Files
seaweedfs/weed/s3api/bucket_size_metrics_test.go
T
Chris Lu 7364f148bd fix(s3/shell): factor EC volumes into bucket size metrics and collection.list (#9182)
* fix(s3/shell): include EC volumes in bucket size metrics and collection.list

S3 bucket size metrics exported to Prometheus (and fed through
stats.UpdateBucketSizeMetrics) are computed by
collectCollectionInfoFromTopology, which only walked diskInfo.VolumeInfos.
As soon as a volume was encoded to EC it dropped out of every aggregate,
so Grafana showed bucket sizes shrinking while physical disk usage kept
climbing. The shell helper collectCollectionInfo — used by collection.list
and s3.bucket.quota.enforce — had the same gap, with the EC branch left as
a commented-out TODO.

Fold EC shards into both paths using the same approach the admin dashboard
already uses (PR #9093):

- PhysicalSize / Size sum across shard holders: EC shards are node-local
  (not replicas), so per-node TotalSize() and MinusParityShards().TotalSize()
  sum to the whole-volume physical and logical sizes respectively.
- FileCount is deduped via max across reporters (every shard holder reports
  the same .ecx count; a slow node with a not-yet-loaded .ecx reports 0 and
  must not pin the aggregate).
- DeleteCount is summed (each delete tombstones exactly one node's .ecj).
- VolumeCount increments once per unique EC volume id.

Adds regression tests covering pure-EC, mixed regular+EC, and the
slow-reporter FileCount dedupe case.

Refs #9086

* Address PR review feedback: EC size helpers, composite key, VolumeCount dedupe

- Add EcShardsTotalSize / EcShardsDataSize helpers in the erasure_coding
  package that walk the shard bitmap directly instead of materializing a
  ShardsInfo and copying it via MinusParityShards(). Keeps the
  DataShardsCount dependency encapsulated in one place and avoids the
  per-shard allocation/copy overhead in the metrics hot path.
- Switch shell collectCollectionInfo ecVolumes map to a composite
  {collection, volumeId} key, matching the bucket_size_metrics collector
  and defending against any cross-collection volume id aliasing.
- Dedupe VolumeCount in shell addToCollection by volume id so regular
  volumes aren't counted once per replica presence. Aligns the shell's
  collection.list output with the S3 metrics collector and the EC branch,
  all of which now report logical volume counts.
- Add unit tests for the new helpers and for the regular-volume
  VolumeCount dedupe.

* Parameterize EcShardsDataSize with dataShards for custom EC ratios

Add a dataShards parameter to EcShardsDataSize so forks with per-volume
ratio metadata (e.g. the enterprise data_shards field carried on an
extended VolumeEcShardInformationMessage) can pass the configured value
and get accurate logical sizes under custom EC policies like 6+3 or 16+6.
Passing 0 or a negative value falls back to the upstream DataShardsCount
default, which is correct for the fixed 10+4 layout — so OSS callers in
s3api and shell pass 0 and keep their current behavior.

Added table cases covering the custom 6+3 and 16+6 paths so the
parameterization is pinned by tests.
2026-04-21 20:17:42 -07:00

