yml/kefu_cloud
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

接入运维真实 metrics：负载、请求错误率与服务探测

Browse Source 
- 新增 GET /admin/ops/metrics 与进程内请求统计中间件
- 运维页展示 CPU/内存/磁盘、24h 曲线与 WS/DB/存储状态
- 去掉未使用的消息队列占位项
This commit is contained in:
yml2213
2026-07-15 15:01:18 +08:00
parent a1336c630b
commit 9251815695
14 changed files with 797 additions and 248 deletions
Download Patch File Download Diff File Expand all files Collapse all files
server/cmd/main.go
+3
View File
@@ -8,6 +8,7 @@ import (
"github.com/gin-gonic/gin"
"kefu-sys/server/internal/config"
"kefu-sys/server/internal/handler"
"kefu-sys/server/internal/metrics"
"kefu-sys/server/internal/middleware"
"kefu-sys/server/internal/model"
"kefu-sys/server/internal/storage"
@@ -43,6 +44,8 @@ func main() {
gin.SetMode(cfg.Server.Mode)
r := gin.Default()
// 全站请求统计（运维监控 24h 曲线）
r.Use(metrics.Middleware())
// CORS
r.Use(func(c *gin.Context) {
server/go.mod
+8
View File
@@ -9,6 +9,7 @@ require (
github.com/golang-jwt/jwt/v5 v5.2.1
github.com/gorilla/websocket v1.5.3
github.com/minio/minio-go/v7 v7.2.1
github.com/shirou/gopsutil/v4 v4.25.1
golang.org/x/crypto v0.51.0
gorm.io/driver/postgres v1.5.9
gorm.io/driver/sqlite v1.5.7
@@ -22,8 +23,10 @@ require (
github.com/cloudwego/base64x v0.1.4 // indirect
github.com/cloudwego/iasm v0.2.0 // indirect
github.com/dustin/go-humanize v1.0.1 // indirect
github.com/ebitengine/purego v0.8.2 // indirect
github.com/gabriel-vasile/mimetype v1.4.3 // indirect
github.com/gin-contrib/sse v0.1.0 // indirect
github.com/go-ole/go-ole v1.2.6 // indirect
github.com/go-playground/locales v0.14.1 // indirect
github.com/go-playground/universal-translator v0.18.1 // indirect
github.com/go-playground/validator/v10 v10.20.0 // indirect
@@ -41,6 +44,7 @@ require (
github.com/klauspost/crc32 v1.3.0 // indirect
github.com/kr/text v0.2.0 // indirect
github.com/leodido/go-urn v1.4.0 // indirect
github.com/lufia/plan9stats v0.0.0-20211012122336-39d0f177ccd0 // indirect
github.com/mattn/go-isatty v0.0.20 // indirect
github.com/mattn/go-sqlite3 v1.14.22 // indirect
github.com/minio/crc64nvme v1.1.1 // indirect
@@ -49,11 +53,15 @@ require (
github.com/modern-go/reflect2 v1.0.2 // indirect
github.com/pelletier/go-toml/v2 v2.3.1 // indirect
github.com/philhofer/fwd v1.2.0 // indirect
github.com/power-devops/perfstat v0.0.0-20210106213030-5aafc221ea8c // indirect
github.com/rogpeppe/go-internal v1.15.0 // indirect
github.com/rs/xid v1.6.0 // indirect
github.com/tinylib/msgp v1.6.1 // indirect
github.com/tklauser/go-sysconf v0.3.12 // indirect
github.com/tklauser/numcpus v0.6.1 // indirect
github.com/twitchyliquid64/golang-asm v0.15.1 // indirect
github.com/ugorji/go/codec v1.2.12 // indirect
