_role_names
+  def traefik_role_names
-    proxy_roles.flat_map(&:name)
+    traefik_roles.flat_map(&:name)
   end
 
-  def proxy_hosts
+  def traefik_hosts
-    proxy_roles.flat_map(&:hosts).uniq
+    traefik_roles.flat_map(&:hosts).uniq
   end
 
   def repository
@@ -174,8 +174,8 @@ class Kamal::Configuration
     Kamal::Configuration::Builder.new(config: self)
   end
 
-  def proxy
+  def traefik
-    Kamal::Configuration::Proxy.new(config: self)
+    raw_config.traefik || {}
   end
 
   def ssh
@@ -187,6 +187,14 @@ class Kamal::Configuration
   end
 
 
+  def healthcheck
+    { "path" => "/up", "port" => 3000, "max_attempts" => 7, "cord" => "/tmp/kamal-cord", "log_lines" => 50 }.merge(raw_config.healthcheck || {})
+  end
+
+  def healthcheck_service
+    [ "healthcheck", service, destination ].compact.join("-")
+  end
+
   def readiness_delay
     raw_config.readiness_delay || 7
   end
@@ -256,7 +264,8 @@ class Kamal::Configuration
       sshkit: sshkit.to_h,
       builder: builder.to_h,
       accessories: raw_config.accessories,
-      logging: logging_args
+      logging: logging_args,
+      healthcheck: healthcheck
     }.compact
   end
 

lib/kamal/configuration/proxy.rb

@@ -1,58 +0,0 @@
-class Kamal::Configuration::Proxy
-  DEFAULT_HTTP_PORT = 80
-  DEFAULT_HTTPS_PORT = 443
-  DEFAULT_IMAGE = "basecamp/parachute:latest"
-
-  delegate :argumentize, :optionize, to: Kamal::Utils
-
-  def initialize(config:)
-    @options = config.raw_config.proxy || {}
-  end
-
-  def image
-    options.fetch("image", DEFAULT_IMAGE)
-  end
-
-  def debug?
-    !!options[:debug]
-  end
-
-  def http_port
-    if options.key?(:http_port)
-      options[:http_port]
-    elsif !automatic_tls?
-      DEFAULT_HTTP_PORT
-    end
-  end
-
-  def https_port
-    if options.key?(:http_port)
-      options[:http_port]
-    elsif automatic_tls?
-      DEFAULT_HTTPS_PORT
-    end
-  end
-
-  def container_name
-    "parachute_#{http_port}_#{https_port}"
-  end
-
-  def docker_options_args
-    optionize(options.fetch("options", {}))
-  end
-
-  def publish_args
-    argumentize "--publish", *("#{http_port}:80" if http_port), *("#{https_port}:80" if https_port)
-  end
-
-  def deploy_options
-    options.fetch(:deploy, {})
-  end
-
-  def deploy_command_args
-    optionize deploy_options
-  end
-
-  private
-    attr_accessor :options
-end

lib/kamal/configuration/role.rb

@@ -1,4 +1,5 @@
 class Kamal::Configuration::Role
+  CORD_FILE = "cord"
   delegate :argumentize, :optionize, to: Kamal::Utils
 
   attr_accessor :name
@@ -34,7 +35,7 @@ class Kamal::Configuration::Role
   end
 
   def labels
-    default_labels.merge(custom_labels)
+    default_labels.merge(traefik_labels).merge(custom_labels)
   end
 
   def label_args
@@ -68,11 +69,37 @@ class Kamal::Configuration::Role
   end
 
 
-  def running_proxy?
+  def health_check_args(cord: true)
-    if specializations["proxy"].nil?
+    if health_check_cmd.present?
+      if cord && uses_cord?
+        optionize({ "health-cmd" => health_check_cmd_with_cord, "health-interval" => health_check_interval })
+          .concat(cord_volume.docker_args)
+      else
+        optionize({ "health-cmd" => health_check_cmd, "health-interval" => health_check_interval })
+      end
+    else
+      []
+    end
+  end
+
+  def health_check_cmd
+    health_check_options["cmd"] || http_health_check(port: health_check_options["port"], path: health_check_options["path"])
+  end
+
+  def health_check_cmd_with_cord
+    "(#{health_check_cmd}) && (stat #{cord_container_file} > /dev/null || exit 1)"
+  end
+
+  def health_check_interval
+    health_check_options["interval"] || "1s"
+  end
+
+
+  def running_traefik?
+    if specializations["traefik"].nil?
       primary?
     else
-      specializations["proxy"]
+      specializations["traefik"]
     end
   end
 
@@ -81,6 +108,35 @@ class Kamal::Configuration::Role
   end
 
 
+  def uses_cord?
+    running_traefik? && cord_volume && health_check_cmd.present?
+  end
+
+  def cord_host_directory
+    File.join config.run_directory_as_docker_volume, "cords", [ container_prefix, config.run_id ].join("-")
+  end
+
+  def cord_volume
+    if (cord = health_check_options["cord"])
+      @cord_volume ||= Kamal::Configuration::Volume.new \
+        host_path: File.join(config.run_directory, "cords", [ container_prefix, config.run_id ].join("-")),
+        container_path: cord
+    end
+  end
+
+  def cord_host_file
+    File.join cord_volume.host_path, CORD_FILE
+  end
+
+  def cord_container_directory
+    health_check_options.fetch("cord", nil)
+  end
+
+  def cord_container_file
+    File.join cord_volume.container_path, CORD_FILE
+  end
+
+
   def container_name(version = nil)
     [ container_prefix, version || config.version ].compact.join("-")
   end
@@ -95,7 +151,7 @@ class Kamal::Configuration::Role
   end
 
   def assets?
-    asset_path.present? && running_proxy?
+    asset_path.present? && running_traefik?
   end
 
   def asset_volume(version = nil)
@@ -146,6 +202,27 @@ class Kamal::Configuration::Role
       { "service" => config.service, "role" => name, "destination" => config.destination }
     end
 
+    def traefik_labels
+      if running_traefik?
+        {
+          # Setting a service property ensures that the generated service name will be consistent between versions
+          "traefik.http.services.#{traefik_service}.loadbalancer.server.scheme" => "http",
+
+          "traefik.http.routers.#{traefik_service}.rule" => "PathPrefix(`/`)",
+          "traefik.http.routers.#{traefik_service}.priority" => "2",
+          "traefik.http.middlewares.#{traefik_service}-retry.retry.attempts" => "5",
+          "traefik.http.middlewares.#{traefik_service}-retry.retry.initialinterval" => "500ms",
+          "traefik.http.routers.#{traefik_service}.middlewares" => "#{traefik_service}-retry@docker"
+        }
+      else
+        {}
+      end
+    end
+
+    def traefik_service
+      container_prefix
+    end
+
     def custom_labels
       Hash.new.tap do |labels|
         labels.merge!(config.labels) if config.labels.present?
@@ -171,4 +248,16 @@ class Kamal::Configuration::Role
         config: config.env,
         secrets_file: File.join(config.host_env_directory, "roles", "#{container_prefix}.env")
     end
+
+    def http_health_check(port:, path:)
+      "curl -f #{URI.join("http://localhost:#{port}", path)} || exit 1" if path.present? || port.present?
+    end
+
+    def health_check_options
+      @health_check_options ||= begin
+        options = specializations["healthcheck"] || {}
+        options = config.healthcheck.merge(options) if running_traefik?
+        options
+      end
+    end
 end

lib/kamal/version.rb

@@ -1,3 +1,3 @@
 module Kamal
-  VERSION = "1.5.2"
+  VERSION = "1.6.0"
 end

test/cli/app_test.rb

@@ -11,20 +11,28 @@ class CliAppTest < CliTestCase
   end
 
   test "boot will rename if same version is already running" do
+    Object.any_instance.stubs(:sleep)
     run_command("details") # Preheat Kamal const
 
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
       .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", raise_on_non_zero_exit: false)
       .returns("12345678") # running version
 
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
+      .returns("running") # health check
+
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
       .with(:sh, "-c", "'docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''{{.ID}}'\\'') ; docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting'", "|", :head, "-1", "|", "while read line; do echo ${line#app-web-}; done", raise_on_non_zero_exit: false)
       .returns("123") # old version
 
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{.NetworkSettings.IPAddress}}{{range $k, $v := .NetworkSettings.Ports}}{{printf \":%s\" $k}}{{break}}{{end}}'", "|", :sed, "-e", "'s/\\/tcp$//'")
+      .with(:docker, :inspect, "-f '{{ range .Mounts }}{{printf \"%s %s\\n\" .Source .Destination}}{{ end }}'", "app-web-123", "|", :awk, "'$2 == \"/tmp/kamal-cord\" {print $1}'", raise_on_non_zero_exit: false)
-      .returns("172.1.0.2:80")
+      .returns("cordfile") # old version
-      .at_least_once
+
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-123$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
+      .returns("unhealthy") # old version unhealthy
 
     run_command("boot").tap do |output|
       assert_match /Renaming container .* to .* as already deployed on 1.1.1.1/, output # Rename
@@ -57,16 +65,22 @@ class CliAppTest < CliTestCase
   end
 
   test "boot with assets" do
-    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", raise_on_non_zero_exit: false)
+    Object.any_instance.stubs(:sleep)
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", raise_on_non_zero_exit: false)
       .returns("12345678") # running version
 
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
+      .returns("running") # health check
+
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
       .with(:sh, "-c", "'docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''{{.ID}}'\\'') ; docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting'", "|", :head, "-1", "|", "while read line; do echo ${line#app-web-}; done", raise_on_non_zero_exit: false)
       .returns("123").twice # old version
 
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{.NetworkSettings.IPAddress}}{{range $k, $v := .NetworkSettings.Ports}}{{printf \":%s\" $k}}{{break}}{{end}}'", "|", :sed, "-e", "'s/\\/tcp$//'")
+      .with(:docker, :inspect, "-f '{{ range .Mounts }}{{printf \"%s %s\\n\" .Source .Destination}}{{ end }}'", "app-web-123", "|", :awk, "'$2 == \"/tmp/kamal-cord\" {print $1}'", raise_on_non_zero_exit: false)
-      .returns("172.1.0.2:80").at_least_once
+      .returns("") # old version
 
     run_command("boot", config: :with_assets).tap do |output|
       assert_match "docker tag dhh/app:latest dhh/app:latest", output
@@ -79,17 +93,23 @@ class CliAppTest < CliTestCase
   end
 
   test "boot with host tags" do
+    Object.any_instance.stubs(:sleep)
+
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
     .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", raise_on_non_zero_exit: false)
     .returns("12345678") # running version
 
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
+      .returns("running") # health check
+
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
       .with(:sh, "-c", "'docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''{{.ID}}'\\'') ; docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting'", "|", :head, "-1", "|", "while read line; do echo ${line#app-web-}; done", raise_on_non_zero_exit: false)
       .returns("123") # old version
 
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{.NetworkSettings.IPAddress}}{{range $k, $v := .NetworkSettings.Ports}}{{printf \":%s\" $k}}{{break}}{{end}}'", "|", :sed, "-e", "'s/\\/tcp$//'")
+      .with(:docker, :inspect, "-f '{{ range .Mounts }}{{printf \"%s %s\\n\" .Source .Destination}}{{ end }}'", "app-web-123", "|", :awk, "'$2 == \"/tmp/kamal-cord\" {print $1}'", raise_on_non_zero_exit: false)
-      .returns("172.1.0.2:80").at_least_once
+      .returns("") # old version
 
     run_command("boot", config: :with_env_tags).tap do |output|
       assert_match "docker tag dhh/app:latest dhh/app:latest", output
@@ -99,11 +119,21 @@ class CliAppTest < CliTestCase
   end
 
   test "boot with web barrier opened" do
-    SSHKit::Backend::Abstract.any_instance.stubs(:capture_with_info).returns("")
+    Object.any_instance.stubs(:sleep)
+
+    SSHKit::Backend::Abstract.any_instance.stubs(:capture_with_info).returns("123") # old version
 
