Files
seaweedFS/weed/shell/command_cluster_status.go
Chris Lu e10f11b480 opt: reduce ShardsInfo memory usage with bitmap and sorted slice (#7974)
* opt: reduce ShardsInfo memory usage with bitmap and sorted slice

- Replace map[ShardId]*ShardInfo with sorted []ShardInfo slice
- Add ShardBits (uint32) bitmap for O(1) existence checks
- Use binary search for O(log n) lookups by shard ID
- Maintain sorted order for efficient iteration
- Add comprehensive unit tests and benchmarks

Memory savings:
- Map overhead: ~48 bytes per entry eliminated
- Pointers: 8 bytes per entry eliminated
- Total: ~56 bytes per shard saved

Performance improvements:
- Has(): O(1) using bitmap
- Size(): O(log n) using binary search (was O(1), acceptable tradeoff)
- Count(): O(1) using popcount on bitmap
- Iteration: Faster due to cache locality

* refactor: add methods to ShardBits type

- Add Has(), Set(), Clear(), and Count() methods to ShardBits
- Simplify ShardsInfo methods by using ShardBits methods
- Improves code readability and encapsulation

* opt: use ShardBits directly in ShardsCountFromVolumeEcShardInformationMessage

Avoid creating a full ShardsInfo object just to count shards.
Directly cast vi.EcIndexBits to ShardBits and use Count() method.

* opt: use strings.Builder in ShardsInfo.String() for efficiency

* refactor: change AsSlice to return []ShardInfo (values instead of pointers)

This completes the memory optimization by avoiding unnecessary pointer slices and potential allocations.

* refactor: rename ShardsCountFromVolumeEcShardInformationMessage to GetShardCount

* fix: prevent deadlock in Add and Subtract methods

Copy shards data from 'other' before releasing its lock to avoid
potential deadlock when a.Add(b) and b.Add(a) are called concurrently.

The previous implementation held other's lock while calling si.Set/Delete,
which acquires si's lock. This could deadlock if two goroutines tried to
add/subtract each other concurrently.

* opt: avoid unnecessary locking in constructor functions

ShardsInfoFromVolume and ShardsInfoFromVolumeEcShardInformationMessage
now build shards slice and bitmap directly without calling Set(), which
acquires a lock on every call. Since the object is local and not yet
shared, locking is unnecessary and adds overhead.

This improves performance during object construction.

* fix: rename 'copy' variable to avoid shadowing built-in function

The variable name 'copy' in TestShardsInfo_Copy shadowed the built-in
copy() function, which is confusing and bad practice. Renamed to 'siCopy'.

* opt: use math/bits.OnesCount32 and reorganize types

1. Replace manual popcount loop with math/bits.OnesCount32 for better
   performance and idiomatic Go code
2. Move ShardSize type definition to ec_shards_info.go for better code
   organization since it's primarily used there

* refactor: Set() now accepts ShardInfo for future extensibility

Changed Set(id ShardId, size ShardSize) to Set(shard ShardInfo) to
support future additions to ShardInfo without changing the API.

This makes the code more extensible as new fields can be added to
ShardInfo (e.g., checksum, location, etc.) without breaking the Set API.

* refactor: move ShardInfo and ShardSize to separate file

Created ec_shard_info.go to hold the basic shard types (ShardInfo and
ShardSize) for better code organization and separation of concerns.

* refactor: add ShardInfo constructor and helper functions

Added NewShardInfo() constructor and IsValid() method to better
encapsulate ShardInfo creation and validation. Updated code to use
the constructor for cleaner, more maintainable code.

* fix: update remaining Set() calls to use NewShardInfo constructor

Fixed compilation errors in storage and shell packages where Set() calls
were not updated to use the new NewShardInfo() constructor.

* fix: remove unreachable code in filer backup commands

Removed unreachable return statements after infinite loops in
filer_backup.go and filer_meta_backup.go to fix compilation errors.

* fix: rename 'new' variable to avoid shadowing built-in

Renamed 'new' to 'result' in MinusParityShards, Plus, and Minus methods
to avoid shadowing Go's built-in new() function.

