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
|
package storage
import (
"encoding/json"
"sync"
"time"
"github.com/prometheus/client_golang/prometheus"
"github.com/prometheus/client_golang/prometheus/promauto"
"github.com/seaweedfs/seaweedfs/telemetry/proto"
)
type PrometheusStorage struct {
// Prometheus metrics
totalClusters prometheus.Gauge
activeClusters prometheus.Gauge
volumeServerCount *prometheus.GaugeVec
totalDiskBytes *prometheus.GaugeVec
totalVolumeCount *prometheus.GaugeVec
filerCount *prometheus.GaugeVec
brokerCount *prometheus.GaugeVec
clusterInfo *prometheus.GaugeVec
telemetryReceived prometheus.Counter
// In-memory storage for API endpoints (if needed)
mu sync.RWMutex
instances map[string]*telemetryData
stats map[string]interface{}
}
// telemetryData is an internal struct that includes the received timestamp
type telemetryData struct {
*proto.TelemetryData
ReceivedAt time.Time `json:"received_at"`
}
func NewPrometheusStorage() *PrometheusStorage {
return &PrometheusStorage{
totalClusters: promauto.NewGauge(prometheus.GaugeOpts{
Name: "seaweedfs_telemetry_total_clusters",
Help: "Total number of unique SeaweedFS clusters (last 30 days)",
}),
activeClusters: promauto.NewGauge(prometheus.GaugeOpts{
Name: "seaweedfs_telemetry_active_clusters",
Help: "Number of active SeaweedFS clusters (last 7 days)",
}),
volumeServerCount: promauto.NewGaugeVec(prometheus.GaugeOpts{
Name: "seaweedfs_telemetry_volume_servers",
Help: "Number of volume servers per cluster",
}, []string{"cluster_id", "version", "os", "deployment"}),
totalDiskBytes: promauto.NewGaugeVec(prometheus.GaugeOpts{
Name: "seaweedfs_telemetry_disk_bytes",
Help: "Total disk usage in bytes per cluster",
}, []string{"cluster_id", "version", "os", "deployment"}),
totalVolumeCount: promauto.NewGaugeVec(prometheus.GaugeOpts{
Name: "seaweedfs_telemetry_volume_count",
Help: "Total number of volumes per cluster",
}, []string{"cluster_id", "version", "os", "deployment"}),
filerCount: promauto.NewGaugeVec(prometheus.GaugeOpts{
Name: "seaweedfs_telemetry_filer_count",
Help: "Number of filer servers per cluster",
}, []string{"cluster_id", "version", "os", "deployment"}),
brokerCount: promauto.NewGaugeVec(prometheus.GaugeOpts{
Name: "seaweedfs_telemetry_broker_count",
Help: "Number of broker servers per cluster",
}, []string{"cluster_id", "version", "os", "deployment"}),
clusterInfo: promauto.NewGaugeVec(prometheus.GaugeOpts{
Name: "seaweedfs_telemetry_cluster_info",
Help: "Cluster information (always 1, labels contain metadata)",
}, []string{"cluster_id", "version", "os", "deployment", "features"}),
telemetryReceived: promauto.NewCounter(prometheus.CounterOpts{
Name: "seaweedfs_telemetry_reports_received_total",
Help: "Total number of telemetry reports received",
}),
instances: make(map[string]*telemetryData),
stats: make(map[string]interface{}),
}
}
func (s *PrometheusStorage) StoreTelemetry(data *proto.TelemetryData) error {
s.mu.Lock()
defer s.mu.Unlock()
// Update Prometheus metrics
labels := prometheus.Labels{
"cluster_id": data.ClusterId,
"version": data.Version,
"os": data.Os,
"deployment": data.Deployment,
}
s.volumeServerCount.With(labels).Set(float64(data.VolumeServerCount))
s.totalDiskBytes.With(labels).Set(float64(data.TotalDiskBytes))
s.totalVolumeCount.With(labels).Set(float64(data.TotalVolumeCount))
s.filerCount.With(labels).Set(float64(data.FilerCount))
s.brokerCount.With(labels).Set(float64(data.BrokerCount))
// Features as JSON string for the label
featuresJSON, _ := json.Marshal(data.Features)
