Bump version for 1.7.0

Configuration validation

Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    kamal (1.5.2)
+    kamal (1.7.0)
       activesupport (>= 7.0)
       base64 (~> 0.2)
       bcrypt_pbkdf (~> 1.0)
@@ -11,6 +11,7 @@ PATH
       net-ssh (~> 7.0)
       sshkit (>= 1.22.2, < 2.0)
       thor (~> 1.2)
+      x25519 (~> 1.0, >= 1.0.10)
       zeitwerk (~> 2.5)
 
 GEM
@@ -165,6 +166,7 @@ GEM
       concurrent-ruby (~> 1.0)
     unicode-display_width (2.5.0)
     webrick (1.8.1)
+    x25519 (1.0.10)
     zeitwerk (2.6.12)
 
 PLATFORMS

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 a [custom proxy](https://github.com/basecamp/kamal-proxy) 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).
 

bin/docs

@@ -0,0 +1,134 @@
+#!/usr/bin/env ruby
+require "stringio"
+
+def usage
+  puts "Usage: #{$0} <kamal_site_repo>"
+  exit 1
+end
+
+usage if ARGV.size != 1
+
+kamal_site_repo = ARGV[0]
+
+if !File.directory?(kamal_site_repo)
+  puts "Error: #{kamal_site_repo} is not a directory"
+  exit 1
+end
+
+DOCS = {
+  "accessory" => "Accessories",
+  "boot" => "Booting",
+  "builder" => "Builders",
+  "configuration" => "Configuration overview",
+  "env" => "Environment variables",
+  "healthcheck" => "Healthchecks",
+  "logging" => "Logging",
+  "registry" => "Docker Registry",
+  "role" => "Roles",
+  "servers" => "Servers",
+  "ssh" => "SSH",
+  "sshkit" => "SSHKit",
+  "traefik" => "Traefik"
+}
+
+class DocWriter
+  attr_reader :from_file, :to_file, :key, :heading, :body, :output, :in_yaml
+
+  def initialize(from_file, to_dir)
+    @from_file = from_file
+    @key = File.basename(from_file, ".yml")
+    @to_file = File.join(to_dir, "#{linkify(DOCS[key])}.md")
+    @body = File.readlines(from_file)
+    @heading = body.shift.chomp("\n")
+    @output = nil
+  end
+
+  def write
+    puts "Writing #{to_file}"
+    generate_markdown
+    File.write(to_file, output.string)
+  end
+
+  private
+    def generate_markdown
+      @output = StringIO.new
+
+      generate_header
+
+      place = :in_section
+
+      loop do
+        line = body.shift&.chomp("\n")
+        break if line.nil?
+
+        case place
+        when :new_section, :in_section
+          if line.empty?
+            output.puts
+            place = :new_section
+          elsif line =~ /^ *#/
+            generate_line(line, place: place)
+            place = :in_section
+          else
+            output.puts "```yaml"
+            output.print line
+            place = :in_yaml
+          end
+        when :in_yaml
+          if line =~ /^ *#/
+            output.puts "```"
+            generate_line(line, place: :new_section)
+            place = :in_section
+          else
+            output.puts
+            output.print line
+          end
+        end
+      end
+
+      output.puts "\n```" if place == :in_yaml
+    end
+
+    def generate_header
+      output.puts "---"
+      output.puts "title: #{heading[2..-1]}"
+      output.puts "---"
+      output.puts
+      output.puts heading
+      output.puts
+    end
+
+    def generate_line(line, place: :in_section)
+      line = line.gsub(/^ *#\s?/, "")
+
+      if line =~ /(.*)kamal docs ([a-z]*)(.*)/
+        line = "#{$1}[#{DOCS[$2]}](../#{linkify(DOCS[$2])})#{$3}"
+      end
+
+      if line =~ /(.*)https:\/\/kamal-deploy.org([a-z\/-]*)(.*)/
+        line = "#{$1}[#{titlify($2.split("/").last)}](#{$2})#{$3}"
+      end
+
+      if place == :new_section
+        output.puts "## [#{line}](##{linkify(line)})"
+      else
+        output.puts line
+      end
+    end
+
+    def linkify(text)
+      text.downcase.gsub(" ", "-")
+    end
+
+    def titlify(text)
+      text.capitalize.gsub("-", " ")
+    end
+end
+
+from_dir = File.join(File.dirname(__FILE__), "../lib/kamal/configuration/docs")
+to_dir = File.join(kamal_site_repo, "docs/configuration")
+Dir.glob("#{from_dir}/*") do |from_file|
+  key = File.basename(from_file, ".yml")
+
+  DocWriter.new(from_file, to_dir).write
+end

kamal.gemspec

@@ -18,6 +18,7 @@ Gem::Specification.new do |spec|
   spec.add_dependency "dotenv", "~> 2.8"
   spec.add_dependency "zeitwerk", "~> 2.5"
   spec.add_dependency "ed25519", "~> 1.2"
+  spec.add_dependency "x25519", "~> 1.0", ">= 1.0.10"
   spec.add_dependency "bcrypt_pbkdf", "~> 1.0"
   spec.add_dependency "concurrent-ruby", "~> 1.2"
   spec.add_dependency "base64", "~> 0.2"

lib/kamal.rb

@@ -1,8 +1,10 @@
 module Kamal
+  class ConfigurationError < StandardError; end
 end
 
 require "active_support"
 require "zeitwerk"
+require "yaml"
 
 loader = Zeitwerk::Loader.for_gem
 loader.ignore(File.join(__dir__, "kamal", "sshkit_with_ext.rb"))

lib/kamal/cli.rb

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

lib/kamal/cli/accessory.rb

@@ -149,23 +149,25 @@ class Kamal::Cli::Accessory < Kamal::Cli::Base
   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 :grep_options, aliases: "-o", desc: "Additional options supplied to grep"
   option :follow, aliases: "-f", desc: "Follow logs on primary server (or specific host set by --hosts)"
   def logs(name)
     with_accessory(name) do |accessory, hosts|
       grep = options[:grep]
+      grep_options = options[:grep_options]
 
       if options[:follow]
         run_locally do
           info "Following logs on #{hosts}..."
-          info accessory.follow_logs(grep: grep)
+          info accessory.follow_logs(grep: grep, grep_options: grep_options)
-          exec accessory.follow_logs(grep: grep)
+          exec accessory.follow_logs(grep: grep, grep_options: grep_options)
         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(hosts) do
-          puts capture_with_info(*accessory.logs(since: since, lines: lines, grep: grep))
+          puts capture_with_info(*accessory.logs(since: since, lines: lines, grep: grep, grep_options: grep_options))
         end
       end
     end

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
@@ -186,12 +166,15 @@ class Kamal::Cli::App < Kamal::Cli::Base
   option :since, aliases: "-s", desc: "Show lines 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 lines to show from each server"
   option :grep, aliases: "-g", desc: "Show lines with grep match only (use this to fetch specific requests by id)"
+  option :grep_options, aliases: "-o", desc: "Additional options supplied to grep"
   option :follow, aliases: "-f", desc: "Follow log on primary server (or specific host set by --hosts)"
   def logs
     # FIXME: Catch when app containers aren't running
 
     grep = options[:grep]
+    grep_options = options[:grep_options]
     since = options[:since]
+
     if options[:follow]
       lines = options[:lines].presence || ((since || grep) ? nil : 10) # Default to 10 lines if since or grep isn't set
 
@@ -202,8 +185,8 @@ class Kamal::Cli::App < Kamal::Cli::Base
         role = KAMAL.roles_on(KAMAL.primary_host).first
 
         app = KAMAL.app(role: role, host: host)
-        info app.follow_logs(host: KAMAL.primary_host, lines: lines, grep: grep)
+        info app.follow_logs(host: KAMAL.primary_host, lines: lines, grep: grep, grep_options: grep_options)
-        exec app.follow_logs(host: KAMAL.primary_host, lines: lines, grep: grep)
+        exec app.follow_logs(host: KAMAL.primary_host, lines: lines, grep: grep, grep_options: grep_options)
       end
     else
       lines = options[:lines].presence || ((since || grep) ? nil : 100) # Default to 100 lines if since or grep isn't set
@@ -213,7 +196,7 @@ class Kamal::Cli::App < Kamal::Cli::Base
 
         roles.each do |role|
           begin
-            puts_by_host host, capture_with_info(*KAMAL.app(role: role, host: host).logs(since: since, lines: lines, grep: grep))
+            puts_by_host host, capture_with_info(*KAMAL.app(role: role, host: host).logs(since: since, lines: lines, grep: grep, grep_options: grep_options))
           rescue SSHKit::Command::Failed
             puts_by_host host, "Nothing found"
           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

@@ -35,22 +35,25 @@ class Kamal::Cli::Build < Kamal::Cli::Base
 
     run_locally do
       begin
-        KAMAL.with_verbosity(:debug) do
+        context_hosts = capture_with_info(*KAMAL.builder.context_hosts).split("\n")
-          Dir.chdir(KAMAL.config.builder.build_directory) { execute *push }
+
+        if context_hosts != KAMAL.builder.config_context_hosts
+          warn "Context hosts have changed, so re-creating builder, was: #{context_hosts.join(", ")}], now: #{KAMAL.builder.config_context_hosts.join(", ")}"
+          cli.remove
+          cli.create
         end
       rescue SSHKit::Command::Failed => e
-        if e.message =~ /(no builder)|(no such file or directory)/
         warn "Missing compatible builder, so creating a new one first"
-
+        if e.message =~ /(context not found|no builder)/
-          if cli.create
+          cli.create
-            KAMAL.with_verbosity(:debug) do
-              Dir.chdir(KAMAL.config.builder.build_directory) { execute *push }
-            end
-          end
         else
           raise
         end
       end
+
+      KAMAL.with_verbosity(:debug) do
+        Dir.chdir(KAMAL.config.builder.build_directory) { execute *push }
+      end
     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

@@ -25,7 +25,7 @@ class Kamal::Cli::Main < Kamal::Cli::Base
       invoke_options = deploy_options
 
       say "Log into image registry...", :magenta
-      invoke "kamal:cli:registry:login", [], invoke_options
+      invoke "kamal:cli:registry:login", [], invoke_options.merge(skip_local: options[:skip_push])
 
       if options[:skip_push]
         say "Pull app image...", :magenta
@@ -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
@@ -126,6 +126,18 @@ class Kamal::Cli::Main < Kamal::Cli::Base
     end
   end
 
+  desc "docs", "Show Kamal documentation for configuration setting"
+  def docs(section = nil)
+    case section
+    when NilClass
+      puts Kamal::Configuration.validation_doc
+    else
+      puts Kamal::Configuration.const_get(section.titlecase.to_sym).validation_doc
+    end
+  rescue NameError
+    puts "No documentation found for #{section}"
+  end
+
   desc "init", "Create config stub in config/deploy.yml and env stub in .env"
   option :bundle, type: :boolean, default: false, desc: "Add Kamal to the Gemfile and create a bin/kamal binstub"
   def init
@@ -189,15 +201,15 @@ 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)
+        invoke "kamal:cli:registry:logout", [], options.without(:confirmed).merge(skip_local: true)
       end
     end
   end
@@ -231,8 +243,8 @@ 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

@@ -1,164 +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 "update", "Update from Traefik to kamal-proxy, for when moving from Kamal v1 to Kamal v2"
-  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 update
-    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
-            info "Updating proxy from Traefik to kamal-proxy on #{host}..."
-            execute *KAMAL.auditor.record("Updated proxy from Traefik to kamal-proxy"), verbosity: :debug
-            execute *KAMAL.registry.login
-
-            info "Stopping and removing Traefik on #{host}..."
-            execute *KAMAL.proxy.stop(name: "traefik"), raise_on_non_zero_exit: false
-            execute *KAMAL.proxy.remove_container(filter: "label=org.opencontainers.image.title=traefik")
-            execute *KAMAL.proxy.remove_image(filter: "label=org.opencontainers.image.title=traefik")
-
-            info "Stopping and removing kamal-proxy on #{host}, if running..."
-            execute *KAMAL.proxy.stop, raise_on_non_zero_exit: false
-            execute *KAMAL.proxy.remove_container
-
-            info "Starting kamal-proxy on #{host}..."
-            execute *KAMAL.proxy.run
-
-            KAMAL.roles_on(host).select(&:running_proxy?).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
-              endpoint = capture_with_info(*app.container_endpoint(version: version)).strip
-              raise Kamal::Cli::BootError, "Failed to get endpoint for #{role} on #{host}, is the app container running?" if endpoint.empty?
-
-              info "Deploying #{endpoint} for role `#{role}` on #{host}..."
-              execute *KAMAL.proxy.deploy(role.container_prefix, target: endpoint)
-            end
-          end
-          run_hook "post-proxy-reboot", hosts: host_list
-        end
-      end
-    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/registry.rb

@@ -1,18 +1,17 @@
 class Kamal::Cli::Registry < Kamal::Cli::Base
   desc "login", "Log in to registry locally and remotely"
+  option :skip_local, aliases: "-L", type: :boolean, default: false, desc: "Skip local login"
+  option :skip_remote, aliases: "-R", type: :boolean, default: false, desc: "Skip remote login"
   def login
-    run_locally    { execute *KAMAL.registry.login }
+    run_locally    { execute *KAMAL.registry.login } unless options[:skip_local]
-    on(KAMAL.hosts) { execute *KAMAL.registry.login }
+    on(KAMAL.hosts) { execute *KAMAL.registry.login } unless options[:skip_remote]
-  # FIXME: This rescue needed?
-  rescue ArgumentError => e
-    puts e.message
   end
 
-  desc "logout", "Log out of registry remotely"
+  desc "logout", "Log out of registry locally and remotely"
+  option :skip_local, aliases: "-L", type: :boolean, default: false, desc: "Skip local login"
+  option :skip_remote, aliases: "-R", type: :boolean, default: false, desc: "Skip remote login"
   def logout
-    on(KAMAL.hosts) { execute *KAMAL.registry.logout }
+    run_locally    { execute *KAMAL.registry.logout } unless options[:skip_local]
-  # FIXME: This rescue needed?
+    on(KAMAL.hosts) { execute *KAMAL.registry.logout } unless options[:skip_remote]
-  rescue ArgumentError => e
-    puts e.message
   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/docker-setup.sample

@@ -1,4 +1,4 @@
-#!/usr/bin/env ruby
+#!/bin/sh
 
 # A sample docker-setup hook
 #

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 +1,122 @@
+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 :grep_options, aliases: "-o", desc: "Additional options supplied to grep"
+  option :follow, aliases: "-f", desc: "Follow logs on primary server (or specific host set by --hosts)"
+  def logs
+    grep = options[:grep]
+    grep_options = options[:grep_options]
+
+    if options[:follow]
+      run_locally do
+        info "Following logs on #{KAMAL.primary_host}..."
+        info KAMAL.traefik.follow_logs(host: KAMAL.primary_host, grep: grep, grep_options: grep_options)
+        exec KAMAL.traefik.follow_logs(host: KAMAL.primary_host, grep: grep, grep_options: grep_options)
+      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, grep_options: grep_options)), 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/accessory.rb

@@ -36,17 +36,17 @@ class Kamal::Commands::Accessory < Kamal::Commands::Base
   end
 