github.com/yusufpapurcu/wmi v1.2.4 // indirect
github.com/zeebo/xxh3 v1.1.0 // indirect
go.yaml.in/yaml/v3 v3.0.4 // indirect
golang.org/x/arch v0.8.0 // indirect
server/go.sum
+22
View File
@@ -18,12 +18,16 @@ github.com/disintegration/imaging v1.6.2 h1:w1LecBlG2Lnp8B3jk5zSuNqd7b4DXhcjwek1
github.com/disintegration/imaging v1.6.2/go.mod h1:44/5580QXChDfwIclfc/PCwrr44amcmDAg8hxG0Ewe4=
github.com/dustin/go-humanize v1.0.1 h1:GzkhY7T5VNhEkwH0PVJgjz+fX1rhBrR7pRT3mDkpeCY=
github.com/dustin/go-humanize v1.0.1/go.mod h1:Mu1zIs6XwVuF/gI1OepvI0qD18qycQx+mFykh5fBlto=
github.com/ebitengine/purego v0.8.2 h1:jPPGWs2sZ1UgOSgD2bClL0MJIqu58nOmIcBuXr62z1I=
github.com/ebitengine/purego v0.8.2/go.mod h1:iIjxzd6CiRiOG0UyXP+V1+jWqUXVjPKLAI0mRfJZTmQ=
github.com/gabriel-vasile/mimetype v1.4.3 h1:in2uUcidCuFcDKtdcBxlR0rJ1+fsokWf+uqxgUFjbI0=
github.com/gabriel-vasile/mimetype v1.4.3/go.mod h1:d8uq/6HKRL6CGdk+aubisF/M5GcPfT7nKyLpA0lbSSk=
github.com/gin-contrib/sse v0.1.0 h1:Y/yl/+YNO8GZSjAhjMsSuLt29uWRFHdHYUb5lYOV9qE=
github.com/gin-contrib/sse v0.1.0/go.mod h1:RHrZQHXnP2xjPF+u1gW/2HnVO7nvIa9PG3Gm+fLHvGI=
github.com/gin-gonic/gin v1.10.0 h1:nTuyha1TYqgedzytsKYqna+DfLos46nTv2ygFy86HFU=
github.com/gin-gonic/gin v1.10.0/go.mod h1:4PMNQiOhvDRa013RKVbsiNwoyezlm2rm0uX/T7kzp5Y=
github.com/go-ole/go-ole v1.2.6 h1:/Fpf6oFPoeFik9ty7siob0G6Ke8QvQEuVcuChpwXzpY=
github.com/go-ole/go-ole v1.2.6/go.mod h1:pprOEPIfldk/42T2oK7lQ4v4JSDwmV0As9GaiUsvbm0=
github.com/go-playground/assert/v2 v2.2.0 h1:JvknZsQTYeFEAhQwI4qEt9cyV5ONwRHC+lYKSsYSR8s=
github.com/go-playground/assert/v2 v2.2.0/go.mod h1:VDjEfimB/XKnb+ZQfWdccd7VUvScMdVu0Titje2rxJ4=
github.com/go-playground/locales v0.14.1 h1:EWaQ/wswjilfKLTECiXz7Rh+3BjFhfDFKv/oXslEjJA=
@@ -36,6 +40,7 @@ github.com/goccy/go-json v0.10.2 h1:CrxCmQqYDkv1z7lO7Wbh2HN93uovUHgrECaO5ZrCXAU=
github.com/goccy/go-json v0.10.2/go.mod h1:6MelG93GURQebXPDq3khkgXZkazVtN9CRI+MGFi0w8I=
github.com/golang-jwt/jwt/v5 v5.2.1 h1:OuVbFODueb089Lh128TAcimifWaLhJwVflnrgM17wHk=
github.com/golang-jwt/jwt/v5 v5.2.1/go.mod h1:pqrtFR0X4osieyHYxtmOUWsAWrfe1Q5UVIyoH402zdk=
github.com/google/go-cmp v0.5.6/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
github.com/google/go-cmp v0.7.0 h1:wk8382ETsv4JYUZwIsn6YpYiWiBsYLSJiTsyBybVuN8=
github.com/google/go-cmp v0.7.0/go.mod h1:pXiqmnSA92OHEEa9HXL2W4E7lf9JzCmGVUdgjX3N/iU=
github.com/google/gofuzz v1.0.0/go.mod h1:dBl0BpW6vV/+mYPU4Po3pmUjxk6FQPldtuIdl/M65Eg=
@@ -72,6 +77,8 @@ github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
github.com/kr/text v0.2.0/go.mod h1:eLer722TekiGuMkidMxC/pM04lWEeraHUUmBw8l2grE=
github.com/leodido/go-urn v1.4.0 h1:WT9HwE9SGECu3lg4d/dIA+jxlljEa1/ffXKmRjqdmIQ=
github.com/leodido/go-urn v1.4.0/go.mod h1:bvxc+MVxLKB4z00jd1z+Dvzr47oO32F/QSNjSBOlFxI=