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{.NetworkSettings.IPAddress}}{{range $k, $v := .NetworkSettings.Ports}}{{printf \":%s\" $k}}{{break}}{{end}}'", "|", :sed, "-e", "'s/\\/tcp$//'")
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
-      .returns("172.1.0.2:80").at_least_once
+      .returns("running").at_least_once # web health check passing
+
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-123$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
+      .returns("unhealthy").at_least_once # web health check failing
+
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-workers-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
+      .returns("running").at_least_once # workers health check
 
     run_command("boot", config: :with_roles, host: nil).tap do |output|
       assert_match "Waiting for the first healthy web container before booting workers on 1.1.1.3...", output
@@ -116,35 +146,22 @@ class CliAppTest < CliTestCase
   test "boot with web barrier closed" do
     Thread.report_on_exception = false
 
-    SSHKit::Backend::Abstract.any_instance.stubs(:capture_with_info).returns("")
+    Object.any_instance.stubs(:sleep)
+
+    SSHKit::Backend::Abstract.any_instance.stubs(:capture_with_info).returns("123") # old version
 
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", raise_on_non_zero_exit: false)
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
-      .returns("abcdef123456")
+      .returns("unhealthy").at_least_once # web health check failing
-      .twice # web container id
-
-    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-workers-latest$", "--quiet", raise_on_non_zero_exit: false)
-      .returns("abcdef123456")
-      .twice # worker container id
-
-    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info).with { |*args| args[0..1] == [ :sh, "-c" ] }.returns("123").at_least_once
-
-    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{.NetworkSettings.IPAddress}}{{range $k, $v := .NetworkSettings.Ports}}{{printf \":%s\" $k}}{{break}}{{end}}'", "|", :sed, "-e", "'s/\\/tcp$//'")
-      .returns("172.1.0.2:80")
-      .at_least_once
-
-    SSHKit::Backend::Abstract.any_instance.stubs(:execute).returns("")
-
-    SSHKit::Backend::Abstract.any_instance.expects(:execute).with(:docker, :exec, "parachute_80_", :parachute, :deploy, "app-web", "--target", "\"172.1.0.2:80\"").raises(SSHKit::Command::Failed, "Deploy failed").at_least_once
 
     stderred do
-      run_command("boot", config: :with_roles, host: nil, allowed_error_message: "Deploy failed").tap do |output|
+      run_command("boot", config: :with_roles, host: nil, allow_execute_error: true).tap do |output|
         assert_match "Waiting for the first healthy web container before booting workers on 1.1.1.3...", output
         assert_match "Waiting for the first healthy web container before booting workers on 1.1.1.4...", output
         assert_match "First web container is unhealthy, not booting workers on 1.1.1.3", output
         assert_match "First web container is unhealthy, not booting workers on 1.1.1.4", output
+        assert_match "Running docker container ls --all --filter name=^app-web-latest$ --quiet | xargs docker stop on 1.1.1.1", output
+        assert_match "Running docker container ls --all --filter name=^app-web-latest$ --quiet | xargs docker stop on 1.1.1.2", output
       end
     end
   ensure
@@ -152,17 +169,6 @@ class CliAppTest < CliTestCase
   end
 
   test "start" do
-    # SSHKit::Backend::Abstract.any_instance.stubs(:capture_with_info).returns("")
-
-    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:sh, "-c", "'docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''{{.ID}}'\\'') ; docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting'", "|", :head, "-1", "|", "while read line; do echo ${line#app-web-}; done", raise_on_non_zero_exit: false)
-      .returns("123") # current version
-
-    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-123$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{.NetworkSettings.IPAddress}}{{range $k, $v := .NetworkSettings.Ports}}{{printf \":%s\" $k}}{{break}}{{end}}'", "|", :sed, "-e", "'s/\\/tcp$//'")
-      .returns("172.1.0.2:80")
-      .at_least_once
-
     run_command("start").tap do |output|
       assert_match "docker start app-web-999", output
     end
@@ -327,15 +333,25 @@ class CliAppTest < CliTestCase
   end
 
   private
-    def run_command(*command, config: :with_accessories, host: "1.1.1.1", allowed_error_message: nil)
+    def run_command(*command, config: :with_accessories, host: "1.1.1.1", allow_execute_error: false)
       stdouted do
         Kamal::Cli::App.start([ *command, "-c", "test/fixtures/deploy_#{config}.yml", *([ "--hosts", host ] if host) ])
       rescue SSHKit::Runner::ExecuteError => e
-        raise e unless allowed_error_message && e.message.include?(allowed_error_message)
+        raise e unless allow_execute_error
       end
     end
 
     def stub_running
+      Object.any_instance.stubs(:sleep)
+
       SSHKit::Backend::Abstract.any_instance.stubs(:capture_with_info).returns("123") # old version
+
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-latest$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
+        .returns("running") # health check
+
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-123$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
+        .returns("unhealthy") # health check
     end
 end

test/cli/build_test.rb

@@ -9,10 +9,18 @@ class CliBuildTest < CliTestCase
   end
 
   test "push" do
-    with_build_directory do
+    with_build_directory do |build_directory|
       Kamal::Commands::Hook.any_instance.stubs(:hook_exists?).returns(true)
       hook_variables = { version: 999, service_version: "app@999", hosts: "1.1.1.1,1.1.1.2,1.1.1.3,1.1.1.4", command: "build", subcommand: "push" }
 
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:git, "-C", anything, :"rev-parse", :HEAD)
+        .returns(Kamal::Git.revision)
+
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:git, "-C", anything, :status, "--porcelain")
+        .returns("")
+
       run_command("push", "--verbose").tap do |output|
         assert_hook_ran "pre-build", output, **hook_variables
         assert_match /Cloning repo into build directory/, output
@@ -23,28 +31,33 @@ class CliBuildTest < CliTestCase
     end
   end
 
-  test "push reseting clone" do
+  test "push resetting clone" do
-    with_build_directory do
+    with_build_directory do |build_directory|
       stub_setup
-      build_dir = "#{Dir.tmpdir}/kamal-clones/app-#{pwd_sha}/kamal/"
 
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute).with(:docker, "--version", "&&", :docker, :buildx, "version")
+      SSHKit::Backend::Abstract.any_instance.expects(:execute).with(:docker, "--version", "&&", :docker, :buildx, "version")
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute).with(:docker, :buildx, :create, "--use", "--name", "kamal-app-multiarch")
 
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute).with(:mkdir, "-p", "#{Dir.tmpdir}/kamal-clones/app-#{pwd_sha}")
+      SSHKit::Backend::Abstract.any_instance.expects(:execute)
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute)
         .with(:git, "-C", "#{Dir.tmpdir}/kamal-clones/app-#{pwd_sha}", :clone, Dir.pwd)
         .raises(SSHKit::Command::Failed.new("fatal: destination path 'kamal' already exists and is not an empty directory"))
         .then
         .returns(true)
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute).with(:git, "-C", build_dir, :remote, "set-url", :origin, Dir.pwd)
+      SSHKit::Backend::Abstract.any_instance.expects(:execute).with(:git, "-C", build_directory, :remote, "set-url", :origin, Dir.pwd)
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute).with(:git, "-C", build_dir, :fetch, :origin)
+      SSHKit::Backend::Abstract.any_instance.expects(:execute).with(:git, "-C", build_directory, :fetch, :origin)
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute).with(:git, "-C", build_dir, :reset, "--hard", Kamal::Git.revision)
+      SSHKit::Backend::Abstract.any_instance.expects(:execute).with(:git, "-C", build_directory, :reset, "--hard", Kamal::Git.revision)
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute).with(:git, "-C", build_dir, :clean, "-fdx")
+      SSHKit::Backend::Abstract.any_instance.expects(:execute).with(:git, "-C", build_directory, :clean, "-fdx")
 
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute)
+      SSHKit::Backend::Abstract.any_instance.expects(:execute)
         .with(:docker, :buildx, :build, "--push", "--platform", "linux/amd64,linux/arm64", "--builder", "kamal-app-multiarch", "-t", "dhh/app:999", "-t", "dhh/app:latest", "--label", "service=\"app\"", "--file", "Dockerfile", ".")
 
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:git, "-C", anything, :"rev-parse", :HEAD)
+        .returns(Kamal::Git.revision)
+
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:git, "-C", anything, :status, "--porcelain")
+        .returns("")
+
       run_command("push", "--verbose").tap do |output|
         assert_match /Cloning repo into build directory/, output
         assert_match /Resetting local clone/, output
@@ -64,25 +77,65 @@ class CliBuildTest < CliTestCase
     end
   end
 
-  test "push without builder" do
+  test "push with corrupt clone" do
-    with_build_directory do
+    with_build_directory do |build_directory|
       stub_setup
 
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute)
+      SSHKit::Backend::Abstract.any_instance.expects(:execute).with(:docker, "--version", "&&", :docker, :buildx, "version")
+
+      SSHKit::Backend::Abstract.any_instance.expects(:execute)
+        .with(:git, "-C", "#{Dir.tmpdir}/kamal-clones/app-#{pwd_sha}", :clone, Dir.pwd)
+        .raises(SSHKit::Command::Failed.new("fatal: destination path 'kamal' already exists and is not an empty directory"))
+        .then
+        .returns(true)
+        .twice
+
+      SSHKit::Backend::Abstract.any_instance.expects(:execute).with(:git, "-C", build_directory, :remote, "set-url", :origin, Dir.pwd)
+        .raises(SSHKit::Command::Failed.new("fatal: not a git repository"))
+
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:git, "-C", anything, :"rev-parse", :HEAD)
+        .returns(Kamal::Git.revision)
+
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:git, "-C", anything, :status, "--porcelain")
+        .returns("")
+
+      Dir.stubs(:chdir)
+
+      run_command("push", "--verbose") do |output|
+        assert_match /Cloning repo into build directory `#{build_directory}`\.\.\..*Cloning repo into build directory `#{build_directory}`\.\.\./, output
+        assert_match "Resetting local clone as `#{build_directory}` already exists...", output
+        assert_match "Error preparing clone: Failed to clone repo: fatal: not a git repository, deleting and retrying...", output
+      end
+    end
+  end
+
+  test "push without builder" do
+    with_build_directory do |build_directory|
+      stub_setup
+
+      SSHKit::Backend::Abstract.any_instance.expects(:execute)
         .with(:docker, "--version", "&&", :docker, :buildx, "version")
 
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute)
+      SSHKit::Backend::Abstract.any_instance.expects(:execute)
         .with(:docker, :buildx, :create, "--use", "--name", "kamal-app-multiarch")
 
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute)
+      SSHKit::Backend::Abstract.any_instance.expects(:execute)
         .with { |*args| args[0..1] == [ :docker, :buildx ] }
         .raises(SSHKit::Command::Failed.new("no builder"))
         .then
         .returns(true)
 
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute).with { |*args| args.first.start_with?("git") }
+      SSHKit::Backend::Abstract.any_instance.expects(:execute).with { |*args| args.first.start_with?("git") }
 
