feat(platform): registry/uploads/backup PVC 容量可配置

2026-09-24 23:10:40 +08:00
1 parent 720ab81bf8
commit e75448a118
@@ -49,6 +49,9 @@ func cmdManifests(args []string, stdout, stderr io.Writer) int {
 	backupPVC := fs.String("backup-pvc", "felis-backups", "name of the world-archive PVC this bundle renders in the Minecraft namespace and advertises to the backup/restore executors via FELIS_BACKUP_PVC (default: felis-backups; pass an empty value to render none, leaving backup/restore answering 503)")
 	worldsHostPath := fs.String("worlds-host-path", "", "node directory the reaper reads worlds from: each world PVC resolves as <path>/<pvc>, or as the stock local-path directory <path>/<pv-name>_<ns>_<pvc-name> (k3s storage root: /var/lib/rancher/k3s/storage); enables the reaper CronJob (requires --archive-local-path and a non-empty --backup-pvc)")
 	archiveLocalPath := fs.String("archive-local-path", "", "path the backup PVC is mounted at in the reaper CronJob; MUST equal felis.toml [archive] local_path")
+	registryStorage := fs.String("registry-storage", "", "capacity the registry PVC requests (default 10Gi; k3s local-path does not enforce it)")
+	uploadsStorage := fs.String("uploads-storage", "", "capacity the uploads PVC requests (default 5Gi; k3s local-path does not enforce it)")
+	backupStorage := fs.String("backup-storage", "", "capacity the world-archive PVC requests (default 10Gi; k3s local-path does not enforce it)")
 	reaperNode := fs.String("reaper-node", "", "node that holds --worlds-host-path: pins the reaper CronJob's pod there via nodeSelector kubernetes.io/hostname (multi-node clusters need this, or the reaper may schedule where the hostPath is empty)")
 	var velocityCIDRs multiFlag
 	fs.Var(&velocityCIDRs, "velocity-cidr", "CIDR of a Velocity proxy host allowed to reach game port 25565 (repeatable, REQUIRED)")
@@ -139,7 +142,7 @@ func cmdManifests(args []string, stdout, stderr io.Writer) int {
 			"(pass --worlds-host-path and --archive-local-path — the archive PVC defaults to felis-backups — to enable it)")
 	}

-	out, err := platform.RenderYAML(platform.Params{
+	params := platform.Params{
 		ControlNamespace:   *controlNS,
 		MinecraftNamespace: *minecraftNS,
 		BuildNamespace:     *buildNS,
@@ -157,7 +160,16 @@ func cmdManifests(args []string, stdout, stderr io.Writer) int {