 
-  def logs(since: nil, lines: nil, grep: nil)
+  def logs(since: nil, lines: nil, grep: nil, grep_options: nil)
     pipe \
       docker(:logs, service_name, (" --since #{since}" if since), (" --tail #{lines}" if lines), "--timestamps", "2>&1"),
-      ("grep '#{grep}'" if grep)
+      ("grep '#{grep}'#{" #{grep_options}" if grep_options}" if grep)
   end
 
-  def follow_logs(grep: nil)
+  def follow_logs(grep: nil, grep_options: nil)
     run_over_ssh \
       pipe \
         docker(:logs, service_name, "--timestamps", "--tail", "10", "--follow", "2>&1"),
-        (%(grep "#{grep}") if grep)
+        (%(grep "#{grep}"#{" #{grep_options}" if grep_options}) if grep)
   end
 
 

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/app/logging.rb

@@ -1,17 +1,17 @@
 module Kamal::Commands::App::Logging
-  def logs(version: nil, since: nil, lines: nil, grep: nil)
+  def logs(version: nil, since: nil, lines: nil, grep: nil, grep_options: nil)
     pipe \
       version ? container_id_for_version(version) : current_running_container_id,
       "xargs docker logs#{" --since #{since}" if since}#{" --tail #{lines}" if lines} 2>&1",
-      ("grep '#{grep}'" if grep)
+      ("grep '#{grep}'#{" #{grep_options}" if grep_options}" if grep)
   end
 
-  def follow_logs(host:, lines: nil, grep: nil)
+  def follow_logs(host:, lines: nil, grep: nil, grep_options: nil)
     run_over_ssh \
       pipe(
         current_running_container_id,
         "xargs docker logs --timestamps#{" --tail #{lines}" if lines} --follow 2>&1",
-        (%(grep "#{grep}") if grep)
+        (%(grep "#{grep}"#{" #{grep_options}" if grep_options}) if grep)
       ),
       host: host
   end

lib/kamal/commands/builder.rb

@@ -1,7 +1,8 @@
 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 :create, :remove, :push, :clean, :pull, :info, :context_hosts, :config_context_hosts, :validate_image,
+    to: :target
 
   include Clone
 
@@ -10,18 +11,23 @@ class Kamal::Commands::Builder < Kamal::Commands::Base
   end
 
   def target
-    case
+    if config.builder.multiarch?
-    when !config.builder.multiarch? && !config.builder.cached?
+      if config.builder.remote?
-      native
+        if config.builder.local?
-    when !config.builder.multiarch? && config.builder.cached?
-      native_cached
-    when config.builder.local? && config.builder.remote?
           multiarch_remote
-    when config.builder.remote?
+        else
           native_remote
+        end
       else
         multiarch
       end
+    else
+      if config.builder.cached?
+        native_cached
+      else
+        native
+      end
+    end
   end
 
   def native

lib/kamal/commands/builder/base.rb

@@ -2,6 +2,8 @@
 class Kamal::Commands::Builder::Base < Kamal::Commands::Base
   class BuilderError < StandardError; end
 
+  ENDPOINT_DOCKER_HOST_INSPECT = "'{{.Endpoints.docker.Host}}'"
+
   delegate :argumentize, to: Kamal::Utils
   delegate :args, :secrets, :dockerfile, :target, :local_arch, :local_host, :remote_arch, :remote_host, :cache_from, :cache_to, :ssh, to: :builder_config
 
@@ -30,6 +32,13 @@ class Kamal::Commands::Builder::Base < Kamal::Commands::Base
       )
   end
 
+  def context_hosts
+    :true
+  end
+
+  def config_context_hosts
+    []
+  end
 
   private
     def build_tags
@@ -74,4 +83,8 @@ class Kamal::Commands::Builder::Base < Kamal::Commands::Base
     def builder_config
       config.builder
     end
+
+    def context_host(builder_name)
+      docker :context, :inspect, builder_name, "--format", ENDPOINT_DOCKER_HOST_INSPECT
+    end
 end

lib/kamal/commands/builder/multiarch.rb

@@ -22,6 +22,10 @@ class Kamal::Commands::Builder::Multiarch < Kamal::Commands::Builder::Base
       build_context
   end
 
+  def context_hosts
+    docker :buildx, :inspect, builder_name, "> /dev/null"
+  end
+
   private
     def builder_name
       "kamal-#{config.service}-multiarch"

lib/kamal/commands/builder/multiarch/remote.rb

@@ -12,6 +12,16 @@ class Kamal::Commands::Builder::Multiarch::Remote < Kamal::Commands::Builder::Mu
       super
   end
 
+  def context_hosts
+    chain \
+      context_host(builder_name_with_arch(local_arch)),
+      context_host(builder_name_with_arch(remote_arch))
+  end
+
+  def config_context_hosts
+    [ local_host, remote_host ].compact
+  end
+
   private
     def builder_name
       super + "-remote"

lib/kamal/commands/builder/native/cached.rb

@@ -1,6 +1,6 @@
 class Kamal::Commands::Builder::Native::Cached < Kamal::Commands::Builder::Native
   def create
-    docker :buildx, :create, "--use", "--driver=docker-container"
+    docker :buildx, :create, "--name", builder_name, "--use", "--driver=docker-container"
   end
 
   def remove
@@ -13,4 +13,13 @@ class Kamal::Commands::Builder::Native::Cached < Kamal::Commands::Builder::Nativ
       *build_options,
       build_context
   end
+
+  def context_hosts
+    docker :buildx, :inspect, builder_name, "> /dev/null"
+  end
+
+  private
+    def builder_name
+      "kamal-#{config.service}-native-cached"
+    end
 end

lib/kamal/commands/builder/native/remote.rb

@@ -26,6 +26,14 @@ class Kamal::Commands::Builder::Native::Remote < Kamal::Commands::Builder::Nativ
     build_context
   end
 
+  def context_hosts
+    context_host(builder_name_with_arch)
+  end
+
+  def config_context_hosts
+    [ remote_host ]
+  end
+
 
   private
     def builder_name

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/kamal-proxy",
-      *config.logging_args,
-      *proxy_config.docker_options_args,
-      proxy_config.image
-  end
-
-  def start
-    docker :container, :start, container_name
-  end
-
-  def stop(name: container_name)
-    docker :container, :stop, name
-  end
-
-  def start_or_run
-    combine start, run, by: "||"
-  end
-
-  def deploy(service, target:)
-    optionize({ target: target })
-    docker :exec, container_name, "kamal-proxy", :deploy, service, *optionize({ target: target }), *proxy_config.deploy_command_args
-  end
-
-  def remove(service, target:)
-    docker :exec, container_name, "kamal-proxy", :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(filter: container_filter)
-    docker :container, :prune, "--force", "--filter", filter
-  end
-
-  def remove_image(filter: image_filter)
-    docker :image, :prune, "--all", "--force", "--filter", filter
-  end
-
-  private
-    def container_filter
-      "label=org.opencontainers.image.title=kamal-proxy"
-    end
-
-    def image_filter
-      "label=org.opencontainers.image.title=kamal-proxy"
-    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/registry.rb

@@ -3,21 +3,12 @@ class Kamal::Commands::Registry < Kamal::Commands::Base
 
   def login
     docker :login,
-      registry["server"],
+      registry.server,
-      "-u", sensitive(Kamal::Utils.escape_shell_value(lookup("username"))),
+      "-u", sensitive(Kamal::Utils.escape_shell_value(registry.username)),
-      "-p", sensitive(Kamal::Utils.escape_shell_value(lookup("password")))
+      "-p", sensitive(Kamal::Utils.escape_shell_value(registry.password))
   end
 
   def logout
-    docker :logout, registry["server"]
+    docker :logout, registry.server
-  end
-
-  private
-    def lookup(key)
-      if registry[key].is_a?(Array)
-        ENV.fetch(registry[key].first).dup
-      else
-        registry[key]
-      end
   end
 end

lib/kamal/commands/traefik.rb

@@ -0,0 +1,85 @@
+class Kamal::Commands::Traefik < Kamal::Commands::Base
+  delegate :argumentize, :optionize, to: Kamal::Utils
+  delegate :port, :publish?, :labels, :env, :image, :options, :args, to: :"config.traefik"
+
+  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, grep_options: nil)
+    pipe \
+      docker(:logs, "traefik", (" --since #{since}" if since), (" --tail #{lines}" if lines), "--timestamps", "2>&1"),
+      ("grep '#{grep}'#{" #{grep_options}" if grep_options}" if grep)
+  end
+
+  def follow_logs(host:, grep: nil, grep_options: nil)
+    run_over_ssh pipe(
+      docker(:logs, "traefik", "--timestamps", "--tail", "10", "--follow", "2>&1"),
+      (%(grep "#{grep}"#{" #{grep_options}" if grep_options}) 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 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 if publish?
+    end
+
+    def label_args
+      argumentize "--label", labels
+    end
+
+    def env_args
+      env.args
+    end
+
+    def docker_options_args
+      optionize(options)
+    end
+
+    def cmd_option_args
+      optionize args, with: "="
+    end
+end

lib/kamal/configuration.rb

@@ -1,15 +1,19 @@
 require "active_support/ordered_options"
 require "active_support/core_ext/string/inquiry"
 require "active_support/core_ext/module/delegation"
+require "active_support/core_ext/hash/keys"
 require "pathname"
 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, :labels, :stop_wait_time, :hooks_path, to: :raw_config, allow_nil: true
   delegate :argumentize, :optionize, to: Kamal::Utils
 
   attr_reader :destination, :raw_config
+  attr_reader :accessories, :boot, :builder, :env, :healthcheck, :logging, :traefik, :servers, :ssh, :sshkit, :registry
+
+  include Validation
 
   class << self
     def create_from(config_file:, destination: nil, version: nil)
@@ -42,7 +46,29 @@ class Kamal::Configuration
     @raw_config = ActiveSupport::InheritableOptions.new(raw_config)
     @destination = destination
     @declared_version = version
-    valid? if validate
+
+    validate! raw_config, example: validation_yml.symbolize_keys, context: ""
+
+    # Eager load config to validate it, these are first as they have dependencies later on
+    @servers = Servers.new(config: self)
+    @registry = Registry.new(config: self)
+
+    @accessories = @raw_config.accessories&.keys&.collect { |name| Accessory.new(name, config: self) } || []
+    @boot = Boot.new(config: self)
+    @builder = Builder.new(config: self)
+    @env = Env.new(config: @raw_config.env || {})
+
+    @healthcheck = Healthcheck.new(healthcheck_config: @raw_config.healthcheck)
+    @logging = Logging.new(logging_config: @raw_config.logging)
+    @traefik = Traefik.new(config: self)
+    @ssh = Ssh.new(config: self)
+    @sshkit = Sshkit.new(config: self)
+
+    ensure_destination_if_required
+    ensure_required_keys_present
+    ensure_valid_kamal_version
+    ensure_retain_containers_valid
+    ensure_valid_service_name
   end
 
 
@@ -71,17 +97,13 @@ class Kamal::Configuration
 
 
   def roles
-    @roles ||= role_names.collect { |role_name| Role.new(role_name, config: self) }
+    servers.roles
   end
 
   def role(name)
     roles.detect { |r| r.name == name.to_s }
   end
 
-  def accessories
-    @accessories ||= raw_config.accessories&.keys&.collect { |name| Kamal::Configuration::Accessory.new(name, config: self) } || []
-  end
-
   def accessory(name)
     accessories.detect { |a| a.name == name.to_s }
   end
@@ -107,20 +129,20 @@ 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
-    [ raw_config.registry["server"], image ].compact.join("/")
+    [ registry.server, image ].compact.join("/")
   end
 
   def absolute_image
@@ -157,36 +179,14 @@ class Kamal::Configuration
   end
 
   def logging_args
-    if logging.present?
+    logging.args
-      optionize({ "log-driver" => logging["driver"] }.compact) +
-        argumentize("--log-opt", logging["options"])
-    else
-      argumentize("--log-opt", { "max-size" => "10m" })
-    end
   end
 
 
-  def boot
+  def healthcheck_service
-    Kamal::Configuration::Boot.new(config: self)
+    [ "healthcheck", service, destination ].compact.join("-")
   end
 
-  def builder
-    Kamal::Configuration::Builder.new(config: self)
-  end
-
-  def proxy
-    Kamal::Configuration::Proxy.new(config: self)
-  end
-
-  def ssh
-    Kamal::Configuration::Ssh.new(config: self)
-  end
-
-  def sshkit
-    Kamal::Configuration::Sshkit.new(config: self)
-  end
-
-
   def readiness_delay
     raw_config.readiness_delay || 7
   end
@@ -221,13 +221,9 @@ class Kamal::Configuration
     File.join(run_directory, "env")
   end
 
-  def env
-    raw_config.env || {}
-  end
-
   def env_tags
     @env_tags ||= if (tags = raw_config.env["tags"])
-      tags.collect { |name, config| Kamal::Configuration::Env::Tag.new(name, config: config) }
+      tags.collect { |name, config| Env::Tag.new(name, config: config) }
     else
       []
     end
@@ -238,10 +234,6 @@ class Kamal::Configuration
   end
 
 
-  def valid?
-    ensure_destination_if_required && ensure_required_keys_present && ensure_valid_kamal_version && ensure_retain_containers_valid && ensure_valid_service_name
-  end
-
   def to_h
     {
       roles: role_names,
@@ -256,11 +248,11 @@ class Kamal::Configuration
       sshkit: sshkit.to_h,
       builder: builder.to_h,
       accessories: raw_config.accessories,
-      logging: logging_args
+      logging: logging_args,
+      healthcheck: healthcheck.to_h
     }.compact
   end
 
-
   private
     # Will raise ArgumentError if any required config keys are missing
     def ensure_destination_if_required
@@ -273,29 +265,21 @@ class Kamal::Configuration
 
     def ensure_required_keys_present
       %i[ service image registry servers ].each do |key|
-        raise ArgumentError, "Missing required configuration for #{key}" unless raw_config[key].present?
+        raise Kamal::ConfigurationError, "Missing required configuration for #{key}" unless raw_config[key].present?
       end
 
-      if raw_config.registry["username"].blank?
+      unless role(primary_role_name).present?
-        raise ArgumentError, "You must specify a username for the registry in config/deploy.yml"
+        raise Kamal::ConfigurationError, "The primary_role #{primary_role_name} isn't defined"
-      end
-
-      if raw_config.registry["password"].blank?
-        raise ArgumentError, "You must specify a password for the registry in config/deploy.yml (or set the ENV variable if that's used)"
-      end
-
-      unless role_names.include?(primary_role_name)
-        raise ArgumentError, "The primary_role #{primary_role_name} isn't defined"
       end
 
       if primary_role.hosts.empty?
-        raise ArgumentError, "No servers specified for the #{primary_role.name} primary_role"
+        raise Kamal::ConfigurationError, "No servers specified for the #{primary_role.name} primary_role"
       end
 
       unless allow_empty_roles?
         roles.each do |role|
           if role.hosts.empty?
-            raise ArgumentError, "No servers specified for the #{role.name} role. You can ignore this with allow_empty_roles: true"
+            raise Kamal::ConfigurationError, "No servers specified for the #{role.name} role. You can ignore this with allow_empty_roles: true"
           end
         end
       end
@@ -304,21 +288,21 @@ class Kamal::Configuration
     end
 
     def ensure_valid_service_name
-      raise ArgumentError, "Service name can only include alphanumeric characters, hyphens, and underscores" unless raw_config[:service] =~ /^[a-z0-9_-]+$/i
+      raise Kamal::ConfigurationError, "Service name can only include alphanumeric characters, hyphens, and underscores" unless raw_config[:service] =~ /^[a-z0-9_-]+$/i
 
       true
     end
 
     def ensure_valid_kamal_version
       if minimum_version && Gem::Version.new(minimum_version) > Gem::Version.new(Kamal::VERSION)
-        raise ArgumentError, "Current version is #{Kamal::VERSION}, minimum required is #{minimum_version}"
+        raise Kamal::ConfigurationError, "Current version is #{Kamal::VERSION}, minimum required is #{minimum_version}"
       end
 
       true
     end
 
     def ensure_retain_containers_valid
-      raise ArgumentError, "Must retain at least 1 container" if retain_containers < 1
+      raise Kamal::ConfigurationError, "Must retain at least 1 container" if retain_containers < 1
 
       true
     end

lib/kamal/configuration/accessory.rb

@@ -1,30 +1,39 @@
 class Kamal::Configuration::Accessory
+  include Kamal::Configuration::Validation
+
   delegate :argumentize, :optionize, to: Kamal::Utils
 
