feat(server): Add maxPodLogsToRender setting (argoproj#14617)

Signed-off-by: lukasz <lukasz.luszczynski@getyourguide.com>
Co-authored-by: Ishita Sequeira <46771830+ishitasequeira@users.noreply.github.com>

docs/operator-manual/argocd-cm.yaml

@@ -320,6 +320,10 @@ data:
   # cluster.inClusterEnabled indicates whether to allow in-cluster server address. This is enabled by default.
   cluster.inClusterEnabled: "true"
 
+  # The maximum number of pod logs to render in UI. If the application has more than this number of pods, the logs will not be rendered.
+  # This is to prevent the UI from becoming unresponsive when rendering a large number of logs. Default is 10.
+  server.maxPodLogsToRender: 10
+
   # Application pod logs RBAC enforcement enables control over who can and who can't view application pod logs.
   # When you enable the switch, pod logs will be visible only to admin role by default. Other roles/users will not be able to view them via cli and UI.
   # When you enable the switch, viewing pod logs for other roles/users will require explicit RBAC allow policies (allow get on logs subresource).

server/application/application.go

@@ -65,7 +65,6 @@ import (
 type AppResourceTreeFn func(ctx context.Context, app *appv1.Application) (*appv1.ApplicationTree, error)
 
 const (
-	maxPodLogsToRender                 = 10
 	backgroundPropagationPolicy string = "background"
 	foregroundPropagationPolicy string = "foreground"
 )
@@ -1579,8 +1578,13 @@ func (s *Server) PodLogs(q *application.ApplicationPodLogsQuery, ws application.
 		return nil
 	}
 
-	if len(pods) > maxPodLogsToRender {
-		return errors.New("Max pods to view logs are reached. Please provide more granular query.")
+	maxPodLogsToRender, err := s.settingsMgr.GetMaxPodLogsToRender()
+	if err != nil {
+		return fmt.Errorf("error getting MaxPodLogsToRender config: %w", err)
+	}
+
+	if int64(len(pods)) > maxPodLogsToRender {
+		return status.Error(codes.InvalidArgument, "max pods to view logs are reached. Please provide more granular query")
 	}
 
 	var streams []chan logEntry

server/application/application_test.go

@@ -132,10 +132,10 @@ func newTestAppServer(t *testing.T, objects ...runtime.Object) *Server {
 		_ = enf.SetBuiltinPolicy(assets.BuiltinPolicyCSV)
 		enf.SetDefaultRole("role:admin")
 	}
-	return newTestAppServerWithEnforcerConfigure(f, t, objects...)
+	return newTestAppServerWithEnforcerConfigure(f, t, map[string]string{}, objects...)
 }
 
-func newTestAppServerWithEnforcerConfigure(f func(*rbac.Enforcer), t *testing.T, objects ...runtime.Object) *Server {
+func newTestAppServerWithEnforcerConfigure(f func(*rbac.Enforcer), t *testing.T, additionalConfig map[string]string, objects ...runtime.Object) *Server {
 	kubeclientset := fake.NewSimpleClientset(&v1.ConfigMap{
 		ObjectMeta: metav1.ObjectMeta{
 			Namespace: testNamespace,
@@ -144,6 +144,7 @@ func newTestAppServerWithEnforcerConfigure(f func(*rbac.Enforcer), t *testing.T,
 				"app.kubernetes.io/part-of": "argocd",
 			},
 		},
+		Data: additionalConfig,
 	}, &v1.Secret{
 		ObjectMeta: metav1.ObjectMeta{
 			Name:      "argocd-secret",
@@ -752,7 +753,7 @@ func TestNoAppEnumeration(t *testing.T) {
 		}
 	})
 	testDeployment := kube.MustToUnstructured(&deployment)
-	appServer := newTestAppServerWithEnforcerConfigure(f, t, testApp, testHelmApp, testDeployment)
+	appServer := newTestAppServerWithEnforcerConfigure(f, t, map[string]string{}, testApp, testHelmApp, testDeployment)
 
 	noRoleCtx := context.Background()
 	// nolint:staticcheck
@@ -1272,7 +1273,7 @@ g, group-49, role:test3
 `
 		_ = enf.SetUserPolicy(policy)
 	}
-	appServer := newTestAppServerWithEnforcerConfigure(f, t, objects...)
+	appServer := newTestAppServerWithEnforcerConfigure(f, t, map[string]string{}, objects...)
 
 	res, err := appServer.List(ctx, &application.ApplicationQuery{})
 
