1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
|
package topology
import (
"errors"
"fmt"
"github.com/chrislusf/seaweedfs/weed/storage/types"
"math/rand"
"sync"
"time"
"github.com/chrislusf/raft"
"github.com/chrislusf/seaweedfs/weed/glog"
"github.com/chrislusf/seaweedfs/weed/pb/master_pb"
"github.com/chrislusf/seaweedfs/weed/sequence"
"github.com/chrislusf/seaweedfs/weed/storage"
"github.com/chrislusf/seaweedfs/weed/storage/needle"
"github.com/chrislusf/seaweedfs/weed/storage/super_block"
"github.com/chrislusf/seaweedfs/weed/util"
)
type Topology struct {
vacuumLockCounter int64
NodeImpl
collectionMap *util.ConcurrentReadMap
ecShardMap map[needle.VolumeId]*EcShardLocations
ecShardMapLock sync.RWMutex
pulse int64
volumeSizeLimit uint64
replicationAsMin bool
Sequence sequence.Sequencer
chanFullVolumes chan storage.VolumeInfo
Configuration *Configuration
RaftServer raft.Server
}
func NewTopology(id string, seq sequence.Sequencer, volumeSizeLimit uint64, pulse int, replicationAsMin bool) *Topology {
t := &Topology{}
t.id = NodeId(id)
t.nodeType = "Topology"
t.NodeImpl.value = t
t.diskUsages = newDiskUsages()
t.children = make(map[NodeId]Node)
t.collectionMap = util.NewConcurrentReadMap()
t.ecShardMap = make(map[needle.VolumeId]*EcShardLocations)
t.pulse = int64(pulse)
t.volumeSizeLimit = volumeSizeLimit
t.replicationAsMin = replicationAsMin
t.Sequence = seq
t.chanFullVolumes = make(chan storage.VolumeInfo)
t.Configuration = &Configuration{}
return t
}
func (t *Topology) IsLeader() bool {
if t.RaftServer != nil {
if t.RaftServer.State() == raft.Leader {
return true
}
}
return false
}
func (t *Topology) Leader() (string, error) {
l := ""
for count := 0; count < 3; count++ {
if t.RaftServer != nil {
l = t.RaftServer.Leader()
} else {
return "", errors.New("Raft Server not ready yet!")
}
if l != "" {
break
} else {
time.Sleep(time.Duration(5+count) * time.Second)
}
}
return l, nil
}
func (t *Topology) Lookup(collection string, vid needle.VolumeId) (dataNodes []*DataNode) {
// maybe an issue if lots of collections?
if collection == "" {
for _, c := range t.collectionMap.Items() {
if list := c.(*Collection).Lookup(vid); list != nil {
return list
}
}
} else {
if c, ok := t.collectionMap.Find(collection); ok {
return c.(*Collection).Lookup(vid)
}
}
if locations, found := t.LookupEcShards(vid); found {
for _, loc := range locations.Locations {
dataNodes = append(dataNodes, loc...)
}
return dataNodes
}
return nil
}
func (t *Topology) NextVolumeId() (needle.VolumeId, error) {
vid := t.GetMaxVolumeId()
next := vid.Next()
if _, err := t.RaftServer.Do(NewMaxVolumeIdCommand(next)); err != nil {
return 0, err
}
return next, nil
}
func (t *Topology) HasWritableVolume(option *VolumeGrowOption) bool {
vl := t.GetVolumeLayout(option.Collection, option.ReplicaPlacement, option.Ttl, option.DiskType)
return vl.GetActiveVolumeCount(option) > 0
}
func (t *Topology) PickForWrite(count uint64, option *VolumeGrowOption) (string, uint64, *DataNode, error) {
vid, count, datanodes, err := t.GetVolumeLayout(option.Collection, option.ReplicaPlacement, option.Ttl, option.DiskType).PickForWrite(count, option)
if err != nil {
return "", 0, nil, fmt.Errorf("failed to find writable volumes for collection:%s replication:%s ttl:%s error: %v", option.Collection, option.ReplicaPlacement.String(), option.Ttl.String(), err)
}
if datanodes.Length() == 0 {
return "", 0, nil, fmt.Errorf("no writable volumes available for collection:%s replication:%s ttl:%s", option.Collection, option.ReplicaPlacement.String(), option.Ttl.String())
}
fileId := t.Sequence.NextFileId(count)
return needle.NewFileId(*vid, fileId, rand.Uint32()).String(), count, datanodes.Head(), nil
}
func (t *Topology) GetVolumeLayout(collectionName string, rp *super_block.ReplicaPlacement, ttl *needle.TTL, diskType types.DiskType) *VolumeLayout {