-  attr_accessor :name, :specifics
+  attr_reader :name, :accessory_config, :env
 
   def initialize(name, config:)
-    @name, @config, @specifics = name.inquiry, config, config.raw_config["accessories"][name]
+    @name, @config, @accessory_config = name.inquiry, config, config.raw_config["accessories"][name]
+
+    validate! \
+      accessory_config,
+      example: validation_yml["accessories"]["mysql"],
+      context: "accessories/#{name}",
+      with: Kamal::Configuration::Validator::Accessory
+
+    @env = Kamal::Configuration::Env.new \
+      config: accessory_config.fetch("env", {}),
+      secrets_file: File.join(config.host_env_directory, "accessories", "#{service_name}.env"),
+      context: "accessories/#{name}/env"
   end
 
   def service_name
-    specifics["service"] || "#{config.service}-#{name}"
+    accessory_config["service"] || "#{config.service}-#{name}"
   end
 
   def image
-    specifics["image"]
+    accessory_config["image"]
   end
 
   def hosts
-    if (specifics.keys & [ "host", "hosts", "roles" ]).size != 1
-      raise ArgumentError, "Specify one of `host`, `hosts` or `roles` for accessory `#{name}`"
-    end
-
     hosts_from_host || hosts_from_hosts || hosts_from_roles
   end
 
   def port
-    if port = specifics["port"]&.to_s
+    if port = accessory_config["port"]&.to_s
       port.include?(":") ? port : "#{port}:#{port}"
     end
   end
@@ -34,32 +43,26 @@ class Kamal::Configuration::Accessory
   end
 
   def labels
-    default_labels.merge(specifics["labels"] || {})
+    default_labels.merge(accessory_config["labels"] || {})
   end
 
   def label_args
     argumentize "--label", labels
   end
 
-  def env
-    Kamal::Configuration::Env.from_config \
-      config: specifics.fetch("env", {}),
-      secrets_file: File.join(config.host_env_directory, "accessories", "#{service_name}.env")
-  end
-
   def env_args
     env.args
   end
 
   def files
-    specifics["files"]&.to_h do |local_to_remote_mapping|
+    accessory_config["files"]&.to_h do |local_to_remote_mapping|
       local_file, remote_file = local_to_remote_mapping.split(":")
       [ expand_local_file(local_file), expand_remote_file(remote_file) ]
     end || {}
   end
 
   def directories
-    specifics["directories"]&.to_h do |host_to_container_mapping|
+    accessory_config["directories"]&.to_h do |host_to_container_mapping|
       host_path, container_path = host_to_container_mapping.split(":")
       [ expand_host_path(host_path), container_path ]
     end || {}
@@ -74,7 +77,7 @@ class Kamal::Configuration::Accessory
   end
 
   def option_args
-    if args = specifics["options"]
+    if args = accessory_config["options"]
       optionize args
     else
       []
@@ -82,7 +85,7 @@ class Kamal::Configuration::Accessory
   end
 
   def cmd
-    specifics["cmd"]
+    accessory_config["cmd"]
   end
 
   private
@@ -116,18 +119,18 @@ class Kamal::Configuration::Accessory
     end
 
     def specific_volumes
-      specifics["volumes"] || []
+      accessory_config["volumes"] || []
     end
 
     def remote_files_as_volumes
-      specifics["files"]&.collect do |local_to_remote_mapping|
+      accessory_config["files"]&.collect do |local_to_remote_mapping|
         _, remote_file = local_to_remote_mapping.split(":")
         "#{service_data_directory + remote_file}:#{remote_file}"
       end || []
     end
 
     def remote_directories_as_volumes
-      specifics["directories"]&.collect do |host_to_container_mapping|
+      accessory_config["directories"]&.collect do |host_to_container_mapping|
         host_path, container_path = host_to_container_mapping.split(":")
         [ expand_host_path(host_path), container_path ].join(":")
       end || []
@@ -146,30 +149,16 @@ class Kamal::Configuration::Accessory
     end
 
     def hosts_from_host
-      if specifics.key?("host")
+      [ accessory_config["host"] ] if accessory_config.key?("host")
-        host = specifics["host"]
-        if host
-          [ host ]
-        else
-          raise ArgumentError, "Missing host for accessory `#{name}`"
-        end
-      end
     end
 
     def hosts_from_hosts
-      if specifics.key?("hosts")
+      accessory_config["hosts"] if accessory_config.key?("hosts")
-        hosts = specifics["hosts"]
-        if hosts.is_a?(Array)
-          hosts
-        else
-          raise ArgumentError, "Hosts should be an Array for accessory `#{name}`"
-        end
-      end
     end
 
     def hosts_from_roles
-      if specifics.key?("roles")
+      if accessory_config.key?("roles")
-        specifics["roles"].flat_map { |role| config.role(role).hosts }
+        accessory_config["roles"].flat_map { |role| config.role(role).hosts }
       end
     end
 end

lib/kamal/configuration/boot.rb

@@ -1,20 +1,25 @@
 class Kamal::Configuration::Boot
+  include Kamal::Configuration::Validation
+
+  attr_reader :boot_config, :host_count
+
   def initialize(config:)
-    @options = config.raw_config.boot || {}
+    @boot_config = config.raw_config.boot || {}
     @host_count = config.all_hosts.count
+    validate! boot_config
   end
 
   def limit
-    limit = @options["limit"]
+    limit = boot_config["limit"]
 
     if limit.to_s.end_with?("%")
-      [ @host_count * limit.to_i / 100, 1 ].max
+      [ host_count * limit.to_i / 100, 1 ].max
     else
       limit
     end
   end
 
   def wait
-    @options["wait"]
+    boot_config["wait"]
   end
 end

lib/kamal/configuration/builder.rb

@@ -1,73 +1,79 @@
 class Kamal::Configuration::Builder
-  def initialize(config:)
+  include Kamal::Configuration::Validation
-    @options = config.raw_config.builder || {}
-    @image = config.image
-    @server = config.registry["server"]
-    @service = config.service
-    @destination = config.destination
 
-    valid?
+  attr_reader :config, :builder_config
+  delegate :image, :service, to: :config
+  delegate :server, to: :"config.registry"
+
+  def initialize(config:)
+    @config = config
+    @builder_config = config.raw_config.builder || {}
+    @image = config.image
+    @server = config.registry.server
+    @service = config.service
+
+    validate! builder_config, with: Kamal::Configuration::Validator::Builder
   end
 
   def to_h
-    @options
+    builder_config
   end
 
   def multiarch?
-    @options["multiarch"] != false
+    builder_config["multiarch"] != false
   end
 
   def local?
-    !!@options["local"]
+    !!builder_config["local"]
   end
 
   def remote?
-    !!@options["remote"]
+    !!builder_config["remote"]
   end
 
   def cached?
-    !!@options["cache"]
+    !!builder_config["cache"]
   end
 
   def args
-    @options["args"] || {}
+    builder_config["args"] || {}
   end
 
   def secrets
-    @options["secrets"] || []
+    builder_config["secrets"] || []
   end
 
   def dockerfile
-    @options["dockerfile"] || "Dockerfile"
+    builder_config["dockerfile"] || "Dockerfile"
   end
 
   def target
-    @options["target"]
+    builder_config["target"]
   end
 
   def context
-    @options["context"] || "."
+    builder_config["context"] || "."
   end
 
   def local_arch
-    @options["local"]["arch"] if local?
+    builder_config["local"]["arch"] if local?
   end
 
   def local_host
-    @options["local"]["host"] if local?
+    builder_config["local"]["host"] if local?
   end
 
   def remote_arch
-    @options["remote"]["arch"] if remote?
+    builder_config["remote"]["arch"] if remote?
   end
 
   def remote_host
-    @options["remote"]["host"] if remote?
+    builder_config["remote"]["host"] if remote?
   end
 
   def cache_from
     if cached?
-      case @options["cache"]["type"]
+      case builder_config["cache"]["type"]
       when "gha"
         cache_from_config_for_gha
       when "registry"
@@ -78,7 +84,7 @@ class Kamal::Configuration::Builder
 
   def cache_to
     if cached?
-      case @options["cache"]["type"]
+      case builder_config["cache"]["type"]
       when "gha"
         cache_to_config_for_gha
       when "registry"
@@ -88,15 +94,15 @@ class Kamal::Configuration::Builder
   end
 
   def ssh
-    @options["ssh"]
+    builder_config["ssh"]
   end
 
   def git_clone?
-    Kamal::Git.used? && @options["context"].nil?
+    Kamal::Git.used? && builder_config["context"].nil?
   end
 
   def clone_directory
-    @clone_directory ||= File.join Dir.tmpdir, "kamal-clones", [ @service, pwd_sha ].compact.join("-")
+    @clone_directory ||= File.join Dir.tmpdir, "kamal-clones", [ service, pwd_sha ].compact.join("-")
   end
 
   def build_directory
@@ -109,18 +115,12 @@ class Kamal::Configuration::Builder
   end
 
   private
-    def valid?
-      if @options["cache"] && @options["cache"]["type"]
-        raise ArgumentError, "Invalid cache type: #{@options["cache"]["type"]}" unless [ "gha", "registry" ].include?(@options["cache"]["type"])
-      end
-    end
-
     def cache_image
-      @options["cache"]&.fetch("image", nil) || "#{@image}-build-cache"
+      builder_config["cache"]&.fetch("image", nil) || "#{image}-build-cache"
     end
 
     def cache_image_ref
-      [ @server, cache_image ].compact.join("/")
+      [ server, cache_image ].compact.join("/")
     end
 
     def cache_from_config_for_gha
@@ -132,11 +132,11 @@ class Kamal::Configuration::Builder
     end
 
     def cache_to_config_for_gha
-      [ "type=gha", @options["cache"]&.fetch("options", nil) ].compact.join(",")
+      [ "type=gha", builder_config["cache"]&.fetch("options", nil) ].compact.join(",")
     end
 
     def cache_to_config_for_registry
-      [ "type=registry", @options["cache"]&.fetch("options", nil), "ref=#{cache_image_ref}" ].compact.join(",")
+      [ "type=registry", builder_config["cache"]&.fetch("options", nil), "ref=#{cache_image_ref}" ].compact.join(",")
     end
 
     def repo_basename

lib/kamal/configuration/docs/accessory.yml

@@ -0,0 +1,90 @@
+# Accessories
+#
+# Accessories can be booted on a single host, a list of hosts, or on specific roles.
+# The hosts do not need to be defined in the Kamal servers configuration.
+#
+# Accessories are managed separately from the main service - they are not updated
+# when you deploy and they do not have zero-downtime deployments.
+#
+# Run `kamal accessory boot <accessory>` to boot an accessory.
+# See `kamal accessory --help` for more information.
+
+# Configuring accessories
+#
+# First define the accessory in the `accessories`
+accessories:
+  mysql:
+
+    # Service name
+    #
+    # This is used in the service label and defaults to `<service>-<accessory>`
+    # where `<service>` is the main service name from the root configuration
+    service: mysql
+
+    # Image
+    #
+    # The Docker image to use, prefix with a registry if not using Docker hub
+    image: mysql:8.0
+
+    # Accessory hosts
+    #
+    # Specify one  of `host`, `hosts` or `roles`
+    host: mysql-db1
+    hosts:
+      - mysql-db1
+      - mysql-db2
+    roles:
+      - mysql
+
+    # Custom command
+    #
+    # You can set a custom command to run in the container, if you do not want to use the default
+    cmd: "bin/mysqld"
+
+    # Port mappings
+    #
+    # See https://docs.docker.com/network/, especially note the warning about the security
+    # implications of exposing ports publicly.
+    port: "127.0.0.1:3306:3306"
+
+    # Labels
+    labels:
+      app: myapp
+
+    # Options
+    # These are passed to the Docker run command in the form `--<name> <value>`
+    options:
+      restart: always
+      cpus: 2
+
+    # Environment variables
+    # See kamal docs env for more information
+    env:
+      ...
+
+    # Copying files
+    #
+    # You can specify files to mount into the container.
+    # The format is `local:remote` where `local` is the path to the file on the local machine
+    # and `remote` is the path to the file in the container.
+    #
+    # They will be uploaded from the local repo to the host and then mounted.
+    #
+    # ERB files will be evaluated before being copied.
+    files:
+      - config/my.cnf.erb:/etc/mysql/my.cnf
+      - config/myoptions.cnf:/etc/mysql/myoptions.cnf
+
+    # Directories
+    #
+    # You can specify directories to mount into the container. They will be created on the host
+    # before being mounted
+    directories:
+      - mysql-logs:/var/log/mysql
+
+    # Volumes
+    #
+    # Any other volumes to mount, in addition to the files and directories.
+    # They are not created or copied before mounting
+    volumes:
+      - /path/to/mysql-logs:/var/log/mysql

lib/kamal/configuration/docs/boot.yml

@@ -0,0 +1,19 @@
+# Booting
+#
+# When deploying to large numbers of hosts, you might prefer not to restart your services on every host at the same time.
+#
+# Kamal’s default is to boot new containers on all hosts in parallel. But you can control this with the boot configuration.
+
+# Fixed group sizes
+#
+# Here we boot 2 hosts at a time with a 10 second gap between each group.
+boot:
+  limit: 2
+  wait: 10
+
+# Percentage of hosts
+#
+# Here we boot 25% of the hosts at a time with a 2 second gap between each group.
+boot:
+  limit: 25%
+  wait: 2