-      SSHKit::Backend::Abstract.any_instance.stubs(:execute).with(:mkdir, "-p", "#{Dir.tmpdir}/kamal-clones/app-#{pwd_sha}")
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:git, "-C", anything, :"rev-parse", :HEAD)
+        .returns(Kamal::Git.revision)
+
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:git, "-C", anything, :status, "--porcelain")
+        .returns("")
 
       run_command("push").tap do |output|
         assert_match /WARN Missing compatible builder, so creating a new one first/, output
@@ -183,7 +236,7 @@ class CliBuildTest < CliTestCase
       build_directory = File.join Dir.tmpdir, "kamal-clones", "app-#{pwd_sha}", "kamal"
       FileUtils.mkdir_p build_directory
       FileUtils.touch File.join build_directory, "Dockerfile"
-      yield
+      yield build_directory + "/"
     ensure
       FileUtils.rm_rf build_directory
     end

test/cli/env_test.rb

@@ -4,11 +4,14 @@ class CliEnvTest < CliTestCase
   test "push" do
     run_command("push").tap do |output|
       assert_match "Running /usr/bin/env mkdir -p .kamal/env/roles on 1.1.1.1", output
+      assert_match "Running /usr/bin/env mkdir -p .kamal/env/traefik on 1.1.1.1", output
       assert_match "Running /usr/bin/env mkdir -p .kamal/env/accessories on 1.1.1.1", output
       assert_match "Running /usr/bin/env mkdir -p .kamal/env/roles on 1.1.1.1", output
+      assert_match "Running /usr/bin/env mkdir -p .kamal/env/traefik on 1.1.1.2", output
       assert_match "Running /usr/bin/env mkdir -p .kamal/env/accessories on 1.1.1.1", output
       assert_match ".kamal/env/roles/app-web.env", output
       assert_match ".kamal/env/roles/app-workers.env", output
+      assert_match ".kamal/env/traefik/traefik.env", output
       assert_match ".kamal/env/accessories/app-redis.env", output
     end
   end
@@ -19,6 +22,8 @@ class CliEnvTest < CliTestCase
       assert_match "Running /usr/bin/env rm -f .kamal/env/roles/app-web.env on 1.1.1.2", output
       assert_match "Running /usr/bin/env rm -f .kamal/env/roles/app-workers.env on 1.1.1.3", output
       assert_match "Running /usr/bin/env rm -f .kamal/env/roles/app-workers.env on 1.1.1.4", output
+      assert_match "Running /usr/bin/env rm -f .kamal/env/traefik/traefik.env on 1.1.1.1", output
+      assert_match "Running /usr/bin/env rm -f .kamal/env/traefik/traefik.env on 1.1.1.2", output
       assert_match "Running /usr/bin/env rm -f .kamal/env/accessories/app-redis.env on 1.1.1.1", output
       assert_match "Running /usr/bin/env rm -f .kamal/env/accessories/app-redis.env on 1.1.1.2", output
       assert_match "Running /usr/bin/env rm -f .kamal/env/accessories/app-mysql.env on 1.1.1.3", output

test/cli/main_test.rb

@@ -24,7 +24,7 @@ class CliMainTest < CliTestCase
     # deploy
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:registry:login", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:build:pull", [], invoke_options)
-    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:proxy:boot", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:traefik:boot", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:stale_containers", [], invoke_options.merge(stop: true))
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:boot", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:prune:all", [], invoke_options)
@@ -36,7 +36,7 @@ class CliMainTest < CliTestCase
       assert_match /Acquiring the deploy lock/, output
       assert_match /Log into image registry/, output
       assert_match /Pull app image/, output
-      assert_match /Ensure proxy is running/, output
+      assert_match /Ensure Traefik is running/, output
       assert_match /Detect stale containers/, output
       assert_match /Prune old containers and images/, output
       assert_match /Releasing the deploy lock/, output
@@ -48,7 +48,7 @@ class CliMainTest < CliTestCase
 
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:registry:login", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:build:deliver", [], invoke_options)
-    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:proxy:boot", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:traefik:boot", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:stale_containers", [], invoke_options.merge(stop: true))
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:boot", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:prune:all", [], invoke_options)
@@ -61,7 +61,7 @@ class CliMainTest < CliTestCase
       assert_match /Log into image registry/, output
       assert_match /Build and push app image/, output
       assert_hook_ran "pre-deploy", output, **hook_variables
-      assert_match /Ensure proxy is running/, output
+      assert_match /Ensure Traefik is running/, output
       assert_match /Detect stale containers/, output
       assert_match /Prune old containers and images/, output
       assert_hook_ran "post-deploy", output, **hook_variables, runtime: true
@@ -73,7 +73,7 @@ class CliMainTest < CliTestCase
 
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:registry:login", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:build:pull", [], invoke_options)
-    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:proxy:boot", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:traefik:boot", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:stale_containers", [], invoke_options.merge(stop: true))
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:boot", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:prune:all", [], invoke_options)
@@ -82,7 +82,7 @@ class CliMainTest < CliTestCase
       assert_match /Acquiring the deploy lock/, output
       assert_match /Log into image registry/, output
       assert_match /Pull app image/, output
-      assert_match /Ensure proxy is running/, output
+      assert_match /Ensure Traefik is running/, output
       assert_match /Detect stale containers/, output
       assert_match /Prune old containers and images/, output
       assert_match /Releasing the deploy lock/, output
@@ -108,6 +108,14 @@ class CliMainTest < CliTestCase
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_debug)
       .with(:stat, ".kamal/locks/app", ">", "/dev/null", "&&", :cat, ".kamal/locks/app/details", "|", :base64, "-d")
 
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:git, "-C", anything, :"rev-parse", :HEAD)
+      .returns(Kamal::Git.revision)
+
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:git, "-C", anything, :status, "--porcelain")
+      .returns("")
+
     assert_raises(Kamal::Cli::LockError) do
       run_command("deploy")
     end
@@ -129,6 +137,14 @@ class CliMainTest < CliTestCase
       .with { |*arg| arg[0..1] == [ :mkdir, ".kamal/locks/app" ] }
       .raises(SocketError, "getaddrinfo: nodename nor servname provided, or not known")
 
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:git, "-C", anything, :"rev-parse", :HEAD)
+      .returns(Kamal::Git.revision)
+
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:git, "-C", anything, :status, "--porcelain")
+      .returns("")
+
     assert_raises(SSHKit::Runner::ExecuteError) do
       run_command("deploy")
     end
@@ -153,7 +169,7 @@ class CliMainTest < CliTestCase
 
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:registry:login", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:build:deliver", [], invoke_options)
-    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:proxy:boot", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:traefik:boot", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:stale_containers", [], invoke_options.merge(stop: true))
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:boot", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:prune:all", [], invoke_options)
@@ -163,12 +179,27 @@ class CliMainTest < CliTestCase
     end
   end
 
+  test "deploy without healthcheck if primary host doesn't have traefik" do
+    invoke_options = { "config_file" => "test/fixtures/deploy_workers_only.yml", "version" => "999", "skip_hooks" => false }
+
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:healthcheck:perform", [], invoke_options).never
+
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:registry:login", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:build:deliver", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:traefik:boot", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:stale_containers", [], invoke_options.merge(stop: true))
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:boot", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:prune:all", [], invoke_options)
+
+    run_command("deploy", config_file: "deploy_workers_only")
+  end
+
   test "deploy with missing secrets" do
     invoke_options = { "config_file" => "test/fixtures/deploy_with_secrets.yml", "version" => "999", "skip_hooks" => false }
 
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:registry:login", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:build:deliver", [], invoke_options)
-    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:proxy:boot", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:traefik:boot", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:stale_containers", [], invoke_options.merge(stop: true))
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:boot", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:prune:all", [], invoke_options)
@@ -220,6 +251,7 @@ class CliMainTest < CliTestCase
   end
 
   test "rollback good version" do
+    Object.any_instance.stubs(:sleep)
     [ "web", "workers" ].each do |role|
       SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
         .with(:docker, :container, :ls, "--all", "--filter", "name=^app-#{role}-123$", "--quiet", raise_on_non_zero_exit: false)
@@ -230,11 +262,18 @@ class CliMainTest < CliTestCase
       SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
         .with(:sh, "-c", "'docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=#{role} --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''{{.ID}}'\\'') ; docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=#{role} --filter status=running --filter status=restarting'", "|", :head, "-1", "|", "while read line; do echo ${line#app-#{role}-}; done", raise_on_non_zero_exit: false)
         .returns("version-to-rollback\n").at_least_once
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:docker, :container, :ls, "--all", "--filter", "name=^app-#{role}-123$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
+        .returns("running").at_least_once # health check
     end
 
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-123$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{.NetworkSettings.IPAddress}}{{range $k, $v := .NetworkSettings.Ports}}{{printf \":%s\" $k}}{{break}}{{end}}'", "|", :sed, "-e", "'s/\\/tcp$//'")
+      .with(:docker, :inspect, "-f '{{ range .Mounts }}{{printf \"%s %s\\n\" .Source .Destination}}{{ end }}'", "app-web-version-to-rollback", "|", :awk, "'$2 == \"/tmp/kamal-cord\" {print $1}'", raise_on_non_zero_exit: false)
-      .returns("172.1.0.2:80").at_least_once
+      .returns("corddirectory").at_least_once # health check
+
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-version-to-rollback$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
+      .returns("unhealthy").at_least_once # health check
 
     Kamal::Commands::Hook.any_instance.stubs(:hook_exists?).returns(true)
     hook_variables = { version: 123, service_version: "app@123", hosts: "1.1.1.1,1.1.1.2,1.1.1.3,1.1.1.4", command: "rollback" }
@@ -251,6 +290,8 @@ class CliMainTest < CliTestCase
   test "rollback without old version" do
     Kamal::Cli::Main.any_instance.stubs(:container_available?).returns(true)
 
+    Kamal::Cli::Healthcheck::Poller.stubs(:sleep)
+
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
       .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-123$", "--quiet", raise_on_non_zero_exit: false)
       .returns("").at_least_once
@@ -258,8 +299,8 @@ class CliMainTest < CliTestCase
       .with(:sh, "-c", "'docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''{{.ID}}'\\'') ; docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting'", "|", :head, "-1", "|", "while read line; do echo ${line#app-web-}; done", raise_on_non_zero_exit: false)
       .returns("").at_least_once
     SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-123$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{.NetworkSettings.IPAddress}}{{range $k, $v := .NetworkSettings.Ports}}{{printf \":%s\" $k}}{{break}}{{end}}'", "|", :sed, "-e", "'s/\\/tcp$//'")
+      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-123$", "--quiet", "|", :xargs, :docker, :inspect, "--format", "'{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}'")
-      .returns("127.1.0.4:80").at_least_once
+      .returns("running").at_least_once # health check
 
     run_command("rollback", "123").tap do |output|
       assert_match "docker run --detach --restart unless-stopped --name app-web-123", output
@@ -268,7 +309,7 @@ class CliMainTest < CliTestCase
   end
 
   test "details" do
-    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:proxy:details")
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:traefik:details")
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:app:details")
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:accessory:details", [ "all" ])
 
