[feature] Add more default securityContext

controllers/factory/builders.go

@@ -528,13 +528,21 @@ func addStrictSecuritySettingsToPod(p *v1.PodSecurityContext, enableStrictSecuri
 	if !enableStrictSecurity || p != nil {
 		return p
 	}
-	return &v1.PodSecurityContext{
+	securityContext := v1.PodSecurityContext{
 		RunAsNonRoot: pointer.Bool(true),
 		// '65534' refers to 'nobody' in all the used default images like alpine, busybox
 		RunAsUser:  pointer.Int64(65534),
 		RunAsGroup: pointer.Int64(65534),
 		FSGroup:    pointer.Int64(65534),
+		SeccompProfile: &v1.SeccompProfile{
+			Type: v1.SeccompProfileTypeRuntimeDefault,
+		},
+	}
+	if k8stools.IsFSGroupChangePolicySupported() {
+		onRootMismatch := v1.FSGroupChangeOnRootMismatch
+		securityContext.FSGroupChangePolicy = &onRootMismatch
 	}
+	return &securityContext
 }
 
 func addStrictSecuritySettingsToContainers(containers []v1.Container, enableStrictSecurity bool) []v1.Container {
@@ -547,6 +555,11 @@ func addStrictSecuritySettingsToContainers(containers []v1.Container, enableStri
 			container.SecurityContext = &v1.SecurityContext{
 				ReadOnlyRootFilesystem:   pointer.Bool(true),
 				AllowPrivilegeEscalation: pointer.Bool(false),
+				Capabilities: &v1.Capabilities{
+					Drop: []v1.Capability{
+						"ALL",
+					},
+				},
 			}
 		}
 	}

controllers/factory/builders_test.go

@@ -14,6 +14,7 @@ import (
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 	"k8s.io/apimachinery/pkg/runtime"
 	"k8s.io/apimachinery/pkg/types"
+	"k8s.io/apimachinery/pkg/version"
 	"k8s.io/utils/pointer"
 )
 
@@ -638,6 +639,7 @@ func TestAddStrictSecuritySettingsToPod(t *testing.T) {
 		podSecurityPolicy    *v1.PodSecurityContext
 		enableStrictSecurity bool
 		exp                  *v1.PodSecurityContext
+		kubeletVersion       version.Info
 	}
 	tests := []struct {
 		name     string
@@ -649,11 +651,13 @@ func TestAddStrictSecuritySettingsToPod(t *testing.T) {
 			args: args{
 				enableStrictSecurity: true,
 				exp: &v1.PodSecurityContext{
-					RunAsNonRoot: pointer.Bool(true),
-					RunAsUser:    pointer.Int64(65534),
-					RunAsGroup:   pointer.Int64(65534),
-					FSGroup:      pointer.Int64(65534),
+					RunAsNonRoot:        pointer.Bool(true),
+					RunAsUser:           pointer.Int64(65534),
+					RunAsGroup:          pointer.Int64(65534),
+					FSGroup:             pointer.Int64(65534),
+					FSGroupChangePolicy: (*v1.PodFSGroupChangePolicy)(pointer.StringPtr("OnRootMismatch")),
 				},
+				kubeletVersion: version.Info{Major: "1", Minor: "27"},
 			},
 		},
 		{
@@ -677,6 +681,7 @@ func TestAddStrictSecuritySettingsToPod(t *testing.T) {
 		},
 	}
 	for _, tt := range tests {
+		k8stools.SetKubernetesVersionWithDefaults(&tt.args.kubeletVersion, 0, 0)
 		res := addStrictSecuritySettingsToPod(tt.args.podSecurityPolicy, tt.args.enableStrictSecurity)
 		if diff := deep.Equal(res, tt.args.exp); len(diff) > 0 {
 			t.Fatalf("got unexpected result: %v, expect: %v", res, tt.args.exp)

