feat(log): 落地统一日志底座与系统日志运维能力

2026-02-12 16:27:29 +08:00
parent a5f29019d9
commit fff1d54858
48 changed files with 4265 additions and 65 deletions
--- a/backend/internal/handler/admin/ops_runtime_logging_handler_test.go
+++ b/backend/internal/handler/admin/ops_runtime_logging_handler_test.go
@@ -0,0 +1,173 @@
+package admin
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+
+	"github.com/Wei-Shaw/sub2api/internal/config"
+	"github.com/Wei-Shaw/sub2api/internal/pkg/logger"
+	"github.com/Wei-Shaw/sub2api/internal/server/middleware"
+	"github.com/Wei-Shaw/sub2api/internal/service"
+	"github.com/gin-gonic/gin"
+)
+
+type testSettingRepo struct {
+	values map[string]string
+}
+
+func newTestSettingRepo() *testSettingRepo {
+	return &testSettingRepo{values: map[string]string{}}
+}
+
+func (s *testSettingRepo) Get(ctx context.Context, key string) (*service.Setting, error) {
+	v, err := s.GetValue(ctx, key)
+	if err != nil {
+		return nil, err
+	}
+	return &service.Setting{Key: key, Value: v}, nil
+}
+func (s *testSettingRepo) GetValue(ctx context.Context, key string) (string, error) {
+	v, ok := s.values[key]
+	if !ok {
+		return "", service.ErrSettingNotFound
+	}
+	return v, nil
+}
+func (s *testSettingRepo) Set(ctx context.Context, key, value string) error {
+	s.values[key] = value
+	return nil
+}
+func (s *testSettingRepo) GetMultiple(ctx context.Context, keys []string) (map[string]string, error) {
+	out := make(map[string]string, len(keys))
+	for _, k := range keys {
+		if v, ok := s.values[k]; ok {
+			out[k] = v
+		}
+	}
+	return out, nil
+}
+func (s *testSettingRepo) SetMultiple(ctx context.Context, settings map[string]string) error {
+	for k, v := range settings {
+		s.values[k] = v
+	}
+	return nil
+}
+func (s *testSettingRepo) GetAll(ctx context.Context) (map[string]string, error) {
+	out := make(map[string]string, len(s.values))
+	for k, v := range s.values {
+		out[k] = v
+	}
+	return out, nil
+}
+func (s *testSettingRepo) Delete(ctx context.Context, key string) error {
+	delete(s.values, key)
+	return nil
+}
+
+func newOpsRuntimeRouter(handler *OpsHandler, withUser bool) *gin.Engine {
+	gin.SetMode(gin.TestMode)
+	r := gin.New()
+	if withUser {
+		r.Use(func(c *gin.Context) {
+			c.Set(string(middleware.ContextKeyUser), middleware.AuthSubject{UserID: 7})
+			c.Next()
+		})
+	}
+	r.GET("/runtime/logging", handler.GetRuntimeLogConfig)
+	r.PUT("/runtime/logging", handler.UpdateRuntimeLogConfig)
+	r.POST("/runtime/logging/reset", handler.ResetRuntimeLogConfig)
+	return r
+}
+
+func newRuntimeOpsService(t *testing.T) *service.OpsService {
+	t.Helper()
+	if err := logger.Init(logger.InitOptions{
+		Level:       "info",
+		Format:      "json",
+		ServiceName: "sub2api",
+		Environment: "test",
+		Output: logger.OutputOptions{
+			ToStdout: false,
+			ToFile:   false,
+		},
+	}); err != nil {
+		t.Fatalf("init logger: %v", err)
+	}
+
+	settingRepo := newTestSettingRepo()
+	cfg := &config.Config{
+		Ops: config.OpsConfig{Enabled: true},
+		Log: config.LogConfig{
+			Level:           "info",
+			Caller:          true,
+			StacktraceLevel: "error",
+			Sampling: config.LogSamplingConfig{
+				Enabled:    false,
+				Initial:    100,
+				Thereafter: 100,
+			},
+		},
+	}
