discovery: add rails service name detector

pkg/collector/corechecks/servicediscovery/module/impl_linux.go

@@ -353,6 +353,8 @@ func (s *discovery) getServiceInfo(proc *process.Process) (*serviceInfo, error)
  }
 
  contextMap := make(usm.DetectorContextMap)
+ contextMap[usm.ServiceProc] = proc
+
  fs := usm.NewSubDirFS(root)
  ctx := usm.NewDetectionContext(cmdline, env, fs)
  ctx.Pid = int(proc.Pid)

pkg/collector/corechecks/servicediscovery/usm/ruby.go

@@ -0,0 +1,104 @@
+// Unless explicitly stated otherwise all files in this repository are licensed
+// under the Apache License Version 2.0.
+// This product includes software developed at Datadog (https://www.datadoghq.com/).
+// Copyright 2024-present Datadog, Inc.
+
+package usm
+
+import (
+ "errors"
+ "fmt"
+ "io"
+ "io/fs"
+ "regexp"
+ "strings"
+
+ "github.com/shirou/gopsutil/v3/process"
+
+ "github.com/DataDog/datadog-agent/pkg/util/log"
+)
+
+var (
+ moduleRegexp = regexp.MustCompile(`module\s+([A-Z][a-zA-Z0-9_]*)`)
+ matchFirstCap = regexp.MustCompile("(.)([A-Z][a-z]+)")
+ matchAllCap = regexp.MustCompile("([a-z0-9])([A-Z])")
+)
+
+type railsDetector struct {
+ ctx DetectionContext
+}
+
+func newRailsDetector(ctx DetectionContext) detector {
+ return &railsDetector{ctx}
+}
+
+// detect checks if the service is a Rails application by looking for a
+// `config/application.rb` file generated by `rails new` when a new rails
+// project is created. This file should contain a `module` declaration with the
+// application name.
+func (r railsDetector) detect(_ []string) (ServiceMetadata, bool) {
+ var proc *process.Process
+
+ if procEntry, ok := r.ctx.ContextMap[ServiceProc]; ok {
+ if p, ok := procEntry.(*process.Process); ok {
+ proc = p
+ } else {
+ log.Errorf("could not get process object in rails detector: got type %T", procEntry)
+ }
+ }
+
+ cwd, err := proc.Cwd()
+ if err != nil {
+ log.Debugf("could not get cwd of process: %s", err)
+ return ServiceMetadata{}, false
+ }
+
+ absFile := abs("config/application.rb", cwd)
+ if _, err := fs.Stat(r.ctx.fs, absFile); err != nil {
+ return ServiceMetadata{}, false
+ }
+
+ name, err := r.findRailsApplicationName(absFile)
+ if err != nil {
+ log.Debugf("could not find ruby application name: %s", err)
+ return ServiceMetadata{}, false
+ }
+
+ return NewServiceMetadata(railsUnderscore(name)), true
+}
+
+// findRailsApplicationName scans the `config/application.rb` file to find the
+// Rails application name.
+func (r railsDetector) findRailsApplicationName(filename string) (string, error) {
+ file, err := r.ctx.fs.Open(filename)
+ if err != nil {
+ return "", fmt.Errorf("could not open application.rb: %w", err)
+ }
+ defer file.Close()
+
+ reader, err := SizeVerifiedReader(file)
+ if err != nil {
+ return "", fmt.Errorf("skipping application.rb (%q): %w", filename, err)
+ }
+
+ bytes, err := io.ReadAll(reader)
+ if err != nil {
+ return "", fmt.Errorf("unable to read application.rb (%q): %w", filename, err)
+ }
+
+ matches := moduleRegexp.FindSubmatch(bytes)
+ if len(matches) < 2 {
+ // No match found
+ return "", errors.New("could not find Ruby module name")
+ }
+
+ return string(matches[1]), nil
+}
+
+// railsUnderscore converts a PascalCasedWord to a snake_cased_word.
+// It keeps uppercase acronyms together when converting (e.g. "HTTPServer" -> "http_server").
+func railsUnderscore(pascalCasedWord string) string {
+ snake := matchFirstCap.ReplaceAllString(pascalCasedWord, "${1}_${2}")
+ snake = matchAllCap.ReplaceAllString(snake, "${1}_${2}")
+ return strings.ToLower(snake)
+}