* fix: update remaining test files to use NewShardInfo constructor

Fixed Set() calls in command_volume_list_test.go and
ec_rebalance_slots_test.go to use NewShardInfo() constructor.
2026-01-06 00:09:52 -08:00

440 lines
12 KiB
Go

package shell
import (
"context"
"flag"
"fmt"
"math"
"strings"
"sync"
"github.com/dustin/go-humanize"
"github.com/dustin/go-humanize/english"
"github.com/seaweedfs/seaweedfs/weed/operation"
"github.com/seaweedfs/seaweedfs/weed/pb"
"github.com/seaweedfs/seaweedfs/weed/pb/master_pb"
"github.com/seaweedfs/seaweedfs/weed/pb/volume_server_pb"
"github.com/seaweedfs/seaweedfs/weed/storage/erasure_coding"
"github.com/seaweedfs/seaweedfs/weed/storage/needle"
"io"
)
func init() {
Commands = append(Commands, &commandClusterStatus{})
}
// Map of volume_id -> [volume replicas] with stat details.
type VolumeReplicaStats struct {
Id string
VolumeId uint32
Files uint64
FilesDeleted uint64
TotalSize uint64
}
type RegularVolumeStats map[uint32][]*VolumeReplicaStats
type commandClusterStatus struct{}
type ClusterStatusPrinter struct {
writer io.Writer
writerMu sync.Mutex
humanize bool
maxParallelization int
locked bool
collections []string
topology *master_pb.TopologyInfo
volumeSizeLimitMb uint64
regularVolumeStats RegularVolumeStats
}
func (c *commandClusterStatus) Name() string {
return "cluster.status"
}
func (c *commandClusterStatus) Help() string {
return `outputs a quick overview of the cluster status`
}
func (c *commandClusterStatus) HasTag(CommandTag) bool {
return false
}
func (c *commandClusterStatus) Do(args []string, commandEnv *CommandEnv, writer io.Writer) (err error) {
flags := flag.NewFlagSet(c.Name(), flag.ContinueOnError)
humanize := flags.Bool("humanize", true, "human-readable output")
includeFiles := flags.Bool("files", false, "include detailed file metrics, from all volume servers")
maxParallelization := flags.Int("maxParallelization", DefaultMaxParallelization, "run up to X tasks in parallel, whenever possible")
if err = flags.Parse(args); err != nil {
return err
}
collections, err := ListCollectionNames(commandEnv, true, true)
if err != nil {
return err
}
topology, volumeSizeLimitMb, err := collectTopologyInfo(commandEnv, 0)
if err != nil {
return err
}
sp := &ClusterStatusPrinter{
writer: writer,
humanize: *humanize,
maxParallelization: *maxParallelization,
locked: commandEnv.isLocked(),
collections: collections,
topology: topology,
volumeSizeLimitMb: volumeSizeLimitMb,
}
if *includeFiles {
if err := sp.loadFileStats(commandEnv); err != nil {
return err
}
}
sp.Print()
return nil
}
func (sp *ClusterStatusPrinter) uint64(n uint64) string {
if !sp.humanize {
return fmt.Sprintf("%d", n)
}
return humanize.Comma(int64(n))
}
func (sp *ClusterStatusPrinter) int(n int) string {
return sp.uint64(uint64(n))
}
func (sp *ClusterStatusPrinter) uint64Plural(n uint64, str string) string {
if !sp.humanize {
return fmt.Sprintf("%s(s)", str)
}
uin := math.MaxInt
if n < math.MaxInt {
uin = int(n)
}
return english.PluralWord(int(uin), str, "")
}
func (sp *ClusterStatusPrinter) plural(n int, str string) string {
return sp.uint64Plural(uint64(n), str)
}
func (sp *ClusterStatusPrinter) bytes(b uint64) string {
if !sp.humanize {
return fmt.Sprintf("%d %s", b, sp.plural(int(b), "byte"))
}
return fmt.Sprintf("%s", humanize.Bytes(b))
}
func (sp *ClusterStatusPrinter) uint64Ratio(a, b uint64) string {
var p float64
if b != 0 {
p = float64(a) / float64(b)
}
if !sp.humanize {
return fmt.Sprintf("%.02f", p)
}
return fmt.Sprintf("%s", humanize.FtoaWithDigits(p, 2))
}
func (sp *ClusterStatusPrinter) intRatio(a, b int) string {
return sp.uint64Ratio(uint64(a), uint64(b))
}
func (sp *ClusterStatusPrinter) uint64Pct(a, b uint64) string {
var p float64
if b != 0 {
p = 100 * float64(a) / float64(b)
}
if !sp.humanize {
return fmt.Sprintf("%.02f%%", p)
