Add pprof endpoints to the monitoring server if enabled (#5562)

The documentation claims this should work, but it currently doesn't.

changelog/fragments/1726740792-pprof-endpoint.yaml

@@ -0,0 +1,32 @@
+# Kind can be one of:
+# - breaking-change: a change to previously-documented behavior
+# - deprecation: functionality that is being removed in a later release
+# - bug-fix: fixes a problem in a previous version
+# - enhancement: extends functionality but does not break or fix existing behavior
+# - feature: new functionality
+# - known-issue: problems that we are aware of in a given version
+# - security: impacts on the security of a product or a user’s deployment.
+# - upgrade: important information for someone upgrading from a prior version
+# - other: does not fit into any of the other categories
+kind: bug-fix
+
+# Change summary; a 80ish characters long description of the change.
+summary: Add pprof endpoints to the monitoring server if enabled
+
+# Long description; in case the summary is not enough to describe the change
+# this field accommodate a description without length limits.
+# NOTE: This field will be rendered only for breaking-change and known-issue kinds at the moment.
+#description:
+
+# Affected component; usually one of "elastic-agent", "fleet-server", "filebeat", "metricbeat", "auditbeat", "all", etc.
+component: elastic-agent
+
+# PR URL; optional; the PR number that added the changeset.
+# If not present is automatically filled by the tooling finding the PR where this changelog fragment has been added.
+# NOTE: the tooling supports backports, so it's able to fill the original PR number instead of the backport PR number.
+# Please provide it if you are adding a fragment for a different PR.
+#pr: https://github.com/owner/repo/1234
+
+# Issue URL; optional; the GitHub issue related to this changeset (either closes or is part of).
+# If not present is automatically filled by the tooling with the issue linked to the PR number.
+#issue: https://github.com/owner/repo/1234

internal/pkg/agent/application/monitoring/server.go

@@ -6,6 +6,7 @@ package monitoring
 
 import (
 	"net/http"
+	_ "net/http/pprof" //nolint:gosec // this is only conditionally exposed
 	"net/url"
 	"os"
 	"path/filepath"
@@ -76,6 +77,11 @@ func exposeMetricsEndpoint(
 			r.Handle("/liveness", createHandler(livenessHandler(coord)))
 		}
 
+		if isPprofEnabled(cfg) {
+			// importing net/http/pprof adds the handlers to the right paths on the default Mux, so we just defer to it here
+			r.PathPrefix("/debug/pprof/").Handler(http.DefaultServeMux)
+		}
+
 		mux := http.NewServeMux()
 		mux.Handle("/", r)
 
@@ -131,3 +137,7 @@ func isHttpUrl(s string) bool {
 func isProcessStatsEnabled(cfg *monitoringCfg.MonitoringConfig) bool {
 	return cfg != nil && cfg.HTTP.Enabled
 }
+
+func isPprofEnabled(cfg *monitoringCfg.MonitoringConfig) bool {
+	return cfg != nil && cfg.Pprof != nil && cfg.Pprof.Enabled
+}

internal/pkg/agent/application/monitoring/server_test.go

@@ -98,18 +98,18 @@ func TestHTTPReloadEnableBehavior(t *testing.T) {
 			t.Logf("starting server...")
 			serverReloader.Start()
 			if testCase.httpOnAtInit {
-				waitOnReturnCode(t, http.StatusOK, "?failon=failed", serverReloader)
+				waitOnReturnCode(t, http.StatusOK, "liveness", "?failon=failed", serverReloader)
 			} else {
-				waitOnReturnCode(t, http.StatusNotFound, "?failon=failed", serverReloader)
+				waitOnReturnCode(t, http.StatusNotFound, "liveness", "?failon=failed", serverReloader)
 			}
 
 			err = serverReloader.Reload(testCase.secondConfig)
 			require.NoError(t, err)
 
 			if testCase.httpOnAfterReload {
-				waitOnReturnCode(t, http.StatusOK, "?failon=failed", serverReloader)
+				waitOnReturnCode(t, http.StatusOK, "liveness", "?failon=failed", serverReloader)
 			} else {
-				waitOnReturnCode(t, http.StatusNotFound, "?failon=failed", serverReloader)
+				waitOnReturnCode(t, http.StatusNotFound, "liveness", "?failon=failed", serverReloader)
 			}
 
 		})
@@ -132,24 +132,51 @@ func TestBasicLivenessConfig(t *testing.T) {
 	t.Logf("starting server...")
 	serverReloader.Start()
 
-	waitOnReturnCode(t, http.StatusInternalServerError, "?failon=degraded", serverReloader)
+	waitOnReturnCode(t, http.StatusInternalServerError, "liveness", "?failon=degraded", serverReloader)
 
-	waitOnReturnCode(t, http.StatusOK, "?failon=failed", serverReloader)
+	waitOnReturnCode(t, http.StatusOK, "liveness", "?failon=failed", serverReloader)
 
 	t.Logf("stopping server...")
 	err = serverReloader.Stop()
 	require.NoError(t, err)
 
 }
 
-func waitOnReturnCode(t *testing.T, expectedReturnCode int, formValue string, rel *reload.ServerReloader) {
+func TestPprofEnabled(t *testing.T) {
+	_ = logp.DevelopmentSetup()
+	testAPIConfig := api.Config{}
+	testConfig := config.MonitoringConfig{
+		Enabled: true,
+		HTTP: &config.MonitoringHTTPConfig{
+			Enabled: true,
+			Port:    0,
+		},
+		Pprof: &config.PprofConfig{
+			Enabled: true,
+		},
+	}
+	serverReloader, err := NewServer(logp.L(), testAPIConfig, nil, nil, fakeCoordCfg, "linux", &testConfig)
+	require.NoError(t, err)
+
+	t.Logf("starting server...")
+	serverReloader.Start()
+
+	waitOnReturnCode(t, http.StatusOK, "debug/pprof/", "", serverReloader)
+
+	t.Logf("stopping server...")
+	err = serverReloader.Stop()
+	require.NoError(t, err)
+
+}
+
+func waitOnReturnCode(t *testing.T, expectedReturnCode int, path string, formValue string, rel *reload.ServerReloader) {
 	ctx, cancel := context.WithTimeout(context.Background(), time.Second*30)
 	defer cancel()
 	client := &http.Client{}
 	fastEventually(t, func() bool {
-		path := fmt.Sprintf("http://%s/liveness%s", rel.Addr().String(), formValue)
-		t.Logf("checking %s", path)
-		req, err := http.NewRequestWithContext(ctx, "GET", path, nil)
+		url := fmt.Sprintf("http://%s/%s%s", rel.Addr().String(), path, formValue)
+		t.Logf("checking %s", url)
+		req, err := http.NewRequestWithContext(ctx, "GET", url, nil)
 		require.NoError(t, err)
 
 		resp, err := client.Do(req)