@@ -437,9 +478,9 @@ class CliMainTest < CliTestCase
 
   test "remove with confirmation" do
     run_command("remove", "-y", config_file: "deploy_with_accessories").tap do |output|
-      assert_match /docker container stop parachute/, output
+      assert_match /docker container stop traefik/, output
-      assert_match /docker container prune --force --filter label=org.opencontainers.image.title=parachute/, output
+      assert_match /docker container prune --force --filter label=org.opencontainers.image.title=Traefik/, output
-      assert_match /docker image prune --all --force --filter label=org.opencontainers.image.title=parachute/, output
+      assert_match /docker image prune --all --force --filter label=org.opencontainers.image.title=Traefik/, output
 
       assert_match /docker ps --quiet --filter label=service=app | xargs docker stop/, output
       assert_match /docker container prune --force --filter label=service=app/, output

test/cli/proxy_test.rb

@@ -1,94 +0,0 @@
-require_relative "cli_test_case"
-
-class CliProxyTest < CliTestCase
-  test "boot" do
-    run_command("boot").tap do |output|
-      assert_match "docker login", output
-      assert_match "docker run --name parachute_80_ --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --volume parachute_80_:/root/.config/parachute --log-opt max-size=\"10m\" #{Kamal::Configuration::Proxy::DEFAULT_IMAGE}", output
-    end
-  end
-
-  test "reboot" do
-    Kamal::Commands::Registry.any_instance.expects(:login).twice
-
-    run_command("reboot", "-y").tap do |output|
-      assert_match "docker container stop parachute", output
-      assert_match "docker container prune --force --filter label=org.opencontainers.image.title=parachute", output
-      assert_match "docker run --name parachute_80_ --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --volume parachute_80_:/root/.config/parachute --log-opt max-size=\"10m\" #{Kamal::Configuration::Proxy::DEFAULT_IMAGE}", output
-    end
-  end
-
-  test "reboot --rolling" do
-    run_command("reboot", "--rolling", "-y").tap do |output|
-      assert_match "Running docker container prune --force --filter label=org.opencontainers.image.title=parachute on 1.1.1.1", output
-    end
-  end
-
-  test "start" do
-    run_command("start").tap do |output|
-      assert_match "docker container start parachute", output
-    end
-  end
-
-  test "stop" do
-    run_command("stop").tap do |output|
-      assert_match "docker container stop parachute", output
-    end
-  end
-
-  test "restart" do
-    Kamal::Cli::Proxy.any_instance.expects(:stop)
-    Kamal::Cli::Proxy.any_instance.expects(:start)
-
-    run_command("restart")
-  end
-
-  test "details" do
-    run_command("details").tap do |output|
-      assert_match "docker ps --filter name=^parachute_80_$", output
-    end
-  end
-
-  test "logs" do
-    SSHKit::Backend::Abstract.any_instance.stubs(:capture)
-      .with(:docker, :logs, "parachute_80_", " --tail 100", "--timestamps", "2>&1")
-      .returns("Log entry")
-
-    run_command("logs").tap do |output|
-      assert_match "Proxy Host: 1.1.1.1", output
-      assert_match "Log entry", output
-    end
-  end
-
-  test "logs with follow" do
-    SSHKit::Backend::Abstract.any_instance.stubs(:exec)
-      .with("ssh -t root@1.1.1.1 -p 22 'docker logs parachute_80_ --timestamps --tail 10 --follow 2>&1'")
-
-    assert_match "docker logs parachute_80_ --timestamps --tail 10 --follow", run_command("logs", "--follow")
-  end
-
-  test "remove" do
-    Kamal::Cli::Proxy.any_instance.expects(:stop)
-    Kamal::Cli::Proxy.any_instance.expects(:remove_container)
-    Kamal::Cli::Proxy.any_instance.expects(:remove_image)
-
-    run_command("remove")
-  end
-
-  test "remove_container" do
-    run_command("remove_container").tap do |output|
-      assert_match "docker container prune --force --filter label=org.opencontainers.image.title=parachute", output
-    end
-  end
-
-  test "remove_image" do
-    run_command("remove_image").tap do |output|
-      assert_match "docker image prune --all --force --filter label=org.opencontainers.image.title=parachute", output
-    end
-  end
-
-  private
-    def run_command(*command)
-      stdouted { Kamal::Cli::Proxy.start([ *command, "-c", "test/fixtures/deploy_with_accessories.yml" ]) }
-    end
-end

test/cli/prune_test.rb

@@ -18,10 +18,12 @@ class CliPruneTest < CliTestCase
   test "containers" do
     run_command("containers").tap do |output|
       assert_match /docker ps -q -a --filter label=service=app --filter status=created --filter status=exited --filter status=dead | tail -n +6 | while read container_id; do docker rm $container_id; done on 1.1.1.\d/, output
+      assert_match /docker container prune --force --filter label=service=healthcheck-app on 1.1.1.\d/, output
      end
 
     run_command("containers", "--retain", "10").tap do |output|
       assert_match /docker ps -q -a --filter label=service=app --filter status=created --filter status=exited --filter status=dead | tail -n +11 | while read container_id; do docker rm $container_id; done on 1.1.1.\d/, output
+      assert_match /docker container prune --force --filter label=service=healthcheck-app on 1.1.1.\d/, output
     end
 
     assert_raises(RuntimeError, "retain must be at least 1") do

test/cli/traefik_test.rb

@@ -0,0 +1,96 @@
+require_relative "cli_test_case"
+
+class CliTraefikTest < CliTestCase
+  test "boot" do
+    run_command("boot").tap do |output|
+      assert_match "docker login", output
+      assert_match "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{Kamal::Commands::Traefik::DEFAULT_IMAGE} --providers.docker --log.level=\"DEBUG\"", output
+    end
+  end
+
+  test "reboot" do
+    Kamal::Commands::Registry.any_instance.expects(:login).twice
+
+    run_command("reboot", "-y").tap do |output|
+      assert_match "docker container stop traefik", output
+      assert_match "docker container prune --force --filter label=org.opencontainers.image.title=Traefik", output
+      assert_match "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{Kamal::Commands::Traefik::DEFAULT_IMAGE} --providers.docker --log.level=\"DEBUG\"", output
+    end
+  end
+
+  test "reboot --rolling" do
+    Object.any_instance.stubs(:sleep)
+
+    run_command("reboot", "--rolling", "-y").tap do |output|
+      assert_match "Running docker container prune --force --filter label=org.opencontainers.image.title=Traefik on 1.1.1.1", output
+    end
+  end
+
+  test "start" do
+    run_command("start").tap do |output|
+      assert_match "docker container start traefik", output
+    end
+  end
+
+  test "stop" do
+    run_command("stop").tap do |output|
+      assert_match "docker container stop traefik", output
+    end
+  end
+
+  test "restart" do
+    Kamal::Cli::Traefik.any_instance.expects(:stop)
+    Kamal::Cli::Traefik.any_instance.expects(:start)
+
+    run_command("restart")
+  end
+
+  test "details" do
+    run_command("details").tap do |output|
+      assert_match "docker ps --filter name=^traefik$", output
+    end
+  end
+
+  test "logs" do
+    SSHKit::Backend::Abstract.any_instance.stubs(:capture)
+      .with(:docker, :logs, "traefik", " --tail 100", "--timestamps", "2>&1")
+      .returns("Log entry")
+
+    run_command("logs").tap do |output|
+      assert_match "Traefik Host: 1.1.1.1", output
+      assert_match "Log entry", output
+    end
+  end
+
+  test "logs with follow" do
+    SSHKit::Backend::Abstract.any_instance.stubs(:exec)
+      .with("ssh -t root@1.1.1.1 -p 22 'docker logs traefik --timestamps --tail 10 --follow 2>&1'")
+
+    assert_match "docker logs traefik --timestamps --tail 10 --follow", run_command("logs", "--follow")
+  end
+
+  test "remove" do
+    Kamal::Cli::Traefik.any_instance.expects(:stop)
+    Kamal::Cli::Traefik.any_instance.expects(:remove_container)
+    Kamal::Cli::Traefik.any_instance.expects(:remove_image)
+
+    run_command("remove")
+  end
+
+  test "remove_container" do
+    run_command("remove_container").tap do |output|
+      assert_match "docker container prune --force --filter label=org.opencontainers.image.title=Traefik", output
+    end
+  end
+
+  test "remove_image" do
+    run_command("remove_image").tap do |output|
+      assert_match "docker image prune --all --force --filter label=org.opencontainers.image.title=Traefik", output
+    end
+  end
+
+  private
+    def run_command(*command)
+      stdouted { Kamal::Cli::Traefik.start([ *command, "-c", "test/fixtures/deploy_with_accessories.yml" ]) }
+    end
+end

test/commander_test.rb

@@ -136,18 +136,18 @@ class CommanderTest < ActiveSupport::TestCase
     assert_equal [ "1.1.1.3", "1.1.1.4", "1.1.1.1", "1.1.1.2" ], @kamal.hosts
   end
 
-  test "proxy hosts should observe filtered roles" do
+  test "traefik hosts should observe filtered roles" do
     configure_with(:deploy_with_aliases)
 
     @kamal.specific_roles = [ "web_tokyo" ]
-    assert_equal [ "1.1.1.3", "1.1.1.4" ], @kamal.proxy_hosts
+    assert_equal [ "1.1.1.3", "1.1.1.4" ], @kamal.traefik_hosts
   end
 
-  test "proxy hosts should observe filtered hosts" do
+  test "traefik hosts should observe filtered hosts" do
     configure_with(:deploy_with_aliases)
 
     @kamal.specific_hosts = [ "1.1.1.4" ]
-    assert_equal [ "1.1.1.4" ], @kamal.proxy_hosts
+    assert_equal [ "1.1.1.4" ], @kamal.traefik_hosts
   end
 
   private

test/commands/app_test.rb

@@ -14,13 +14,13 @@ class CommandsAppTest < ActiveSupport::TestCase
 
   test "run" do
     assert_equal \
-      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --log-opt max-size=\"10m\" --label service=\"app\" --label role=\"web\" --label destination dhh/app:999",
+      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --health-cmd \"(curl -f http://localhost:3000/up || exit 1) && (stat /tmp/kamal-cord/cord > /dev/null || exit 1)\" --health-interval \"1s\" --volume $(pwd)/.kamal/cords/app-web-12345678901234567890123456789012:/tmp/kamal-cord --log-opt max-size=\"10m\" --label service=\"app\" --label role=\"web\" --label destination --label traefik.http.services.app-web.loadbalancer.server.scheme=\"http\" --label traefik.http.routers.app-web.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.app-web.priority=\"2\" --label traefik.http.middlewares.app-web-retry.retry.attempts=\"5\" --label traefik.http.middlewares.app-web-retry.retry.initialinterval=\"500ms\" --label traefik.http.routers.app-web.middlewares=\"app-web-retry@docker\" dhh/app:999",
       new_command.run.join(" ")
   end
 
   test "run with hostname" do
     assert_equal \
-      "docker run --detach --restart unless-stopped --name app-web-999 --hostname myhost -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --log-opt max-size=\"10m\" --label service=\"app\" --label role=\"web\" --label destination dhh/app:999",
+      "docker run --detach --restart unless-stopped --name app-web-999 --hostname myhost -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --health-cmd \"(curl -f http://localhost:3000/up || exit 1) && (stat /tmp/kamal-cord/cord > /dev/null || exit 1)\" --health-interval \"1s\" --volume $(pwd)/.kamal/cords/app-web-12345678901234567890123456789012:/tmp/kamal-cord --log-opt max-size=\"10m\" --label service=\"app\" --label role=\"web\" --label destination --label traefik.http.services.app-web.loadbalancer.server.scheme=\"http\" --label traefik.http.routers.app-web.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.app-web.priority=\"2\" --label traefik.http.middlewares.app-web-retry.retry.attempts=\"5\" --label traefik.http.middlewares.app-web-retry.retry.initialinterval=\"500ms\" --label traefik.http.routers.app-web.middlewares=\"app-web-retry@docker\" dhh/app:999",
       new_command.run(hostname: "myhost").join(" ")
   end
 
