From 00811d4a183294b1848b64de207bf06b979935f2 Mon Sep 17 00:00:00 2001
From: Chris Lu <chris.lu@gmail.com>
Date: Sun, 18 Oct 2026 20:37:58 -0700
Subject: [PATCH] shell: volume.check.duplicate_ids audit command (#11763)

---
 .../command_volume_check_duplicate_ids.go     | 101 ++++++++++++++++++
 1 file changed, 101 insertions(+)
 create mode 100644 weed/shell/command_volume_check_duplicate_ids.go

diff --git a/weed/shell/command_volume_check_duplicate_ids.go b/weed/shell/command_volume_check_duplicate_ids.go
new file mode 100644
index 0000000000..dd463630ef
--- /dev/null
+++ b/weed/shell/command_volume_check_duplicate_ids.go
@@ -0,0 +1,101 @@
+package shell
+
+import (
+	"fmt"
+	"io"
+	"slices"
+
+	"github.com/seaweedfs/seaweedfs/weed/storage/needle"
+	"github.com/seaweedfs/seaweedfs/weed/storage/super_block"
+)
+
+func init() {
+	Commands = append(Commands, &commandVolumeCheckDuplicateIds{})
+}
+
+type commandVolumeCheckDuplicateIds struct {
+}
+
+func (c *commandVolumeCheckDuplicateIds) Name() string {
+	return "volume.check.duplicate_ids"
+}
+
+func (c *commandVolumeCheckDuplicateIds) Help() string {
+	return `find volume ids registered under more than one layout
+
+	volume.check.duplicate_ids
+
+	Volume ids are global. The same id under two collections, replica
+	placements, TTLs, or disk types means the allocator reused an id, for
+	example after master state loss. See volume.growth.resume.
+
+`
+}
+
+func (c *commandVolumeCheckDuplicateIds) HasTag(CommandTag) bool {
+	return false
+}
+
+func (c *commandVolumeCheckDuplicateIds) Do(args []string, commandEnv *CommandEnv, writer io.Writer) (err error) {
+	topoInfo, _, err := collectTopologyInfo(commandEnv, 0)
+	if err != nil {
+		return err
+	}
+
+	// vid -> layout key -> locations
+	seen := make(map[needle.VolumeId]map[string][]string)
+	for _, dc := range topoInfo.DataCenterInfos {
+		for _, r := range dc.RackInfos {
+			for _, dn := range r.DataNodeInfos {
+				for diskKey, di := range dn.DiskInfos {
+					for _, vi := range di.VolumeInfos {
+						rp, _ := super_block.NewReplicaPlacementFromByte(byte(vi.ReplicaPlacement))
+						key := fmt.Sprintf("collection=%q replication=%s ttl=%s disk=%s",
+							vi.Collection, rp.String(), needle.LoadTTLFromUint32(vi.Ttl), vi.DiskType)
+						layouts := seen[needle.VolumeId(vi.Id)]
+						if layouts == nil {
+							layouts = make(map[string][]string)
+							seen[needle.VolumeId(vi.Id)] = layouts
+						}
+						layouts[key] = append(layouts[key], fmt.Sprintf("%s %s", dn.Id, diskKey))
+					}
+				}
+			}
+		}
+	}
+
+	duplicates := 0
+	for _, vid := range sortedKeys(seen) {
+		layouts := seen[vid]
+		if len(layouts) < 2 {
+			continue
+		}
+		duplicates++
+		fmt.Fprintf(writer, "volume %d is registered under %d layouts:\n", vid, len(layouts))
+		for _, key := range sortedKeysString(layouts) {
+			fmt.Fprintf(writer, "  %-60s %v\n", key, layouts[key])
+		}
+	}
+	if duplicates == 0 {
+		fmt.Fprintf(writer, "no duplicate volume ids found\n")
+	}
+	return nil
+}
+
+func sortedKeys[V any](m map[needle.VolumeId]V) []needle.VolumeId {
+	keys := make([]needle.VolumeId, 0, len(m))
+	for k := range m {
+		keys = append(keys, k)
+	}
+	slices.Sort(keys)
+	return keys
+}
+
+func sortedKeysString[V any](m map[string]V) []string {
+	keys := make([]string, 0, len(m))
+	for k := range m {
+		keys = append(keys, k)
+	}
+	slices.Sort(keys)
+	return keys
+}