github.com/lufia/plan9stats v0.0.0-20211012122336-39d0f177ccd0 h1:6E+4a0GO5zZEnZ81pIr0yLvtUWk2if982qA3F3QD6H4=
github.com/lufia/plan9stats v0.0.0-20211012122336-39d0f177ccd0/go.mod h1:zJYVVT2jmtg6P3p1VtQj7WsuWi/y4VnjVBn7F8KPB3I=
github.com/mattn/go-isatty v0.0.20 h1:xfD0iDuEKnDkl03q4limB+vH+GxLEtL/jb4xVJSWWEY=
github.com/mattn/go-isatty v0.0.20/go.mod h1:W+V8PltTTMOvKvAeJH7IuucS94S2C6jfK/D7dTCTo3Y=
github.com/mattn/go-sqlite3 v1.14.22 h1:2gZY6PC6kBnID23Tichd1K+Z0oS6nE/XwU+Vz/5o4kU=
@@ -93,10 +100,14 @@ github.com/philhofer/fwd v1.2.0 h1:e6DnBTl7vGY+Gz322/ASL4Gyp1FspeMvx1RNDoToZuM=
github.com/philhofer/fwd v1.2.0/go.mod h1:RqIHx9QI14HlwKwm98g9Re5prTQ6LdeRQn+gXJFxsJM=
github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
github.com/power-devops/perfstat v0.0.0-20210106213030-5aafc221ea8c h1:ncq/mPwQF4JjgDlrVEn3C11VoGHZN7m8qihwgMEtzYw=
github.com/power-devops/perfstat v0.0.0-20210106213030-5aafc221ea8c/go.mod h1:OmDBASR4679mdNQnz2pUhc2G8CO2JrUAVFDRBDP/hJE=
github.com/rogpeppe/go-internal v1.15.0 h1:D0RCU5rMAp+SpgkiNdrjfJ+LX4J1M32V2NeCY7EJ6hc=
github.com/rogpeppe/go-internal v1.15.0/go.mod h1:DrUVZyrJU+txYW5/1kwtXQSMFio52ZOxX7yM1VHvnxs=
github.com/rs/xid v1.6.0 h1:fV591PaemRlL6JfRxGDEPl69wICngIQ3shQtzfy2gxU=
github.com/rs/xid v1.6.0/go.mod h1:7XoLgs4eV+QndskICGsho+ADou8ySMSjJKDIan90Nz0=
github.com/shirou/gopsutil/v4 v4.25.1 h1:QSWkTc+fu9LTAWfkZwZ6j8MSUk4A2LV7rbH0ZqmLjXs=
github.com/shirou/gopsutil/v4 v4.25.1/go.mod h1:RoUCUpndaJFtT+2zsZzzmhvbfGoDCJ7nFXKJf8GqJbI=
github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
github.com/stretchr/objx v0.4.0/go.mod h1:YvHI0jy2hoMjB+UWwv71VJQ9isScKT/TqJzVSSt89Yw=
github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpEOglKo=
@@ -111,10 +122,16 @@ github.com/stretchr/testify v1.11.1 h1:7s2iGBzp5EwR7/aIZr8ao5+dra3wiQyKjjFuvgVKu
github.com/stretchr/testify v1.11.1/go.mod h1:wZwfW3scLgRK+23gO65QZefKpKQRnfz6sD981Nm4B6U=
github.com/tinylib/msgp v1.6.1 h1:ESRv8eL3u+DNHUoSAAQRE50Hm162zqAnBoGv9PzScPY=
github.com/tinylib/msgp v1.6.1/go.mod h1:RSp0LW9oSxFut3KzESt5Voq4GVWyS+PSulT77roAqEA=
github.com/tklauser/go-sysconf v0.3.12 h1:0QaGUFOdQaIVdPgfITYzaTegZvdCjmYO52cSFAEVmqU=
github.com/tklauser/go-sysconf v0.3.12/go.mod h1:Ho14jnntGE1fpdOqQEEaiKRpvIavV0hSfmBq8nJbHYI=
github.com/tklauser/numcpus v0.6.1 h1:ng9scYS7az0Bk4OZLvrNXNSAO2Pxr1XXRAPyjhIx+Fk=
github.com/tklauser/numcpus v0.6.1/go.mod h1:1XfjsgE2zo8GVw7POkMbHENHzVg3GzmoZ9fESEdAacY=
github.com/twitchyliquid64/golang-asm v0.15.1 h1:SU5vSMR7hnwNxj24w34ZyCi/FmDZTkS4MhqMhdFk5YI=
github.com/twitchyliquid64/golang-asm v0.15.1/go.mod h1:a1lVb/DtPvCB8fslRZhAngC2+aY1QWCk3Cedj/Gdt08=
github.com/ugorji/go/codec v1.2.12 h1:9LC83zGrHhuUA9l16C9AHXAqEV/2wBQ4nkvumAE65EE=