lib/kamal/configuration/docs/builder.yml

@@ -0,0 +1,107 @@
+# Builder
+#
+# The builder configuration controls how the application is built with `docker build` or `docker buildx build`
+#
+# If no configuration is specified, Kamal will:
+# 1. Create a buildx context called `kamal-<service>-multiarch`
+# 2. Use `docker buildx build` to build a multiarch image for linux/amd64,linux/arm64 with that context
+#
+# See https://kamal-deploy.org/docs/configuration/builder-examples/ for more information
+
+# Builder options
+#
+# Options go under the builder key in the root configuration.
+builder:
+
+  # Multiarch
+  #
+  # Enables multiarch builds, defaults to `true`
+  multiarch: false
+
+  # Local configuration
+  #
+  # The build configuration for local builds, only used if multiarch is enabled (the default)
+  #
+  # If there is no remote configuration, by default we build for amd64 and arm64.
+  # If you only want to build for one architecture, you can specify it here.
+  # The docker socket is optional and uses the default docker host socket when not specified
+  local:
+    arch: amd64
+    host: /var/run/docker.sock
+
+  # Remote configuration
+  #
+  # The build configuration for remote builds, also only used if multiarch is enabled.
+  # The arch is required and can be either amd64 or arm64.
+  remote:
+    arch: arm64
+    host: ssh://docker@docker-builder
+
+  # Builder cache
+  #
+  # The type must be either 'gha' or 'registry'
+  #
+  # The image is only used for registry cache
+  cache:
+    type: registry
+    options: mode=max
+    image: kamal-app-build-cache
+
+  # Build context
+  #
+  # If this is not set, then a local git clone of the repo is used.
+  # This ensures a clean build with no uncommitted changes.
+  #
+  # To use the local checkout instead you can set the context to `.`, or a path to another directory.
+  context: .
+
+  # Dockerfile
+  #
+  # The Dockerfile to use for building, defaults to `Dockerfile`
+  dockerfile: Dockerfile.production
+
+  # Build target
+  #
+  # If not set, then the default target is used
+  target: production
+
+  # Build Arguments
+  #
+  # Any additional build arguments, passed to `docker build` with `--build-arg <key>=<value>`
+  args:
+    ENVIRONMENT: production
+
+  # Referencing build arguments
+  #
+  # ```shell
+  # ARG RUBY_VERSION
+  # FROM ruby:$RUBY_VERSION-slim as base
+  # ```
+
+  # Build secrets
+  #
+  # Values are read from the environment.
+  #
+  secrets:
+    - SECRET1
+    - SECRET2
+
+  # Referencing Build Secrets
+  #
+  # ```shell
+  # # Copy Gemfiles
+  # COPY Gemfile Gemfile.lock ./
+  #
+  # # Install dependencies, including private repositories via access token
+  # # Then remove bundle cache with exposed GITHUB_TOKEN)
+  # RUN --mount=type=secret,id=GITHUB_TOKEN \
+  #   BUNDLE_GITHUB__COM=x-access-token:$(cat /run/secrets/GITHUB_TOKEN) \
+  #   bundle install && \
+  #   rm -rf /usr/local/bundle/cache
+  # ```
+
+
+  # SSH
+  #
+  # SSH agent socket or keys to expose to the build
+  ssh: default=$SSH_AUTH_SOCK

lib/kamal/configuration/docs/configuration.yml

@@ -0,0 +1,157 @@
+# Kamal Configuration
+#
+# Configuration is read from the `config/deploy.yml`
+#
+# When running commands, you can specify a destination with the `-d` flag,
+# e.g. `kamal deploy -d staging`
+#
+# In this case the configuration will also be read from `config/deploy.staging.yml`
+# and merged with the base configuration.
+#
+# The available configuration options are explained below.
+
+# The service name
+# This is a required value. It is used as the container name prefix.
+service: myapp
+
+# The Docker image name
+#
+# The image will be pushed to the configured registry.
+image: my-image
+
+# Labels
+#
+# Additional labels to add to the container
+labels:
+  my-label: my-value
+
+# Additional volumes to mount into the container
+volumes:
+  - /path/on/host:/path/in/container:ro
+
+# Registry
+#
+# The Docker registry configuration, see kamal docs registry
+registry:
+  ...
+
+# Servers
+#
+# The servers to deploy to, optionally with custom roles, see kamal docs servers
+servers:
+  ...
+
+# Environment variables
+#
+# See kamal docs env
+env:
+  ...
+
+# Asset Bridging
+#
+# Used for asset bridging across deployments, default to `nil`
+#
+# If there are changes to CSS or JS files, we may get requests
+# for the old versions on the new container and vice-versa.
+#
+# To avoid 404s we can specify an asset path.
+# Kamal will replace that path in the container with a mapped
+# volume containing both sets of files.
+# This requires that file names change when the contents change
+# (e.g. by including a hash of the contents in the name).
+
+# To configure this, set the path to the assets:
+asset_path: /path/to/assets
+
+# Path to hooks, defaults to `.kamal/hooks`
+# See https://kamal-deploy.org/docs/hooks for more information
+hooks_path: /user_home/kamal/hooks
+
+# Require destinations
+#
+# Whether deployments require a destination to be specified, defaults to `false`
+require_destination: true
+
+# The primary role
+#
+# This defaults to `web`, but if you have no web role, you can change this
+primary_role: workers
+
+# Allowing empty roles
+#
+# Whether roles with no servers are allowed. Defaults to `false`.
+allow_empty_roles: false
+
+# Stop wait time
+#
+# How long we wait for a container to stop before killing it, defaults to 30 seconds
+stop_wait_time: 60
+
+# Retain containers
+#
+# How many old containers and images we retain, defaults to 5
+retain_containers: 3
+
+# Minimum version
+#
+# The minimum version of Kamal required to deploy this configuration, defaults to nil
+minimum_version: 1.3.0
+
+# Readiness delay
+#
+# Seconds to wait for a container to boot after is running, default 7
+# This only applies to containers that do not specify a healthcheck
+readiness_delay: 4
+
+# Run directory
+#
+# Directory to store kamal runtime files in on the host, default `.kamal`
+run_directory: /etc/kamal
+
+# SSH options
+#
+# See kamal docs ssh
+ssh:
+  ...
+
+# Builder options
+#
+# See kamal docs builder
+builder:
+  ...
+
+# Accessories
+#
+# Additionals services to run in Docker, see kamal docs accessory
+accessories:
+  ...
+
+# Traefik
+#
+# The Traefik proxy is used for zero-downtime deployments, see kamal docs traefik
+traefik:
+  ...
+
+# SSHKit
+#
+# See kamal docs sshkit
+sshkit:
+  ...
+
+# Boot options
+#
+# See kamal docs boot
+boot:
+  ...
+
+# Healthcheck
+#
+# Configuring healthcheck commands, intervals and timeouts, see kamal docs healthcheck
+healthcheck:
+  ...
+
+# Logging
+#
+# Docker logging configuration, see kamal docs logging
+logging:
+  ...

lib/kamal/configuration/docs/env.yml

@@ -0,0 +1,72 @@
+# Environment variables
+#
+# Environment variables can be set directory in the Kamal configuration or
+# for loaded from a .env file, for secrets that should not be checked into Git.
+
+# Reading environment variables from the configuration
+#
+# Environment variables can be set directly in the configuration file.
+#
+# These are passed to the docker run command when deploying.
+env:
+  DATABASE_HOST: mysql-db1
+  DATABASE_PORT: 3306
+
+# Using .env file to load required environment variables
+#
+# Kamal uses dotenv to automatically load environment variables set in the .env file present
+# in the application root.
+#
+# This file can be used to set variables like KAMAL_REGISTRY_PASSWORD or database passwords.
+# But for this reason you must ensure that .env files are not checked into Git or included
+# in your Dockerfile! The format is just key-value like:
+# ```
+# KAMAL_REGISTRY_PASSWORD=pw
+# DB_PASSWORD=secret123
+# ```
+# See https://kamal-deploy.org/docs/commands/envify/ for how to use generated .env files.
+#
+# To pass the secrets you should list them under the `secret` key. When you do this the
+# other variables need to be moved under the `clear` key.
+#
+# Unlike clear valies, secrets are not passed directly to the container,
+# but are stored in an env file on the host
+# The file is not updated when deploying, only when running `kamal envify` or `kamal env push`.
+env:
+  clear:
+    DB_USER: app
+  secret:
+    - DB_PASSWORD
+
+# Tags
+#
+# Tags are used to add extra env variables to specific hosts.
+# See kamal docs servers for how to tag hosts.
+#
+# Tags are only allowed in the top level env configuration (i.e not under a role specific env).
+#
+# The env variables can be specified with secret and clear values as explained above.
+env:
+  tags:
+    <tag1>:
+      MYSQL_USER: monitoring
+    <tag2>:
+      clear:
+        MYSQL_USER: readonly
+      secret:
+        - MYSQL_PASSWORD
+
+# Example configuration
+env:
+  clear:
+    MYSQL_USER: app
+  secret:
+    - MYSQL_PASSWORD
+  tags:
+    monitoring:
+      MYSQL_USER: monitoring
+    replica:
+      clear:
+        MYSQL_USER: readonly
+      secret:
+        - READONLY_PASSWORD

lib/kamal/configuration/docs/healthcheck.yml

@@ -0,0 +1,59 @@
+# Healthcheck configuration
+#
+# On roles that are running Traefik, Kamal will supply a default healthcheck to `docker run`.
+# For other roles, by default no healthcheck is supplied.
+#
+# If no healthcheck is supplied and the image does not define one, they we wait for the container
+# to reach a running state and then pause for the readiness delay.
+#
+# The default healthcheck is `curl -f http://localhost:<port>/<path>`, so it assumes that `curl`
+# is available within the container.
+
+# Healthcheck options
+#
+# These go under the `healthcheck` key in the root or role configuration.
+healthcheck:
+
+  # Command
+  #
+  # The command to run, defaults to `curl -f http://localhost:<port>/<path>` on roles running Traefik
+  cmd: "curl -f http://localhost"
+
+  # Interval
+  #
+  # The Docker healthcheck interval, defaults to `1s`
+  interval: 10s
+
+  # Max attempts
+  #
+  # The maximum number of times we poll the container to see if it is healthy, defaults to `7`
+  # Each check is separated by an increasing interval starting with 1 second.
+  max_attempts: 3
+
+  # Port
+  #
+  # The port to use in the healthcheck, defaults to `3000`
+  port: "80"
+
+  # Path
+  #
+  # The path to use in the healthcheck, defaults to `/up`
+  path: /health
+
+  # Cords for zero-downtime deployments
+  #
+  # The cord file is used for zero-downtime deployments. The healthcheck is augmented with a check
+  # for the existance of the file. This allows us to delete the file and force the container to
+  # become unhealthy, causing Traefik to stop routing traffic to it.
+  #
+  # Kamal mounts a volume at this location and creates the file before starting the container.
+  # You can set the value to `false` to disable the cord file, but this loses the zero-downtime
+  # guarantee.
+  #
+  # The default value is `/tmp/kamal-cord`
+  cord: /cord
+
+  # Log lines
+  #
+  # Number of lines to log from the container when the healthcheck fails, defaults to `50`
+  log_lines: 100

lib/kamal/configuration/docs/logging.yml

@@ -0,0 +1,21 @@
+# Custom logging configuration
+#
+# Set these to control the Docker logging driver and options.
+
+# Logging settings
+#
+# These go under the logging key in the configuration file.
+#
+# This can be specified in the root level or for a specific role.
+logging:
+
+  # Driver
+  #
+  # The logging driver to use, passed to Docker via `--log-driver`
+  driver: json-file
+
+  # Options
+  #
+  # Any logging options to pass to the driver, passed to Docker via `--log-opt`
+  options:
+    max-size: 100m

lib/kamal/configuration/docs/registry.yml

@@ -0,0 +1,49 @@
+# Registry
+#
+# The default registry is Docker Hub, but you can change it using registry/server:
+#
+# A reference to secret (in this case DOCKER_REGISTRY_TOKEN) will look up the secret
+# in the local environment.
+
+registry:
+  server: registry.digitalocean.com
+  username:
+    - DOCKER_REGISTRY_TOKEN
+  password:
+    - DOCKER_REGISTRY_TOKEN
+
+# Using AWS ECR as the container registry
+# You will need to have the aws CLI installed locally for this to work.
+# AWS ECR’s access token is only valid for 12hrs. In order to not have to manually regenerate the token every time, you can use ERB in the deploy.yml file to shell out to the aws cli command, and obtain the token:
+
+registry:
+  server: <your aws account id>.dkr.ecr.<your aws region id>.amazonaws.com
+  username: AWS
+  password: <%= %x(aws ecr get-login-password) %>
+
+# Using GCP Artifact Registry as the container registry
+# To sign into Artifact Registry, you would need to
+# [create a service account](https://cloud.google.com/iam/docs/service-accounts-create#creating)
+# and [set up roles and permissions](https://cloud.google.com/artifact-registry/docs/access-control#permissions).
+# Normally, assigning a roles/artifactregistry.writer role should be sufficient.
+#
+# Once the service account is ready, you need to generate and download a JSON key, base64 encode it and add to .env:
+#
+# ```shell
+# echo "KAMAL_REGISTRY_PASSWORD=$(base64 -i /path/to/key.json)" | tr -d "\\n"  >> .env
+# ```
+# Use the env variable as password along with _json_key_base64 as username.
+# Here’s the final configuration:
+
+registry:
+  server: <your registry region>-docker.pkg.dev
+  username: _json_key_base64
+  password:
+    - KAMAL_REGISTRY_PASSWORD
+
+# Validating the configuration
+#
+# You can validate the configuration by running:
+# ```shell
+# kamal registry login
+# ```

lib/kamal/configuration/docs/role.yml

@@ -0,0 +1,52 @@
+# Roles
+#
+# Roles are used to configure different types of servers in the deployment.
+# The most common use for this is to run a web servers and job servers.
+#
+# Kamal expects there to be a `web` role, unless you set a different `primary_role`
+# in the root configuration.
+
+# Role configuration
+#
+# Roles are specified under the servers key
+servers:
+
+  # Simple role configuration
+  #
+  #
+  # This can be a list of hosts, if you don't need custom configuration for the role.
+  #
+  # You can set tags on the hosts for custom env variables (see kamal docs env)
+  web:
+    - 172.1.0.1
+    - 172.1.0.2: experiment1
+    - 172.1.0.2: [ experiment1, experiment2 ]
+
+  # Custom role configuration
+  #
+  # When there are other options to set, the list of hosts goes under the `hosts` key
+  #
+  # By default only the primary role uses Traefik, but you can set `traefik` to change
+  # it.
+  #
+  # You can also set a custom cmd to run in the container, and overwrite other settings
+  # from the root configuration.
+  workers:
+    hosts:
+      - 172.1.0.3
+      - 172.1.0.4: experiment1
+    traefik: true
+    cmd: "bin/jobs"
+    options:
+      memory: 2g
+      cpus: 4
+    healthcheck:
+      ...
+    logging:
+      ...
+    labels:
+      my-label: workers
+    env:
+      ...
+    asset_path: /public
+

lib/kamal/configuration/docs/servers.yml

@@ -0,0 +1,27 @@
+# Servers
+#
+# Servers are split into different roles, with each role having its own configuration.
+#
+# For simpler deployments though where all servers are identical, you can just specify a list of servers
+# They will be implicitly assigned to the `web` role.
+servers:
+  - 172.0.0.1
+  - 172.0.0.2
+  - 172.0.0.3
+
+# Tagging servers
+#
+# Servers can be tagged, with the tags used to add custom env variables (see kamal docs env).
+servers:
+  - 172.0.0.1
+  - 172.0.0.2: experiments
+  - 172.0.0.3: [ experiments, three ]
+
+# Roles
+#
+# For more complex deployments (e.g. if you are running job hosts), you can specify roles, and configure each separately (see kamal docs role)
+servers:
+  web:
+    ...
+  workers:
+    ...

