[metricbeat] - Allow metricsets to report their status via v2 protocol

CHANGELOG.next.asciidoc

@@ -52,6 +52,7 @@ https://github.com/elastic/beats/compare/v8.8.1\...main[Check the HEAD diff]
 
 - Setting period for counter cache for Prometheus remote_write at least to 60sec {pull}38553[38553]
 - Add support of Graphite series 1.1.0+ tagging extension for statsd module. {pull}39619[39619]
+- Allow metricsets to report their status via control v2 protocol. {pull}40025[40025]
 - Remove fallback to the node limit for the `kubernetes.pod.cpu.usage.limit.pct` and `kubernetes.pod.memory.usage.limit.pct` metrics calculation
 
 *Osquerybeat*

metricbeat/helper/http_test.go

@@ -19,7 +19,7 @@ package helper
 
 import (
 	"fmt"
-	"io/ioutil"
+	"io"
 	"net"
 	"net/http"
 	"net/http/httptest"
@@ -31,6 +31,7 @@ import (
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 
+	"github.com/elastic/beats/v7/libbeat/management/status"
 	"github.com/elastic/beats/v7/metricbeat/helper/dialer"
 	"github.com/elastic/beats/v7/metricbeat/mb"
 	"github.com/elastic/beats/v7/metricbeat/mb/parse"
@@ -55,7 +56,7 @@ func TestGetAuthHeaderFromToken(t *testing.T) {
 	for _, test := range tests {
 		t.Run(test.Name, func(t *testing.T) {
 			content := []byte(test.Content)
-			tmpfile, err := ioutil.TempFile("", "token")
+			tmpfile, err := os.CreateTemp("", "token")
 			if err != nil {
 				t.Fatal(err)
 			}
@@ -236,14 +237,14 @@ func TestOverUnixSocket(t *testing.T) {
 			fmt.Fprintf(w, "ehlo!")
 		})
 
-		go http.Serve(l, mux)
+		go http.Serve(l, mux) //nolint:all // Ignore the error
 
 		return l
 	}
 
 	for title, c := range cases {
 		t.Run(title, func(t *testing.T) {
-			tmpDir, err := ioutil.TempDir("", "testsocket")
+			tmpDir, err := os.MkdirTemp("", "testsocket")
 			require.NoError(t, err)
 			defer os.RemoveAll(tmpDir)
 
@@ -262,7 +263,7 @@ func TestOverUnixSocket(t *testing.T) {
 			r, err := h.FetchResponse()
 			require.NoError(t, err)
 			defer r.Body.Close()
-			content, err := ioutil.ReadAll(r.Body)
+			content, err := io.ReadAll(r.Body)
 			require.NoError(t, err)
 			assert.Equal(t, []byte("ehlo!"), content)
 		})
@@ -327,3 +328,5 @@ func (*dummyModule) Config() mb.ModuleConfig {
 func (*dummyModule) UnpackConfig(interface{}) error {
 	return nil
 }
+func (dummyModule) UpdateStatus(_ status.Status, _ string)    {}
+func (dummyModule) SetStatusReporter(_ status.StatusReporter) {}

metricbeat/mb/mb.go

@@ -27,6 +27,7 @@ import (
 	"net/url"
 	"time"
 
+	"github.com/elastic/beats/v7/libbeat/management/status"
 	"github.com/elastic/beats/v7/metricbeat/helper/dialer"
 	conf "github.com/elastic/elastic-agent-libs/config"
 	"github.com/elastic/elastic-agent-libs/logp"
@@ -62,9 +63,11 @@ const (
 
 // Module is the common interface for all Module implementations.
 type Module interface {
-	Name() string                      // Name returns the name of the Module.
-	Config() ModuleConfig              // Config returns the ModuleConfig used to create the Module.
-	UnpackConfig(to interface{}) error // UnpackConfig unpacks the raw module config to the given object.
+	Name() string                                           // Name returns the name of the Module.
+	Config() ModuleConfig                                   // Config returns the ModuleConfig used to create the Module.
+	UnpackConfig(to interface{}) error                      // UnpackConfig unpacks the raw module config to the given object.
+	UpdateStatus(status status.Status, msg string)          // UpdateStatus updates the status of the module. Reflected on elastic-agent.
+	SetStatusReporter(statusReporter status.StatusReporter) // SetStatusReporter updates the status reporter for the given module.
 }
 