github.com/ugorji/go/codec v1.2.12/go.mod h1:UNopzCgEMSXjBc6AOMqYvWC1ktqTAfzJZUZgYf6w6lg=
github.com/yusufpapurcu/wmi v1.2.4 h1:zFUKzehAFReQwLys1b/iSMl+JQGSCSjtVqQn9bBrPo0=
github.com/yusufpapurcu/wmi v1.2.4/go.mod h1:SBZ9tNy3G9/m5Oi98Zks0QjeHVDvuK0qfxQmPyzfmi0=
github.com/zeebo/assert v1.3.0 h1:g7C04CbJuIDKNPFHmsk4hwZDO5O+kntRxzaUoNXj+IQ=
github.com/zeebo/assert v1.3.0/go.mod h1:Pq9JiuJQpG8JLJdtkwrJESF0Foym2/D9XMU5ciN/wJ0=
github.com/zeebo/xxh3 v1.1.0 h1:s7DLGDK45Dyfg7++yxI0khrfwq9661w9EN78eP/UZVs=
@@ -133,7 +150,11 @@ golang.org/x/net v0.53.0 h1:d+qAbo5L0orcWAr0a9JweQpjXF19LMXJE8Ey7hwOdUA=
golang.org/x/net v0.53.0/go.mod h1:JvMuJH7rrdiCfbeHoo3fCQU24Lf5JJwT9W3sJFulfgs=
golang.org/x/sync v0.20.0 h1:e0PTpb7pjO8GAtTs2dQ6jYa5BWYlMuX047Dco/pItO4=
golang.org/x/sync v0.20.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
golang.org/x/sys v0.0.0-20190916202348-b4ddaad3f8a3/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
golang.org/x/sys v0.0.0-20201204225414-ed752295db88/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
golang.org/x/sys v0.8.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
golang.org/x/sys v0.11.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
golang.org/x/sys v0.44.0 h1:ildZl3J4uzeKP07r2F++Op7E9B29JRUy+a27EibtBTQ=
golang.org/x/sys v0.44.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
@@ -141,6 +162,7 @@ golang.org/x/text v0.3.6/go.mod h1:5Zoc/QRtKVWzQhOtBMvqHzDpF6irO9z98xDceosuGiQ=
golang.org/x/text v0.37.0 h1:Cqjiwd9eSg8e0QAkyCaQTNHFIIzWtidPahFWR83rTrc=
golang.org/x/text v0.37.0/go.mod h1:a5sjxXGs9hsn/AJVwuElvCAo9v8QYLzvavO5z2PiM38=
golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
google.golang.org/protobuf v1.36.10 h1:AYd7cD/uASjIL6Q9LiTjz8JLcrh/88q5UObnmY3aOOE=
google.golang.org/protobuf v1.36.10/go.mod h1:HTf+CrKn2C3g5S8VImy6tdcUvCska2kB7j23XfzDpco=
gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
server/internal/handler/admin.go
+7 -2
View File
@@ -10,11 +10,16 @@ import (
"golang.org/x/crypto/bcrypt"
"kefu-sys/server/internal/middleware"
"kefu-sys/server/internal/model"
"kefu-sys/server/internal/storage"
)
type AdminHandler struct{}
type AdminHandler struct {
store storage.ObjectStorage
}
func NewAdminHandler() *AdminHandler { return &AdminHandler{} }
func NewAdminHandler(store storage.ObjectStorage) *AdminHandler {
return &AdminHandler{store: store}
}
func (h *AdminHandler) Stats(c *gin.Context) {
var tenantTotal, activeTotal, suspendedTotal, expiringTotal, newThisMonth int64
server/internal/handler/ops_metrics.go
+230
View File
@@ -0,0 +1,230 @@
package handler
import (
"context"
"fmt"
"runtime"
"time"
"github.com/gin-gonic/gin"
"github.com/shirou/gopsutil/v4/cpu"
"github.com/shirou/gopsutil/v4/disk"