pkg/collector/corechecks/servicediscovery/usm/ruby_test.go

@@ -0,0 +1,48 @@
+// Unless explicitly stated otherwise all files in this repository are licensed
+// under the Apache License Version 2.0.
+// This product includes software developed at Datadog (https://www.datadoghq.com/).
+// Copyright 2024-present Datadog, Inc.
+
+//go:build linux
+
+package usm
+
+import (
+ "path/filepath"
+ "testing"
+
+ "github.com/stretchr/testify/assert"
+ "github.com/stretchr/testify/require"
+)
+
+func TestGenerateNameFromRailsApplicationRb(t *testing.T) {
+ tests := []struct {
+ name string
+ path string
+ expected string
+ }{
+ {
+ name: "name is found",
+ path: "./testdata/application.rb",
+ expected: "rails_hello",
+ },
+ {
+ name: "name not found",
+ path: "./testdata/application_invalid.rb",
+ expected: "",
+ },
+ }
+ full, err := filepath.Abs("testdata/root")
+ require.NoError(t, err)
+ for _, tt := range tests {
+ t.Run(tt.name, func(t *testing.T) {
+ instance := &railsDetector{ctx: DetectionContext{
+ fs: NewSubDirFS(full),
+ contextMap: make(DetectorContextMap),
+ }}
+ value, ok := instance.findRailsApplicationName(tt.path)
+ assert.Equal(t, len(tt.expected) > 0, ok)
+ assert.Equal(t, tt.expected, railsUnderscore(value))
+ })
+ }
+}

pkg/collector/corechecks/servicediscovery/usm/service.go

@@ -35,6 +35,8 @@ const (
  NodePackageJSONPath = iota
  // ServiceSubFS The SubdirFS instance package.json path is valid in.
  ServiceSubFS = iota
+ // The pointer to the Process instance of the service
+ ServiceProc = iota
 )
 
 const (
@@ -188,8 +190,9 @@ var languageDetectors = map[language.Language]detectorCreatorFn{
 // Map executables that usually have additional process context of what's
 // running, to context detectors
 var executableDetectors = map[string]detectorCreatorFn{
- "sudo": newSimpleDetector,
  "gunicorn": newGunicornDetector,
+ "puma": newRailsDetector,
+ "sudo": newSimpleDetector,
 }
 
 func serviceNameInjected(envs envs.Variables) bool {

pkg/collector/corechecks/servicediscovery/usm/testdata/root/testdata/application.rb

@@ -0,0 +1,38 @@
+require_relative "boot"
+
+require "rails"
+# Pick the frameworks you want:
+require "active_model/railtie"
+# require "active_job/railtie"
+require "active_record/railtie"
+# require "active_storage/engine"
+require "action_controller/railtie"
+# require "action_mailer/railtie"
+# require "action_mailbox/engine"
+# require "action_text/engine"
+require "action_view/railtie"
+# require "action_cable/engine"
+require "sprockets/railtie"
+require "rails/test_unit/railtie"
+
+# Require the gems listed in Gemfile, including any gems
+# you've limited to :test, :development, or :production.
+Bundler.require(*Rails.groups)
+
+module RailsHello
+ class Application < Rails::Application
+ # Initialize configuration defaults for originally generated Rails version.
+ config.load_defaults 6.1
+
+ # Configuration for the application, engines, and railties goes here.
+ #
+ # These settings can be overridden in specific environments using the files
+ # in config/environments, which are processed later.
+ #
+ # config.time_zone = "Central Time (US & Canada)"
+ # config.eager_load_paths << Rails.root.join("extras")
+
+ # Don't generate system test files.
+ config.generators.system_tests = nil
+ end
+end

pkg/collector/corechecks/servicediscovery/usm/testdata/root/testdata/application_invalid.rb

@@ -0,0 +1,2 @@
+class SomeRubyClass
+end

test/new-e2e/tests/discovery/linux_test.go

@@ -33,7 +33,13 @@ type linuxTestSuite struct {
  e2e.BaseSuite[environments.Host]
 }
 
