fix(admin): list all masters and dedupe EC file counts in dashboard (#9093)

* fix(admin): list all masters and dedupe EC file counts in dashboard

Dashboard -> Master Nodes only ever showed the currently connected master
because getMasterNodesStatus hard-coded a single entry. Replace it with a
RaftListClusterServers call that returns every master in the raft group and
tags the real leader, falling back to the current master only if the raft
call fails.

Buckets -> Object Store Buckets could render 0 objects for a bucket backed
by an EC volume. Every shard holder reports the same whole-volume
file_count (read from the replicated .ecx), so the first-seen value wins;
if that first node had not yet finished loading .ecx it reported 0 and
pinned the aggregate at 0. Take the max across reporting nodes instead.

The dashboard header total_files also dropped after volumes were converted
to erasure coding because getTopologyViaGRPC never folded EC file_count
into topology.TotalFiles. Aggregate it with the same max/sum dedupe.

* fix(admin): address PR review comments

- bound RaftListClusterServers with a 3s timeout so the dashboard endpoint
  cannot hang on a stalled master
- pre-validate raft addresses with net.SplitHostPort before calling
  pb.GrpcAddressToServerAddress, which otherwise glog.Fatalf's on a
  malformed entry and would crash the admin process
- when raft is unreachable, mark the fallback master as not-leader rather
  than claiming leadership the code cannot verify
- warn when summed EC delete_count exceeds file_count while folding into
  topology.TotalFiles, matching collectCollectionStats

* fix(admin): distinguish empty raft response from RPC failure

When RaftListClusterServers returns successfully with no servers, raft is
not initialized (standalone/non-raft cluster), so the single fallback
master is the leader. Only treat the fallback as a non-leader when the
RPC actually failed.

* fix(admin): remove misleading Objects column from S3 buckets page

The bucket "Objects" column displayed needle counts from volume
collection stats, not actual S3 object counts. This is confusing
because a single S3 object can span multiple needles (multipart
uploads, versions) and the count is inaccurate for EC volumes.

Remove the ObjectCount field from S3Bucket, the Objects table column,
the sort-by-objects handler, the detail-view row, and both CSV export
references.

* fix(admin): correct cell indexes in fallback bucket CSV export

After the Objects column was removed, the fallback CSV exporter in
admin.js still used stale cell indexes: cells[1] mapped to Owner
(not Created), cells[2] to Created (not Size), cells[3] to Logical
Size (not Quota). Align all indexes with the current table column
order and include Owner, Logical Size, and Physical Size.

weed/admin/dash/admin_server.go

@@ -543,13 +543,6 @@ func (s *AdminServer) sortBuckets(buckets []S3Bucket, sortBy, sortOrder string)
 				}
 				return a.CreatedAt.Before(b.CreatedAt)
 			}
-		case "objects":
-			if a.ObjectCount != b.ObjectCount {
-				if desc {
-					return a.ObjectCount > b.ObjectCount
-				}
-				return a.ObjectCount < b.ObjectCount
-			}
 		case "logical_size":
 			if a.LogicalSize != b.LogicalSize {
 				if desc {
@@ -646,14 +639,11 @@ func (s *AdminServer) GetS3Buckets() ([]S3Bucket, error) {
 				// Determine collection name for this bucket
 				collectionName := getCollectionName(filerConfig.FilerGroup, bucketName)
 
-				// Get size and object count from collection data
 				var physicalSize int64
 				var logicalSize int64
-				var objectCount int64
 				if collectionData, exists := collectionMap[collectionName]; exists {
 					physicalSize = collectionData.PhysicalSize
 					logicalSize = collectionData.LogicalSize
-					objectCount = collectionData.FileCount
 				}
 