"github.com/shirou/gopsutil/v4/mem"
"kefu-sys/server/internal/metrics"
"kefu-sys/server/internal/middleware"
"kefu-sys/server/internal/model"
"kefu-sys/server/internal/storage"
"kefu-sys/server/internal/ws"
)
// OpsMetrics 平台运维实时指标（当前 API 实例视角）。
// GET /api/admin/ops/metrics
func (h *AdminHandler) OpsMetrics(c *gin.Context) {
now := time.Now()
// —— 系统负载 ——
load := collectSystemLoad()
// —— 请求统计 ——
reqSeries, errSeries, hourLabels, winReq, winErr := metrics.Default.Snapshot24h()
totalReq, totalErr, total5xx, startedAt := metrics.Default.Totals()
var errNow float64
if winReq > 0 {
errNow = float64(winErr) / float64(winReq) * 100
}
var qps float64
uptimeSec := now.Sub(startedAt).Seconds()
if uptimeSec > 0 {
qps = float64(totalReq) / uptimeSec
}
// —— 服务探测 ——
services := collectServiceHealth(h.store)
// —— 整体状态 ——
overall := "normal"
for _, s := range services {
if st, _ := s["status"].(string); st == "error" {
overall = "error"
break
}
if st, _ := s["status"].(string); st == "warning" && overall == "normal" {
overall = "warning"
}
}
if load.CPUPercent >= 90 || load.MemPercent >= 90 || load.DiskPercent >= 95 {
if overall == "normal" {
overall = "warning"
}
}
if errNow >= 5 {
overall = "error"
} else if errNow >= 2 && overall == "normal" {
overall = "warning"
}
wsTotal, wsAgents, wsVisitors := ws.DefaultHub.Stats()
middleware.JSON(c, gin.H{
"overall": overall,
"updated_at": now.Format(time.RFC3339),
"instance": "api", // 当前进程实例
"uptime_sec": int64(uptimeSec),
"services": services,
"load": gin.H{
"cpu_percent": load.CPUPercent,
"mem_percent": load.MemPercent,
"disk_percent": load.DiskPercent,
"cpu_cores": load.CPUCores,
"mem_total_mb": load.MemTotalMB,
"mem_used_mb": load.MemUsedMB,
"disk_total_gb": load.DiskTotalGB,
"disk_used_gb": load.DiskUsedGB,
"go_goroutines": runtime.NumGoroutine(),
"go_heap_mb": load.GoHeapMB,
},
"api": gin.H{
"requests_last_24h": reqSeries,
"error_rate_last_24h": errSeries,
"hours": hourLabels,
"qps": round2(qps),
"error_rate_now": round2(errNow),
"window_requests": winReq,
"window_errors": winErr,
"total_requests": totalReq,
"total_errors": totalErr,
"total_5xx": total5xx,
"in_flight": metrics.Default.InFlight(),
},
"websocket": gin.H{
"connections": wsTotal,
"agents": wsAgents,
"visitors": wsVisitors,
},
})
}
type systemLoad struct {
CPUPercent float64
MemPercent float64
DiskPercent float64
CPUCores int
MemTotalMB uint64
MemUsedMB uint64
DiskTotalGB float64
DiskUsedGB float64
GoHeapMB float64
}
func collectSystemLoad() systemLoad {
out := systemLoad{
CPUCores: runtime.NumCPU(),
}
var ms runtime.MemStats
runtime.ReadMemStats(&ms)
out.GoHeapMB = round2(float64(ms.HeapAlloc) / 1024 / 1024)
if percents, err := cpu.Percent(120*time.Millisecond, false); err == nil && len(percents) > 0 {