+	return service.NewOpsService(nil, settingRepo, cfg, nil, nil, nil, nil, nil, nil, nil, nil)
+}
+
+func TestOpsRuntimeLoggingHandler_GetConfig(t *testing.T) {
+	h := NewOpsHandler(newRuntimeOpsService(t))
+	r := newOpsRuntimeRouter(h, false)
+
+	w := httptest.NewRecorder()
+	req := httptest.NewRequest(http.MethodGet, "/runtime/logging", nil)
+	r.ServeHTTP(w, req)
+	if w.Code != http.StatusOK {
+		t.Fatalf("status=%d, want 200", w.Code)
+	}
+}
+
+func TestOpsRuntimeLoggingHandler_UpdateUnauthorized(t *testing.T) {
+	h := NewOpsHandler(newRuntimeOpsService(t))
+	r := newOpsRuntimeRouter(h, false)
+
+	body := `{"level":"debug","enable_sampling":false,"sampling_initial":100,"sampling_thereafter":100,"caller":true,"stacktrace_level":"error","retention_days":30}`
+	w := httptest.NewRecorder()
+	req := httptest.NewRequest(http.MethodPut, "/runtime/logging", bytes.NewBufferString(body))
+	req.Header.Set("Content-Type", "application/json")
+	r.ServeHTTP(w, req)
+	if w.Code != http.StatusUnauthorized {
+		t.Fatalf("status=%d, want 401", w.Code)
+	}
+}
+
+func TestOpsRuntimeLoggingHandler_UpdateAndResetSuccess(t *testing.T) {
+	h := NewOpsHandler(newRuntimeOpsService(t))
+	r := newOpsRuntimeRouter(h, true)
+
+	payload := map[string]any{
+		"level":               "debug",
+		"enable_sampling":     false,
+		"sampling_initial":    100,
+		"sampling_thereafter": 100,
+		"caller":              true,
+		"stacktrace_level":    "error",
+		"retention_days":      30,
+	}
+	raw, _ := json.Marshal(payload)
+
+	w := httptest.NewRecorder()
+	req := httptest.NewRequest(http.MethodPut, "/runtime/logging", bytes.NewReader(raw))
+	req.Header.Set("Content-Type", "application/json")
+	r.ServeHTTP(w, req)
+	if w.Code != http.StatusOK {
+		t.Fatalf("update status=%d, want 200, body=%s", w.Code, w.Body.String())
+	}
+
+	w = httptest.NewRecorder()
+	req = httptest.NewRequest(http.MethodPost, "/runtime/logging/reset", nil)
+	r.ServeHTTP(w, req)
+	if w.Code != http.StatusOK {
+		t.Fatalf("reset status=%d, want 200, body=%s", w.Code, w.Body.String())
+	}
+}
--- a/backend/internal/handler/admin/ops_settings_handler.go
+++ b/backend/internal/handler/admin/ops_settings_handler.go
@@ -4,6 +4,7 @@ import (
 	"net/http"

 	"github.com/Wei-Shaw/sub2api/internal/pkg/response"
+	"github.com/Wei-Shaw/sub2api/internal/server/middleware"
 	"github.com/Wei-Shaw/sub2api/internal/service"
 	"github.com/gin-gonic/gin"
 )
@@ -101,6 +102,84 @@ func (h *OpsHandler) UpdateAlertRuntimeSettings(c *gin.Context) {
 	response.Success(c, updated)
 }

+// GetRuntimeLogConfig returns runtime log config (DB-backed).
+// GET /api/v1/admin/ops/runtime/logging
+func (h *OpsHandler) GetRuntimeLogConfig(c *gin.Context) {
+	if h.opsService == nil {
+		response.Error(c, http.StatusServiceUnavailable, "Ops service not available")
+		return
+	}
+	if err := h.opsService.RequireMonitoringEnabled(c.Request.Context()); err != nil {
+		response.ErrorFrom(c, err)
+		return
+	}
+
+	cfg, err := h.opsService.GetRuntimeLogConfig(c.Request.Context())
+	if err != nil {
+		response.Error(c, http.StatusInternalServerError, "Failed to get runtime log config")
+		return
+	}
+	response.Success(c, cfg)
+}
+
+// UpdateRuntimeLogConfig updates runtime log config and applies changes immediately.