 		ServerEgressDenyCIDRs:  []string(serverDenyCIDRs),
 		ServerEgressAllowCIDRs: []string(serverAllowCIDRs),
-	})
+
+		RegistryStorage: *registryStorage,
+		UploadsStorage:  *uploadsStorage,
+		BackupStorage:   *backupStorage,
+	}
+	if err := params.Validate(); err != nil {
+		fmt.Fprintf(stderr, "felis manifests: %v\n", err)
+		return 2
+	}
+	out, err := platform.RenderYAML(params)
 	if err != nil {
 		fmt.Fprintf(stderr, "felis manifests: render: %v\n", err)
 		return 1
@@ -219,3 +219,27 @@ func TestManifestsReaperNodePin(t *testing.T) {
 		t.Errorf("--reaper-node without --worlds-host-path: exit = %d, want 2", code)
 	}
 }
+
+// TestManifestsStorageSizes proves the PVC size flags reach the rendered claims
+// and a size the API server would reject fails before anything is applied.
+func TestManifestsStorageSizes(t *testing.T) {
+	var out, errBuf bytes.Buffer
+	code := run([]string{"manifests", "--felis-image", "reg/felis:test", "--velocity-cidr", "10.0.0.5/32",
+		"--registry-storage", "40Gi", "--uploads-storage", "8Gi"}, &out, &errBuf)
+	if code != 0 {
+		t.Fatalf("exit code = %d, stderr = %s", code, errBuf.String())
+	}
+	for _, want := range []string{"storage: 40Gi", "storage: 8Gi"} {
+		if !strings.Contains(out.String(), want) {
+			t.Errorf("bundle lacks %q", want)
+		}
+	}
+
+	out.Reset()
+	errBuf.Reset()
+	code = run([]string{"manifests", "--felis-image", "reg/felis:test", "--velocity-cidr", "10.0.0.5/32",
+		"--registry-storage", "lots"}, &out, &errBuf)
+	if code == 0 || !strings.Contains(errBuf.String(), "registry storage") {
+		t.Errorf("--registry-storage lots: exit %d, stderr %q; want a refusal naming the flag", code, errBuf.String())
+	}
+}
@@ -77,6 +77,10 @@
 #                     worlds idle beyond the retention window, and grants the reaper's
 #                     uid (1000) traverse access to that root — k3s ships it 0700
 #                     root:root (default: unset = no reaper)
+#   FELIS_REGISTRY_STORAGE / FELIS_UPLOADS_STORAGE / FELIS_BACKUP_STORAGE capacity the
+#                     registry, uploads and world-archive PVCs request on first install
+#                     (defaults: 10Gi, 5Gi, 10Gi). An existing claim keeps its size; on
+#                     k3s local-path the number is not enforced, see troubleshooting §9
 #   PKG_LOCK_TIMEOUT seconds to wait for package-manager locks (default: 900)
 #   APT_LOCK_TIMEOUT legacy alias for PKG_LOCK_TIMEOUT
 set -Eeuo pipefail
@@ -134,6 +138,10 @@ FELIS_PANEL_NODEPORT="${FELIS_PANEL_NODEPORT:-30443}"
 # construction — a mismatch would leave tarLocal's absolute archive refs unresolvable.
 FELIS_BACKUP_PVC="${FELIS_BACKUP_PVC:-felis-backups}"
 FELIS_ARCHIVE_LOCAL_PATH="${FELIS_ARCHIVE_LOCAL_PATH:-/var/lib/felis/archives}"
+# PVC capacities; empty keeps `felis manifests`' defaults.
+FELIS_REGISTRY_STORAGE="${FELIS_REGISTRY_STORAGE:-}"
+FELIS_UPLOADS_STORAGE="${FELIS_UPLOADS_STORAGE:-}"
+FELIS_BACKUP_STORAGE="${FELIS_BACKUP_STORAGE:-}"
 # Retention is opt-in because it DELETES worlds (after a verified archive): point this at the
 # node directory the world volumes live under. On the k3s this installer provisions that is
 # /var/lib/rancher/k3s/storage — the reaper resolves each PVC's local-path directory exactly
@@ -2828,6 +2836,15 @@ deploy_bundle() {
    fi
    manifest_args+=(--worlds-host-path "$FELIS_WORLDS_HOST_PATH" --archive-local-path "$FELIS_ARCHIVE_LOCAL_PATH")
  fi
+  local size
+  size="$(pvc_size "$CONTROL_NS" registry "$FELIS_REGISTRY_STORAGE" FELIS_REGISTRY_STORAGE)"
+  if [ -n "$size" ]; then manifest_args+=(--registry-storage "$size"); fi
+  size="$(pvc_size "$CONTROL_NS" felis-uploads "$FELIS_UPLOADS_STORAGE" FELIS_UPLOADS_STORAGE)"
+  if [ -n "$size" ]; then manifest_args+=(--uploads-storage "$size"); fi
+  if [ -n "$FELIS_BACKUP_PVC" ]; then
+    size="$(pvc_size "$MINECRAFT_NS" "$FELIS_BACKUP_PVC" "$FELIS_BACKUP_STORAGE" FELIS_BACKUP_STORAGE)"
+    if [ -n "$size" ]; then manifest_args+=(--backup-storage "$size"); fi
+  fi
  "$HOST_BIN" manifests "${manifest_args[@]}" | kube apply -f -
  restart_existing_control_plane "$had_api" "$had_operator"

@@ -2841,6 +2858,24 @@ deploy_bundle() {
  done
 }

+# pvc_size <namespace> <claim> <wanted> <env name> prints the size to render the claim
+# with: its current request when it exists, else the wanted size (empty = the renderer's
+# default). A claim's request can only grow, and only on a storage class that allows
+# expansion (k3s local-path does not), so re-applying a different size would fail the
+# whole apply; a mismatch is reported and left to the operator.
+pvc_size() {
+  local ns="$1" claim="$2" want="$3" env="$4" have
+  have="$(kube -n "$ns" get pvc "$claim" -o jsonpath='{.spec.resources.requests.storage}' 2>/dev/null || true)"
+  if [ -z "$have" ]; then
+    printf '%s' "$want"
+    return 0
+  fi
+  if [ -n "$want" ] && [ "$want" != "$have" ]; then
+    warn "PVC ${ns}/${claim} already requests ${have}; keeping it (${env}=${want} applies to a new claim; grow this one with kubectl patch where its storage class allows expansion)"
+  fi
+  printf '%s' "$have"
+}
+
 restart_existing_control_plane() {
  local had_api="$1" had_operator="$2"
  [ "$had_api$had_operator" != "00" ] || return 0
@@ -1,5 +1,11 @@
 package platform