out.CPUPercent = round2(percents[0])
}
if vm, err := mem.VirtualMemory(); err == nil {
out.MemPercent = round2(vm.UsedPercent)
out.MemTotalMB = vm.Total / 1024 / 1024
out.MemUsedMB = vm.Used / 1024 / 1024
}
if du, err := disk.Usage("/"); err == nil {
out.DiskPercent = round2(du.UsedPercent)
out.DiskTotalGB = round2(float64(du.Total) / 1024 / 1024 / 1024)
out.DiskUsedGB = round2(float64(du.Used) / 1024 / 1024 / 1024)
}
return out
}
func collectServiceHealth(store storage.ObjectStorage) []gin.H {
services := make([]gin.H, 0, 5)
// API：本进程存活
services = append(services, gin.H{
"name": "API服务",
"status": "normal",
"detail": "进程运行中",
"latency_ms": 0,
})
// WebSocket Hub
wsTotal, wsAgents, wsVisitors := ws.DefaultHub.Stats()
services = append(services, gin.H{
"name": "WebSocket",
"status": "normal",
"detail": fmt.Sprintf("连接 %d（坐席 %d / 访客 %d）", wsTotal, wsAgents, wsVisitors),
"connections": wsTotal,
})
// 数据库
dbStatus, dbDetail, dbLatency := probeDatabase()
services = append(services, gin.H{
"name": "数据库",
"status": dbStatus,
"detail": dbDetail,
"latency_ms": dbLatency,
})
// 对象存储
stStatus, stDetail, stLatency := probeStorage(store)
services = append(services, gin.H{
"name": "对象存储",
"status": stStatus,
"detail": stDetail,
"latency_ms": stLatency,
})
return services
}
func probeDatabase() (status, detail string, latencyMs int64) {
if model.DB == nil {
return "error", "数据库未初始化", 0
}
sqlDB, err := model.DB.DB()
if err != nil {
return "error", "获取连接失败: " + err.Error(), 0
}
ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
defer cancel()
start := time.Now()
if err := sqlDB.PingContext(ctx); err != nil {
return "error", "Ping 失败: " + err.Error(), time.Since(start).Milliseconds()
}
lat := time.Since(start).Milliseconds()
status = "normal"
detail = fmt.Sprintf("Ping 正常 · %dms", lat)
if lat >= 200 {
status = "warning"
detail = fmt.Sprintf("Ping 偏慢 · %dms", lat)
}
return status, detail, lat
}
func probeStorage(store storage.ObjectStorage) (status, detail string, latencyMs int64) {
if store == nil {
return "warning", "未配置对象存储", 0
}
ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
defer cancel()
start := time.Now()
if err := store.HealthCheck(ctx); err != nil {
return "error", "不可达: " + err.Error(), time.Since(start).Milliseconds()
}
lat := time.Since(start).Milliseconds()
status = "normal"
detail = fmt.Sprintf("Bucket 可达 · %dms", lat)
if lat >= 500 {
status = "warning"
detail = fmt.Sprintf("探测偏慢 · %dms", lat)
}
return status, detail, lat
}
func round2(v float64) float64 {
return float64(int(v*100+0.5)) / 100
}
server/internal/handler/router.go
+2 -1
View File
@@ -13,7 +13,7 @@ func SetupRoutes(r *gin.Engine, store storage.ObjectStorage, storageCfg config.S