}
return fmt.Sprintf("%s%%", humanize.FtoaWithDigits(p, 2))
}
func (sp *ClusterStatusPrinter) intPct(a, b int) string {
return sp.uint64Pct(uint64(a), uint64(b))
}
func (sp *ClusterStatusPrinter) write(format string, a ...any) {
sp.writerMu.Lock()
defer sp.writerMu.Unlock()
format = strings.TrimRight(format, " ")
if len(format) == 0 {
format = "\n"
}
fmt.Fprintf(sp.writer, format, a...)
last := format[len(format)-1:]
if last != "\n" && last != "\r" {
fmt.Fprint(sp.writer, "\n")
}
}
func (sp *ClusterStatusPrinter) Print() {
sp.write("")
sp.printClusterInfo()
sp.printVolumeInfo()
sp.printStorageInfo()
sp.printFilesInfo()
}
// TODO: collect stats for EC volumes as well
func (sp *ClusterStatusPrinter) loadFileStats(commandEnv *CommandEnv) error {
sp.regularVolumeStats = RegularVolumeStats{}
var mu sync.Mutex
var progressTotal, progressDone uint64
ewg := NewErrorWaitGroup(sp.maxParallelization)
for _, dci := range sp.topology.DataCenterInfos {
for _, ri := range dci.RackInfos {
for _, dni := range ri.DataNodeInfos {
for _, d := range dni.DiskInfos {
mu.Lock()
progressTotal += uint64(len(d.VolumeInfos))
mu.Unlock()
for _, v := range d.VolumeInfos {
ewg.Add(func() error {
// Collect regular volume stats
err := operation.WithVolumeServerClient(false, pb.NewServerAddressWithGrpcPort(dni.Id, int(dni.GrpcPort)), commandEnv.option.GrpcDialOption, func(volumeServerClient volume_server_pb.VolumeServerClient) error {
resp, reqErr := volumeServerClient.VolumeStatus(context.Background(), &volume_server_pb.VolumeStatusRequest{
VolumeId: uint32(v.Id),
})
if reqErr != nil {
return reqErr
}
mu.Lock()
defer mu.Unlock()
if resp != nil {
if _, ok := sp.regularVolumeStats[v.Id]; !ok {
sp.regularVolumeStats[v.Id] = []*VolumeReplicaStats{}
}
sp.regularVolumeStats[v.Id] = append(sp.regularVolumeStats[v.Id], &VolumeReplicaStats{
Id: dni.Id,
VolumeId: v.Id,
Files: resp.FileCount,
FilesDeleted: resp.FileDeletedCount,
TotalSize: resp.VolumeSize,
})
}
progressDone++
return nil
})
if err != nil {
return err
}
mu.Lock()
sp.write("collecting file stats: %s \r", sp.uint64Pct(progressDone, progressTotal))
mu.Unlock()
return nil
})
}
}
}
}
}
err := ewg.Wait()
sp.write("")
return err
}
func (sp *ClusterStatusPrinter) printClusterInfo() {
dcs := len(sp.topology.DataCenterInfos)
racks := 0
nodes := 0
disks := 0
for _, dci := range sp.topology.DataCenterInfos {
racks += len(dci.RackInfos)
for _, ri := range dci.RackInfos {
for _, dni := range ri.DataNodeInfos {
nodes++
disks += len(dni.DiskInfos)
}
}
}
status := "unlocked"
if sp.locked {
status = "LOCKED"
}
sp.write("cluster:")
sp.write("\tid: %s", sp.topology.Id)
sp.write("\tstatus: %s", status)
sp.write("\tnodes: %s", sp.int(nodes))
sp.write("\ttopology: %s %s, %s %s on %s %s",
sp.int(dcs), sp.plural(dcs, "DC"),
sp.int(disks), sp.plural(disks, "disk"),
sp.int(racks), sp.plural(racks, "rack"))
sp.write("")
}
func (sp *ClusterStatusPrinter) printVolumeInfo() {
collections := len(sp.collections)
var maxVolumes uint64
volumeIds := map[needle.VolumeId]bool{}
ecVolumeIds := map[needle.VolumeId]bool{}
var replicas, roReplicas, rwReplicas, ecShards int
for _, dci := range sp.topology.DataCenterInfos {
for _, ri := range dci.RackInfos {
for _, dni := range ri.DataNodeInfos {
for _, di := range dni.DiskInfos {
maxVolumes += uint64(di.MaxVolumeCount)
for _, vi := range di.VolumeInfos {
vid := needle.VolumeId(vi.Id)
volumeIds[vid] = true
replicas++
if vi.ReadOnly {
roReplicas++
} else {
rwReplicas++
}
}
for _, eci := range di.EcShardInfos {
vid := needle.VolumeId(eci.Id)
ecVolumeIds[vid] = true
ecShards += erasure_coding.GetShardCount(eci)
}
}
}
}
}
volumes := len(volumeIds)
ecVolumes := len(ecVolumeIds)
totalVolumes := volumes + ecVolumes
sp.write("volumes:")
sp.write("\ttotal: %s %s, %s %s",
sp.int(totalVolumes), sp.plural(totalVolumes, "volume"),