+// PUT /api/v1/admin/ops/runtime/logging
+func (h *OpsHandler) UpdateRuntimeLogConfig(c *gin.Context) {
+	if h.opsService == nil {
+		response.Error(c, http.StatusServiceUnavailable, "Ops service not available")
+		return
+	}
+	if err := h.opsService.RequireMonitoringEnabled(c.Request.Context()); err != nil {
+		response.ErrorFrom(c, err)
+		return
+	}
+
+	var req service.OpsRuntimeLogConfig
+	if err := c.ShouldBindJSON(&req); err != nil {
+		response.BadRequest(c, "Invalid request body")
+		return
+	}
+
+	subject, ok := middleware.GetAuthSubjectFromContext(c)
+	if !ok || subject.UserID <= 0 {
+		response.Error(c, http.StatusUnauthorized, "Unauthorized")
+		return
+	}
+
+	updated, err := h.opsService.UpdateRuntimeLogConfig(c.Request.Context(), &req, subject.UserID)
+	if err != nil {
+		response.Error(c, http.StatusBadRequest, err.Error())
+		return
+	}
+	response.Success(c, updated)
+}
+
+// ResetRuntimeLogConfig removes runtime override and falls back to env/yaml baseline.
+// POST /api/v1/admin/ops/runtime/logging/reset
+func (h *OpsHandler) ResetRuntimeLogConfig(c *gin.Context) {
+	if h.opsService == nil {
+		response.Error(c, http.StatusServiceUnavailable, "Ops service not available")
+		return
+	}
+	if err := h.opsService.RequireMonitoringEnabled(c.Request.Context()); err != nil {
+		response.ErrorFrom(c, err)
+		return
+	}
+
+	subject, ok := middleware.GetAuthSubjectFromContext(c)
+	if !ok || subject.UserID <= 0 {
+		response.Error(c, http.StatusUnauthorized, "Unauthorized")
+		return
+	}
+
+	updated, err := h.opsService.ResetRuntimeLogConfig(c.Request.Context(), subject.UserID)
+	if err != nil {
+		response.Error(c, http.StatusBadRequest, err.Error())
+		return
+	}
+	response.Success(c, updated)
+}
+
 // GetAdvancedSettings returns Ops advanced settings (DB-backed).
 // GET /api/v1/admin/ops/advanced-settings
 func (h *OpsHandler) GetAdvancedSettings(c *gin.Context) {
--- a/backend/internal/handler/admin/ops_system_log_handler.go
+++ b/backend/internal/handler/admin/ops_system_log_handler.go
@@ -0,0 +1,174 @@
+package admin
+
+import (
+	"net/http"
+	"strconv"
+	"strings"
+	"time"
+
+	"github.com/Wei-Shaw/sub2api/internal/pkg/response"
+	"github.com/Wei-Shaw/sub2api/internal/server/middleware"
+	"github.com/Wei-Shaw/sub2api/internal/service"
+	"github.com/gin-gonic/gin"
+)
+
+type opsSystemLogCleanupRequest struct {
+	StartTime string `json:"start_time"`
+	EndTime   string `json:"end_time"`
+
+	Level           string `json:"level"`
+	Component       string `json:"component"`
+	RequestID       string `json:"request_id"`
+	ClientRequestID string `json:"client_request_id"`
+	UserID          *int64 `json:"user_id"`
+	AccountID       *int64 `json:"account_id"`
+	Platform        string `json:"platform"`
+	Model           string `json:"model"`
+	Query           string `json:"q"`
+}
+
+// ListSystemLogs returns indexed system logs.