@@ -28,7 +28,31 @@ class CommandsAppTest < ActiveSupport::TestCase
     @config[:volumes] = [ "/local/path:/container/path" ]
 
     assert_equal \
-      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --log-opt max-size=\"10m\" --volume /local/path:/container/path --label service=\"app\" --label role=\"web\" --label destination dhh/app:999",
+      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --health-cmd \"(curl -f http://localhost:3000/up || exit 1) && (stat /tmp/kamal-cord/cord > /dev/null || exit 1)\" --health-interval \"1s\" --volume $(pwd)/.kamal/cords/app-web-12345678901234567890123456789012:/tmp/kamal-cord --log-opt max-size=\"10m\" --volume /local/path:/container/path --label service=\"app\" --label role=\"web\" --label destination --label traefik.http.services.app-web.loadbalancer.server.scheme=\"http\" --label traefik.http.routers.app-web.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.app-web.priority=\"2\" --label traefik.http.middlewares.app-web-retry.retry.attempts=\"5\" --label traefik.http.middlewares.app-web-retry.retry.initialinterval=\"500ms\" --label traefik.http.routers.app-web.middlewares=\"app-web-retry@docker\" dhh/app:999",
+      new_command.run.join(" ")
+  end
+
+  test "run with custom healthcheck path" do
+    @config[:healthcheck] = { "path" => "/healthz" }
+
+    assert_equal \
+      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --health-cmd \"(curl -f http://localhost:3000/healthz || exit 1) && (stat /tmp/kamal-cord/cord > /dev/null || exit 1)\" --health-interval \"1s\" --volume $(pwd)/.kamal/cords/app-web-12345678901234567890123456789012:/tmp/kamal-cord --log-opt max-size=\"10m\" --label service=\"app\" --label role=\"web\" --label destination --label traefik.http.services.app-web.loadbalancer.server.scheme=\"http\" --label traefik.http.routers.app-web.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.app-web.priority=\"2\" --label traefik.http.middlewares.app-web-retry.retry.attempts=\"5\" --label traefik.http.middlewares.app-web-retry.retry.initialinterval=\"500ms\" --label traefik.http.routers.app-web.middlewares=\"app-web-retry@docker\" dhh/app:999",
+      new_command.run.join(" ")
+  end
+
+  test "run with custom healthcheck command" do
+    @config[:healthcheck] = { "cmd" => "/bin/up" }
+
+    assert_equal \
+      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --health-cmd \"(/bin/up) && (stat /tmp/kamal-cord/cord > /dev/null || exit 1)\" --health-interval \"1s\" --volume $(pwd)/.kamal/cords/app-web-12345678901234567890123456789012:/tmp/kamal-cord --log-opt max-size=\"10m\" --label service=\"app\" --label role=\"web\" --label destination --label traefik.http.services.app-web.loadbalancer.server.scheme=\"http\" --label traefik.http.routers.app-web.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.app-web.priority=\"2\" --label traefik.http.middlewares.app-web-retry.retry.attempts=\"5\" --label traefik.http.middlewares.app-web-retry.retry.initialinterval=\"500ms\" --label traefik.http.routers.app-web.middlewares=\"app-web-retry@docker\" dhh/app:999",
+      new_command.run.join(" ")
+  end
+
+  test "run with role-specific healthcheck options" do
+    @config[:servers] = { "web" => { "hosts" => [ "1.1.1.1" ], "healthcheck" => { "cmd" => "/bin/healthy" } } }
+
+    assert_equal \
+      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --health-cmd \"(/bin/healthy) && (stat /tmp/kamal-cord/cord > /dev/null || exit 1)\" --health-interval \"1s\" --volume $(pwd)/.kamal/cords/app-web-12345678901234567890123456789012:/tmp/kamal-cord --log-opt max-size=\"10m\" --label service=\"app\" --label role=\"web\" --label destination --label traefik.http.services.app-web.loadbalancer.server.scheme=\"http\" --label traefik.http.routers.app-web.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.app-web.priority=\"2\" --label traefik.http.middlewares.app-web-retry.retry.attempts=\"5\" --label traefik.http.middlewares.app-web-retry.retry.initialinterval=\"500ms\" --label traefik.http.routers.app-web.middlewares=\"app-web-retry@docker\" dhh/app:999",
       new_command.run.join(" ")
   end
 
@@ -43,7 +67,7 @@ class CommandsAppTest < ActiveSupport::TestCase
     @config[:logging] = { "driver" => "local", "options" => { "max-size" => "100m", "max-file" => "3" } }
 
     assert_equal \
-      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --log-driver \"local\" --log-opt max-size=\"100m\" --log-opt max-file=\"3\" --label service=\"app\" --label role=\"web\" --label destination dhh/app:999",
+      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --health-cmd \"(curl -f http://localhost:3000/up || exit 1) && (stat /tmp/kamal-cord/cord > /dev/null || exit 1)\" --health-interval \"1s\" --volume $(pwd)/.kamal/cords/app-web-12345678901234567890123456789012:/tmp/kamal-cord --log-driver \"local\" --log-opt max-size=\"100m\" --log-opt max-file=\"3\" --label service=\"app\" --label role=\"web\" --label destination --label traefik.http.services.app-web.loadbalancer.server.scheme=\"http\" --label traefik.http.routers.app-web.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.app-web.priority=\"2\" --label traefik.http.middlewares.app-web-retry.retry.attempts=\"5\" --label traefik.http.middlewares.app-web-retry.retry.initialinterval=\"500ms\" --label traefik.http.routers.app-web.middlewares=\"app-web-retry@docker\" dhh/app:999",
       new_command.run.join(" ")
   end
 
@@ -52,7 +76,7 @@ class CommandsAppTest < ActiveSupport::TestCase
     @config[:servers] = { "web" => { "hosts" => [ "1.1.1.1" ], "logging" => { "driver" => "local", "options" => { "max-size" => "100m" } } } }
 
     assert_equal \
-      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --log-driver \"local\" --log-opt max-size=\"100m\" --log-opt max-file=\"3\" --label service=\"app\" --label role=\"web\" --label destination dhh/app:999",
+      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --health-cmd \"(curl -f http://localhost:3000/up || exit 1) && (stat /tmp/kamal-cord/cord > /dev/null || exit 1)\" --health-interval \"1s\" --volume $(pwd)/.kamal/cords/app-web-12345678901234567890123456789012:/tmp/kamal-cord --log-driver \"local\" --log-opt max-size=\"100m\" --log-opt max-file=\"3\" --label service=\"app\" --label role=\"web\" --label destination --label traefik.http.services.app-web.loadbalancer.server.scheme=\"http\" --label traefik.http.routers.app-web.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.app-web.priority=\"2\" --label traefik.http.middlewares.app-web-retry.retry.attempts=\"5\" --label traefik.http.middlewares.app-web-retry.retry.initialinterval=\"500ms\" --label traefik.http.routers.app-web.middlewares=\"app-web-retry@docker\" dhh/app:999",
       new_command.run.join(" ")
   end
 
@@ -61,7 +85,7 @@ class CommandsAppTest < ActiveSupport::TestCase
     @config[:env]["tags"] = { "tag1" => { "ENV1" => "value1" } }
 
     assert_equal \
-      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --env ENV1=\"value1\" --log-opt max-size=\"10m\" --label service=\"app\" --label role=\"web\" --label destination dhh/app:999",
+      "docker run --detach --restart unless-stopped --name app-web-999 -e KAMAL_CONTAINER_NAME=\"app-web-999\" -e KAMAL_VERSION=\"999\" --env-file .kamal/env/roles/app-web.env --env ENV1=\"value1\" --health-cmd \"(curl -f http://localhost:3000/up || exit 1) && (stat /tmp/kamal-cord/cord > /dev/null || exit 1)\" --health-interval \"1s\" --volume $(pwd)/.kamal/cords/app-web-12345678901234567890123456789012:/tmp/kamal-cord --log-opt max-size=\"10m\" --label service=\"app\" --label role=\"web\" --label destination --label traefik.http.services.app-web.loadbalancer.server.scheme=\"http\" --label traefik.http.routers.app-web.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.app-web.priority=\"2\" --label traefik.http.middlewares.app-web-retry.retry.attempts=\"5\" --label traefik.http.middlewares.app-web-retry.retry.initialinterval=\"500ms\" --label traefik.http.routers.app-web.middlewares=\"app-web-retry@docker\" dhh/app:999",
       new_command.run.join(" ")
   end
 
@@ -374,6 +398,20 @@ class CommandsAppTest < ActiveSupport::TestCase
     assert_equal "rm -f .kamal/env/roles/app-web.env", new_command.remove_env_file.join(" ")
   end
 
+  test "cord" do
+    assert_equal "docker inspect -f '{{ range .Mounts }}{{printf \"%s %s\\n\" .Source .Destination}}{{ end }}' app-web-123 | awk '$2 == \"/tmp/kamal-cord\" {print $1}'", new_command.cord(version: 123).join(" ")
+  end
+
+  test "tie cord" do
+    assert_equal "mkdir -p . ; touch cordfile", new_command.tie_cord("cordfile").join(" ")
+    assert_equal "mkdir -p corddir ; touch corddir/cordfile", new_command.tie_cord("corddir/cordfile").join(" ")
+    assert_equal "mkdir -p /corddir ; touch /corddir/cordfile", new_command.tie_cord("/corddir/cordfile").join(" ")
+  end
+
+  test "cut cord" do
+    assert_equal "rm -r corddir", new_command.cut_cord("corddir").join(" ")
+  end
+
   test "extract assets" do
     assert_equal [
       :mkdir, "-p", ".kamal/assets/extracted/app-web-999", "&&",

test/commands/hook_test.rb

@@ -7,7 +7,8 @@ class CommandsHookTest < ActiveSupport::TestCase
     freeze_time
 
     @config = {
-      service: "app", image: "dhh/app", registry: { "username" => "dhh", "password" => "secret" }, servers: [ "1.1.1.1" ]
+      service: "app", image: "dhh/app", registry: { "username" => "dhh", "password" => "secret" }, servers: [ "1.1.1.1" ],
+      traefik: { "args" => { "accesslog.format" => "json", "metrics.prometheus.buckets" => "0.1,0.3,1.2,5.0" } }
     }
 
     @performer = `whoami`.strip

test/commands/lock_test.rb

@@ -3,7 +3,8 @@ require "test_helper"
 class CommandsLockTest < ActiveSupport::TestCase
   setup do
     @config = {
-      service: "app", image: "dhh/app", registry: { "username" => "dhh", "password" => "secret" }, servers: [ "1.1.1.1" ]
+      service: "app", image: "dhh/app", registry: { "username" => "dhh", "password" => "secret" }, servers: [ "1.1.1.1" ],
+      traefik: { "args" => { "accesslog.format" => "json", "metrics.prometheus.buckets" => "0.1,0.3,1.2,5.0" } }
     }
   end
 