controllers/factory/k8stools/version.go

@@ -69,6 +69,16 @@ func IsHPAV2BetaSupported() bool {
 	return false
 }
 
+// IsFSGroupChangePolicySupported checks if `fsGroupChangePolicy` is supported,
+// Supported since 1.20
+// https://kubernetes.io/blog/2020/12/14/kubernetes-release-1.20-fsgroupchangepolicy-fsgrouppolicy/#allow-users-to-skip-recursive-permission-changes-on-mount
+func IsFSGroupChangePolicySupported() bool {
+	if ServerMajorVersion == 1 && ServerMinorVersion >= 20 {
+		return true
+	}
+	return false
+}
+
 // NewHPAEmptyObject returns HorizontalPodAutoscaler object for given kubernetes version
 func NewHPAEmptyObject(opts ...func(obj client.Object)) client.Object {
 	var hpa client.Object = &v2beta2.HorizontalPodAutoscaler{}

docs/CHANGELOG.MD

@@ -34,7 +34,8 @@
 - [vmcluster](https://docs.victoriametrics.com/operator/api.html#vmagent): add [example config](https://github.com/VictoriaMetrics/operator/blob/master/config/examples/vmcluster_with_additional_claim.yaml) for cluster with custom storage claims.
 - [vmrule](https://docs.victoriametrics.com/operator/api.html#vmrule): support `update_entries_limit` field in rules, refer to [alerting rules](https://docs.victoriametrics.com/vmalert.html#alerting-rules). See [this PR](https://github.com/VictoriaMetrics/operator/pull/691) for details.
 - [vmrule](https://docs.victoriametrics.com/operator/api.html#vmrule): support `keep_firing_for` field in rules, refer to [alerting rules](https://docs.victoriametrics.com/vmalert.html#alerting-rules). See [this PR](https://github.com/VictoriaMetrics/operator/pull/711) for details.
-- [vmoperator parameters](https://docs.victoriametrics.com/operator/vars.html): Add option `VM_ENABLESTRICTSECURITY` and enable strict security context by default. See [this issue](https://github.com/VictoriaMetrics/operator/issues/637) and [this PR](https://github.com/VictoriaMetrics/operator/pull/692/) for details.
+- [vmoperator parameters](https://docs.victoriametrics.com/operator/vars.html): Add option `VM_ENABLESTRICTSECURITY` and enable strict security context by default. See [this issue](https://github.com/VictoriaMetrics/operator/issues/637), [this](https://github.com/VictoriaMetrics/operator/pull/692/) and [this](https://github.com/VictoriaMetrics/operator/pull/712) PR for details.
+
 
 <a name="v0.35.1"></a>
 ## [v0.35.1](https://github.com/VictoriaMetrics/operator/releases/tag/v0.35.1) - 12 Jul 2023

internal/config/config.go

@@ -254,9 +254,20 @@ type BaseOperatorConf struct {
 	// 2. RunAsUser/RunAsGroup/FSGroup: 65534
 	// '65534' refers to 'nobody' in all the used default images like alpine, busybox.
 	// If you're using customize image, please make sure '65534' is a valid uid in there or specify SecurityContext.
+	// 3. FSGroupChangePolicy: &onRootMismatch
+	// If KubeVersion>=1.20, use `FSGroupChangePolicy="onRootMismatch"` to skip the recursive permission change
+	// when the root of the volume already has the correct permissions
+	// 4. SeccompProfile:
+	//      type: RuntimeDefault
+	// Use `RuntimeDefault` seccomp profile by default, which is defined by the container runtime,
+	// instead of using the Unconfined (seccomp disabled) mode.
+	//
 	// Default container SecurityContext include:
 	// 1. AllowPrivilegeEscalation: false
 	// 2. ReadOnlyRootFilesystem: true
+	// 3. Capabilities:
+	//      drop:
+	//        - all
 	EnableStrictSecurity bool `default:"true"`
 }