1
0
Fork 0
mirror of https://github.com/chrislusf/seaweedfs synced 2025-08-16 17:12:46 +02:00
seaweedfs/weed/topology/disk.go
Chris Lu 891a2fb6eb
Admin: misc improvements on admin server and workers. EC now works. (#7055)
* initial design

* added simulation as tests

* reorganized the codebase to move the simulation framework and tests into their own dedicated package

* integration test. ec worker task

* remove "enhanced" reference

* start master, volume servers, filer

Current Status
 Master: Healthy and running (port 9333)
 Filer: Healthy and running (port 8888)
 Volume Servers: All 6 servers running (ports 8080-8085)
🔄 Admin/Workers: Will start when dependencies are ready

* generate write load

* tasks are assigned

* admin start wtih grpc port. worker has its own working directory

* Update .gitignore

* working worker and admin. Task detection is not working yet.

* compiles, detection uses volumeSizeLimitMB from master

* compiles

* worker retries connecting to admin

* build and restart

* rendering pending tasks

* skip task ID column

* sticky worker id

* test canScheduleTaskNow

* worker reconnect to admin

* clean up logs

* worker register itself first

* worker can run ec work and report status

but:
1. one volume should not be repeatedly worked on.
2. ec shards needs to be distributed and source data should be deleted.

* move ec task logic

* listing ec shards

* local copy, ec. Need to distribute.

* ec is mostly working now

* distribution of ec shards needs improvement
* need configuration to enable ec

* show ec volumes

* interval field UI component

* rename

* integration test with vauuming

* garbage percentage threshold

* fix warning

* display ec shard sizes

* fix ec volumes list

* Update ui.go

* show default values

* ensure correct default value

* MaintenanceConfig use ConfigField

* use schema defined defaults

* config

* reduce duplication

* refactor to use BaseUIProvider

* each task register its schema

* checkECEncodingCandidate use ecDetector

* use vacuumDetector

* use volumeSizeLimitMB

* remove

remove

* remove unused

* refactor

* use new framework

* remove v2 reference

* refactor

* left menu can scroll now

* The maintenance manager was not being initialized when no data directory was configured for persistent storage.

* saving config

* Update task_config_schema_templ.go

* enable/disable tasks

* protobuf encoded task configurations

* fix system settings

* use ui component

* remove logs

* interface{} Reduction

* reduce interface{}

* reduce interface{}

* avoid from/to map

* reduce interface{}

* refactor

* keep it DRY

* added logging

* debug messages

* debug level

* debug

* show the log caller line

* use configured task policy

* log level

* handle admin heartbeat response

* Update worker.go

* fix EC rack and dc count

* Report task status to admin server

* fix task logging, simplify interface checking, use erasure_coding constants

* factor in empty volume server during task planning

* volume.list adds disk id

* track disk id also

* fix locking scheduled and manual scanning

* add active topology

* simplify task detector

* ec task completed, but shards are not showing up

* implement ec in ec_typed.go

* adjust log level

* dedup

* implementing ec copying shards and only ecx files

* use disk id when distributing ec shards

🎯 Planning: ActiveTopology creates DestinationPlan with specific TargetDisk
📦 Task Creation: maintenance_integration.go creates ECDestination with DiskId
🚀 Task Execution: EC task passes DiskId in VolumeEcShardsCopyRequest
💾 Volume Server: Receives disk_id and stores shards on specific disk (vs.store.Locations[req.DiskId])
📂 File System: EC shards and metadata land in the exact disk directory planned