lib/kamal/configuration/docs/ssh.yml

@@ -0,0 +1,46 @@
+# SSH configuration
+#
+# Kamal uses SSH to connect run commands on your hosts.
+# By default it will attempt to connect to the root user on port 22
+#
+# If you are using non-root user, you may need to bootstrap your servers manually, before using them with Kamal. On Ubuntu, you’d do:
+#
+# ```shell
+# sudo apt update
+# sudo apt upgrade -y
+# sudo apt install -y docker.io curl git
+# sudo usermod -a -G docker app
+# ```
+
+
+# SSH options
+#
+# The options are specified under the ssh key in the configuration file.
+ssh:
+
+  # The SSH user
+  #
+  # Defaults to `root`
+  #
+  user: app
+
+  # The SSH port
+  #
+  # Defaults to 22
+  port: "2222"
+
+  # Proxy host
+  #
+  # Specified in the form <host> or <user>@<host>
+  proxy: root@proxy-host
+
+  # Proxy command
+  #
+  # A custom proxy command, required for older versions of SSH
+  proxy_command: "ssh -W %h:%p user@proxy"
+
+  # Log level
+  #
+  # Defaults to `fatal`. Set this to debug if you are having
+  # SSH connection issues.
+  log_level: debug

lib/kamal/configuration/docs/sshkit.yml

@@ -0,0 +1,23 @@
+# SSHKit
+#
+# [SSHKit](https://github.com/capistrano/sshkit) is the SSH toolkit used by Kamal.
+#
+# The default settings should be sufficient for most use cases, but
+# when connecting to a large number of hosts you may need to adjust
+
+# SSHKit options
+#
+# The options are specified under the sshkit key in the configuration file.
+sshkit:
+
+  # Max concurrent starts
+  #
+  # Creating SSH connections concurrently can be an issue when deploying to many servers.
+  # By default Kamal will limit concurrent connection starts to 30 at a time.
+  max_concurrent_starts: 10
+
+  # Pool idle timeout
+  #
+  # Kamal sets a long idle timeout of 900 seconds on connections to try to avoid
+  # re-connection storms after an idle period, like building an image or waiting for CI.
+  pool_idle_timeout: 300

lib/kamal/configuration/docs/traefik.yml

@@ -0,0 +1,62 @@
+# Traefik
+#
+# Traefik is a reverse proxy, used by Kamal for zero-downtime deployments.
+#
+# We start an instance on the hosts in it's own container.
+#
+# During a deployment:
+# 1. We start a new container which Traefik automatically detects due to the labels we have applied
+# 2. Traefik starts routing traffic to the new container
+# 3. We force the old container to fail it's healthcheck, causing Traefik to stop routing traffic to it
+# 4. We stop the old container
+
+# Traefik settings
+#
+# Traekik is configured in the root configuration under `traefik`.
+traefik:
+
+  # Image
+  #
+  # The Traefik image to use, defaults to `traefik:v2.10`
+  image: traefik:v2.9
+
+  # Host port
+  #
+  # The host port to publish the Traefik container on, defaults to `80`
+  host_port: "8080"
+
+  # Disabling publishing
+  #
+  # To avoid publishing the Traefik container, set this to `false`
+  publish: false
+
+  # Labels
+  #
+  # Additional labels to apply to the Traefik container
+  labels:
+    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"
+
+  # Arguments
+  #
+  # Additional arguments to pass to the Traefik container
+  args:
+    entryPoints.http.address: ":80"
+    entryPoints.http.forwardedHeaders.insecure: true
+    accesslog: true
+    accesslog.format: json
+
+  # Options
+  #
+  # Additional options to pass to `docker run`
+  options:
+    cpus: 2
+
+  # Environment variables
+  #
+  # See kamal docs env
+  env:
+    ...

lib/kamal/configuration/env.rb

@@ -1,18 +1,15 @@
 class Kamal::Configuration::Env
-  attr_reader :secrets_keys, :clear, :secrets_file
+  include Kamal::Configuration::Validation
+
+  attr_reader :secrets_keys, :clear, :secrets_file, :context
   delegate :argumentize, to: Kamal::Utils
 
-  def self.from_config(config:, secrets_file: nil)
+  def initialize(config:, secrets_file: nil, context: "env")
-    secrets_keys = config.fetch("secret", [])
+    @clear = config.fetch("clear", config.key?("secret") || config.key?("tags") ? {} : config)
-    clear = config.fetch("clear", config.key?("secret") || config.key?("tags") ? {} : config)
+    @secrets_keys = config.fetch("secret", [])
-
-    new clear: clear, secrets_keys: secrets_keys, secrets_file: secrets_file
-  end
-
-  def initialize(clear:, secrets_keys:, secrets_file:)
-    @clear = clear
-    @secrets_keys = secrets_keys
     @secrets_file = secrets_file
+    @context = context
+    validate! config, context: context, with: Kamal::Configuration::Validator::Env
   end
 
   def args
@@ -33,8 +30,7 @@ class Kamal::Configuration::Env
 
   def merge(other)
     self.class.new \
-      clear: @clear.merge(other.clear),
+      config: { "clear" => clear.merge(other.clear), "secret" => secrets_keys | other.secrets_keys },
-      secrets_keys: @secrets_keys | other.secrets_keys,
+      secrets_file: secrets_file || other.secrets_file
-      secrets_file: secrets_file
   end
 end

lib/kamal/configuration/env/tag.rb

@@ -7,6 +7,6 @@ class Kamal::Configuration::Env::Tag
   end
 
   def env
-    Kamal::Configuration::Env.from_config(config: config)
+    Kamal::Configuration::Env.new(config: config)
   end
 end

lib/kamal/configuration/healthcheck.rb

@@ -0,0 +1,63 @@
+class Kamal::Configuration::Healthcheck
+  include Kamal::Configuration::Validation
+
+  attr_reader :healthcheck_config
+
+  def initialize(healthcheck_config:, context: "healthcheck")
+    @healthcheck_config = healthcheck_config || {}
+    validate! @healthcheck_config, context: context
+  end
+
+  def merge(other)
+    self.class.new healthcheck_config: healthcheck_config.deep_merge(other.healthcheck_config)
+  end
+
+  def cmd
+    healthcheck_config.fetch("cmd", http_health_check)
+  end
+
+  def port
+    healthcheck_config.fetch("port", 3000)
+  end
+
+  def path
+    healthcheck_config.fetch("path", "/up")
+  end
+
+  def max_attempts
+    healthcheck_config.fetch("max_attempts", 7)
+  end
+
+  def interval
+    healthcheck_config.fetch("interval", "1s")
+  end
+
+  def cord
+    healthcheck_config.fetch("cord", "/tmp/kamal-cord")
+  end
+
+  def log_lines
+    healthcheck_config.fetch("log_lines", 50)
+  end
+
+  def set_port_or_path?
+    healthcheck_config["port"].present? || healthcheck_config["path"].present?
+  end
+
+  def to_h
+    {
+      "cmd" => cmd,
+      "interval" => interval,
+      "max_attempts" => max_attempts,
+      "port" => port,
+      "path" => path,
+      "cord" => cord,
+      "log_lines" => log_lines
+    }
+  end
+
+  private
+    def http_health_check
+      "curl -f #{URI.join("http://localhost:#{port}", path)} || exit 1" if path.present? || port.present?
+    end
+end

lib/kamal/configuration/logging.rb

@@ -0,0 +1,33 @@
+class Kamal::Configuration::Logging
+  delegate :optionize, :argumentize, to: Kamal::Utils
+
+  include Kamal::Configuration::Validation
+
+  attr_reader :logging_config
+
+  def initialize(logging_config:, context: "logging")
+    @logging_config = logging_config || {}
+    validate! @logging_config, context: context
+  end
+
+  def driver
+    logging_config["driver"]
+  end
+
+  def options
+    logging_config.fetch("options", {})
+  end
+
+  def merge(other)
+    self.class.new logging_config: logging_config.deep_merge(other.logging_config)
+  end
+
+  def args
+    if driver.present? || options.present?
+      optionize({ "log-driver" => driver }.compact) +
+        argumentize("--log-opt", options)
+    else
+      argumentize("--log-opt", { "max-size" => "10m" })
+    end
+  end
+end

lib/kamal/configuration/proxy.rb

@@ -1,50 +0,0 @@
-class Kamal::Configuration::Proxy
-  DEFAULT_HTTP_PORT = 80
-  DEFAULT_HTTPS_PORT = 443
-  DEFAULT_IMAGE = "basecamp/kamal-proxy: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
-    options.fetch(:http_port, DEFAULT_HTTP_PORT)
-  end
-
-  def https_port
-    options.fetch(:http_port, DEFAULT_HTTPS_PORT)
-  end
-
-  def container_name
-    "kamal-proxy"
-  end
-
-  def docker_options_args
-    optionize(options.fetch("options", {}))
-  end
-
-  def publish_args
-    argumentize "--publish", [ *("#{http_port}:#{DEFAULT_HTTP_PORT}" if http_port), *("#{https_port}:#{DEFAULT_HTTPS_PORT}" 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/registry.rb

@@ -0,0 +1,31 @@
+class Kamal::Configuration::Registry
+  include Kamal::Configuration::Validation
+
+  attr_reader :registry_config
+
+  def initialize(config:)
+    @registry_config = config.raw_config.registry || {}
+    validate! registry_config, with: Kamal::Configuration::Validator::Registry
+  end
+
+  def server
+    registry_config["server"]
+  end
+
+  def username
+    lookup("username")
+  end
+
+  def password
+    lookup("password")
+  end
+
+  private
+    def lookup(key)
+      if registry_config[key].is_a?(Array)
+        ENV.fetch(registry_config[key].first).dup
+      else
+        registry_config[key]
+      end
+    end
+end

lib/kamal/configuration/role.rb

@@ -1,12 +1,33 @@
 class Kamal::Configuration::Role
+  include Kamal::Configuration::Validation
+
+  CORD_FILE = "cord"
   delegate :argumentize, :optionize, to: Kamal::Utils
 
-  attr_accessor :name
+  attr_reader :name, :config, :specialized_env, :specialized_logging, :specialized_healthcheck
+
   alias to_s name
 
   def initialize(name, config:)
     @name, @config = name.inquiry, config
-    @tagged_hosts ||= extract_tagged_hosts_from_config
+    validate! \
+      specializations,
+      example: validation_yml["servers"]["workers"],
+      context: "servers/#{name}",
+      with: Kamal::Configuration::Validator::Role
+
+    @specialized_env = Kamal::Configuration::Env.new \
+      config: specializations.fetch("env", {}),
+      secrets_file: File.join(config.host_env_directory, "roles", "#{container_prefix}.env"),
+      context: "servers/#{name}/env"
+
+    @specialized_logging = Kamal::Configuration::Logging.new \
+      logging_config: specializations.fetch("logging", {}),
+      context: "servers/#{name}/logging"
+
+    @specialized_healthcheck = Kamal::Configuration::Healthcheck.new \
+      healthcheck_config: specializations.fetch("healthcheck", {}),
+      context: "servers/#{name}/healthcheck"
   end
 
   def primary_host
@@ -34,7 +55,7 @@ class Kamal::Configuration::Role
   end
 
   def labels
-    default_labels.merge(custom_labels)
+    default_labels.merge(traefik_labels).merge(custom_labels)
   end
 
   def label_args
@@ -42,21 +63,17 @@ class Kamal::Configuration::Role
   end
 
   def logging_args
-    args = config.logging || {}
+    logging.args
-    args.deep_merge!(specializations["logging"]) if specializations["logging"].present?
-
-    if args.any?
-      optionize({ "log-driver" => args["driver"] }.compact) +
-        argumentize("--log-opt", args["options"])
-    else
-      config.logging_args
   end
+
+  def logging
+    @logging ||= config.logging.merge(specialized_logging)
   end
 
 
   def env(host)
     @envs ||= {}
-    @envs[host] ||= [ base_env, specialized_env, *env_tags(host).map(&:env) ].reduce(:merge)
+    @envs[host] ||= [ config.env, specialized_env, *env_tags(host).map(&:env) ].reduce(:merge)
   end
 
   def env_args(host)
@@ -68,11 +85,38 @@ class Kamal::Configuration::Role
   end
 
 
-  def running_proxy?
+  def health_check_args(cord: true)
-    if specializations["proxy"].nil?
+    if running_traefik? || healthcheck.set_port_or_path?
+      if cord && uses_cord?
+        optionize({ "health-cmd" => health_check_cmd_with_cord, "health-interval" => healthcheck.interval })
+          .concat(cord_volume.docker_args)
+      else
+        optionize({ "health-cmd" => healthcheck.cmd, "health-interval" => healthcheck.interval })
+      end
+    else
+      []
+    end
+  end
+
+  def healthcheck
+    @healthcheck ||=
+      if running_traefik?
+        config.healthcheck.merge(specialized_healthcheck)
+      else
+        specialized_healthcheck
+      end
+  end
+
+  def health_check_cmd_with_cord
+    "(#{healthcheck.cmd}) && (stat #{cord_container_file} > /dev/null || exit 1)"
+  end
+
+
+  def running_traefik?
+    if specializations["traefik"].nil?
       primary?
     else
-      specializations["proxy"]
+      specializations["traefik"]
     end
   end
 
@@ -81,6 +125,35 @@ class Kamal::Configuration::Role
   end
 
 
+  def uses_cord?
+    running_traefik? && cord_volume && healthcheck.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 = healthcheck.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 +168,7 @@ class Kamal::Configuration::Role
   end
 
   def assets?
-    asset_path.present? && running_proxy?
+    asset_path.present? && running_traefik?
   end
 
   def asset_volume(version = nil)
@@ -114,30 +187,24 @@ class Kamal::Configuration::Role
   end
 
   private
-    attr_accessor :config, :tagged_hosts
+    def tagged_hosts
-
-    def extract_tagged_hosts_from_config
       {}.tap do |tagged_hosts|
         extract_hosts_from_config.map do |host_config|
           if host_config.is_a?(Hash)
-            raise ArgumentError, "Multiple hosts found: #{host_config.inspect}" unless host_config.size == 1
-
             host, tags = host_config.first
             tagged_hosts[host] = Array(tags)
-          elsif host_config.is_a?(String) || host_config.is_a?(Symbol)
+          elsif host_config.is_a?(String)
             tagged_hosts[host_config] = []
-          else
-            raise ArgumentError, "Invalid host config: #{host_config.inspect}"
           end
         end
       end
     end
 
     def extract_hosts_from_config
-      if config.servers.is_a?(Array)
+      if config.raw_config.servers.is_a?(Array)
-        config.servers
+        config.raw_config.servers
       else
-        servers = config.servers[name]
+        servers = config.raw_config.servers[name]
         servers.is_a?(Array) ? servers : Array(servers["hosts"])
       end
     end
@@ -146,29 +213,39 @@ class Kamal::Configuration::Role
       { "service" => config.service, "role" => name, "destination" => config.destination }
     end
 