sp.int(collections), sp.plural(collections, "collection"))
sp.write("\tmax size: %s", sp.bytes(sp.volumeSizeLimitMb*1024*1024))
sp.write("\tregular: %s/%s %s on %s %s, %s writable (%s), %s read-only (%s)",
sp.int(volumes), sp.uint64(maxVolumes), sp.plural(volumes, "volume"),
sp.int(replicas), sp.plural(replicas, "replica"),
sp.int(rwReplicas), sp.intPct(rwReplicas, replicas),
sp.int(roReplicas), sp.intPct(roReplicas, replicas))
sp.write("\tEC: %s EC %s on %s %s (%s shards/volume)",
sp.int(ecVolumes), sp.plural(ecVolumes, "volume"),
sp.int(ecShards), sp.plural(ecShards, "shard"),
sp.intRatio(ecShards, ecVolumes))
sp.write("")
}
func (sp *ClusterStatusPrinter) printStorageInfo() {
perVolumeSize := map[needle.VolumeId]uint64{}
perEcVolumeSize := map[needle.VolumeId]uint64{}
var rawVolumeSize, rawEcVolumeSize uint64
for _, dci := range sp.topology.DataCenterInfos {
for _, ri := range dci.RackInfos {
for _, dni := range ri.DataNodeInfos {
for _, di := range dni.DiskInfos {
for _, vi := range di.VolumeInfos {
vid := needle.VolumeId(vi.Id)
perVolumeSize[vid] = vi.Size
rawVolumeSize += vi.Size
}
for _, eci := range di.EcShardInfos {
vid := needle.VolumeId(eci.Id)
var size uint64
for _, ss := range eci.ShardSizes {
size += uint64(ss)
}
perEcVolumeSize[vid] += size
rawEcVolumeSize += size
}
}
}
}
}
// normalize EC logical volume sizes given shard settings
for vid := range perEcVolumeSize {
perEcVolumeSize[vid] = perEcVolumeSize[vid] * erasure_coding.DataShardsCount / erasure_coding.TotalShardsCount
}
var volumeSize, ecVolumeSize uint64
for _, s := range perVolumeSize {
volumeSize += s
}
for _, s := range perEcVolumeSize {
ecVolumeSize += s
}
totalSize := volumeSize + ecVolumeSize
sp.write("storage:")
sp.write("\ttotal: %s", sp.bytes(totalSize))
sp.write("\tregular volumes: %s", sp.bytes(volumeSize))
sp.write("\tEC volumes: %s", sp.bytes(ecVolumeSize))
sp.write("\traw: %s on volume replicas, %s on EC shards", sp.bytes(rawVolumeSize), sp.bytes(rawEcVolumeSize))
sp.write("")
}
func (sp *ClusterStatusPrinter) printFilesInfo() {
if len(sp.regularVolumeStats) == 0 {
return
}
var regularFilesTotal, regularFilesDeleted, regularFilesSize uint64
var regularFilesTotalRaw, regularFilesDeletedRaw, regularFilesSizeRaw uint64
for _, replicaStats := range sp.regularVolumeStats {
rc := uint64(len(replicaStats))
var volumeFilesTotal, volumeFilesSize, volumeFilesDeleted uint64
for _, rs := range replicaStats {
regularFilesTotalRaw += rs.Files
regularFilesSizeRaw += rs.TotalSize
regularFilesDeletedRaw += rs.FilesDeleted
volumeFilesTotal += rs.Files
volumeFilesSize += rs.TotalSize
volumeFilesDeleted += rs.FilesDeleted
}
regularFilesTotal += (volumeFilesTotal / rc)
regularFilesSize += (volumeFilesSize / rc)
regularFilesDeleted += (volumeFilesDeleted / rc)
}
regularFiles := regularFilesTotal - regularFilesDeleted
regularFilesRaw := regularFilesTotalRaw - regularFilesDeletedRaw
var avgFileSize uint64
if regularFilesTotal != 0 {
avgFileSize = regularFilesSize / regularFilesTotal
}
sp.write("files:")
sp.write("\tregular: %s %s, %s readable (%s), %s deleted (%s), avg %s per file",
sp.uint64(regularFilesTotal), sp.uint64Plural(regularFilesTotal, "file"),
sp.uint64(regularFiles), sp.uint64Pct(regularFiles, regularFilesTotal),
sp.uint64(regularFilesDeleted), sp.uint64Pct(regularFilesDeleted, regularFilesTotal),
sp.bytes(avgFileSize))
sp.write("\tregular raw: %s %s, %s readable (%s), %s deleted (%s), %s total",
sp.uint64(regularFilesTotalRaw), sp.uint64Plural(regularFilesTotalRaw, "file"),
sp.uint64(regularFilesRaw), sp.uint64Pct(regularFilesRaw, regularFilesTotalRaw),
sp.uint64(regularFilesDeletedRaw), sp.uint64Pct(regularFilesDeletedRaw, regularFilesTotalRaw),
sp.bytes(regularFilesSizeRaw))
sp.write("\tEC: [no data]")
sp.write("\tEC raw: [no data]")
sp.write("")
}