110902ad4b
Add HealthService with Liveness (no-op) and Readiness (DB+Redis ping
with per-component timeout) checks. Expose three endpoints:
- /healthz : new liveness endpoint, zero-dependency, always 200
- /ready : new readiness endpoint, returns 503 with details on dep
failure; suitable for K8s readinessProbe and load balancers
- /health : preserved for backward compatibility, equivalent to
/healthz
Switch primary docker-compose healthcheck to /ready so the container
is only marked healthy once DB+Redis are reachable. Standalone/dev/
local compose files keep /health to avoid disrupting existing setups.
Tests: unit tests cover liveness, readiness with both deps healthy,
each dep failing independently, and per-component timeout enforcement.
62 lines
2.0 KiB
Go
62 lines
2.0 KiB
Go
package routes
|
||
|
||
import (
|
||
"context"
|
||
"net/http"
|
||
"time"
|
||
|
||
"github.com/Wei-Shaw/sub2api/internal/service"
|
||
"github.com/gin-gonic/gin"
|
||
)
|
||
|
||
// readinessHandlerTimeout 限定 readiness 端点对外的最大返回耗时。
|
||
// HealthService 内部对每个组件再有独立超时，所以这里给宽一点即可。
|
||
const readinessHandlerTimeout = 3 * time.Second
|
||
|
||
// RegisterCommonRoutes 注册通用路由（健康检查、状态等）。
|
||
//
|
||
// 健康端点的语义分层：
|
||
// - /healthz : liveness 探针。零依赖、永远 200。容器/进程探活专用。
|
||
// - /ready : readiness 探针。检查 DB+Redis；任一失败返回 503。
|
||
// - /health : 历史端点，等价于 /healthz，保留向后兼容。
|
||
//
|
||
// dashboard 用的"业务健康分"由 ops_health_score 单独提供，与本路由无关。
|
||
func RegisterCommonRoutes(r *gin.Engine, healthService *service.HealthService) {
|
||
// Liveness：仅证明进程在响应。
|
||
livenessHandler := func(c *gin.Context) {
|
||
_ = healthService.Liveness()
|
||
c.JSON(http.StatusOK, gin.H{"status": "ok"})
|
||
}
|
||
r.GET("/healthz", livenessHandler)
|
||
r.GET("/health", livenessHandler) // 向后兼容旧的 docker-compose healthcheck
|
||
|
||
// Readiness：检查关键依赖。失败时返回 503 但仍带详情，便于排障。
|
||
r.GET("/ready", func(c *gin.Context) {
|
||
ctx, cancel := context.WithTimeout(c.Request.Context(), readinessHandlerTimeout)
|
||
defer cancel()
|
||
report := healthService.Readiness(ctx)
|
||
status := http.StatusOK
|
||
if !report.OK {
|
||
status = http.StatusServiceUnavailable
|
||
}
|
||
c.JSON(status, report)
|
||
})
|
||
|
||
// Claude Code 遥测日志（忽略，直接返回200）
|
||
r.POST("/api/event_logging/batch", func(c *gin.Context) {
|
||
c.Status(http.StatusOK)
|
||
})
|
||
|
||
// Setup status endpoint (always returns needs_setup: false in normal mode)
|
||
// This is used by the frontend to detect when the service has restarted after setup
|
||
r.GET("/setup/status", func(c *gin.Context) {
|
||
c.JSON(http.StatusOK, gin.H{
|
||
"code": 0,
|
||
"data": gin.H{
|
||
"needs_setup": false,
|
||
"step": "completed",
|
||
},
|
||
})
|
||
})
|
||
}
|