+    def specializations
+      if config.raw_config.servers.is_a?(Array) || config.raw_config.servers[name].is_a?(Array)
+        {}
+      else
+        config.raw_config.servers[name]
+      end
+    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?
         labels.merge!(specializations["labels"]) if specializations["labels"].present?
       end
     end
-
-    def specializations
-      if config.servers.is_a?(Array) || config.servers[name].is_a?(Array)
-        {}
-      else
-        config.servers[name].except("hosts")
-      end
-    end
-
-    def specialized_env
-      Kamal::Configuration::Env.from_config config: specializations.fetch("env", {})
-    end
-
-    # Secrets are stored in an array, which won't merge by default, so have to do it by hand.
-    def base_env
-      Kamal::Configuration::Env.from_config \
-        config: config.env,
-        secrets_file: File.join(config.host_env_directory, "roles", "#{container_prefix}.env")
-    end
 end

lib/kamal/configuration/servers.rb

@@ -0,0 +1,18 @@
+class Kamal::Configuration::Servers
+  include Kamal::Configuration::Validation
+
+  attr_reader :config, :servers_config, :roles
+
+  def initialize(config:)
+    @config = config
+    @servers_config = config.raw_config.servers
+    validate! servers_config, with: Kamal::Configuration::Validator::Servers
+
+    @roles = role_names.map { |role_name| Kamal::Configuration::Role.new role_name, config: config }
+  end
+
+  private
+    def role_names
+      servers_config.is_a?(Array) ? [ "web" ] : servers_config.keys.sort
+    end
+end

lib/kamal/configuration/ssh.rb

@@ -1,22 +1,27 @@
 class Kamal::Configuration::Ssh
   LOGGER = ::Logger.new(STDERR)
 
+  include Kamal::Configuration::Validation
+
+  attr_reader :ssh_config
+
   def initialize(config:)
-    @config = config.raw_config.ssh || {}
+    @ssh_config = config.raw_config.ssh || {}
+    validate! ssh_config
   end
 
   def user
-    config.fetch("user", "root")
+    ssh_config.fetch("user", "root")
   end
 
   def port
-    config.fetch("port", 22)
+    ssh_config.fetch("port", 22)
   end
 
   def proxy
-    if (proxy = config["proxy"])
+    if (proxy = ssh_config["proxy"])
       Net::SSH::Proxy::Jump.new(proxy.include?("@") ? proxy : "root@#{proxy}")
-    elsif (proxy_command = config["proxy_command"])
+    elsif (proxy_command = ssh_config["proxy_command"])
       Net::SSH::Proxy::Command.new(proxy_command)
     end
   end
@@ -30,13 +35,11 @@ class Kamal::Configuration::Ssh
   end
 
   private
-    attr_accessor :config
-
     def logger
       LOGGER.tap { |logger| logger.level = log_level }
     end
 
     def log_level
-      config.fetch("log_level", :fatal)
+      ssh_config.fetch("log_level", :fatal)
     end
 end

lib/kamal/configuration/sshkit.rb

@@ -1,20 +1,22 @@
 class Kamal::Configuration::Sshkit
+  include Kamal::Configuration::Validation
+
+  attr_reader :sshkit_config
+
   def initialize(config:)
-    @options = config.raw_config.sshkit || {}
+    @sshkit_config = config.raw_config.sshkit || {}
+    validate! sshkit_config
   end
 
   def max_concurrent_starts
-    options.fetch("max_concurrent_starts", 30)
+    sshkit_config.fetch("max_concurrent_starts", 30)
   end
 
   def pool_idle_timeout
-    options.fetch("pool_idle_timeout", 900)
+    sshkit_config.fetch("pool_idle_timeout", 900)
   end
 
   def to_h
-    options
+    sshkit_config
   end
-
-  private
-    attr_accessor :options
 end

lib/kamal/configuration/traefik.rb

@@ -0,0 +1,60 @@
+class Kamal::Configuration::Traefik
+  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"
+  }
+
+  include Kamal::Configuration::Validation
+
+  attr_reader :config, :traefik_config
+
+  def initialize(config:)
+    @config = config
+    @traefik_config = config.raw_config.traefik || {}
+    validate! traefik_config
+  end
+
+  def publish?
+    traefik_config["publish"] != false
+  end
+
+  def labels
+    DEFAULT_LABELS.merge(traefik_config["labels"] || {})
+  end
+
+  def env
+    Kamal::Configuration::Env.new \
+      config: traefik_config.fetch("env", {}),
+      secrets_file: File.join(config.host_env_directory, "traefik", "traefik.env"),
+      context: "traefik/env"
+  end
+
+  def host_port
+    traefik_config.fetch("host_port", CONTAINER_PORT)
+  end
+
+  def options
+    traefik_config.fetch("options", {})
+  end
+
+  def port
+    "#{host_port}:#{CONTAINER_PORT}"
+  end
+
+  def args
+    DEFAULT_ARGS.merge(traefik_config.fetch("args", {}))
+  end
+
+  def image
+    traefik_config.fetch("image", DEFAULT_IMAGE)
+  end
+end

lib/kamal/configuration/validation.rb

@@ -0,0 +1,27 @@
+require "yaml"
+require "active_support/inflector"
+
+module Kamal::Configuration::Validation
+  extend ActiveSupport::Concern
+
+  class_methods do
+    def validation_doc
+      @validation_doc ||= File.read(File.join(File.dirname(__FILE__), "docs", "#{validation_config_key}.yml"))
+    end
+
+    def validation_config_key
+      @validation_config_key ||= name.demodulize.underscore
+    end
+  end
+
+  def validate!(config, example: nil, context: nil, with: Kamal::Configuration::Validator)
+    context ||= self.class.validation_config_key
+    example ||= validation_yml[self.class.validation_config_key]
+
+    with.new(config, example: example, context: context).validate!
+  end
+
+  def validation_yml
+    @validation_yml ||= YAML.load(self.class.validation_doc)
+  end
+end

lib/kamal/configuration/validator.rb

@@ -0,0 +1,140 @@
+class Kamal::Configuration::Validator
+  attr_reader :config, :example, :context
+
+  def initialize(config, example:, context:)
+    @config = config
+    @example = example
+    @context = context
+  end
+
+  def validate!
+    validate_against_example! config, example
+  end
+
+  private
+    def validate_against_example!(validation_config, example)
+      validate_type! validation_config, Hash
+
+      if (unknown_keys = validation_config.keys - example.keys).any?
+        unknown_keys_error unknown_keys
+      end
+
+      validation_config.each do |key, value|
+        with_context(key) do
+          example_value = example[key]
+
+          if example_value == "..."
+            validate_type! value, *(Array if key == :servers), Hash
+          elsif key == "hosts"
+            validate_servers! value
+          elsif example_value.is_a?(Array)
+            validate_array_of! value, example_value.first.class
+          elsif example_value.is_a?(Hash)
+            case key.to_s
+            when "options"
+              validate_type! value, Hash
+            when "args", "labels"
+              validate_hash_of! value, example_value.first[1].class
+            else
+              validate_against_example! value, example_value
+            end
+          else
+            validate_type! value, example_value.class
+          end
+        end
+      end
+    end
+
+
+    def valid_type?(value, type)
+      value.is_a?(type) ||
+        (type == String && stringish?(value)) ||
+        (boolean?(type) && boolean?(value.class))
+    end
+
+    def type_description(type)
+      if type == Integer || type == Array
+        "an #{type.name.downcase}"
+      elsif type == TrueClass || type == FalseClass
+        "a boolean"
+      else
+        "a #{type.name.downcase}"
+      end
+    end
+
+    def boolean?(type)
+      type == TrueClass || type == FalseClass
+    end
+
+    def stringish?(value)
+      value.is_a?(String) || value.is_a?(Symbol) || value.is_a?(Numeric) || value.is_a?(TrueClass) || value.is_a?(FalseClass)
+    end
+
+    def validate_array_of!(array, type)
+      validate_type! array, Array
+
+      array.each_with_index do |value, index|
+        with_context(index) do
+          validate_type! value, type
+        end
+      end
+    end
+
+    def validate_hash_of!(hash, type)
+      validate_type! hash, Hash
+
+      hash.each do |key, value|
+        with_context(key) do
+          validate_type! value, type
+        end
+      end
+    end
+
+    def validate_servers!(servers)
+      validate_type! servers, Array
+
+      servers.each_with_index do |server, index|
+        with_context(index) do
+          validate_type! server, String, Hash
+
+          if server.is_a?(Hash)
+            error "multiple hosts found" unless server.size == 1
+            host, tags = server.first
+
+            with_context(host) do
+              validate_type! tags, String, Array
+              validate_array_of! tags, String if tags.is_a?(Array)
+            end
+          end
+        end
+      end
+    end
+
+    def validate_type!(value, *types)
+      type_error(*types) unless types.any? { |type| valid_type?(value, type) }
+    end
+
+    def error(message)
+      raise Kamal::ConfigurationError, "#{error_context}#{message}"
+    end
+
+    def type_error(*expected_types)
+      error "should be #{expected_types.map { |type| type_description(type) }.join(" or ")}"
+    end
+
+    def unknown_keys_error(unknown_keys)
+      error "unknown #{"key".pluralize(unknown_keys.count)}: #{unknown_keys.join(", ")}"
+    end
+
+    def error_context
+      "#{context}: " if context.present?
+    end
+
+    def with_context(context)
+      old_context = @context
+      @context = [ @context, context ].select(&:present?).join("/")
+      yield
+    ensure
+      @context = old_context
+    end
+end

lib/kamal/configuration/validator/accessory.rb

@@ -0,0 +1,9 @@
+class Kamal::Configuration::Validator::Accessory < Kamal::Configuration::Validator
+  def validate!
+    super
+
+    if (config.keys & [ "host", "hosts", "roles" ]).size != 1
+      error "specify one of `host`, `hosts` or `roles`"
+    end
+  end
+end

lib/kamal/configuration/validator/builder.rb

@@ -0,0 +1,9 @@
+class Kamal::Configuration::Validator::Builder < Kamal::Configuration::Validator
+  def validate!
+    super
+
+    if config["cache"] && config["cache"]["type"]
+      error "Invalid cache type: #{config["cache"]["type"]}" unless [ "gha", "registry" ].include?(config["cache"]["type"])
+    end
+  end
+end

lib/kamal/configuration/validator/env.rb

@@ -0,0 +1,54 @@
+class Kamal::Configuration::Validator::Env < Kamal::Configuration::Validator
+  SPECIAL_KEYS = [ "clear", "secret", "tags" ]
+
+  def validate!
+    if known_keys.any?
+      validate_complex_env!
+    else
+      validate_simple_env!
+    end
+  end
+
+  private
+    def validate_simple_env!
+      validate_hash_of!(config, String)
+    end
+
+    def validate_complex_env!
+      unknown_keys_error unknown_keys if unknown_keys.any?
+
+      with_context("clear") { validate_hash_of!(config["clear"], String) } if config.key?("clear")
+      with_context("secret") { validate_array_of!(config["secret"], String) } if config.key?("secret")
+      validate_tags! if config.key?("tags")
+    end
+
+    def known_keys
+      @known_keys ||= config.keys & SPECIAL_KEYS
+    end
+
+    def unknown_keys
+      @unknown_keys ||= config.keys - SPECIAL_KEYS
+    end
+
+    def validate_tags!
+      if context == "env"
+        with_context("tags") do
+          validate_type! config["tags"], Hash
+
+          config["tags"].each do |tag, value|
+            with_context(tag) do
+              validate_type! value, Hash
+
+              Kamal::Configuration::Validator::Env.new(
+                value,
+                example: example["tags"].values[1],
+                context: context
+              ).validate!
+            end
+          end
+        end
+      else
+        error "tags are only allowed in the root env"
+      end
+    end
+end

lib/kamal/configuration/validator/registry.rb

@@ -0,0 +1,25 @@
+class Kamal::Configuration::Validator::Registry < Kamal::Configuration::Validator
+  STRING_OR_ONE_ITEM_ARRAY_KEYS = [ "username", "password" ]
+
+  def validate!
+    validate_against_example! \
+      config.except(*STRING_OR_ONE_ITEM_ARRAY_KEYS),
+      example.except(*STRING_OR_ONE_ITEM_ARRAY_KEYS)
+
+    validate_string_or_one_item_array! "username"
+    validate_string_or_one_item_array! "password"
+  end
+
+  private
+    def validate_string_or_one_item_array!(key)
+      with_context(key) do
+        value = config[key]
+
+        error "is required" unless value.present?
+
+        unless value.is_a?(String) || (value.is_a?(Array) && value.size == 1 && value.first.is_a?(String))
+          error "should be a string or an array with one string (for secret lookup)"
+        end
+      end
+    end
+end

lib/kamal/configuration/validator/role.rb

@@ -0,0 +1,11 @@
+class Kamal::Configuration::Validator::Role < Kamal::Configuration::Validator
+  def validate!
+    validate_type! config, Array, Hash
+
+    if config.is_a?(Array)
+      validate_servers! "servers", config
+    else
+      super
+    end
+  end
+end

lib/kamal/configuration/validator/servers.rb

@@ -0,0 +1,7 @@
+class Kamal::Configuration::Validator::Servers < Kamal::Configuration::Validator
+  def validate!
+    validate_type! config, Array, Hash
+
+    validate_servers! config if config.is_a?(Array)
+  end
+end

lib/kamal/version.rb

@@ -1,3 +1,3 @@
 module Kamal
-  VERSION = "1.5.2"
+  VERSION = "1.7.0"
 end

test/cli/accessory_test.rb

@@ -116,6 +116,20 @@ class CliAccessoryTest < CliTestCase
     assert_match "docker logs app-mysql  --tail 100 --timestamps 2>&1", run_command("logs", "mysql")
   end
 
+  test "logs with grep" do
+    SSHKit::Backend::Abstract.any_instance.stubs(:exec)
+      .with("ssh -t root@1.1.1.3 'docker logs app-mysql --timestamps 2>&1 | grep \'hey\''")
+
+    assert_match "docker logs app-mysql --timestamps 2>&1 | grep 'hey'", run_command("logs", "mysql", "--grep", "hey")
+  end
+
+  test "logs with grep and grep options" do
+    SSHKit::Backend::Abstract.any_instance.stubs(:exec)
+      .with("ssh -t root@1.1.1.3 'docker logs app-mysql --timestamps 2>&1 | grep \'hey\' -C 2'")
+
+    assert_match "docker logs app-mysql --timestamps 2>&1 | grep 'hey' -C 2", run_command("logs", "mysql", "--grep", "hey", "--grep-options", "-C 2")
+  end
+
   test "logs with follow" do
     SSHKit::Backend::Abstract.any_instance.stubs(:exec)
       .with("ssh -t root@1.1.1.3 -p 22 'docker logs app-mysql --timestamps --tail 10 --follow 2>&1'")
@@ -123,6 +137,20 @@ class CliAccessoryTest < CliTestCase
     assert_match "docker logs app-mysql --timestamps --tail 10 --follow 2>&1", run_command("logs", "mysql", "--follow")
   end
 