 				// Get quota information from entry
@@ -695,7 +685,6 @@ func (s *AdminServer) GetS3Buckets() ([]S3Bucket, error) {
 					CreatedAt:          createdAt,
 					LogicalSize:        logicalSize,
 					PhysicalSize:       physicalSize,
-					ObjectCount:        objectCount,
 					LastModified:       lastModified,
 					Quota:              quota,
 					QuotaEnabled:       quotaEnabled,
@@ -750,7 +739,6 @@ func (s *AdminServer) GetBucketDetails(bucketName string) (*BucketDetails, error
 	} else if data, ok := stats[collectionName]; ok {
 		details.Bucket.LogicalSize = data.LogicalSize
 		details.Bucket.PhysicalSize = data.PhysicalSize
-		details.Bucket.ObjectCount = data.FileCount
 	}
 
 	err = s.WithFilerClient(func(client filer_pb.SeaweedFilerClient) error {
@@ -1672,11 +1660,13 @@ type collectionStats struct {
 	FileCount    int64
 }
 
-// ecVolumeCounts is used to correctly combine EC volume counts reported by
-// multiple nodes. Every node holding any shard of an EC volume reports the
-// same file_count (total entries in the replicated .ecx), so we dedupe it
-// per volume id. In contrast, a needle delete is applied on exactly one
-// shard holder, so each node reports its own local tombstone count and the
+// ecVolumeCounts combines EC volume counts reported by multiple nodes.
+// Every node holding any shard of an EC volume reports the same file_count
+// (total entries in the replicated .ecx), so we dedupe it per volume id by
+// taking the max — a node that has not yet finished loading .ecx would
+// otherwise pin the aggregate at 0 and zero out the bucket object count.
+// In contrast, a needle delete is recorded locally on the shard holder
+// that served it, so each node reports its own tombstone count and the
 // true delete total is the sum across nodes.
 type ecVolumeCounts struct {
 	collection  string
@@ -1723,11 +1713,19 @@ func collectCollectionStats(topologyInfo *master_pb.TopologyInfo) map[string]col
 						data.LogicalSize += int64(shards.MinusParityShards().TotalSize())
 						collectionMap[collection] = data
 
+						// fileCount is volume-wide (same .ecx on every shard
+						// holder) so take the max to dedupe — a node that has
+						// not yet finished loading .ecx reports 0 and must not
+						// pin the aggregate. deleteCount is node-local and is
+						// summed across shard holders.
 						agg, ok := ecVolumeAgg[ecShardInfo.Id]
 						if !ok {
-							agg = &ecVolumeCounts{collection: collection, fileCount: ecShardInfo.FileCount}
+							agg = &ecVolumeCounts{collection: collection}
 							ecVolumeAgg[ecShardInfo.Id] = agg
 						}
+						if ecShardInfo.FileCount > agg.fileCount {
+							agg.fileCount = ecShardInfo.FileCount
+						}
 						agg.deleteCount += ecShardInfo.DeleteCount
 					}
 				}
@@ -1736,16 +1734,13 @@ func collectCollectionStats(topologyInfo *master_pb.TopologyInfo) map[string]col
 	}
 
 	// Fold EC per-volume counts into the collection totals. fileCount is
-	// already deduped (set once per volume), deleteCount is the sum of
-	// local tombstones across every node holding shards of the volume.
+	// deduped via max across every node reporting shards for the volume;
+	// deleteCount is summed across the same nodes.
 	for vid, agg := range ecVolumeAgg {
 		data := collectionMap[agg.collection]
 		if agg.fileCount >= agg.deleteCount {
 			data.FileCount += int64(agg.fileCount - agg.deleteCount)
 		} else {
-			// Should not happen in steady state — indicates a node reporting
-			// a stale fileCount, a skewed heartbeat, or a delete-counter bug.
-			// Defend the UI by skipping the add and surface the anomaly.
 			glog.Warningf("ec volume %d in collection %q: summed delete_count=%d exceeds file_count=%d; skipping object count",
 				vid, agg.collection, agg.deleteCount, agg.fileCount)
 		}