test/commands/proxy_test.rb

@@ -1,126 +0,0 @@
-require "test_helper"
-
-class CommandsProxyTest < ActiveSupport::TestCase
-  setup do
-    @config = {
-      service: "app", image: "dhh/app", registry: { "username" => "dhh", "password" => "secret" }, servers: [ "1.1.1.1" ]
-    }
-
-    ENV["EXAMPLE_API_KEY"] = "456"
-  end
-
-  teardown do
-    ENV.delete("EXAMPLE_API_KEY")
-  end
-
-  test "run" do
-    assert_equal \
-      "docker run --name parachute_80_ --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --volume parachute_80_:/root/.config/parachute --log-opt max-size=\"10m\" #{Kamal::Configuration::Proxy::DEFAULT_IMAGE}",
-      new_command.run.join(" ")
-  end
-
-  test "run with ports configured" do
-    assert_equal \
-      "docker run --name parachute_80_ --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --volume parachute_80_:/root/.config/parachute --log-opt max-size=\"10m\" #{Kamal::Configuration::Proxy::DEFAULT_IMAGE}",
-      new_command.run.join(" ")
-  end
-
-  test "run without configuration" do
-    @config.delete(:proxy)
-
-    assert_equal \
-      "docker run --name parachute_80_ --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --volume parachute_80_:/root/.config/parachute --log-opt max-size=\"10m\" #{Kamal::Configuration::Proxy::DEFAULT_IMAGE}",
-      new_command.run.join(" ")
-  end
-
-  test "run with logging config" do
-    @config[:logging] = { "driver" => "local", "options" => { "max-size" => "100m", "max-file" => "3" } }
-
-    assert_equal \
-      "docker run --name parachute_80_ --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --volume parachute_80_:/root/.config/parachute --log-driver \"local\" --log-opt max-size=\"100m\" --log-opt max-file=\"3\" #{Kamal::Configuration::Proxy::DEFAULT_IMAGE}",
-      new_command.run.join(" ")
-  end
-
-  test "proxy start" do
-    assert_equal \
-      "docker container start parachute_80_",
-      new_command.start.join(" ")
-  end
-
-  test "proxy stop" do
-    assert_equal \
-      "docker container stop parachute_80_",
-      new_command.stop.join(" ")
-  end
-
-  test "proxy info" do
-    assert_equal \
-      "docker ps --filter name=^parachute_80_$",
-      new_command.info.join(" ")
-  end
-
-  test "proxy logs" do
-    assert_equal \
-      "docker logs parachute_80_ --timestamps 2>&1",
-      new_command.logs.join(" ")
-  end
-
-  test "proxy logs since 2h" do
-    assert_equal \
-      "docker logs parachute_80_  --since 2h --timestamps 2>&1",
-      new_command.logs(since: "2h").join(" ")
-  end
-
-  test "proxy logs last 10 lines" do
-    assert_equal \
-      "docker logs parachute_80_  --tail 10 --timestamps 2>&1",
-      new_command.logs(lines: 10).join(" ")
-  end
-
-  test "proxy logs with grep hello!" do
-    assert_equal \
-      "docker logs parachute_80_ --timestamps 2>&1 | grep 'hello!'",
-      new_command.logs(grep: "hello!").join(" ")
-  end
-
-  test "proxy remove container" do
-    assert_equal \
-      "docker container prune --force --filter label=org.opencontainers.image.title=parachute",
-      new_command.remove_container.join(" ")
-  end
-
-  test "proxy remove image" do
-    assert_equal \
-      "docker image prune --all --force --filter label=org.opencontainers.image.title=parachute",
-      new_command.remove_image.join(" ")
-  end
-
-  test "proxy follow logs" do
-    assert_equal \
-      "ssh -t root@1.1.1.1 -p 22 'docker logs parachute_80_ --timestamps --tail 10 --follow 2>&1'",
-      new_command.follow_logs(host: @config[:servers].first)
-  end
-
-  test "proxy follow logs with grep hello!" do
-    assert_equal \
-      "ssh -t root@1.1.1.1 -p 22 'docker logs parachute_80_ --timestamps --tail 10 --follow 2>&1 | grep \"hello!\"'",
-      new_command.follow_logs(host: @config[:servers].first, grep: "hello!")
-  end
-
-  test "deploy" do
-    assert_equal \
-      "docker exec parachute_80_ parachute deploy service --target \"172.1.0.2:80\"",
-      new_command.deploy("service", target: "172.1.0.2:80").join(" ")
-  end
-
-  test "remove" do
-    assert_equal \
-      "docker exec parachute_80_ parachute remove service --target \"172.1.0.2:80\"",
-      new_command.remove("service", target: "172.1.0.2:80").join(" ")
-  end
-
-  private
-    def new_command
-      Kamal::Commands::Proxy.new(Kamal::Configuration.new(@config, version: "123"))
-    end
-end

test/commands/prune_test.rb

@@ -3,7 +3,8 @@ require "test_helper"
 class CommandsPruneTest < ActiveSupport::TestCase
   setup do
     @config = {
-      service: "app", image: "dhh/app", registry: { "username" => "dhh", "password" => "secret" }, servers: [ "1.1.1.1" ]
+      service: "app", image: "dhh/app", registry: { "username" => "dhh", "password" => "secret" }, servers: [ "1.1.1.1" ],
+      traefik: { "args" => { "accesslog.format" => "json", "metrics.prometheus.buckets" => "0.1,0.3,1.2,5.0" } }
     }
   end
 
@@ -29,6 +30,12 @@ class CommandsPruneTest < ActiveSupport::TestCase
       new_command.app_containers(retain: 3).join(" ")
   end
 
+  test "healthcheck containers" do
+    assert_equal \
+      "docker container prune --force --filter label=service=healthcheck-app",
+      new_command.healthcheck_containers.join(" ")
+  end
+
   private
     def new_command
       Kamal::Commands::Prune.new(Kamal::Configuration.new(@config, version: "123"))

test/commands/server_test.rb

@@ -3,7 +3,8 @@ require "test_helper"
 class CommandsServerTest < ActiveSupport::TestCase
   setup do
     @config = {
-      service: "app", image: "dhh/app", registry: { "username" => "dhh", "password" => "secret" }, servers: [ "1.1.1.1" ]
+      service: "app", image: "dhh/app", registry: { "username" => "dhh", "password" => "secret" }, servers: [ "1.1.1.1" ],
+      traefik: { "args" => { "accesslog.format" => "json", "metrics.prometheus.buckets" => "0.1,0.3,1.2,5.0" } }
     }
   end
 

test/commands/traefik_test.rb

@@ -0,0 +1,198 @@
+require "test_helper"
+
+class CommandsTraefikTest < ActiveSupport::TestCase
+  setup do
+    @image = "traefik:test"
+
+    @config = {
+      service: "app", image: "dhh/app", registry: { "username" => "dhh", "password" => "secret" }, servers: [ "1.1.1.1" ],
+      traefik: { "image" => @image, "args" => { "accesslog.format" => "json", "api.insecure" => true, "metrics.prometheus.buckets" => "0.1,0.3,1.2,5.0" } }
+    }
+
+    ENV["EXAMPLE_API_KEY"] = "456"
+  end
+
+  teardown do
+    ENV.delete("EXAMPLE_API_KEY")
+  end
+
+  test "run" do
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+
+    @config[:traefik]["host_port"] = "8080"
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 8080:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+
+    @config[:traefik]["publish"] = false
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+  end
+
+  test "run with ports configured" do
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+
+    @config[:traefik]["options"] = { "publish" => %w[9000:9000 9001:9001] }
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" --publish \"9000:9000\" --publish \"9001:9001\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+  end
+
+  test "run with volumes configured" do
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+
+    @config[:traefik]["options"] = { "volume" => %w[./letsencrypt/acme.json:/letsencrypt/acme.json] }
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" --volume \"./letsencrypt/acme.json:/letsencrypt/acme.json\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+  end
+
+  test "run with several options configured" do
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+
+    @config[:traefik]["options"] = { "volume" => %w[./letsencrypt/acme.json:/letsencrypt/acme.json], "publish" => %w[8080:8080], "memory" => "512m" }
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" --volume \"./letsencrypt/acme.json:/letsencrypt/acme.json\" --publish \"8080:8080\" --memory \"512m\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+  end
+
+  test "run with labels configured" do
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+
+    @config[:traefik]["labels"] = { "traefik.http.routers.dashboard.service" => "api@internal", "traefik.http.routers.dashboard.middlewares" => "auth" }
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" --label traefik.http.routers.dashboard.service=\"api@internal\" --label traefik.http.routers.dashboard.middlewares=\"auth\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+  end
+
+  test "run with env configured" do
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+
+    @config[:traefik]["env"] = { "secret" => %w[EXAMPLE_API_KEY] }
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+  end
+
+  test "run without configuration" do
+    @config.delete(:traefik)
+
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{Kamal::Commands::Traefik::DEFAULT_IMAGE} --providers.docker --log.level=\"DEBUG\"",
+      new_command.run.join(" ")
+  end
+
+  test "run with logging config" do
+    @config[:logging] = { "driver" => "local", "options" => { "max-size" => "100m", "max-file" => "3" } }
+
+    assert_equal \
+      "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-driver \"local\" --log-opt max-size=\"100m\" --log-opt max-file=\"3\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"DEBUG\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+  end
+
+  test "run with default args overriden" do
+    @config[:traefik]["args"]["log.level"] = "ERROR"
+
+    assert_equal \
+    "docker run --name traefik --detach --restart unless-stopped --publish 80:80 --volume /var/run/docker.sock:/var/run/docker.sock --env-file .kamal/env/traefik/traefik.env --log-opt max-size=\"10m\" --label traefik.http.routers.catchall.entryPoints=\"http\" --label traefik.http.routers.catchall.rule=\"PathPrefix(\\`/\\`)\" --label traefik.http.routers.catchall.service=\"unavailable\" --label traefik.http.routers.catchall.priority=\"1\" --label traefik.http.services.unavailable.loadbalancer.server.port=\"0\" #{@image} --providers.docker --log.level=\"ERROR\" --accesslog.format=\"json\" --api.insecure --metrics.prometheus.buckets=\"0.1,0.3,1.2,5.0\"",
+      new_command.run.join(" ")
+  end
+
+  test "traefik start" do
+    assert_equal \
+      "docker container start traefik",
+      new_command.start.join(" ")
+  end
+
+  test "traefik stop" do
+    assert_equal \
+      "docker container stop traefik",
+      new_command.stop.join(" ")
+  end
+
+  test "traefik info" do
+    assert_equal \
+      "docker ps --filter name=^traefik$",
+      new_command.info.join(" ")
+  end
+
+  test "traefik logs" do
+    assert_equal \
+      "docker logs traefik --timestamps 2>&1",
+      new_command.logs.join(" ")
+  end
+
+  test "traefik logs since 2h" do
+    assert_equal \
+      "docker logs traefik  --since 2h --timestamps 2>&1",
+      new_command.logs(since: "2h").join(" ")
+  end
+
+  test "traefik logs last 10 lines" do
+    assert_equal \
+      "docker logs traefik  --tail 10 --timestamps 2>&1",
+      new_command.logs(lines: 10).join(" ")
+  end
+
+  test "traefik logs with grep hello!" do
+    assert_equal \
+      "docker logs traefik --timestamps 2>&1 | grep 'hello!'",
+      new_command.logs(grep: "hello!").join(" ")
+  end
+
+  test "traefik remove container" do
+    assert_equal \
+      "docker container prune --force --filter label=org.opencontainers.image.title=Traefik",
+      new_command.remove_container.join(" ")
+  end
+
+  test "traefik remove image" do
+    assert_equal \
+      "docker image prune --all --force --filter label=org.opencontainers.image.title=Traefik",
+      new_command.remove_image.join(" ")
+  end
+
+  test "traefik follow logs" do
+    assert_equal \
+      "ssh -t root@1.1.1.1 -p 22 'docker logs traefik --timestamps --tail 10 --follow 2>&1'",
+      new_command.follow_logs(host: @config[:servers].first)
+  end
+
+  test "traefik follow logs with grep hello!" do
+    assert_equal \
+      "ssh -t root@1.1.1.1 -p 22 'docker logs traefik --timestamps --tail 10 --follow 2>&1 | grep \"hello!\"'",
+      new_command.follow_logs(host: @config[:servers].first, grep: "hello!")
+  end
+
+  test "secrets io" do
+    @config[:traefik]["env"] = { "secret" => %w[EXAMPLE_API_KEY] }
+
+    assert_equal "EXAMPLE_API_KEY=456\n", new_command.env.secrets_io.string
+  end
+
+  test "make_env_directory" do
+    assert_equal "mkdir -p .kamal/env/traefik", new_command.make_env_directory.join(" ")
+  end
+
+  test "remove_env_file" do
+    assert_equal "rm -f .kamal/env/traefik/traefik.env", new_command.remove_env_file.join(" ")
+  end
+
+  private
+    def new_command
+      Kamal::Commands::Traefik.new(Kamal::Configuration.new(@config, version: "123"))
+    end
+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", "destination" ], @config.role(:web).label_args
+    assert_equal [ "--label", "service=\"app\"", "--label", "role=\"web\"", "--label", "destination", "--label", "traefik.http.services.app-web.loadbalancer.server.scheme=\"http\"", "--label", "traefik.http.routers.app-web.rule=\"PathPrefix(\\`/\\`)\"", "--label", "traefik.http.routers.app-web.priority=\"2\"", "--label", "traefik.http.middlewares.app-web-retry.retry.attempts=\"5\"", "--label", "traefik.http.middlewares.app-web-retry.retry.initialinterval=\"500ms\"", "--label", "traefik.http.routers.app-web.middlewares=\"app-web-retry@docker\"" ], @config.role(:web).label_args
   end
 
   test "custom labels" do
@@ -56,6 +56,19 @@ class ConfigurationRoleTest < ActiveSupport::TestCase
     assert_equal "70", @config_with_roles.role(:workers).labels["my.custom.label"]
   end
 