@@ -1987,6 +1988,108 @@ func TestLogsGetSelectedPod(t *testing.T) {
 	})
 }
 
+func TestMaxPodLogsRender(t *testing.T) {
+
+	defaultMaxPodLogsToRender, _ := newTestAppServer(t).settingsMgr.GetMaxPodLogsToRender()
+
+	// Case: number of pods to view logs is less than defaultMaxPodLogsToRender
+	podNumber := int(defaultMaxPodLogsToRender - 1)
+	appServer, adminCtx := createAppServerWithMaxLodLogs(t, podNumber)
+
+	t.Run("PodLogs", func(t *testing.T) {
+		err := appServer.PodLogs(&application.ApplicationPodLogsQuery{Name: pointer.String("test")}, &TestPodLogsServer{ctx: adminCtx})
+		statusCode, _ := status.FromError(err)
+		assert.Equal(t, codes.OK, statusCode.Code())
+	})
+
+	// Case: number of pods higher than defaultMaxPodLogsToRender
+	podNumber = int(defaultMaxPodLogsToRender + 1)
+	appServer, adminCtx = createAppServerWithMaxLodLogs(t, podNumber)
+
+	t.Run("PodLogs", func(t *testing.T) {
+		err := appServer.PodLogs(&application.ApplicationPodLogsQuery{Name: pointer.String("test")}, &TestPodLogsServer{ctx: adminCtx})
+		assert.NotNil(t, err)
+		statusCode, _ := status.FromError(err)
+		assert.Equal(t, codes.InvalidArgument, statusCode.Code())
+		assert.Equal(t, "rpc error: code = InvalidArgument desc = max pods to view logs are reached. Please provide more granular query", err.Error())
+	})
+
+	// Case: number of pods to view logs is less than customMaxPodLogsToRender
+	customMaxPodLogsToRender := int64(15)
+	podNumber = int(customMaxPodLogsToRender - 1)
+	appServer, adminCtx = createAppServerWithMaxLodLogs(t, podNumber, customMaxPodLogsToRender)
+
+	t.Run("PodLogs", func(t *testing.T) {
+		err := appServer.PodLogs(&application.ApplicationPodLogsQuery{Name: pointer.String("test")}, &TestPodLogsServer{ctx: adminCtx})
+		statusCode, _ := status.FromError(err)
+		assert.Equal(t, codes.OK, statusCode.Code())
+	})
+
+	// Case: number of pods higher than customMaxPodLogsToRender
+	customMaxPodLogsToRender = int64(15)
+	podNumber = int(customMaxPodLogsToRender + 1)
+	appServer, adminCtx = createAppServerWithMaxLodLogs(t, podNumber, customMaxPodLogsToRender)
+
+	t.Run("PodLogs", func(t *testing.T) {
+		err := appServer.PodLogs(&application.ApplicationPodLogsQuery{Name: pointer.String("test")}, &TestPodLogsServer{ctx: adminCtx})
+		assert.NotNil(t, err)
+		statusCode, _ := status.FromError(err)
+		assert.Equal(t, codes.InvalidArgument, statusCode.Code())
+		assert.Equal(t, "rpc error: code = InvalidArgument desc = max pods to view logs are reached. Please provide more granular query", err.Error())
+	})
+}
+
+// createAppServerWithMaxLodLogs creates a new app server with given number of pods and resources
+func createAppServerWithMaxLodLogs(t *testing.T, podNumber int, maxPodLogsToRender ...int64) (*Server, context.Context) {
+	runtimeObjects := make([]runtime.Object, podNumber+1)
+	resources := make([]appsv1.ResourceStatus, podNumber)
+
+	for i := 0; i < podNumber; i++ {
+		pod := v1.Pod{
+			TypeMeta: metav1.TypeMeta{
+				APIVersion: "v1",
+				Kind:       "Pod",
+			},
+			ObjectMeta: metav1.ObjectMeta{
+				Name:      fmt.Sprintf("pod-%d", i),
+				Namespace: "test",
+			},
+		}
+		resources[i] = appsv1.ResourceStatus{
+			Group:     pod.GroupVersionKind().Group,
+			Kind:      pod.GroupVersionKind().Kind,
+			Version:   pod.GroupVersionKind().Version,
+			Name:      pod.Name,
+			Namespace: pod.Namespace,
+			Status:    "Synced",
+		}
+		runtimeObjects[i] = kube.MustToUnstructured(&pod)
+	}
+
+	testApp := newTestApp(func(app *appsv1.Application) {
+		app.Name = "test"
+		app.Status.Resources = resources
+	})
+	runtimeObjects[podNumber] = testApp
+
+	noRoleCtx := context.Background()
+	// nolint:staticcheck
+	adminCtx := context.WithValue(noRoleCtx, "claims", &jwt.MapClaims{"groups": []string{"admin"}})
+
+	if len(maxPodLogsToRender) > 0 {
+		f := func(enf *rbac.Enforcer) {
+			_ = enf.SetBuiltinPolicy(assets.BuiltinPolicyCSV)
+			enf.SetDefaultRole("role:admin")
+		}
+		formatInt := strconv.FormatInt(maxPodLogsToRender[0], 10)
+		appServer := newTestAppServerWithEnforcerConfigure(f, t, map[string]string{"server.maxPodLogsToRender": formatInt}, runtimeObjects...)
+		return appServer, adminCtx
+	} else {
+		appServer := newTestAppServer(t, runtimeObjects...)
+		return appServer, adminCtx
+	}
+}
+
 // refreshAnnotationRemover runs an infinite loop until it detects and removes refresh annotation or given context is done
 func refreshAnnotationRemover(t *testing.T, ctx context.Context, patched *int32, appServer *Server, appName string, ch chan string) {
 	for ctx.Err() == nil {