infoLabels := prometheus.Labels{
"cluster_id": data.ClusterId,
"version": data.Version,
"os": data.Os,
"deployment": data.Deployment,
"features": string(featuresJSON),
}
s.clusterInfo.With(infoLabels).Set(1)
s.telemetryReceived.Inc()
// Store in memory for API endpoints
s.instances[data.ClusterId] = &telemetryData{
TelemetryData: data,
ReceivedAt: time.Now().UTC(),
}
// Update aggregated stats
s.updateStats()
return nil
}
func (s *PrometheusStorage) GetStats() (map[string]interface{}, error) {
s.mu.RLock()
defer s.mu.RUnlock()
// Return cached stats
result := make(map[string]interface{})
for k, v := range s.stats {
result[k] = v
}
return result, nil
}
func (s *PrometheusStorage) GetInstances(limit int) ([]*telemetryData, error) {
s.mu.RLock()
defer s.mu.RUnlock()
var instances []*telemetryData
count := 0
for _, instance := range s.instances {
if count >= limit {
break
}
instances = append(instances, instance)
count++
}
return instances, nil
}
func (s *PrometheusStorage) GetMetrics(days int) (map[string]interface{}, error) {
s.mu.RLock()
defer s.mu.RUnlock()
// Return current metrics from in-memory storage
// Historical data should be queried from Prometheus directly
cutoff := time.Now().AddDate(0, 0, -days)
var volumeServers []map[string]interface{}
var diskUsage []map[string]interface{}
for _, instance := range s.instances {
if instance.ReceivedAt.After(cutoff) {
volumeServers = append(volumeServers, map[string]interface{}{
"date": instance.ReceivedAt.Format("2006-01-02"),
"value": instance.TelemetryData.VolumeServerCount,
})
diskUsage = append(diskUsage, map[string]interface{}{
"date": instance.ReceivedAt.Format("2006-01-02"),
"value": instance.TelemetryData.TotalDiskBytes,
})
}
}
return map[string]interface{}{
"volume_servers": volumeServers,
"disk_usage": diskUsage,
}, nil
}
func (s *PrometheusStorage) updateStats() {
now := time.Now()
last7Days := now.AddDate(0, 0, -7)
last30Days := now.AddDate(0, 0, -30)
totalInstances := 0
activeInstances := 0
versions := make(map[string]int)
osDistribution := make(map[string]int)
deployments := make(map[string]int)
for _, instance := range s.instances {
if instance.ReceivedAt.After(last30Days) {
totalInstances++
}
if instance.ReceivedAt.After(last7Days) {
activeInstances++
versions[instance.TelemetryData.Version]++
osDistribution[instance.TelemetryData.Os]++
deployments[instance.TelemetryData.Deployment]++
}
}
// Update Prometheus gauges
s.totalClusters.Set(float64(totalInstances))
s.activeClusters.Set(float64(activeInstances))
// Update cached stats for API
s.stats = map[string]interface{}{
"total_instances": totalInstances,
"active_instances": activeInstances,
"versions": versions,
"os_distribution": osDistribution,
"deployments": deployments,
}
}
// CleanupOldInstances removes instances older than the specified duration
func (s *PrometheusStorage) CleanupOldInstances(maxAge time.Duration) {
s.mu.Lock()
defer s.mu.Unlock()
cutoff := time.Now().Add(-maxAge)
for instanceID, instance := range s.instances {
if instance.ReceivedAt.Before(cutoff) {
delete(s.instances, instanceID)
// Remove from Prometheus metrics
labels := prometheus.Labels{
"cluster_id": instance.TelemetryData.ClusterId,
"version": instance.TelemetryData.Version,
"os": instance.TelemetryData.Os,
"deployment": instance.TelemetryData.Deployment,
}
s.volumeServerCount.Delete(labels)
s.totalDiskBytes.Delete(labels)
s.totalVolumeCount.Delete(labels)
s.filerCount.Delete(labels)
s.brokerCount.Delete(labels)
}
}
s.updateStats()
}
|