* Delete original volume from all locations

* clean up existing shard locations

* local encoding and distributing

* Update docker/admin_integration/EC-TESTING-README.md

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

* check volume id range

* simplify

* fix tests

* fix types

* clean up logs and tests

---------

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-07-30 12:38:03 -07:00

289 lines
7.5 KiB
Go

package topology
import (
"fmt"
"sync"
"sync/atomic"
"github.com/seaweedfs/seaweedfs/weed/storage/types"
"github.com/seaweedfs/seaweedfs/weed/util"
"github.com/seaweedfs/seaweedfs/weed/pb/master_pb"
"github.com/seaweedfs/seaweedfs/weed/storage/erasure_coding"
"github.com/seaweedfs/seaweedfs/weed/storage/needle"
"github.com/seaweedfs/seaweedfs/weed/storage"
)
type Disk struct {
NodeImpl
volumes map[needle.VolumeId]storage.VolumeInfo
ecShards map[needle.VolumeId]*erasure_coding.EcVolumeInfo
ecShardsLock sync.RWMutex
}
func NewDisk(diskType string) *Disk {
s := &Disk{}
s.id = NodeId(diskType)
s.nodeType = "Disk"
s.diskUsages = newDiskUsages()
s.volumes = make(map[needle.VolumeId]storage.VolumeInfo, 2)
s.ecShards = make(map[needle.VolumeId]*erasure_coding.EcVolumeInfo, 2)
s.NodeImpl.value = s
return s
}
type DiskUsages struct {
sync.RWMutex
usages map[types.DiskType]*DiskUsageCounts
}
func newDiskUsages() *DiskUsages {
return &DiskUsages{
usages: make(map[types.DiskType]*DiskUsageCounts),
}
}
func (d *DiskUsages) negative() *DiskUsages {
d.RLock()
defer d.RUnlock()
t := newDiskUsages()
for diskType, b := range d.usages {
a := t.getOrCreateDisk(diskType)
a.volumeCount = -b.volumeCount
a.remoteVolumeCount = -b.remoteVolumeCount
a.activeVolumeCount = -b.activeVolumeCount
a.ecShardCount = -b.ecShardCount
a.maxVolumeCount = -b.maxVolumeCount
}
return t
}
func (d *DiskUsages) ToDiskInfo() map[string]*master_pb.DiskInfo {
ret := make(map[string]*master_pb.DiskInfo)
for diskType, diskUsageCounts := range d.usages {
m := &master_pb.DiskInfo{
VolumeCount: diskUsageCounts.volumeCount,
MaxVolumeCount: diskUsageCounts.maxVolumeCount,
FreeVolumeCount: diskUsageCounts.maxVolumeCount - (diskUsageCounts.volumeCount - diskUsageCounts.remoteVolumeCount) - (diskUsageCounts.ecShardCount+1)/erasure_coding.DataShardsCount,
ActiveVolumeCount: diskUsageCounts.activeVolumeCount,
RemoteVolumeCount: diskUsageCounts.remoteVolumeCount,
}
ret[string(diskType)] = m
}
return ret
}
func (d *DiskUsages) FreeSpace() (freeSpace int64) {
d.RLock()
defer d.RUnlock()
for _, diskUsage := range d.usages {
freeSpace += diskUsage.FreeSpace()
}
return
}
func (d *DiskUsages) GetMaxVolumeCount() (maxVolumeCount int64) {
d.RLock()
defer d.RUnlock()
for _, diskUsage := range d.usages {
maxVolumeCount += diskUsage.maxVolumeCount
}
return
}
type DiskUsageCounts struct {
volumeCount int64
remoteVolumeCount int64
activeVolumeCount int64
ecShardCount int64
maxVolumeCount int64
}
func (a *DiskUsageCounts) addDiskUsageCounts(b *DiskUsageCounts) {
atomic.AddInt64(&a.volumeCount, b.volumeCount)
atomic.AddInt64(&a.remoteVolumeCount, b.remoteVolumeCount)
atomic.AddInt64(&a.activeVolumeCount, b.activeVolumeCount)
atomic.AddInt64(&a.ecShardCount, b.ecShardCount)
atomic.AddInt64(&a.maxVolumeCount, b.maxVolumeCount)
}
func (a *DiskUsageCounts) FreeSpace() int64 {
freeVolumeSlotCount := a.maxVolumeCount + a.remoteVolumeCount - a.volumeCount
if a.ecShardCount > 0 {
freeVolumeSlotCount = freeVolumeSlotCount - a.ecShardCount/erasure_coding.DataShardsCount - 1
}
return freeVolumeSlotCount
}
func (a *DiskUsageCounts) minus(b *DiskUsageCounts) *DiskUsageCounts {
return &DiskUsageCounts{
volumeCount: a.volumeCount - b.volumeCount,
remoteVolumeCount: a.remoteVolumeCount - b.remoteVolumeCount,
activeVolumeCount: a.activeVolumeCount - b.activeVolumeCount,
ecShardCount: a.ecShardCount - b.ecShardCount,
maxVolumeCount: a.maxVolumeCount - b.maxVolumeCount,
}
}