util/settings/settings.go

@@ -103,6 +103,8 @@ type ArgoCDSettings struct {
 	InClusterEnabled bool `json:"inClusterEnabled"`
 	// ServerRBACLogEnforceEnable temporary var indicates whether rbac will be enforced on logs
 	ServerRBACLogEnforceEnable bool `json:"serverRBACLogEnforceEnable"`
+	// MaxPodLogsToRender the maximum number of pod logs to render
+	MaxPodLogsToRender int64 `json:"maxPodLogsToRender"`
 	// ExecEnabled indicates whether the UI exec feature is enabled
 	ExecEnabled bool `json:"execEnabled"`
 	// ExecShells restricts which shells are allowed for `exec` and in which order they are tried
@@ -485,6 +487,8 @@ const (
 	inClusterEnabledKey = "cluster.inClusterEnabled"
 	// settingsServerRBACLogEnforceEnable is the key to configure whether logs RBAC enforcement is enabled
 	settingsServerRBACLogEnforceEnableKey = "server.rbac.log.enforce.enable"
+	// MaxPodLogsToRender the maximum number of pod logs to render
+	settingsMaxPodLogsToRender = "server.maxPodLogsToRender"
 	// helmValuesFileSchemesKey is the key to configure the list of supported helm values file schemas
 	helmValuesFileSchemesKey = "helm.valuesFileSchemes"
 	// execEnabledKey is the key to configure whether the UI exec feature is enabled
@@ -788,6 +792,19 @@ func (mgr *SettingsManager) GetServerRBACLogEnforceEnable() (bool, error) {
 	return strconv.ParseBool(argoCDCM.Data[settingsServerRBACLogEnforceEnableKey])
 }
 
+func (mgr *SettingsManager) GetMaxPodLogsToRender() (int64, error) {
+	argoCDCM, err := mgr.getConfigMap()
+	if err != nil {
+		return 10, err
+	}
+
+	if argoCDCM.Data[settingsMaxPodLogsToRender] == "" {
+		return 10, nil
+	}
+
+	return strconv.ParseInt(argoCDCM.Data[settingsMaxPodLogsToRender], 10, 64)
+}
+
 func (mgr *SettingsManager) GetDeepLinks(deeplinkType string) ([]DeepLink, error) {
 	argoCDCM, err := mgr.getConfigMap()
 	if err != nil {
@@ -1457,6 +1474,13 @@ func updateSettingsFromConfigMap(settings *ArgoCDSettings, argoCDCM *apiv1.Confi
 	if settings.PasswordPattern == "" {
 		settings.PasswordPattern = common.PasswordPatten
 	}
+	if maxPodLogsToRenderStr, ok := argoCDCM.Data[settingsMaxPodLogsToRender]; ok {
+		if val, err := strconv.ParseInt(maxPodLogsToRenderStr, 10, 64); err != nil {
+			log.Warnf("Failed to parse '%s' key: %v", settingsMaxPodLogsToRender, err)
+		} else {
+			settings.MaxPodLogsToRender = val
+		}
+	}
 	settings.InClusterEnabled = argoCDCM.Data[inClusterEnabledKey] != "false"
 	settings.ExecEnabled = argoCDCM.Data[execEnabledKey] == "true"
 	execShells := argoCDCM.Data[execShellsKey]