return t.collectionMap.Get(collectionName, func() interface{} {
return NewCollection(collectionName, t.volumeSizeLimit, t.replicationAsMin)
}).(*Collection).GetOrCreateVolumeLayout(rp, ttl, diskType)
}
func (t *Topology) ListCollections(includeNormalVolumes, includeEcVolumes bool) (ret []string) {
mapOfCollections := make(map[string]bool)
for _, c := range t.collectionMap.Items() {
mapOfCollections[c.(*Collection).Name] = true
}
if includeEcVolumes {
t.ecShardMapLock.RLock()
for _, ecVolumeLocation := range t.ecShardMap {
mapOfCollections[ecVolumeLocation.Collection] = true
}
t.ecShardMapLock.RUnlock()
}
for k := range mapOfCollections {
ret = append(ret, k)
}
return ret
}
func (t *Topology) FindCollection(collectionName string) (*Collection, bool) {
c, hasCollection := t.collectionMap.Find(collectionName)
if !hasCollection {
return nil, false
}
return c.(*Collection), hasCollection
}
func (t *Topology) DeleteCollection(collectionName string) {
t.collectionMap.Delete(collectionName)
}
func (t *Topology) DeleteLayout(collectionName string, rp *super_block.ReplicaPlacement, ttl *needle.TTL, diskType types.DiskType) {
collection, found := t.FindCollection(collectionName)
if !found {
return
}
collection.DeleteVolumeLayout(rp, ttl, diskType)
if len(collection.storageType2VolumeLayout.Items()) == 0 {
t.DeleteCollection(collectionName)
}
}
func (t *Topology) RegisterVolumeLayout(v storage.VolumeInfo, dn *DataNode) {
diskType := types.ToDiskType(v.DiskType)
vl := t.GetVolumeLayout(v.Collection, v.ReplicaPlacement, v.Ttl, diskType)
vl.RegisterVolume(&v, dn)
vl.EnsureCorrectWritables(&v)
}
func (t *Topology) UnRegisterVolumeLayout(v storage.VolumeInfo, dn *DataNode) {
glog.Infof("removing volume info: %+v", v)
diskType := types.ToDiskType(v.DiskType)
volumeLayout := t.GetVolumeLayout(v.Collection, v.ReplicaPlacement, v.Ttl, diskType)
volumeLayout.UnRegisterVolume(&v, dn)
if volumeLayout.isEmpty() {
t.DeleteLayout(v.Collection, v.ReplicaPlacement, v.Ttl, diskType)
}
}
func (t *Topology) GetOrCreateDataCenter(dcName string) *DataCenter {
for _, c := range t.Children() {
dc := c.(*DataCenter)
if string(dc.Id()) == dcName {
return dc
}
}
dc := NewDataCenter(dcName)
t.LinkChildNode(dc)
return dc
}
func (t *Topology) SyncDataNodeRegistration(volumes []*master_pb.VolumeInformationMessage, dn *DataNode) (newVolumes, deletedVolumes []storage.VolumeInfo) {
// convert into in memory struct storage.VolumeInfo
var volumeInfos []storage.VolumeInfo
for _, v := range volumes {
if vi, err := storage.NewVolumeInfo(v); err == nil {
volumeInfos = append(volumeInfos, vi)
} else {
glog.V(0).Infof("Fail to convert joined volume information: %v", err)
}
}
// find out the delta volumes
var changedVolumes []storage.VolumeInfo
newVolumes, deletedVolumes, changedVolumes = dn.UpdateVolumes(volumeInfos)
for _, v := range newVolumes {
t.RegisterVolumeLayout(v, dn)
}
for _, v := range deletedVolumes {
t.UnRegisterVolumeLayout(v, dn)
}
for _, v := range changedVolumes {
diskType := types.ToDiskType(v.DiskType)
vl := t.GetVolumeLayout(v.Collection, v.ReplicaPlacement, v.Ttl, diskType)
vl.EnsureCorrectWritables(&v)
}
return
}
func (t *Topology) IncrementalSyncDataNodeRegistration(newVolumes, deletedVolumes []*master_pb.VolumeShortInformationMessage, dn *DataNode) {
var newVis, oldVis []storage.VolumeInfo
for _, v := range newVolumes {
vi, err := storage.NewVolumeInfoFromShort(v)
if err != nil {
glog.V(0).Infof("NewVolumeInfoFromShort %v: %v", v, err)
continue
}
newVis = append(newVis, vi)
}
for _, v := range deletedVolumes {
vi, err := storage.NewVolumeInfoFromShort(v)
if err != nil {
glog.V(0).Infof("NewVolumeInfoFromShort %v: %v", v, err)
continue
}
oldVis = append(oldVis, vi)
}
dn.DeltaUpdateVolumes(newVis, oldVis)
for _, vi := range newVis {
t.RegisterVolumeLayout(vi, dn)
}
for _, vi := range oldVis {
t.UnRegisterVolumeLayout(vi, dn)
}
return
}
|