 // BaseModule implements the Module interface.
@@ -73,9 +76,10 @@ type Module interface {
 // MetricSets, it can embed this type into another struct to satisfy the
 // Module interface requirements.
 type BaseModule struct {
-	name      string
-	config    ModuleConfig
-	rawConfig *conf.C
+	name           string
+	config         ModuleConfig
+	rawConfig      *conf.C
+	statusReporter status.StatusReporter
 }
 
 func (m *BaseModule) String() string {
@@ -95,6 +99,18 @@ func (m *BaseModule) UnpackConfig(to interface{}) error {
 	return m.rawConfig.Unpack(to)
 }
 
+// UpdateStatus updates the status of the module. Reflected on elastic-agent.
+func (m *BaseModule) UpdateStatus(status status.Status, msg string) {
+	if m.statusReporter != nil {
+		m.statusReporter.UpdateStatus(status, msg)
+	}
+}
+
+// SetStatusReporter sets the status repoter of the module.
+func (m *BaseModule) SetStatusReporter(statusReporter status.StatusReporter) {
+	m.statusReporter = statusReporter
+}
+
 // WithConfig re-configures the module with the given raw configuration and returns a
 // copy of the module.
 // Intended to be called from module factories. Note that if metricsets are specified

metricbeat/mb/module/runner.go

@@ -25,6 +25,7 @@ import (
 	"github.com/elastic/beats/v7/libbeat/beat"
 	"github.com/elastic/beats/v7/libbeat/cfgfile"
 	"github.com/elastic/beats/v7/libbeat/common/diagnostics"
+	"github.com/elastic/beats/v7/libbeat/management/status"
 	"github.com/elastic/elastic-agent-libs/monitoring"
 )
 
@@ -123,3 +124,7 @@ func (mr *runner) Diagnostics() []diagnostics.DiagnosticSetup {
 func (mr *runner) String() string {
 	return fmt.Sprintf("%s [metricsets=%d]", mr.mod.Name(), len(mr.mod.metricSets))
 }
+
+func (mr *runner) SetStatusReporter(reporter status.StatusReporter) {
+	mr.mod.SetStatusReporter(reporter)
+}

metricbeat/mb/module/runner_group.go

@@ -23,6 +23,7 @@ import (
 
 	"github.com/elastic/beats/v7/libbeat/cfgfile"
 	"github.com/elastic/beats/v7/libbeat/common/diagnostics"
+	"github.com/elastic/beats/v7/libbeat/management/status"
 )
 
 type runnerGroup struct {
@@ -40,6 +41,14 @@ func newRunnerGroup(runners []cfgfile.Runner) cfgfile.Runner {
 	}
 }
 
+func (rg *runnerGroup) SetStatusReporter(reporter status.StatusReporter) {
+	for _, runner := range rg.runners {
+		if runnerWithStatus, ok := runner.(status.WithStatusReporter); ok {
+			runnerWithStatus.SetStatusReporter(reporter)
+		}
+	}
+}
+
 func (rg *runnerGroup) Start() {
 	rg.startOnce.Do(func() {
 		for _, runner := range rg.runners {

metricbeat/mb/module/wrapper.go

@@ -26,6 +26,7 @@ import (
 
 	"github.com/elastic/beats/v7/libbeat/beat"
 	"github.com/elastic/beats/v7/libbeat/common"
+	"github.com/elastic/beats/v7/libbeat/management/status"
 	"github.com/elastic/beats/v7/metricbeat/mb"
 	conf "github.com/elastic/elastic-agent-libs/config"
 	"github.com/elastic/elastic-agent-libs/logp"
@@ -146,6 +147,7 @@ func (mw *Wrapper) Start(done <-chan struct{}) <-chan beat.Event {
 			registry.Add(metricsPath, msw.Metrics(), monitoring.Full)
 			monitoring.NewString(msw.Metrics(), "starttime").Set(common.Time(time.Now()).String())
 