+import (
+	"fmt"
+
+	"k8s.io/apimachinery/pkg/api/resource"
+)
+
 // Identity constants and the Params that parameterise the install bundle.
 //
 // The label and SA-name constants are PINNED here because three independent
@@ -171,6 +177,35 @@ type Params struct {
 	// must-match. It is reaper-only and has no default; empty (with WorldsHostPath set)
 	// is rejected fail-loud by the generator.
 	ArchiveLocalPath string
+	// RegistryStorage, UploadsStorage and BackupStorage are the capacities the
+	// registry, uploads and world-archive PVCs request ("20Gi"); empty keeps
+	// 10Gi, 5Gi and 10Gi. A PVC's request can only grow, and only on a class that
+	// allows expansion, so the installer keeps an existing PVC's size. On k3s
+	// local-path the request is a label: the volume is a directory on the node's
+	// disk and nothing stops it outgrowing the number; the registry pruner, the
+	// uploads budget (user_uploads_max_bytes) and the archive cap
+	// (max_local_bytes) are what bound them there.
+	RegistryStorage string
+	UploadsStorage  string
+	BackupStorage   string
+}
+
+// Validate reports a Params the renderer cannot turn into objects.
+func (p Params) Validate() error {
+	for _, q := range []struct{ name, v string }{
+		{"registry storage", p.RegistryStorage},
+		{"uploads storage", p.UploadsStorage},
+		{"backup storage", p.BackupStorage},
+	} {
+		if q.v == "" {
+			continue
+		}
+		v, err := resource.ParseQuantity(q.v)
+		if err != nil || v.Sign() <= 0 {
+			return fmt.Errorf("%s %q is not a positive size such as 20Gi", q.name, q.v)
+		}
+	}
+	return nil
 }

 // withDefaults returns a copy of p with zero namespace/registry fields filled.
@@ -198,6 +233,15 @@ func (p Params) withDefaults() Params {
 	if p.RegistryImage == "" {
 		p.RegistryImage = defaultRegistryImage
 	}
+	if p.RegistryStorage == "" {
+		p.RegistryStorage = registryStorageSize
+	}
+	if p.UploadsStorage == "" {
+		p.UploadsStorage = uploadsStorageSize
+	}
+	if p.BackupStorage == "" {
+		p.BackupStorage = backupStorageSize
+	}
 	return p
 }

@@ -1102,7 +1102,7 @@ func registryPVC(p Params) *corev1.PersistentVolumeClaim {
 		Spec: corev1.PersistentVolumeClaimSpec{
 			AccessModes: []corev1.PersistentVolumeAccessMode{corev1.ReadWriteOnce},
 			Resources: corev1.VolumeResourceRequirements{
-				Requests: corev1.ResourceList{corev1.ResourceStorage: resource.MustParse(registryStorageSize)},
+				Requests: corev1.ResourceList{corev1.ResourceStorage: resource.MustParse(p.RegistryStorage)},
 			},
 		},
 	}
@@ -1123,7 +1123,7 @@ func uploadsPVC(p Params) *corev1.PersistentVolumeClaim {
 		Spec: corev1.PersistentVolumeClaimSpec{
 			AccessModes: []corev1.PersistentVolumeAccessMode{corev1.ReadWriteOnce},
 			Resources: corev1.VolumeResourceRequirements{
-				Requests: corev1.ResourceList{corev1.ResourceStorage: resource.MustParse(uploadsStorageSize)},
+				Requests: corev1.ResourceList{corev1.ResourceStorage: resource.MustParse(p.UploadsStorage)},
 			},
 		},
 	}
@@ -1147,7 +1147,7 @@ func backupPVC(p Params) *corev1.PersistentVolumeClaim {
 		Spec: corev1.PersistentVolumeClaimSpec{
 			AccessModes: []corev1.PersistentVolumeAccessMode{corev1.ReadWriteOnce},
 			Resources: corev1.VolumeResourceRequirements{
-				Requests: corev1.ResourceList{corev1.ResourceStorage: resource.MustParse(backupStorageSize)},
+				Requests: corev1.ResourceList{corev1.ResourceStorage: resource.MustParse(p.BackupStorage)},
 			},
 		},
 	}
@@ -1131,3 +1131,37 @@ func TestOperatorMetricsService(t *testing.T) {
 		t.Error("Workloads does not render the operator metrics Service")
 	}
 }
+
+// TestPVCSizes proves the three claim sizes follow Params and default when unset,
+// and that Validate refuses a size the API server would reject.
+func TestPVCSizes(t *testing.T) {
+	sizes := func(p Params) map[string]string {
+		got := map[string]string{}
+		for _, o := range Workloads(p.withDefaults()) {
+			if pvc, ok := o.(*corev1.PersistentVolumeClaim); ok {
+				got[pvc.Name] = pvc.Spec.Resources.Requests.Storage().String()
+			}
+		}
+		return got
+	}
+	p := testParams()
+	p.BackupPVC = "felis-backups"
+	def := sizes(p)
+	if def[registryName] != registryStorageSize || def[uploadsPVCName] != uploadsStorageSize || def["felis-backups"] != backupStorageSize {
+		t.Errorf("default sizes = %v", def)
+	}
+	p.RegistryStorage, p.UploadsStorage, p.BackupStorage = "40Gi", "8Gi", "100Gi"
+	if got := sizes(p); got[registryName] != "40Gi" || got[uploadsPVCName] != "8Gi" || got["felis-backups"] != "100Gi" {
+		t.Errorf("configured sizes = %v", got)
+	}
+	if err := p.Validate(); err != nil {
+		t.Errorf("Validate(%+v) = %v", p, err)
+	}
+	for _, bad := range []string{"lots", "0", "-1Gi"} {
+		q := testParams()
+		q.UploadsStorage = bad
+		if err := q.Validate(); err == nil {
+			t.Errorf("Validate accepted uploads storage %q", bad)
+		}
+	}
+}