customer := NewCustomerHandler()
knowledge := NewKnowledgeHandler()
stats := NewStatisticsHandler()
admin := NewAdminHandler()
admin := NewAdminHandler(store)
channel := NewChannelHandler()
settings := NewSettingsHandler()
staff := NewStaffHandler()
@@ -116,6 +116,7 @@ func SetupRoutes(r *gin.Engine, store storage.ObjectStorage, storageCfg config.S
adminGroup.Use(middleware.PlatformRequired())
{
adminGroup.GET("/stats", admin.Stats)
adminGroup.GET("/ops/metrics", admin.OpsMetrics)
adminGroup.GET("/tenants", admin.ListTenants)
adminGroup.GET("/tenants/:id", admin.GetTenant)
adminGroup.POST("/tenants", admin.CreateTenant)
server/internal/metrics/metrics.go
+140
View File
@@ -0,0 +1,140 @@
package metrics
import (
"sync"
"sync/atomic"
"time"
"github.com/gin-gonic/gin"
)
// Collector 进程内请求统计：近 24 小时按整点小时桶累加。
type Collector struct {
mu sync.Mutex
byHour map[int64]*hourBucket
// 自进程启动以来的累计
totalReq atomic.Int64
totalErr atomic.Int64 // HTTP status >= 400
total5xx atomic.Int64
inFlight atomic.Int64
startedAt time.Time
}
type hourBucket struct {
requests int64
errors int64
status5 int64
}
// Default 全局采集器
var Default = NewCollector()
func NewCollector() *Collector {
return &Collector{
byHour: make(map[int64]*hourBucket),
startedAt: time.Now(),
}
}
func hourKey(t time.Time) int64 {
return t.Unix() / 3600
}
func (c *Collector) Record(status int) {
c.totalReq.Add(1)
isErr := status >= 400
is5 := status >= 500
if isErr {
c.totalErr.Add(1)
}
if is5 {
c.total5xx.Add(1)
}
h := hourKey(time.Now())
c.mu.Lock()
b := c.byHour[h]
if b == nil {
b = &hourBucket{}
c.byHour[h] = b
c.pruneLocked(h)
}
b.requests++
if isErr {
b.errors++
}
if is5 {
b.status5++
}
c.mu.Unlock()
}
func (c *Collector) pruneLocked(nowHour int64) {
cutoff := nowHour - 25
for k := range c.byHour {
if k < cutoff {
delete(c.byHour, k)
}
}
}
// Snapshot24h 返回近 24 个整点小时（含当前小时）的请求量与错误率（0–100）。
// 下标 0 为 23 小时前，23 为当前小时；hours 为当地小时 0–23。
func (c *Collector) Snapshot24h() (requests []int64, errorRates []float64, hours []int, totalReq, totalErr int64) {
now := time.Now()
nowH := hourKey(now)
loc := now.Location()
requests = make([]int64, 24)
errorRates = make([]float64, 24)
hours = make([]int, 24)
c.mu.Lock()
defer c.mu.Unlock()
for i := 0; i < 24; i++ {
hKey := nowH - int64(23-i)
t := time.Unix(hKey*3600, 0).In(loc)
hours[i] = t.Hour()
b := c.byHour[hKey]
var r, e int64
if b != nil {
r, e = b.requests, b.errors
}
requests[i] = r
if r > 0 {
errorRates[i] = float64(e) / float64(r) * 100
} else {
errorRates[i] = 0
}
totalReq += r
totalErr += e
}
return
}
func (c *Collector) Totals() (req, err4, err5 int64, startedAt time.Time) {
return c.totalReq.Load(), c.totalErr.Load(), c.total5xx.Load(), c.startedAt
}