+			msw.module.UpdateStatus(status.Starting, fmt.Sprintf("%s/%s is starting", msw.module.Name(), msw.Name()))
 			msw.run(done, out)
 		}(msw)
 	}
@@ -253,14 +255,20 @@ func (msw *metricSetWrapper) fetch(ctx context.Context, reporter reporter) {
 		err := fetcher.Fetch(reporter.V2())
 		if err != nil {
 			reporter.V2().Error(err)
+			msw.module.UpdateStatus(status.Degraded, fmt.Sprintf("Error fetching data for metricset %s.%s: %s", msw.module.Name(), msw.MetricSet.Name(), err))
 			logp.Err("Error fetching data for metricset %s.%s: %s", msw.module.Name(), msw.Name(), err)
+		} else {
+			msw.module.UpdateStatus(status.Running, "")
 		}
 	case mb.ReportingMetricSetV2WithContext:
 		reporter.StartFetchTimer()
 		err := fetcher.Fetch(ctx, reporter.V2())
 		if err != nil {
 			reporter.V2().Error(err)
+			msw.module.UpdateStatus(status.Degraded, fmt.Sprintf("Error fetching data for metricset %s.%s: %s", msw.module.Name(), msw.MetricSet.Name(), err))
 			logp.Err("Error fetching data for metricset %s.%s: %s", msw.module.Name(), msw.Name(), err)
+		} else {
+			msw.module.UpdateStatus(status.Running, "")
 		}
 	default:
 		panic(fmt.Sprintf("unexpected fetcher type for %v", msw))

metricbeat/mb/testing/modules.go

@@ -53,6 +53,7 @@ that Metricbeat does it and with the same validations.
 		}
 	}
 */
+//nolint:all // It's a test file
 package testing
 
 import (
@@ -61,6 +62,7 @@ import (
 	"testing"
 	"time"
 
+	"github.com/elastic/beats/v7/libbeat/management/status"
 	"github.com/elastic/beats/v7/metricbeat/mb"
 	conf "github.com/elastic/elastic-agent-libs/config"
 	"github.com/elastic/elastic-agent-libs/mapstr"
@@ -72,9 +74,11 @@ type TestModule struct {
 	RawConfig *conf.C
 }
 
-func (m *TestModule) Name() string                      { return m.ModName }
-func (m *TestModule) Config() mb.ModuleConfig           { return m.ModConfig }
-func (m *TestModule) UnpackConfig(to interface{}) error { return m.RawConfig.Unpack(to) }
+func (m *TestModule) Name() string                              { return m.ModName }
+func (m *TestModule) Config() mb.ModuleConfig                   { return m.ModConfig }
+func (m *TestModule) UnpackConfig(to interface{}) error         { return m.RawConfig.Unpack(to) }
+func (m *TestModule) UpdateStatus(_ status.Status, _ string)    {}
+func (m *TestModule) SetStatusReporter(_ status.StatusReporter) {}
 
 func NewTestModule(t testing.TB, config interface{}) *TestModule {
 	c, err := conf.NewConfigFrom(config)

metricbeat/module/elasticsearch/node_stats/data_test.go

@@ -22,6 +22,7 @@ package node_stats
 import (
 	"testing"
 
+	"github.com/elastic/beats/v7/libbeat/management/status"
 	"github.com/elastic/beats/v7/metricbeat/mb"
 	"github.com/elastic/beats/v7/metricbeat/module/elasticsearch"
 )
@@ -60,3 +61,6 @@ func (m mockModule) Config() mb.ModuleConfig {
 func (m mockModule) UnpackConfig(to interface{}) error {
 	return nil
 }
+
+func (m mockModule) UpdateStatus(_ status.Status, _ string)    {}
+func (m mockModule) SetStatusReporter(_ status.StatusReporter) {}