+  test "logs with follow and grep" do
+    SSHKit::Backend::Abstract.any_instance.stubs(:exec)
+      .with("ssh -t root@1.1.1.3 -p 22 'docker logs app-mysql --timestamps --tail 10 --follow 2>&1 | grep \"hey\"'")
+
+    assert_match "docker logs app-mysql --timestamps --tail 10 --follow 2>&1 | grep \"hey\"", run_command("logs", "mysql", "--follow", "--grep", "hey")
+  end
+
+  test "logs with follow, grep, and grep options" do
+    SSHKit::Backend::Abstract.any_instance.stubs(:exec)
+      .with("ssh -t root@1.1.1.3 -p 22 'docker logs app-mysql --timestamps --tail 10 --follow 2>&1 | grep \"hey\" -C 2'")
+
+    assert_match "docker logs app-mysql --timestamps --tail 10 --follow 2>&1 | grep \"hey\" -C 2", run_command("logs", "mysql", "--follow", "--grep", "hey", "--grep-options", "-C 2")
+  end
+
   test "remove with confirmation" do
     Kamal::Cli::Accessory.any_instance.expects(:stop).with("mysql")
     Kamal::Cli::Accessory.any_instance.expects(:remove_container).with("mysql")

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, "kamal-proxy", "kamal-proxy", :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
@@ -284,6 +290,10 @@ class CliAppTest < CliTestCase
       .with("ssh -t root@1.1.1.1 'sh -c 'docker ps --latest --quiet --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 --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1| xargs docker logs --timestamps --tail 10 2>&1'")
 
     assert_match "sh -c 'docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''{{.ID}}'\\'') ; docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs --tail 100 2>&1", run_command("logs")
+
+    assert_match "sh -c 'docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''{{.ID}}'\\'') ; docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs 2>&1 | grep 'hey'", run_command("logs", "--grep", "hey")
+
+    assert_match "sh -c 'docker ps --latest --quiet --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 --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs 2>&1 | grep 'hey' -C 2", run_command("logs", "--grep", "hey", "--grep-options", "-C 2")
   end
 
   test "logs with follow" do
@@ -293,6 +303,20 @@ class CliAppTest < CliTestCase
     assert_match "sh -c '\\''docker ps --latest --quiet --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 --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting'\\'' | head -1 | xargs docker logs --timestamps --tail 10 --follow 2>&1", run_command("logs", "--follow")
   end
 
+  test "logs with follow and grep" do
+    SSHKit::Backend::Abstract.any_instance.stubs(:exec)
+      .with("ssh -t root@1.1.1.1 -p 22 'sh -c '\\''docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''\\'\\'''\\''{{.ID}}'\\''\\'\\'''\\'') ; docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting'\\'' | head -1 | xargs docker logs --timestamps --follow 2>&1 | grep \"hey\"'")
+
+    assert_match "sh -c '\\''docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''\\'\\'''\\''{{.ID}}'\\''\\'\\'''\\'') ; docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting'\\'' | head -1 | xargs docker logs --timestamps --follow 2>&1 | grep \"hey\"", run_command("logs", "--follow", "--grep", "hey")
+  end
+
+  test "logs with follow, grep and grep options" do
+    SSHKit::Backend::Abstract.any_instance.stubs(:exec)
+      .with("ssh -t root@1.1.1.1 -p 22 'sh -c '\\''docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''\\'\\'''\\''{{.ID}}'\\''\\'\\'''\\'') ; docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting'\\'' | head -1 | xargs docker logs --timestamps --follow 2>&1 | grep \"hey\" -C 2'")
+
+    assert_match "sh -c '\\''docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''\\'\\'''\\''{{.ID}}'\\''\\'\\'''\\'') ; docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting'\\'' | head -1 | xargs docker logs --timestamps --follow 2>&1 | grep \"hey\" -C 2", run_command("logs", "--follow", "--grep", "hey", "--grep-options", "-C 2")
+  end
+
   test "version" do
     run_command("version").tap do |output|
       assert_match "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", output
@@ -327,15 +351,25 @@ 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

@@ -21,6 +21,10 @@ class CliBuildTest < CliTestCase
         .with(:git, "-C", anything, :status, "--porcelain")
         .returns("")
 
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+        .with(:docker, :buildx, :inspect, "kamal-app-multiarch", "> /dev/null")
+        .returns("")
+
       run_command("push", "--verbose").tap do |output|
         assert_hook_ran "pre-build", output, **hook_variables
         assert_match /Cloning repo into build directory/, output
@@ -121,11 +125,9 @@ class CliBuildTest < CliTestCase
       SSHKit::Backend::Abstract.any_instance.expects(:execute)
         .with(:docker, :buildx, :create, "--use", "--name", "kamal-app-multiarch")
 
-      SSHKit::Backend::Abstract.any_instance.expects(:execute)
+      SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-        .with { |*args| args[0..1] == [ :docker, :buildx ] }
+        .with(:docker, :buildx, :inspect, "kamal-app-multiarch", "> /dev/null")
         .raises(SSHKit::Command::Failed.new("no builder"))
-        .then
-        .returns(true)
 
       SSHKit::Backend::Abstract.any_instance.expects(:execute).with { |*args| args.first.start_with?("git") }
 
@@ -137,6 +139,9 @@ class CliBuildTest < CliTestCase
         .with(:git, "-C", anything, :status, "--porcelain")
         .returns("")
 
+      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", ".")
+
       run_command("push").tap do |output|
         assert_match /WARN Missing compatible builder, so creating a new one first/, output
       end

test/cli/cli_test_case.rb

@@ -36,6 +36,9 @@ class CliTestCase < ActiveSupport::TestCase
         .with { |arg1, arg2| arg1 == :mkdir && arg2 == ".kamal/locks/app" }
       SSHKit::Backend::Abstract.any_instance.stubs(:execute)
         .with { |arg1, arg2| arg1 == :rm && arg2 == ".kamal/locks/app/details" }
+      SSHKit::Backend::Abstract.any_instance.stubs(:capture_with_info)
+        .with { |*args| args[0..2] == [ :docker, :buildx, :inspect ] }
+        .returns("")
     end
 
     def assert_hook_ran(hook, output, version:, service_version:, hosts:, command:, subcommand: nil, runtime: false)

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

@@ -22,9 +22,9 @@ class CliMainTest < CliTestCase
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:main:envify", [], invoke_options)
     Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:accessory:boot", [ "all" ], invoke_options)
     # 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:registry:login", [], invoke_options.merge(skip_local: true))
     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
@@ -46,9 +46,9 @@ class CliMainTest < CliTestCase
   test "deploy" do
     invoke_options = { "config_file" => "test/fixtures/deploy_simple.yml", "version" => "999", "skip_hooks" => false, "verbose" => true }
 
-    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:registry:login", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:registry:login", [], invoke_options.merge(skip_local: false))
     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
@@ -71,9 +71,9 @@ class CliMainTest < CliTestCase
   test "deploy with skip_push" do
     invoke_options = { "config_file" => "test/fixtures/deploy_simple.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:registry:login", [], invoke_options.merge(skip_local: true))
     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
@@ -116,6 +116,10 @@ class CliMainTest < CliTestCase
       .with(:git, "-C", anything, :status, "--porcelain")
       .returns("")
 
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:docker, :buildx, :inspect, "kamal-app-multiarch", "> /dev/null")
+      .returns("")
+
     assert_raises(Kamal::Cli::LockError) do
       run_command("deploy")
     end
@@ -145,16 +149,20 @@ class CliMainTest < CliTestCase
       .with(:git, "-C", anything, :status, "--porcelain")
       .returns("")
 
+    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
+      .with(:docker, :buildx, :inspect, "kamal-app-multiarch", "> /dev/null")
+      .returns("")
+
     assert_raises(SSHKit::Runner::ExecuteError) do
       run_command("deploy")
     end
   end
 
   test "deploy errors during outside section leave remove lock" do
-    invoke_options = { "config_file" => "test/fixtures/deploy_simple.yml", "version" => "999", "skip_hooks" => false }
+    invoke_options = { "config_file" => "test/fixtures/deploy_simple.yml", "version" => "999", "skip_hooks" => false, :skip_local => false }
 
     Kamal::Cli::Main.any_instance.expects(:invoke)
-      .with("kamal:cli:registry:login", [], invoke_options)
+      .with("kamal:cli:registry:login", [], invoke_options.merge(skip_local: false))
       .raises(RuntimeError)
 
     assert_not KAMAL.holding_lock?
@@ -167,9 +175,9 @@ class CliMainTest < CliTestCase
   test "deploy with skipped hooks" do
     invoke_options = { "config_file" => "test/fixtures/deploy_simple.yml", "version" => "999", "skip_hooks" => true }
 
-    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:registry:login", [], invoke_options)
+    Kamal::Cli::Main.any_instance.expects(:invoke).with("kamal:cli:registry:login", [], invoke_options.merge(skip_local: false))
     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)
@@ -179,12 +187,27 @@ 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.merge(skip_local: false))
+    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:registry:login", [], invoke_options.merge(skip_local: false))
     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)
@@ -236,6 +259,7 @@ 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$", "--quiet", raise_on_non_zero_exit: false)
@@ -246,11 +270,18 @@ class CliMainTest < CliTestCase
       SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
         .with(:sh, "-c", "'docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=#{role} --filter status=running --filter status=restarting --filter ancestor=$(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" }
@@ -267,6 +298,8 @@ 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$", "--quiet", raise_on_non_zero_exit: false)
       .returns("").at_least_once
@@ -274,8 +307,8 @@ class CliMainTest < CliTestCase
       .with(:sh, "-c", "'docker ps --latest --format '\\''{{.Names}}'\\'' --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(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
@@ -284,7 +317,7 @@ 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" ])
 
@@ -347,19 +380,6 @@ class CliMainTest < CliTestCase
     end
   end
 
-  test "config with aliases" do
-    run_command("config", config_file: "deploy_with_aliases").tap do |output|
-      config = YAML.load(output)
-
-      assert_equal [ "web", "web_tokyo", "workers", "workers_tokyo" ], config[:roles]
-      assert_equal [ "1.1.1.1", "1.1.1.2", "1.1.1.3", "1.1.1.4" ], config[:hosts]
-      assert_equal "999", config[:version]
-      assert_equal "registry.digitalocean.com/dhh/app", config[:repository]
-      assert_equal "registry.digitalocean.com/dhh/app:999", config[:absolute_image]
-      assert_equal "app-999", config[:service_with_version]
-    end
-  end
-
   test "init" do
     Pathname.any_instance.expects(:exist?).returns(false).times(3)
     Pathname.any_instance.stubs(:mkpath)
@@ -412,11 +432,10 @@ class CliMainTest < CliTestCase
   end
 
   test "envify" do
-    Pathname.any_instance.expects(:exist?).returns(true).times(3)
+    with_test_dot_env_erb(contents: "HELLO=<%= 'world' %>") do
-    File.expects(:read).with(".env.erb").returns("HELLO=<%= 'world' %>")
-    File.expects(:write).with(".env", "HELLO=world", perm: 0600)
-
       run_command("envify")
+      assert_equal("HELLO=world", File.read(".env"))
+    end
   end
 
   test "envify with blank line trimming" do
@@ -427,19 +446,17 @@ class CliMainTest < CliTestCase
       <% end  -%>
     EOF
 
-    Pathname.any_instance.expects(:exist?).returns(true).times(3)
+    with_test_dot_env_erb(contents: file) do
-    File.expects(:read).with(".env.erb").returns(file.strip)
-    File.expects(:write).with(".env", "HELLO=world\nKEY=value\n", perm: 0600)
-
       run_command("envify")
+      assert_equal("HELLO=world\nKEY=value\n", File.read(".env"))
+    end
   end
 
   test "envify with destination" do
-    Pathname.any_instance.expects(:exist?).returns(true).times(4)
+    with_test_dot_env_erb(contents: "HELLO=<%= 'world' %>", file: ".env.world.erb") do
-    File.expects(:read).with(".env.world.erb").returns("HELLO=<%= 'world' %>")
-    File.expects(:write).with(".env.world", "HELLO=world", perm: 0600)
-
       run_command("envify", "-d", "world", config_file: "deploy_for_dest")
+      assert_equal "HELLO=world", File.read(".env.world")
+    end
   end
 
   test "envify with skip_push" do
@@ -453,9 +470,9 @@ 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 kamal-proxy/, output
+      assert_match /docker container stop traefik/, output
-      assert_match /docker container prune --force --filter label=org.opencontainers.image.title=kamal-proxy/, 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=kamal-proxy/, 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
@@ -475,6 +492,24 @@ class CliMainTest < CliTestCase
     end
   end
 
+  test "docs" do
+    run_command("docs").tap do |output|
+      assert_match "# Kamal Configuration", output
+    end
+  end
+
+  test "docs subsection" do
+    run_command("docs", "accessory").tap do |output|
+      assert_match "# Accessories", output
+    end
+  end
+
+  test "docs unknown" do
+    run_command("docs", "foo").tap do |output|
+      assert_match "No documentation found for foo", output
+    end
+  end
+
   test "version" do
     version = stdouted { Kamal::Cli::Main.new.version }
     assert_equal Kamal::VERSION, version
@@ -484,4 +519,17 @@ class CliMainTest < CliTestCase
     def run_command(*command, config_file: "deploy_simple")
       stdouted { Kamal::Cli::Main.start([ *command, "-c", "test/fixtures/#{config_file}.yml" ]) }
     end
+
+    def with_test_dot_env_erb(contents:, file: ".env.erb")
+      Dir.mktmpdir do |dir|
+        fixtures_dup = File.join(dir, "test")
+        FileUtils.mkdir_p(fixtures_dup)
+        FileUtils.cp_r("test/fixtures/", fixtures_dup)
+
+        Dir.chdir(dir) do
+          File.write(file, contents)
+          yield
+        end
+      end
+    end
 end

test/cli/proxy_test.rb

@@ -1,116 +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 kamal-proxy --detach --restart unless-stopped --publish 80:80 --publish 443:443 --volume /var/run/docker.sock:/var/run/docker.sock --volume kamal-proxy:/root/.config/kamal-proxy --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 kamal-proxy", output
-      assert_match "docker container prune --force --filter label=org.opencontainers.image.title=kamal-proxy", output
-      assert_match "docker run --name kamal-proxy --detach --restart unless-stopped --publish 80:80 --publish 443:443 --volume /var/run/docker.sock:/var/run/docker.sock --volume kamal-proxy:/root/.config/kamal-proxy --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=kamal-proxy on 1.1.1.1", output
-    end
-  end
-
-  test "start" do
-    run_command("start").tap do |output|
-      assert_match "docker container start kamal-proxy", output
-    end
-  end
-
-  test "stop" do
-    run_command("stop").tap do |output|
-      assert_match "docker container stop kamal-proxy", 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=^kamal-proxy$", output
-    end
-  end
-
-  test "logs" do
-    SSHKit::Backend::Abstract.any_instance.stubs(:capture)
-      .with(:docker, :logs, "kamal-proxy", " --tail 100", "--timestamps", "2>&1")
-      .returns("Log entry")
-
-    run_command("logs").tap do |output|
-      assert_match "Proxy Host: 1.1.1.1", output
-      assert_match "Log entry", output
-    end
-  end
-
-  test "logs with follow" do
-    SSHKit::Backend::Abstract.any_instance.stubs(:exec)
-      .with("ssh -t root@1.1.1.1 -p 22 'docker logs kamal-proxy --timestamps --tail 10 --follow 2>&1'")
-
-    assert_match "docker logs kamal-proxy --timestamps --tail 10 --follow", run_command("logs", "--follow")
-  end
-
-  test "remove" do
-    Kamal::Cli::Proxy.any_instance.expects(:stop)
-    Kamal::Cli::Proxy.any_instance.expects(:remove_container)
-    Kamal::Cli::Proxy.any_instance.expects(:remove_image)
-
-    run_command("remove")
-  end
-
-  test "remove_container" do
-    run_command("remove_container").tap do |output|
-      assert_match "docker container prune --force --filter label=org.opencontainers.image.title=kamal-proxy", output
-    end
-  end
-
-  test "remove_image" do
-    run_command("remove_image").tap do |output|
-      assert_match "docker image prune --all --force --filter label=org.opencontainers.image.title=kamal-proxy", output
-    end
-  end
-
-  test "update" do
-    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with(:docker, :container, :ls, "--all", "--filter", "name=^app-web-123$", "--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
-
-    SSHKit::Backend::Abstract.any_instance.expects(:capture_with_info)
-      .with { |*args| args[0..1] == [ :sh, "-c" ] }
-      .returns("123")
-      .at_least_once
-
-    run_command("update", "-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 image prune --all --force --filter label=org.opencontainers.image.title=traefik", output
-      assert_match "docker container stop kamal-proxy", output
-      assert_match "docker container prune --force --filter label=org.opencontainers.image.title=kamal-proxy", output
-      assert_match "docker run --name kamal-proxy --detach --restart unless-stopped --publish 80:80 --publish 443:443 --volume /var/run/docker.sock:/var/run/docker.sock --volume kamal-proxy:/root/.config/kamal-proxy --log-opt max-size=\"10m\" #{Kamal::Configuration::Proxy::DEFAULT_IMAGE}", output
-      assert_match "docker exec kamal-proxy kamal-proxy deploy app-web --target \"172.1.0.2:80\"", 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/registry_test.rb