func (du *DiskUsages) getOrCreateDisk(diskType types.DiskType) *DiskUsageCounts {
du.Lock()
defer du.Unlock()
t, found := du.usages[diskType]
if found {
return t
}
t = &DiskUsageCounts{}
du.usages[diskType] = t
return t
}
func (d *Disk) String() string {
d.RLock()
defer d.RUnlock()
return fmt.Sprintf("Disk:%s, volumes:%v, ecShards:%v", d.NodeImpl.String(), d.volumes, d.ecShards)
}
func (d *Disk) AddOrUpdateVolume(v storage.VolumeInfo) (isNew, isChanged bool) {
d.Lock()
defer d.Unlock()
return d.doAddOrUpdateVolume(v)
}
func (d *Disk) doAddOrUpdateVolume(v storage.VolumeInfo) (isNew, isChanged bool) {
deltaDiskUsage := &DiskUsageCounts{}
if oldV, ok := d.volumes[v.Id]; !ok {
d.volumes[v.Id] = v
deltaDiskUsage.volumeCount = 1
if v.IsRemote() {
deltaDiskUsage.remoteVolumeCount = 1
}
if !v.ReadOnly {
deltaDiskUsage.activeVolumeCount = 1
}
d.UpAdjustMaxVolumeId(v.Id)
d.UpAdjustDiskUsageDelta(types.ToDiskType(v.DiskType), deltaDiskUsage)
isNew = true
} else {
if oldV.IsRemote() != v.IsRemote() {
if v.IsRemote() {
deltaDiskUsage.remoteVolumeCount = 1
}
if oldV.IsRemote() {
deltaDiskUsage.remoteVolumeCount = -1
}
d.UpAdjustDiskUsageDelta(types.ToDiskType(v.DiskType), deltaDiskUsage)
}
isChanged = d.volumes[v.Id].ReadOnly != v.ReadOnly
d.volumes[v.Id] = v
}
return
}
func (d *Disk) GetVolumes() (ret []storage.VolumeInfo) {
d.RLock()
for _, v := range d.volumes {
ret = append(ret, v)
}
d.RUnlock()
return ret
}
func (d *Disk) GetVolumesById(id needle.VolumeId) (storage.VolumeInfo, error) {
d.RLock()
defer d.RUnlock()
vInfo, ok := d.volumes[id]
if ok {
return vInfo, nil
} else {
return storage.VolumeInfo{}, fmt.Errorf("volumeInfo not found")
}
}
func (d *Disk) DeleteVolumeById(id needle.VolumeId) {
d.Lock()
defer d.Unlock()
delete(d.volumes, id)
}
func (d *Disk) GetDataCenter() *DataCenter {
dn := d.Parent()
rack := dn.Parent()
dcNode := rack.Parent()
dcValue := dcNode.GetValue()
return dcValue.(*DataCenter)
}
func (d *Disk) GetRack() *Rack {
return d.Parent().Parent().(*NodeImpl).value.(*Rack)
}
func (d *Disk) GetTopology() *Topology {
p := d.Parent()
for p.Parent() != nil {
p = p.Parent()
}
t := p.(*Topology)
return t
}
func (d *Disk) ToMap() interface{} {
ret := make(map[string]interface{})
diskUsage := d.diskUsages.getOrCreateDisk(types.ToDiskType(string(d.Id())))
ret["Volumes"] = diskUsage.volumeCount
ret["VolumeIds"] = d.GetVolumeIds()
ret["EcShards"] = diskUsage.ecShardCount
ret["Max"] = diskUsage.maxVolumeCount
ret["Free"] = d.FreeSpace()
return ret
}
func (d *Disk) FreeSpace() int64 {
t := d.diskUsages.getOrCreateDisk(types.ToDiskType(string(d.Id())))
return t.FreeSpace()
}
func (d *Disk) ToDiskInfo() *master_pb.DiskInfo {
diskUsage := d.diskUsages.getOrCreateDisk(types.ToDiskType(string(d.Id())))
// Get disk ID from first volume or EC shard
var diskId uint32
volumes := d.GetVolumes()
ecShards := d.GetEcShards()
if len(volumes) > 0 {
diskId = volumes[0].DiskId
} else if len(ecShards) > 0 {
diskId = ecShards[0].DiskId
}
m := &master_pb.DiskInfo{
Type: string(d.Id()),
VolumeCount: diskUsage.volumeCount,
MaxVolumeCount: diskUsage.maxVolumeCount,
FreeVolumeCount: diskUsage.maxVolumeCount - (diskUsage.volumeCount - diskUsage.remoteVolumeCount) - (diskUsage.ecShardCount+1)/erasure_coding.DataShardsCount,
ActiveVolumeCount: diskUsage.activeVolumeCount,
RemoteVolumeCount: diskUsage.remoteVolumeCount,
DiskId: diskId,
}
for _, v := range volumes {
m.VolumeInfos = append(m.VolumeInfos, v.ToVolumeInformationMessage())
}
for _, ecv := range ecShards {
m.EcShardInfos = append(m.EcShardInfos, ecv.ToVolumeEcShardInformationMessage())
}
return m
}
// GetVolumeIds returns the human readable volume ids limited to count of max 100.
func (d *Disk) GetVolumeIds() string {
d.RLock()
defer d.RUnlock()
ids := make([]int, 0, len(d.volumes))
for k := range d.volumes {
ids = append(ids, int(k))
}
return util.HumanReadableIntsMax(100, ids...)
}