+// GET /api/v1/admin/ops/system-logs
+func (h *OpsHandler) ListSystemLogs(c *gin.Context) {
+	if h.opsService == nil {
+		response.Error(c, http.StatusServiceUnavailable, "Ops service not available")
+		return
+	}
+	if err := h.opsService.RequireMonitoringEnabled(c.Request.Context()); err != nil {
+		response.ErrorFrom(c, err)
+		return
+	}
+
+	page, pageSize := response.ParsePagination(c)
+	if pageSize > 200 {
+		pageSize = 200
+	}
+
+	start, end, err := parseOpsTimeRange(c, "1h")
+	if err != nil {
+		response.BadRequest(c, err.Error())
+		return
+	}
+
+	filter := &service.OpsSystemLogFilter{
+		Page:            page,
+		PageSize:        pageSize,
+		StartTime:       &start,
+		EndTime:         &end,
+		Level:           strings.TrimSpace(c.Query("level")),
+		Component:       strings.TrimSpace(c.Query("component")),
+		RequestID:       strings.TrimSpace(c.Query("request_id")),
+		ClientRequestID: strings.TrimSpace(c.Query("client_request_id")),
+		Platform:        strings.TrimSpace(c.Query("platform")),
+		Model:           strings.TrimSpace(c.Query("model")),
+		Query:           strings.TrimSpace(c.Query("q")),
+	}
+	if v := strings.TrimSpace(c.Query("user_id")); v != "" {
+		id, parseErr := strconv.ParseInt(v, 10, 64)
+		if parseErr != nil || id <= 0 {
+			response.BadRequest(c, "Invalid user_id")
+			return
+		}
+		filter.UserID = &id
+	}
+	if v := strings.TrimSpace(c.Query("account_id")); v != "" {
+		id, parseErr := strconv.ParseInt(v, 10, 64)
+		if parseErr != nil || id <= 0 {
+			response.BadRequest(c, "Invalid account_id")
+			return
+		}
+		filter.AccountID = &id
+	}
+
+	result, err := h.opsService.ListSystemLogs(c.Request.Context(), filter)
+	if err != nil {
+		response.ErrorFrom(c, err)
+		return
+	}
+	response.Paginated(c, result.Logs, int64(result.Total), result.Page, result.PageSize)
+}
+
+// CleanupSystemLogs deletes indexed system logs by filter.
+// POST /api/v1/admin/ops/system-logs/cleanup
+func (h *OpsHandler) CleanupSystemLogs(c *gin.Context) {
+	if h.opsService == nil {
+		response.Error(c, http.StatusServiceUnavailable, "Ops service not available")
+		return
+	}
+	if err := h.opsService.RequireMonitoringEnabled(c.Request.Context()); err != nil {
+		response.ErrorFrom(c, err)
+		return
+	}
+
+	subject, ok := middleware.GetAuthSubjectFromContext(c)
+	if !ok || subject.UserID <= 0 {
+		response.Error(c, http.StatusUnauthorized, "Unauthorized")
+		return
+	}
+
+	var req opsSystemLogCleanupRequest
+	if err := c.ShouldBindJSON(&req); err != nil {
+		response.BadRequest(c, "Invalid request body")
+		return
+	}
+
+	parseTS := func(raw string) (*time.Time, error) {
+		raw = strings.TrimSpace(raw)
+		if raw == "" {
+			return nil, nil
+		}
+		if t, err := time.Parse(time.RFC3339Nano, raw); err == nil {
+			return &t, nil
+		}
+		t, err := time.Parse(time.RFC3339, raw)
+		if err != nil {
+			return nil, err
+		}
+		return &t, nil
+	}
+	start, err := parseTS(req.StartTime)
+	if err != nil {
+		response.BadRequest(c, "Invalid start_time")
+		return
+	}
+	end, err := parseTS(req.EndTime)
+	if err != nil {
+		response.BadRequest(c, "Invalid end_time")
+		return
+	}
+
+	filter := &service.OpsSystemLogCleanupFilter{
+		StartTime:       start,
+		EndTime:         end,
+		Level:           strings.TrimSpace(req.Level),
+		Component:       strings.TrimSpace(req.Component),
+		RequestID:       strings.TrimSpace(req.RequestID),
+		ClientRequestID: strings.TrimSpace(req.ClientRequestID),
+		UserID:          req.UserID,
+		AccountID:       req.AccountID,
+		Platform:        strings.TrimSpace(req.Platform),
+		Model:           strings.TrimSpace(req.Model),
+		Query:           strings.TrimSpace(req.Query),
+	}
+
+	deleted, err := h.opsService.CleanupSystemLogs(c.Request.Context(), filter, subject.UserID)
+	if err != nil {
+		response.ErrorFrom(c, err)
+		return
+	}
+	response.Success(c, gin.H{"deleted": deleted})
+}
+
+// GetSystemLogIngestionHealth returns sink health metrics.
+// GET /api/v1/admin/ops/system-logs/health
+func (h *OpsHandler) GetSystemLogIngestionHealth(c *gin.Context) {
+	if h.opsService == nil {
+		response.Error(c, http.StatusServiceUnavailable, "Ops service not available")
+		return
+	}
+	if err := h.opsService.RequireMonitoringEnabled(c.Request.Context()); err != nil {
+		response.ErrorFrom(c, err)
+		return
+	}
+	response.Success(c, h.opsService.GetSystemLogSinkHealth())
+}