+  test "overwriting default traefik label" do
+    @deploy[:labels] = { "traefik.http.routers.app-web.rule" => "\"Host(\\`example.com\\`) || (Host(\\`example.org\\`) && Path(\\`/traefik\\`))\"" }
+    assert_equal "\"Host(\\`example.com\\`) || (Host(\\`example.org\\`) && Path(\\`/traefik\\`))\"", @config.role(:web).labels["traefik.http.routers.app-web.rule"]
+  end
+
+  test "default traefik label on non-web role" do
+    config = Kamal::Configuration.new(@deploy_with_roles.tap { |c|
+      c[:servers]["beta"] = { "traefik" => "true", "hosts" => [ "1.1.1.5" ] }
+    })
+
+    assert_equal [ "--label", "service=\"app\"", "--label", "role=\"beta\"", "--label", "destination", "--label", "traefik.http.services.app-beta.loadbalancer.server.scheme=\"http\"", "--label", "traefik.http.routers.app-beta.rule=\"PathPrefix(\\`/\\`)\"", "--label", "traefik.http.routers.app-beta.priority=\"2\"", "--label", "traefik.http.middlewares.app-beta-retry.retry.attempts=\"5\"", "--label", "traefik.http.middlewares.app-beta-retry.retry.initialinterval=\"500ms\"", "--label", "traefik.http.routers.app-beta.middlewares=\"app-beta-retry@docker\"" ], config.role(:beta).label_args
+  end
+
   test "env overwritten by role" do
     assert_equal "redis://a/b", @config_with_roles.role(:workers).env("1.1.1.3").clear["REDIS_URL"]
 
@@ -188,6 +201,26 @@ class ConfigurationRoleTest < ActiveSupport::TestCase
     assert_equal ".kamal/env/roles/app-workers.env", @config_with_roles.role(:workers).env("1.1.1.3").secrets_file
   end
 
+  test "uses cord" do
+    assert @config_with_roles.role(:web).uses_cord?
+    assert_not @config_with_roles.role(:workers).uses_cord?
+  end
+
+  test "cord host file" do
+    assert_match %r{.kamal/cords/app-web-[0-9a-f]{32}/cord}, @config_with_roles.role(:web).cord_host_file
+  end
+
+  test "cord volume" do
+    assert_equal "/tmp/kamal-cord", @config_with_roles.role(:web).cord_volume.container_path
+    assert_match %r{.kamal/cords/app-web-[0-9a-f]{32}}, @config_with_roles.role(:web).cord_volume.host_path
+    assert_equal "--volume", @config_with_roles.role(:web).cord_volume.docker_args[0]
+    assert_match %r{\$\(pwd\)/.kamal/cords/app-web-[0-9a-f]{32}:/tmp/kamal-cord}, @config_with_roles.role(:web).cord_volume.docker_args[1]
+  end
+
+  test "cord container file" do
+    assert_equal "/tmp/kamal-cord/cord", @config_with_roles.role(:web).cord_container_file
+  end
+
   test "asset path and volume args" do
     ENV["VERSION"] = "12345"
     assert_nil @config_with_roles.role(:web).asset_volume_args

test/configuration_test.rb

@@ -74,22 +74,22 @@ class ConfigurationTest < ActiveSupport::TestCase
     assert_equal "1.1.1.1", @config_with_roles.primary_host
   end
 
-  test "proxy hosts" do
+  test "traefik hosts" do
-    assert_equal [ "1.1.1.1", "1.1.1.2" ], @config_with_roles.proxy_hosts
+    assert_equal [ "1.1.1.1", "1.1.1.2" ], @config_with_roles.traefik_hosts
 
-    @deploy_with_roles[:servers]["workers"]["proxy"] = true
+    @deploy_with_roles[:servers]["workers"]["traefik"] = true
     config = Kamal::Configuration.new(@deploy_with_roles)
 
-    assert_equal [ "1.1.1.1", "1.1.1.2", "1.1.1.3" ], config.proxy_hosts
+    assert_equal [ "1.1.1.1", "1.1.1.2", "1.1.1.3" ], config.traefik_hosts
   end
 
-  test "filtered proxy hosts" do
+  test "filtered traefik hosts" do
-    assert_equal [ "1.1.1.1", "1.1.1.2" ], @config_with_roles.proxy_hosts
+    assert_equal [ "1.1.1.1", "1.1.1.2" ], @config_with_roles.traefik_hosts
 
-    @deploy_with_roles[:servers]["workers"]["proxy"] = true
+    @deploy_with_roles[:servers]["workers"]["traefik"] = true
     config = Kamal::Configuration.new(@deploy_with_roles)
 
-    assert_equal [ "1.1.1.1", "1.1.1.2", "1.1.1.3" ], config.proxy_hosts
+    assert_equal [ "1.1.1.1", "1.1.1.2", "1.1.1.3" ], config.traefik_hosts
   end
 
   test "version no git repo" do
@@ -154,6 +154,10 @@ class ConfigurationTest < ActiveSupport::TestCase
     assert_equal "app-missing", @config.service_with_version
   end
 
+  test "healthcheck service" do
+    assert_equal "healthcheck-app", @config.healthcheck_service
+  end
+
   test "valid config" do
     assert @config.valid?
     assert @config_with_roles.valid?
@@ -267,7 +271,8 @@ class ConfigurationTest < ActiveSupport::TestCase
         sshkit: {},
         volume_args: [ "--volume", "/local/path:/container/path" ],
         builder: {},
-        logging: [ "--log-opt", "max-size=\"10m\"" ] }
+        logging: [ "--log-opt", "max-size=\"10m\"" ],
+        healthcheck: { "path"=>"/up", "port"=>3000, "max_attempts" => 7, "cord" => "/tmp/kamal-cord", "log_lines" => 50 } }
 
     assert_equal expected_config, @config.to_h
   end
@@ -325,7 +330,7 @@ class ConfigurationTest < ActiveSupport::TestCase
     assert_equal "alternate_web", config.primary_role.name
     assert_equal "1.1.1.4", config.primary_host
     assert config.role(:alternate_web).primary?
-    assert config.role(:alternate_web).running_proxy?
+    assert config.role(:alternate_web).running_traefik?
   end
 
   test "primary role missing" do

test/fixtures/deploy_primary_web_role_override.yml

@@ -2,12 +2,12 @@ service: app
 image: dhh/app
 servers:
   web_chicago:
-    proxy: enabled
+    traefik: enabled
     hosts:
       - 1.1.1.1
       - 1.1.1.2
   web_tokyo:
-    proxy: enabled
+    traefik: enabled
     hosts:
       - 1.1.1.3
       - 1.1.1.4

test/fixtures/deploy_with_aliases.yml

@@ -10,7 +10,7 @@ tokyo_hosts: &tokyo_hosts
 web_common: &web_common
   env:
     ROLE: "web"
-  proxy: true
+  traefik: true
 
 # actual config
 service: app

test/fixtures/deploy_workers_only.yml

@@ -2,7 +2,7 @@ service: app
 image: dhh/app
 servers:
   workers:
-    proxy: false
+    traefik: false
     hosts:
       - 1.1.1.1
       - 1.1.1.2

test/integration/accessory_test.rb

@@ -1,6 +1,6 @@
 require_relative "integration_test"
 
-class IntegrationAccessoryTest < IntegrationTest
+class AccessoryTest < IntegrationTest
   test "boot, stop, start, restart, logs, remove" do
     kamal :envify
 

test/integration/app_test.rb

@@ -1,18 +1,16 @@
 require_relative "integration_test"
 
-class IntegrationAppTest < IntegrationTest
+class AppTest < IntegrationTest
   test "stop, start, boot, logs, images, containers, exec, remove" do
     kamal :envify
 
-    kamal :setup
-
     kamal :deploy
 
     assert_app_is_up
 
     kamal :app, :stop
 
-    assert_app_is_down response_code: "504"
+    assert_app_is_down
 
     kamal :app, :start
 
@@ -26,7 +24,7 @@ class IntegrationAppTest < IntegrationTest
     logs = kamal :app, :logs, capture: true
     assert_match /App Host: vm1/, logs
     assert_match /App Host: vm2/, logs
-    assert_match /GET \/up HTTP\/1.1/, logs
+    assert_match /GET \/ HTTP\/1.1/, logs
 
     images = kamal :app, :images, capture: true
     assert_match /App Host: vm1/, images
@@ -52,6 +50,6 @@ class IntegrationAppTest < IntegrationTest
 
     kamal :app, :remove
 
-    assert_app_is_down response_code: "504"
+    assert_app_is_down
   end
 end

test/integration/broken_deploy_test.rb

@@ -29,5 +29,6 @@ class BrokenDeployTest < IntegrationTest
       assert_match /First web container is unhealthy on vm[12], not booting other roles/, output
       assert_match "First web container is unhealthy, not booting workers on vm3", output
       assert_match "nginx: [emerg] unexpected end of file, expecting \";\" or \"}\" in /etc/nginx/conf.d/default.conf:2", output
+      assert_match 'ERROR {"Status":"unhealthy","FailingStreak":0,"Log":[]}', output
     end
 end

test/integration/docker/deployer/app/.kamal/hooks/post-proxy-reboot

@@ -1,3 +0,0 @@
-#!/bin/sh
-echo "Rebooted proxy on ${KAMAL_HOSTS}"
-mkdir -p /tmp/${TEST_ID} && touch /tmp/${TEST_ID}/post-proxy-reboot

test/integration/docker/deployer/app/.kamal/hooks/post-traefik-reboot

@@ -0,0 +1,3 @@
+#!/bin/sh
+echo "Rebooted Traefik on ${KAMAL_HOSTS}"
+mkdir -p /tmp/${TEST_ID} && touch /tmp/${TEST_ID}/post-traefik-reboot

test/integration/docker/deployer/app/.kamal/hooks/pre-proxy-reboot

@@ -1,3 +0,0 @@
-#!/bin/sh
-echo "Rebooting proxy on ${KAMAL_HOSTS}..."
-mkdir -p /tmp/${TEST_ID} && touch /tmp/${TEST_ID}/pre-proxy-reboot

test/integration/docker/deployer/app/.kamal/hooks/pre-traefik-reboot

@@ -0,0 +1,3 @@
+#!/bin/sh
+echo "Rebooting Traefik on ${KAMAL_HOSTS}..."
+mkdir -p /tmp/${TEST_ID} && touch /tmp/${TEST_ID}/pre-traefik-reboot

test/integration/docker/deployer/app/Dockerfile

@@ -6,5 +6,4 @@ ARG COMMIT_SHA
 RUN echo $COMMIT_SHA > /usr/share/nginx/html/version
 RUN mkdir -p /usr/share/nginx/html/versions && echo "version" > /usr/share/nginx/html/versions/$COMMIT_SHA
 RUN mkdir -p /usr/share/nginx/html/versions && echo "hidden" > /usr/share/nginx/html/versions/.hidden
-RUN echo "Up!" > /usr/share/nginx/html/up
 