-var services = []string{"python-svc", "python-instrumented", "node-json-server", "node-instrumented"}
+var services = []string{
+ "python-svc",
+ "python-instrumented",
+ "node-json-server",
+ "node-instrumented",
+ "rails-svc",
+}
 
 func TestLinuxTestSuite(t *testing.T) {
  agentParams := []func(*agentparams.Params) error{
@@ -125,6 +131,15 @@ func (s *linuxTestSuite) TestServiceDiscoveryCheck() {
  assert.NotZero(c, found.Payload.RSSMemory)
  }
 
+ found = foundMap["rails_hello"]
+ if assert.NotNil(c, found) {
+ assert.Equal(c, "rails_hello", found.Payload.ServiceName)
+ assert.Equal(c, "rails_hello", found.Payload.GeneratedServiceName)
+ assert.Empty(c, found.Payload.DDService)
+ assert.Empty(c, found.Payload.ServiceNameSource)
+ assert.NotZero(c, found.Payload.RSSMemory)
+ }
+
  assert.Contains(c, foundMap, "json-server")
  }, 3*time.Minute, 10*time.Second)
 }

test/new-e2e/tests/discovery/testdata/provision/provision.sh

@@ -2,6 +2,47 @@
 
 set -e
 
+install_systemd_unit() {
+ while [[ $# -ge 2 ]]; do
+ case $1 in
+ --workdir)
+ shift
+ workdir="WorkingDirectory=$1"
+ shift
+ ;;
+ *)
+ break
+ ;;
+ esac
+ done
+
+ name=$1
+ command=$2
+ port=$3
+ extraenv=$4
+
+ cat > "/etc/systemd/system/${name}.service" <<- EOM
+[Unit]
+Description=${name}
+After=network.target
+StartLimitIntervalSec=0
+
+[Service]
+Type=simple
+Restart=always
+RestartSec=1
+User=root
+ExecStart=${command}
+Environment="PORT=${port}"
+Environment="NODE_VERSION=20"
+Environment="${extraenv}"
+${workdir}
+
+[Install]
+WantedBy=multi-user.target
+EOM
+}
+
 apt-get update
 apt-get install -y \
  ca-certificates \
@@ -28,43 +69,33 @@ nvm install 20 || nvm install 20 || nvm install 20
 npm install json-server || npm install json-server
 npm install /home/ubuntu/e2e-test/node/instrumented
 
-# Install our own services
-install_systemd_unit () {
- name=$1
- command=$2
- port=$3
- extraenv=$4
-
- cat > "/etc/systemd/system/${name}.service" <<- EOM
-[Unit]
-Description=${name}
-After=network.target
-StartLimitIntervalSec=0
-
-[Service]
-Type=simple
-Restart=always
-RestartSec=1
-User=root
-ExecStart=${command}
-Environment="PORT=${port}"
-Environment="NODE_VERSION=20"
-Environment="${extraenv}"
+# Install Ruby
+## Packages
+apt-get install -y \
+ ruby \
+ ruby-dev \
+ ruby-rails \
+ sqlite3 \
 
-[Install]
-WantedBy=multi-user.target
-EOM
-}
+## Create new Rails project
+pushd /home/ubuntu
+rails new rails-hello --minimal
+bundle install --gemfile=/home/ubuntu/rails-hello/Gemfile
+popd
 
-# Node
+# Install our own services
+## Node
 install_systemd_unit "node-json-server" "$NVM_DIR/nvm-exec npx json-server --port 8084 /home/ubuntu/e2e-test/node/json-server/db.json" "8084" ""
 install_systemd_unit "node-instrumented" "$NVM_DIR/nvm-exec node /home/ubuntu/e2e-test/node/instrumented/server.js" "8085" ""
 
-# Python
+## Python
 install_systemd_unit "python-svc" "/usr/bin/python3 /home/ubuntu/e2e-test/python/server.py" "8082" "DD_SERVICE=python-svc-dd"
 install_systemd_unit "python-instrumented" "/usr/bin/python3 /home/ubuntu/e2e-test/python/instrumented.py" "8083" ""
 
+## Ruby
+install_systemd_unit --workdir "/home/ubuntu/rails-hello" "rails-svc" "rails server" "7777" ""
+
 systemctl daemon-reload
 
-# leave them stopped
+## leave them stopped
 systemctl stop python-svc