@@ -8,12 +8,41 @@ class CliRegistryTest < CliTestCase
     end
   end
 
+  test "login skip local" do
+    run_command("login", "-L").tap do |output|
+      assert_no_match /docker login -u \[REDACTED\] -p \[REDACTED\] as .*@localhost/, output
+      assert_match /docker login -u \[REDACTED\] -p \[REDACTED\] on 1.1.1.\d/, output
+    end
+  end
+
+  test "login skip remote" do
+    run_command("login", "-R").tap do |output|
+      assert_match /docker login -u \[REDACTED\] -p \[REDACTED\] as .*@localhost/, output
+      assert_no_match /docker login -u \[REDACTED\] -p \[REDACTED\] on 1.1.1.\d/, output
+    end
+  end
+
   test "logout" do
     run_command("logout").tap do |output|
+      assert_match /docker logout as .*@localhost/, output
       assert_match /docker logout on 1.1.1.\d/, output
     end
   end
 
+  test "logout skip local" do
+    run_command("logout", "-L").tap do |output|
+      assert_no_match /docker logout as .*@localhost/, output
+      assert_match /docker logout on 1.1.1.\d/, output
+    end
+  end
+
+  test "logout skip remote" do
+    run_command("logout", "-R").tap do |output|
+      assert_match /docker logout as .*@localhost/, output
+      assert_no_match /docker logout on 1.1.1.\d/, output
+    end
+  end
+
   private
     def run_command(*command)
       stdouted { Kamal::Cli::Registry.start([ *command, "-c", "test/fixtures/deploy_with_accessories.yml" ]) }

test/cli/traefik_test.rb

@@ -0,0 +1,110 @@
+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::Configuration::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::Configuration::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 "logs with follow and grep" 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 | grep \"hey\"'")
+
+    assert_match "docker logs traefik --timestamps --tail 10 --follow 2>&1 | grep \"hey\"", run_command("logs", "--follow", "--grep", "hey")
+  end
+
+  test "logs with follow, grep, and grep options" 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 | grep \"hey\" -C 2'")
+
+    assert_match "docker logs traefik --timestamps --tail 10 --follow 2>&1 | grep \"hey\" -C 2", run_command("logs", "--follow", "--grep", "hey", "--grep-options", "-C 2")
+  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)
+    configure_with(:deploy_with_multiple_traefik_roles)
 
     @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)
+    configure_with(:deploy_with_multiple_traefik_roles)
 
-    @kamal.specific_hosts = [ "1.1.1.4" ]
+    @kamal.specific_hosts = [ "1.1.1.2" ]
-    assert_equal [ "1.1.1.4" ], @kamal.proxy_hosts
+    assert_equal [ "1.1.1.2" ], @kamal.traefik_hosts
   end
 
   private

test/commands/accessory_test.rb

@@ -24,7 +24,7 @@ class CommandsAccessoryTest < ActiveSupport::TestCase
           "host" => "1.1.1.6",
           "port" => "6379:6379",
           "labels" => {
-            "cache" => true
+            "cache" => "true"
           },
           "env" => {
             "SOMETHING" => "else"
@@ -125,6 +125,10 @@ class CommandsAccessoryTest < ActiveSupport::TestCase
     assert_equal \
       "docker logs app-mysql  --since 5m  --tail 100 --timestamps 2>&1 | grep 'thing'",
       new_command(:mysql).logs(since: "5m", lines: 100, grep: "thing").join(" ")
+
+    assert_equal \
+      "docker logs app-mysql  --since 5m  --tail 100 --timestamps 2>&1 | grep 'thing' -C 2",
+      new_command(:mysql).logs(since: "5m", lines: 100, grep: "thing", grep_options: "-C 2").join(" ")
   end
 
   test "follow logs" do

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
 
@@ -115,23 +139,45 @@ class CommandsAppTest < ActiveSupport::TestCase
     assert_equal \
       "sh -c 'docker ps --latest --quiet --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 --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs 2>&1",
       new_command.logs.join(" ")
+  end
 
+  test "logs with since" do
     assert_equal \
       "sh -c 'docker ps --latest --quiet --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 --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs --since 5m 2>&1",
       new_command.logs(since: "5m").join(" ")
+  end
 
+  test "logs with lines" do
     assert_equal \
       "sh -c 'docker ps --latest --quiet --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 --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs --tail 100 2>&1",
       new_command.logs(lines: "100").join(" ")
+  end
 
+  test "logs with since and lines" do
     assert_equal \
       "sh -c 'docker ps --latest --quiet --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 --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs --since 5m --tail 100 2>&1",
       new_command.logs(since: "5m", lines: "100").join(" ")
+  end
 
+  test "logs with grep" do
     assert_equal \
       "sh -c 'docker ps --latest --quiet --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 --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs 2>&1 | grep 'my-id'",
       new_command.logs(grep: "my-id").join(" ")
+  end
 
+  test "logs with grep and grep options" do
+    assert_equal \
+      "sh -c 'docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''{{.ID}}'\\'') ; docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs 2>&1 | grep 'my-id' -C 2",
+      new_command.logs(grep: "my-id", grep_options: "-C 2").join(" ")
+  end
+
+  test "logs with since, grep and grep options" do
+    assert_equal \
+      "sh -c 'docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting --filter ancestor=$(docker image ls --filter reference=dhh/app:latest --format '\\''{{.ID}}'\\'') ; docker ps --latest --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs --since 5m 2>&1 | grep 'my-id' -C 2",
+      new_command.logs(since: "5m", grep: "my-id", grep_options: "-C 2").join(" ")
+  end
+
+  test "logs with since and grep" do
     assert_equal \
       "sh -c 'docker ps --latest --quiet --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 --quiet --filter label=service=app --filter label=role=web --filter status=running --filter status=restarting' | head -1 | xargs docker logs --since 5m 2>&1 | grep 'my-id'",
       new_command.logs(since: "5m", grep: "my-id").join(" ")
@@ -374,6 +420,20 @@ 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/builder_test.rb

@@ -158,6 +158,48 @@ class CommandsBuilderTest < ActiveSupport::TestCase
       builder.push.join(" ")
   end
 
+  test "multiarch context hosts" do
+    command = new_builder_command
+    assert_equal "docker buildx inspect kamal-app-multiarch > /dev/null", command.context_hosts.join(" ")
+    assert_equal "", command.config_context_hosts.join(" ")
+  end
+
+  test "native context hosts" do
+    command = new_builder_command(builder: { "multiarch" => false })
+    assert_equal :true, command.context_hosts
+    assert_equal "", command.config_context_hosts.join(" ")
+  end
+
+  test "native cached context hosts" do
+    command = new_builder_command(builder: { "multiarch" => false, "cache" => { "type" => "registry" } })
+    assert_equal "docker buildx inspect kamal-app-native-cached > /dev/null", command.context_hosts.join(" ")
+    assert_equal "", command.config_context_hosts.join(" ")
+  end
+
+  test "native remote context hosts" do
+    command = new_builder_command(builder: { "remote" => { "arch" => "amd64", "host" => "ssh://host" } })
+    assert_equal "docker context inspect kamal-app-native-remote-amd64 --format '{{.Endpoints.docker.Host}}'", command.context_hosts.join(" ")
+    assert_equal [ "ssh://host" ], command.config_context_hosts
+  end
+
+  test "multiarch remote context hosts" do
+    command = new_builder_command(builder: {
+      "remote" => { "arch" => "amd64", "host" => "ssh://host" },
+      "local" => { "arch" => "arm64" }
+    })
+    assert_equal "docker context inspect kamal-app-multiarch-remote-arm64 --format '{{.Endpoints.docker.Host}}' ; docker context inspect kamal-app-multiarch-remote-amd64 --format '{{.Endpoints.docker.Host}}'", command.context_hosts.join(" ")
+    assert_equal [ "ssh://host" ], command.config_context_hosts
+  end
+
+  test "multiarch remote context hosts with local host" do
+    command = new_builder_command(builder: {
+      "remote" => { "arch" => "amd64", "host" => "ssh://host" },
+      "local" => { "arch" => "arm64", "host" => "unix:///var/run/docker.sock" }
+    })
+    assert_equal "docker context inspect kamal-app-multiarch-remote-arm64 --format '{{.Endpoints.docker.Host}}' ; docker context inspect kamal-app-multiarch-remote-amd64 --format '{{.Endpoints.docker.Host}}'", command.context_hosts.join(" ")
+    assert_equal [ "unix:///var/run/docker.sock", "ssh://host" ], command.config_context_hosts
+  end
+
   private
     def new_builder_command(additional_config = {})
       Kamal::Commands::Builder.new(Kamal::Configuration.new(@config.merge(additional_config), version: "123"))

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 kamal-proxy --detach --restart unless-stopped --publish 80:80 --publish 443:443 --volume /var/run/docker.sock:/var/run/docker.sock --volume kamal-proxy:/root/.config/kamal-proxy --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 kamal-proxy --detach --restart unless-stopped --publish 80:80 --publish 443:443 --volume /var/run/docker.sock:/var/run/docker.sock --volume kamal-proxy:/root/.config/kamal-proxy --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 kamal-proxy --detach --restart unless-stopped --publish 80:80 --publish 443:443 --volume /var/run/docker.sock:/var/run/docker.sock --volume kamal-proxy:/root/.config/kamal-proxy --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 kamal-proxy --detach --restart unless-stopped --publish 80:80 --publish 443:443 --volume /var/run/docker.sock:/var/run/docker.sock --volume kamal-proxy:/root/.config/kamal-proxy --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 kamal-proxy",
-      new_command.start.join(" ")
-  end
-
-  test "proxy stop" do
-    assert_equal \
-      "docker container stop kamal-proxy",
-      new_command.stop.join(" ")
-  end
-
-  test "proxy info" do
-    assert_equal \
-      "docker ps --filter name=^kamal-proxy$",
-      new_command.info.join(" ")
-  end
-
-  test "proxy logs" do
-    assert_equal \
-      "docker logs kamal-proxy --timestamps 2>&1",
-      new_command.logs.join(" ")
-  end
-
-  test "proxy logs since 2h" do
-    assert_equal \
-      "docker logs kamal-proxy  --since 2h --timestamps 2>&1",
-      new_command.logs(since: "2h").join(" ")
-  end
-
-  test "proxy logs last 10 lines" do
-    assert_equal \
-      "docker logs kamal-proxy  --tail 10 --timestamps 2>&1",
-      new_command.logs(lines: 10).join(" ")
-  end
-
-  test "proxy logs with grep hello!" do
-    assert_equal \
-      "docker logs kamal-proxy --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=kamal-proxy",
-      new_command.remove_container.join(" ")
-  end
-
-  test "proxy remove image" do
-    assert_equal \
-      "docker image prune --all --force --filter label=org.opencontainers.image.title=kamal-proxy",
-      new_command.remove_image.join(" ")
-  end
-
-  test "proxy follow logs" do
-    assert_equal \
-      "ssh -t root@1.1.1.1 -p 22 'docker logs kamal-proxy --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 kamal-proxy --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 kamal-proxy kamal-proxy 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 kamal-proxy kamal-proxy 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 +1,204 @@
+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::Configuration::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 logs with grep hello! and grep options" do
+    assert_equal \
+      "docker logs traefik --timestamps 2>&1 | grep 'hello!' -C 2",
+      new_command.logs(grep: "hello!", grep_options: "-C 2").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/accessory_test.rb

@@ -35,7 +35,7 @@ class ConfigurationAccessoryTest < ActiveSupport::TestCase
           "hosts" => [ "1.1.1.6", "1.1.1.7" ],
           "port" => "6379:6379",
           "labels" => {
-            "cache" => true
+            "cache" => "true"
           },
           "env" => {
             "SOMETHING" => "else"
@@ -44,7 +44,7 @@ class ConfigurationAccessoryTest < ActiveSupport::TestCase
             "/var/lib/redis:/data"
           ],
           "options" => {
-            "cpus" => 4,
+            "cpus" => "4",
             "memory" => "2GB"
           }
         },
@@ -54,13 +54,13 @@ class ConfigurationAccessoryTest < ActiveSupport::TestCase
           "roles" => [ "web" ],
           "port" => "4321:4321",
           "labels" => {
-            "cache" => true
+            "cache" => "true"
           },
           "env" => {
             "STATSD_PORT" => "8126"
           },
           "options" => {
-            "cpus" => 4,
+            "cpus" => "4",
             "memory" => "2GB"
           }
         }
@@ -89,22 +89,20 @@ class ConfigurationAccessoryTest < ActiveSupport::TestCase
 
   test "missing host" do
     @deploy[:accessories]["mysql"]["host"] = nil
-    @config = Kamal::Configuration.new(@deploy)
 
-    assert_raises(ArgumentError) do
+    assert_raises(Kamal::ConfigurationError) do
-      @config.accessory(:mysql).hosts
+      Kamal::Configuration.new(@deploy)
     end
   end
 
   test "setting host, hosts and roles" do
-    @deploy[:accessories]["mysql"]["hosts"] = true
+    @deploy[:accessories]["mysql"]["hosts"] = [ "mysql-db1" ]
-    @deploy[:accessories]["mysql"]["roles"] = true
+    @deploy[:accessories]["mysql"]["roles"] = [ "db" ]
-    @config = Kamal::Configuration.new(@deploy)
 
-    exception = assert_raises(ArgumentError) do
+    exception = assert_raises(Kamal::ConfigurationError) do
-      @config.accessory(:mysql).hosts
+      Kamal::Configuration.new(@deploy)
     end
-    assert_equal "Specify one of `host`, `hosts` or `roles` for accessory `mysql`", exception.message
+    assert_equal "accessories/mysql: specify one of `host`, `hosts` or `roles`", exception.message
   end
 
   test "all hosts" do