212 lines
6.2 KiB
Go

package s3api
import (
"testing"
"github.com/seaweedfs/seaweedfs/weed/pb/master_pb"
)
// TestCollectCollectionInfoFromTopologyEC verifies that EC-encoded volumes
// contribute to per-collection logical/physical size, file count, and volume
// count. Before this fix, encoding a volume to EC caused the bucket size
// metrics exported to Prometheus to drop to zero for that volume.
//
// Layout: one 10+4 EC volume in collection "crm-docs-storage", 14 shards of
// 1000 bytes each split across two nodes.
// - nodeA holds data shards 0..6 (7 * 1000 = 7000)
// - nodeB holds data shards 7..9 (3 * 1000 = 3000) and parity 10..13 (4 * 1000 = 4000)
//
// Expected:
// - PhysicalSize = 14 * 1000 = 14000
// - Size (logical, data shards) = 10 * 1000 = 10000
// - FileCount = 100 total - (2 + 3) local deletes = 95 is NOT what we check; the
// collector reports raw file_count and delete_count as separate gauges, so
// we assert FileCount = 100 (max across reporters) and DeleteCount = 5 (sum).
// - VolumeCount = 1 (one unique EC volume)
func TestCollectCollectionInfoFromTopologyEC(t *testing.T) {
nodeA := &master_pb.DataNodeInfo{
DiskInfos: map[string]*master_pb.DiskInfo{
"disk1": {
EcShardInfos: []*master_pb.VolumeEcShardInformationMessage{
{
Id: 42,
Collection: "crm-docs-storage",
EcIndexBits: (1 << 0) | (1 << 1) | (1 << 2) | (1 << 3) | (1 << 4) | (1 << 5) | (1 << 6),
ShardSizes: []int64{1000, 1000, 1000, 1000, 1000, 1000, 1000},
FileCount: 100,
DeleteCount: 2,
},
},
},
},
}
nodeB := &master_pb.DataNodeInfo{
DiskInfos: map[string]*master_pb.DiskInfo{
"disk1": {
EcShardInfos: []*master_pb.VolumeEcShardInformationMessage{
{
Id: 42,
Collection: "crm-docs-storage",
EcIndexBits: (1 << 7) | (1 << 8) | (1 << 9) | (1 << 10) | (1 << 11) | (1 << 12) | (1 << 13),
ShardSizes: []int64{1000, 1000, 1000, 1000, 1000, 1000, 1000},
FileCount: 100,
DeleteCount: 3,
},
},
},
},
}
topo := &master_pb.TopologyInfo{
DataCenterInfos: []*master_pb.DataCenterInfo{
{
RackInfos: []*master_pb.RackInfo{
{
DataNodeInfos: []*master_pb.DataNodeInfo{nodeA, nodeB},
},
},
},
},
}
got := make(map[string]*CollectionInfo)
collectCollectionInfoFromTopology(topo, got)
info, ok := got["crm-docs-storage"]
if !ok {
t.Fatalf("expected collection crm-docs-storage, got: %v", got)
}
if info.PhysicalSize != 14000 {
t.Errorf("PhysicalSize: got %.0f, want 14000", info.PhysicalSize)
}
if info.Size != 10000 {
t.Errorf("Size (logical): got %.0f, want 10000", info.Size)
}
if info.FileCount != 100 {
t.Errorf("FileCount: got %.0f, want 100 (max across reporters)", info.FileCount)
}
if info.DeleteCount != 5 {
t.Errorf("DeleteCount: got %.0f, want 5 (sum across reporters)", info.DeleteCount)
}
if info.VolumeCount != 1 {
t.Errorf("VolumeCount: got %d, want 1", info.VolumeCount)
}
}
// TestCollectCollectionInfoFromTopologyMixed verifies that regular and EC
// volumes accumulate under the same collection without one clobbering the
// other, which is the state during an in-progress EC conversion.
func TestCollectCollectionInfoFromTopologyMixed(t *testing.T) {
node := &master_pb.DataNodeInfo{
DiskInfos: map[string]*master_pb.DiskInfo{
"disk1": {
VolumeInfos: []*master_pb.VolumeInformationMessage{
{
Id: 1,
Collection: "bucket-mix",
Size: 5000,
FileCount: 50,
DeleteCount: 1,
DeletedByteCount: 100,
},
},
EcShardInfos: []*master_pb.VolumeEcShardInformationMessage{
{
Id: 2,
Collection: "bucket-mix",
EcIndexBits: (1 << 0) | (1 << 1) | (1 << 10), // 2 data + 1 parity
ShardSizes: []int64{3000, 3000, 3000},
FileCount: 80,
DeleteCount: 4,
},
},
},
},
}
topo := &master_pb.TopologyInfo{
DataCenterInfos: []*master_pb.DataCenterInfo{
{
RackInfos: []*master_pb.RackInfo{
{
DataNodeInfos: []*master_pb.DataNodeInfo{node},
},
},
},
},
}
got := make(map[string]*CollectionInfo)
collectCollectionInfoFromTopology(topo, got)
info, ok := got["bucket-mix"]
if !ok {
t.Fatalf("expected collection bucket-mix, got: %v", got)
}
// Regular volume: 5000 physical + logical. EC shards: 9000 physical,
// 6000 logical (data shards 0 and 1).
if info.PhysicalSize != 5000+9000 {
t.Errorf("PhysicalSize: got %.0f, want 14000", info.PhysicalSize)
}
if info.Size != 5000+6000 {
t.Errorf("Size: got %.0f, want 11000", info.Size)
}
if info.FileCount != 50+80 {
t.Errorf("FileCount: got %.0f, want 130", info.FileCount)
}
if info.DeleteCount != 1+4 {
t.Errorf("DeleteCount: got %.0f, want 5", info.DeleteCount)
}
if info.VolumeCount != 2 {
t.Errorf("VolumeCount: got %d, want 2", info.VolumeCount)
}
}
// TestCollectCollectionInfoFromTopologyECFileCountMaxDedupe verifies that a
// slow shard holder reporting file_count=0 (because it has not yet finished
// loading .ecx) does not pin the per-volume FileCount at 0.
func TestCollectCollectionInfoFromTopologyECFileCountMaxDedupe(t *testing.T) {
makeNode := func(bits uint32, sizes []int64, fileCount uint64) *master_pb.DataNodeInfo {
return &master_pb.DataNodeInfo{
DiskInfos: map[string]*master_pb.DiskInfo{
"disk1": {
EcShardInfos: []*master_pb.VolumeEcShardInformationMessage{
{
Id: 11,
Collection: "bucket-b",
EcIndexBits: bits,
ShardSizes: sizes,
FileCount: fileCount,
},
},
},
},
}
}
topo := &master_pb.TopologyInfo{
DataCenterInfos: []*master_pb.DataCenterInfo{
{
RackInfos: []*master_pb.RackInfo{
{
DataNodeInfos: []*master_pb.DataNodeInfo{
makeNode((1<<0)|(1<<1)|(1<<2)|(1<<3)|(1<<4)|(1<<5)|(1<<6), []int64{1, 1, 1, 1, 1, 1, 1}, 0),
makeNode((1<<7)|(1<<8)|(1<<9)|(1<<10)|(1<<11)|(1<<12)|(1<<13), []int64{1, 1, 1, 1, 1, 1, 1}, 6),
},
},
},
},
},
}
got := make(map[string]*CollectionInfo)
collectCollectionInfoFromTopology(topo, got)
info, ok := got["bucket-b"]
if !ok {
t.Fatalf("expected collection bucket-b, got: %v", got)
}
if info.FileCount != 6 {
t.Errorf("FileCount: got %.0f, want 6 (max across reporters)", info.FileCount)
}
}