test/integration/docker/deployer/app/config/deploy.yml

@@ -26,11 +26,14 @@ builder:
   multiarch: false
   args:
     COMMIT_SHA: <%= `git rev-parse HEAD` %>
-proxy:
+healthcheck:
-  image: registry:4443/basecamp/parachute:latest
+  cmd: wget -qO- http://localhost > /dev/null || exit 1
-  http_port: 80
+  max_attempts: 3
-  https_port: 443
+traefik:
-  debug: true
+  args:
+    accesslog: true
+    accesslog.format: json
+  image: registry:4443/traefik:v2.10
 accessories:
   busybox:
     service: custom-busybox

test/integration/docker/deployer/app_with_roles/.kamal/hooks/post-traefik-reboot

@@ -0,0 +1,3 @@
+#!/bin/sh
+echo "Rebooted Traefik on ${KAMAL_HOSTS}"
+mkdir -p /tmp/${TEST_ID} && touch /tmp/${TEST_ID}/post-traefik-reboot

test/integration/docker/deployer/app_with_roles/.kamal/hooks/pre-traefik-reboot

@@ -0,0 +1,3 @@
+#!/bin/sh
+echo "Rebooting Traefik on ${KAMAL_HOSTS}..."
+mkdir -p /tmp/${TEST_ID} && touch /tmp/${TEST_ID}/pre-traefik-reboot

test/integration/docker/deployer/app_with_roles/Dockerfile

@@ -6,4 +6,4 @@ ARG COMMIT_SHA
 RUN echo $COMMIT_SHA > /usr/share/nginx/html/version
 RUN mkdir -p /usr/share/nginx/html/versions && echo "version" > /usr/share/nginx/html/versions/$COMMIT_SHA
 RUN mkdir -p /usr/share/nginx/html/versions && echo "hidden" > /usr/share/nginx/html/versions/.hidden
-RUN echo "Up!" > /usr/share/nginx/html/up
+

test/integration/docker/deployer/app_with_roles/config/deploy.yml

@@ -20,8 +20,14 @@ builder:
   multiarch: false
   args:
     COMMIT_SHA: <%= `git rev-parse HEAD` %>
-proxy:
+healthcheck:
-  image: registry:4443/basecamp/parachute:latest
+  cmd: wget -qO- http://localhost > /dev/null || exit 1
+  max_attempts: 3
+traefik:
+  args:
+    accesslog: true
+    accesslog.format: json
+  image: registry:4443/traefik:v2.10
 accessories:
   busybox:
     service: custom-busybox

test/integration/docker/deployer/setup.sh

@@ -19,7 +19,7 @@ push_image_to_registry_4443() {
 
 install_kamal
 push_image_to_registry_4443 nginx 1-alpine-slim
-push_image_to_registry_4443 basecamp/parachute latest
+push_image_to_registry_4443 traefik v2.10
 push_image_to_registry_4443 busybox 1.36.0
 
 # .ssh is on a shared volume that persists between runs. Clean it up as the

test/integration/integration_test.rb

@@ -44,10 +44,10 @@ class IntegrationTest < ActiveSupport::TestCase
       deployer_exec(:kamal, *commands, **options)
     end
 
-    def assert_app_is_down(response_code: "503")
+    def assert_app_is_down
       response = app_response
-      debug_response_code(response, response_code)
+      debug_response_code(response, "502")
-      assert_equal response_code, response.code
+      assert_equal "502", response.code
     end
 
     def assert_app_is_up(version: nil)
@@ -101,8 +101,8 @@ class IntegrationTest < ActiveSupport::TestCase
     def assert_200(response)
       code = response.code
       if code != "200"
-        puts "Got response code #{code}, here are the proxy logs:"
+        puts "Got response code #{code}, here are the traefik logs:"
-        kamal :proxy, :logs
+        kamal :traefik, :logs
         puts "And here are the load balancer logs"
         docker_compose :logs, :load_balancer
         puts "Tried to get the response code again and got #{app_response.code}"
@@ -129,10 +129,10 @@ class IntegrationTest < ActiveSupport::TestCase
     def debug_response_code(app_response, expected_code)
       code = app_response.code
       if code != expected_code
-        puts "Got response code #{code}, here are the proxy logs:"
+        puts "Got response code #{code}, here are the traefik logs:"
-        kamal :proxy, :logs, raise_on_error: false
+        kamal :traefik, :logs
         puts "And here are the load balancer logs"
-        docker_compose :logs, :load_balancer, raise_on_error: false
+        docker_compose :logs, :load_balancer
         puts "Tried to get the response code again and got #{app_response.code}"
       end
     end

test/integration/lock_test.rb

@@ -1,6 +1,6 @@
 require_relative "integration_test"
 
-class IntegrationLockTest < IntegrationTest
+class LockTest < IntegrationTest
   test "acquire, release, status" do
     kamal :envify
 

test/integration/main_test.rb

@@ -1,15 +1,14 @@
 require_relative "integration_test"
 
-class IntegrationMainTest < IntegrationTest
+class MainTest < IntegrationTest
   test "envify, deploy, redeploy, rollback, details and audit" do
-    kamal :server, :bootstrap
     kamal :envify
     assert_env_files
     remove_local_env_file
 
     first_version = latest_app_version
 
-    assert_app_is_down response_code: "502"
+    assert_app_is_down
 
     kamal :deploy
     assert_app_is_up version: first_version
@@ -29,11 +28,11 @@ class IntegrationMainTest < IntegrationTest
     assert_app_is_up version: first_version
 
     details = kamal :details, capture: true
-    assert_match /Proxy Host: vm1/, details
+    assert_match /Traefik Host: vm1/, details
-    assert_match /Proxy Host: vm2/, details
+    assert_match /Traefik Host: vm2/, details
     assert_match /App Host: vm1/, details
     assert_match /App Host: vm2/, details
-    assert_match /basecamp\/parachute:latest/, details
+    assert_match /traefik:v2.10/, details
     assert_match /registry:4443\/app:#{first_version}/, details
 
     audit = kamal :audit, capture: true
@@ -46,12 +45,11 @@ class IntegrationMainTest < IntegrationTest
   test "app with roles" do
     @app = "app_with_roles"
 
-    kamal :server, :bootstrap
     kamal :envify
 
     version = latest_app_version
 
-    assert_app_is_down response_code: "502"
+    assert_app_is_down
 
     kamal :deploy
 
@@ -81,6 +79,7 @@ class IntegrationMainTest < IntegrationTest
     assert_equal({ user: "root", port: 22, keepalive: true, keepalive_interval: 30, log_level: :fatal }, config[:ssh_options])
     assert_equal({ "multiarch" => false, "args" => { "COMMIT_SHA" => version } }, config[:builder])
     assert_equal [ "--log-opt", "max-size=\"10m\"" ], config[:logging]
+    assert_equal({ "path" => "/up", "port" => 3000, "max_attempts" => 3, "cord"=>"/tmp/kamal-cord", "log_lines" => 50, "cmd"=>"wget -qO- http://localhost > /dev/null || exit 1" }, config[:healthcheck])
   end
 
   test "setup and remove" do

test/integration/proxy_test.rb

@@ -1,66 +0,0 @@
-require_relative "integration_test"
-
-class IntegrationProxyTest < IntegrationTest
-  test "boot, reboot, stop, start, restart, logs, remove" do
-    kamal :server, :bootstrap
-    kamal :envify
-
-    kamal :proxy, :boot
-    assert_proxy_running
-
-    output = kamal :proxy, :reboot, "-y", "--verbose", capture: true
-    assert_proxy_running
-    assert_hooks_ran "pre-proxy-reboot", "post-proxy-reboot"
-    assert_match /Rebooting proxy on vm1,vm2.../, output
-    assert_match /Rebooted proxy on vm1,vm2/, output
-
-    output = kamal :proxy, :reboot, "--rolling", "-y", "--verbose", capture: true
-    assert_proxy_running
-    assert_hooks_ran "pre-proxy-reboot", "post-proxy-reboot"
-    assert_match /Rebooting proxy on vm1.../, output
-    assert_match /Rebooted proxy on vm1/, output
-    assert_match /Rebooting proxy on vm2.../, output
-    assert_match /Rebooted proxy on vm2/, output
-
-    kamal :proxy, :boot
-    assert_proxy_running
-
-    # Check booting when booted doesn't raise an error
-    kamal :proxy, :stop
-    assert_proxy_not_running
-
-    # Check booting when stopped works
-    kamal :proxy, :boot
-    assert_proxy_running
-
-    kamal :proxy, :stop
-    assert_proxy_not_running
-
-    kamal :proxy, :start
-    assert_proxy_running
-
-    kamal :proxy, :restart
-    assert_proxy_running
-
-    logs = kamal :proxy, :logs, capture: true
-    assert_match %r{"level":"INFO","msg":"Server started","http":80,"https":443}, logs
-
-    kamal :proxy, :remove
-    assert_proxy_not_running
-
-    kamal :env, :delete
-  end
-
-  private
-    def assert_proxy_running
-      assert_match %r{registry:4443/basecamp/parachute:latest   "parachute run"}, proxy_details
-    end
-
-    def assert_proxy_not_running
-      assert_no_match %r{registry:4443/basecamp/parachute:latest   "parachute run"}, proxy_details
-    end
-
-    def proxy_details
-      kamal :proxy, :details, capture: true
-    end
-end

test/integration/traefik_test.rb

@@ -0,0 +1,65 @@
+require_relative "integration_test"
+
+class TraefikTest < IntegrationTest
+  test "boot, reboot, stop, start, restart, logs, remove" do
+    kamal :envify
+
+    kamal :traefik, :boot
+    assert_traefik_running
+
+    output = kamal :traefik, :reboot, "-y", "--verbose", capture: true
+    assert_traefik_running
+    assert_hooks_ran "pre-traefik-reboot", "post-traefik-reboot"
+    assert_match /Rebooting Traefik on vm1,vm2.../, output
+    assert_match /Rebooted Traefik on vm1,vm2/, output
+
+    output = kamal :traefik, :reboot, "--rolling", "-y", "--verbose", capture: true
+    assert_traefik_running
+    assert_hooks_ran "pre-traefik-reboot", "post-traefik-reboot"
+    assert_match /Rebooting Traefik on vm1.../, output
+    assert_match /Rebooted Traefik on vm1/, output
+    assert_match /Rebooting Traefik on vm2.../, output
+    assert_match /Rebooted Traefik on vm2/, output
+
+    kamal :traefik, :boot
+    assert_traefik_running
+
+    # Check booting when booted doesn't raise an error
+    kamal :traefik, :stop
+    assert_traefik_not_running
+
+    # Check booting when stopped works
+    kamal :traefik, :boot
+    assert_traefik_running
+
+    kamal :traefik, :stop
+    assert_traefik_not_running
+
+    kamal :traefik, :start
+    assert_traefik_running
+
+    kamal :traefik, :restart
+    assert_traefik_running
+
+    logs = kamal :traefik, :logs, capture: true
+    assert_match /Traefik version [\d.]+ built on/, logs
+
+    kamal :traefik, :remove
+    assert_traefik_not_running
+
+    kamal :env, :delete
+  end
+
+  private
+    def assert_traefik_running
+      assert_match /traefik:v2.10   "\/entrypoint.sh/, traefik_details
+    end
+
+    def assert_traefik_not_running
+      assert_no_match /traefik:v2.10   "\/entrypoint.sh/, traefik_details
+    end
+
+    def traefik_details
+      kamal :traefik, :details, capture: true
+    end
+end