func (c *Collector) InFlight() int64 {
return c.inFlight.Load()
}
// Middleware 记录每个 HTTP 请求的状态码（跳过 WebSocket 升级失败后的统计由 status 决定）。
func Middleware() gin.HandlerFunc {
return func(c *gin.Context) {
// 健康检查可不计入业务错误曲线，但仍记请求量更直观；这里全部计入
Default.inFlight.Add(1)
c.Next()
Default.inFlight.Add(-1)
// WebSocket 升级成功后 status 可能是 101
status := c.Writer.Status()
if status == 0 {
status = 200
}
Default.Record(status)
}
}
server/internal/metrics/metrics_test.go
+32
View File
@@ -0,0 +1,32 @@
package metrics
import (
"testing"
)
func TestRecordAndSnapshot(t *testing.T) {
c := NewCollector()
c.Record(200)
c.Record(200)
c.Record(500)
c.Record(404)
reqs, rates, hours, totalReq, totalErr := c.Snapshot24h()
if len(reqs) != 24 || len(rates) != 24 || len(hours) != 24 {
t.Fatalf("snapshot 长度应为 24: req=%d rates=%d hours=%d", len(reqs), len(rates), len(hours))
}
if totalReq != 4 || totalErr != 2 {
t.Fatalf("窗口统计错误: totalReq=%d totalErr=%d", totalReq, totalErr)
}
// 当前小时桶应有 4 请求
if reqs[23] != 4 {
t.Fatalf("当前小时请求量 = %d, 期望 4", reqs[23])
}
if rates[23] < 49 || rates[23] > 51 {
t.Fatalf("当前小时错误率 = %v, 期望约 50", rates[23])
}
tr, te, t5, _ := c.Totals()
if tr != 4 || te != 2 || t5 != 1 {
t.Fatalf("累计统计错误: req=%d err=%d 5xx=%d", tr, te, t5)
}
}
server/internal/storage/memory.go
+2
View File
@@ -44,6 +44,8 @@ func (m *MemoryStorage) PublicBase() string { return m.base }
func (m *MemoryStorage) EnsureBucket(context.Context) error { return nil }
func (m *MemoryStorage) HealthCheck(context.Context) error { return nil }
func (m *MemoryStorage) Get(key string) ([]byte, error) {
m.mu.RLock()
defer m.mu.RUnlock()
server/internal/storage/s3.go
+5
View File
@@ -40,6 +40,11 @@ func NewS3(cfg config.StorageConfig) (*S3Storage, error) {
return &S3Storage{client: client, bucket: cfg.Bucket, base: base}, nil
}
func (s *S3Storage) HealthCheck(ctx context.Context) error {
_, err := s.client.BucketExists(ctx, s.bucket)
return err
}
func (s *S3Storage) EnsureBucket(ctx context.Context) error {
exists, err := s.client.BucketExists(ctx, s.bucket)
if err != nil {
server/internal/storage/storage.go
+2
View File
@@ -17,6 +17,8 @@ type ObjectStorage interface {
PublicBase() string
// EnsureBucket 开发环境自动建桶
EnsureBucket(ctx context.Context) error
// HealthCheck 探测存储是否可达（运维监控用）
HealthCheck(ctx context.Context) error
}
// PutResult 上传结果
server/internal/ws/ws.go
+15
View File
@@ -83,6 +83,21 @@ func (h *Hub) Run() {
}
}
// Stats 返回当前连接统计（总连接 / 坐席 / 访客）。
func (h *Hub) Stats() (total, agents, visitors int) {
h.mu.RLock()
defer h.mu.RUnlock()
for client := range h.clients {
total++
if client.Kind == "visitor" {
visitors++
} else {
agents++
}
}
return
}
func (h *Hub) send(client *Client, message []byte) {
select {
case client.Send <- message: