feat(ops): 实现上游错误事件记录与查询功能

**新增功能**:
- 新建ops_upstream_error_events表存储上游服务错误详情
- 支持记录上游429/529/5xx错误的详细上下文信息
- 提供按时间范围查询上游错误事件的API

**后端改动**:
1. 模型层(ops_models.go, ops_port.go):
   - 新增UpstreamErrorEvent结构体
   - 扩展Repository接口支持上游错误事件CRUD

2. 仓储层(ops_repo.go):
   - 实现InsertUpstreamErrorEvent写入上游错误
   - 实现GetUpstreamErrorEvents按时间范围查询

3. 服务层(ops_service.go, ops_upstream_context.go):
   - ops_service: 新增GetUpstreamErrorEvents查询方法
   - ops_upstream_context: 封装上游错误上下文构建逻辑

4. Handler层(ops_error_logger.go):
   - 新增GetUpstreamErrorsHandler处理上游错误查询请求

5. Gateway层集成:
   - antigravity_gateway_service.go: 429/529错误时记录上游事件
   - gateway_service.go: OpenAI 429/5xx错误时记录
   - gemini_messages_compat_service.go: Gemini 429/5xx错误时记录
   - openai_gateway_service.go: OpenAI 429/5xx错误时记录
   - ratelimit_service.go: 429限流错误时记录

**数据记录字段**:
- request_id: 关联ops_logs主记录
- platform/model: 上游服务标识
- status_code/error_message: 错误详情
- request_headers/response_body: 调试信息(可选)
- created_at: 错误发生时间
This commit is contained in:
IanShaw027
2026-01-11 15:30:27 +08:00
parent c2962752eb
commit 7ebca553ef
11 changed files with 907 additions and 41 deletions

View File

@@ -543,12 +543,21 @@ func (s *GeminiMessagesCompatService) Forward(ctx context.Context, c *gin.Contex
resp, err = s.httpUpstream.Do(upstreamReq, proxyURL, account.ID, account.Concurrency)
if err != nil {
safeErr := sanitizeUpstreamErrorMessage(err.Error())
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: 0,
Kind: "request_error",
Message: safeErr,
})
if attempt < geminiMaxRetries {
log.Printf("Gemini account %d: upstream request failed, retry %d/%d: %v", account.ID, attempt, geminiMaxRetries, err)
sleepGeminiBackoff(attempt)
continue
}
return nil, s.writeClaudeError(c, http.StatusBadGateway, "upstream_error", "Upstream request failed after retries: "+sanitizeUpstreamErrorMessage(err.Error()))
setOpsUpstreamError(c, 0, safeErr, "")
return nil, s.writeClaudeError(c, http.StatusBadGateway, "upstream_error", "Upstream request failed after retries: "+safeErr)
}
// Special-case: signature/thought_signature validation errors are not transient, but may be fixed by
@@ -558,6 +567,30 @@ func (s *GeminiMessagesCompatService) Forward(ctx context.Context, c *gin.Contex
_ = resp.Body.Close()
if isGeminiSignatureRelatedError(respBody) {
upstreamReqID := resp.Header.Get(requestIDHeader)
if upstreamReqID == "" {
upstreamReqID = resp.Header.Get("x-goog-request-id")
}
upstreamMsg := strings.TrimSpace(extractUpstreamErrorMessage(respBody))
upstreamMsg = sanitizeUpstreamErrorMessage(upstreamMsg)
upstreamDetail := ""
if s.cfg != nil && s.cfg.Gateway.LogUpstreamErrorBody {
maxBytes := s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes
if maxBytes <= 0 {
maxBytes = 2048
}
upstreamDetail = truncateString(string(respBody), maxBytes)
}
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: resp.StatusCode,
UpstreamRequestID: upstreamReqID,
Kind: "signature_error",
Message: upstreamMsg,
Detail: upstreamDetail,
})
var strippedClaudeBody []byte
stageName := ""
switch signatureRetryStage {
@@ -608,6 +641,30 @@ func (s *GeminiMessagesCompatService) Forward(ctx context.Context, c *gin.Contex
s.handleGeminiUpstreamError(ctx, account, resp.StatusCode, resp.Header, respBody)
}
if attempt < geminiMaxRetries {
upstreamReqID := resp.Header.Get(requestIDHeader)
if upstreamReqID == "" {
upstreamReqID = resp.Header.Get("x-goog-request-id")
}
upstreamMsg := strings.TrimSpace(extractUpstreamErrorMessage(respBody))
upstreamMsg = sanitizeUpstreamErrorMessage(upstreamMsg)
upstreamDetail := ""
if s.cfg != nil && s.cfg.Gateway.LogUpstreamErrorBody {
maxBytes := s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes
if maxBytes <= 0 {
maxBytes = 2048
}
upstreamDetail = truncateString(string(respBody), maxBytes)
}
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: resp.StatusCode,
UpstreamRequestID: upstreamReqID,
Kind: "retry",
Message: upstreamMsg,
Detail: upstreamDetail,
})
log.Printf("Gemini account %d: upstream status %d, retry %d/%d", account.ID, resp.StatusCode, attempt, geminiMaxRetries)
sleepGeminiBackoff(attempt)
continue
@@ -633,12 +690,62 @@ func (s *GeminiMessagesCompatService) Forward(ctx context.Context, c *gin.Contex
}
s.handleGeminiUpstreamError(ctx, account, resp.StatusCode, resp.Header, respBody)
if tempMatched {
upstreamReqID := resp.Header.Get(requestIDHeader)
if upstreamReqID == "" {
upstreamReqID = resp.Header.Get("x-goog-request-id")
}
upstreamMsg := strings.TrimSpace(extractUpstreamErrorMessage(respBody))
upstreamMsg = sanitizeUpstreamErrorMessage(upstreamMsg)
upstreamDetail := ""
if s.cfg != nil && s.cfg.Gateway.LogUpstreamErrorBody {
maxBytes := s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes
if maxBytes <= 0 {
maxBytes = 2048
}
upstreamDetail = truncateString(string(respBody), maxBytes)
}
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: resp.StatusCode,
UpstreamRequestID: upstreamReqID,
Kind: "failover",
Message: upstreamMsg,
Detail: upstreamDetail,
})
return nil, &UpstreamFailoverError{StatusCode: resp.StatusCode}
}
if s.shouldFailoverGeminiUpstreamError(resp.StatusCode) {
upstreamReqID := resp.Header.Get(requestIDHeader)
if upstreamReqID == "" {
upstreamReqID = resp.Header.Get("x-goog-request-id")
}
upstreamMsg := strings.TrimSpace(extractUpstreamErrorMessage(respBody))
upstreamMsg = sanitizeUpstreamErrorMessage(upstreamMsg)
upstreamDetail := ""
if s.cfg != nil && s.cfg.Gateway.LogUpstreamErrorBody {
maxBytes := s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes
if maxBytes <= 0 {
maxBytes = 2048
}
upstreamDetail = truncateString(string(respBody), maxBytes)
}
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: resp.StatusCode,
UpstreamRequestID: upstreamReqID,
Kind: "failover",
Message: upstreamMsg,
Detail: upstreamDetail,
})
return nil, &UpstreamFailoverError{StatusCode: resp.StatusCode}
}
return nil, s.writeGeminiMappedError(c, resp.StatusCode, respBody)
upstreamReqID := resp.Header.Get(requestIDHeader)
if upstreamReqID == "" {
upstreamReqID = resp.Header.Get("x-goog-request-id")
}
return nil, s.writeGeminiMappedError(c, account, resp.StatusCode, upstreamReqID, respBody)
}
requestID := resp.Header.Get(requestIDHeader)
@@ -863,6 +970,14 @@ func (s *GeminiMessagesCompatService) ForwardNative(ctx context.Context, c *gin.
resp, err = s.httpUpstream.Do(upstreamReq, proxyURL, account.ID, account.Concurrency)
if err != nil {
safeErr := sanitizeUpstreamErrorMessage(err.Error())
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: 0,
Kind: "request_error",
Message: safeErr,
})
if attempt < geminiMaxRetries {
log.Printf("Gemini account %d: upstream request failed, retry %d/%d: %v", account.ID, attempt, geminiMaxRetries, err)
sleepGeminiBackoff(attempt)
@@ -880,7 +995,8 @@ func (s *GeminiMessagesCompatService) ForwardNative(ctx context.Context, c *gin.
FirstTokenMs: nil,
}, nil
}
return nil, s.writeGoogleError(c, http.StatusBadGateway, "Upstream request failed after retries: "+sanitizeUpstreamErrorMessage(err.Error()))
setOpsUpstreamError(c, 0, safeErr, "")
return nil, s.writeGoogleError(c, http.StatusBadGateway, "Upstream request failed after retries: "+safeErr)
}
if resp.StatusCode >= 400 && s.shouldRetryGeminiUpstreamError(account, resp.StatusCode) {
@@ -899,6 +1015,30 @@ func (s *GeminiMessagesCompatService) ForwardNative(ctx context.Context, c *gin.
s.handleGeminiUpstreamError(ctx, account, resp.StatusCode, resp.Header, respBody)
}
if attempt < geminiMaxRetries {
upstreamReqID := resp.Header.Get(requestIDHeader)
if upstreamReqID == "" {
upstreamReqID = resp.Header.Get("x-goog-request-id")
}
upstreamMsg := strings.TrimSpace(extractUpstreamErrorMessage(respBody))
upstreamMsg = sanitizeUpstreamErrorMessage(upstreamMsg)
upstreamDetail := ""
if s.cfg != nil && s.cfg.Gateway.LogUpstreamErrorBody {
maxBytes := s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes
if maxBytes <= 0 {
maxBytes = 2048
}
upstreamDetail = truncateString(string(respBody), maxBytes)
}
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: resp.StatusCode,
UpstreamRequestID: upstreamReqID,
Kind: "retry",
Message: upstreamMsg,
Detail: upstreamDetail,
})
log.Printf("Gemini account %d: upstream status %d, retry %d/%d", account.ID, resp.StatusCode, attempt, geminiMaxRetries)
sleepGeminiBackoff(attempt)
continue
@@ -962,19 +1102,84 @@ func (s *GeminiMessagesCompatService) ForwardNative(ctx context.Context, c *gin.
}
if tempMatched {
evBody := unwrapIfNeeded(isOAuth, respBody)
upstreamMsg := strings.TrimSpace(extractUpstreamErrorMessage(evBody))
upstreamMsg = sanitizeUpstreamErrorMessage(upstreamMsg)
upstreamDetail := ""
if s.cfg != nil && s.cfg.Gateway.LogUpstreamErrorBody {
maxBytes := s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes
if maxBytes <= 0 {
maxBytes = 2048
}
upstreamDetail = truncateString(string(evBody), maxBytes)
}
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: resp.StatusCode,
UpstreamRequestID: requestID,
Kind: "failover",
Message: upstreamMsg,
Detail: upstreamDetail,
})
return nil, &UpstreamFailoverError{StatusCode: resp.StatusCode}
}
if s.shouldFailoverGeminiUpstreamError(resp.StatusCode) {
evBody := unwrapIfNeeded(isOAuth, respBody)
upstreamMsg := strings.TrimSpace(extractUpstreamErrorMessage(evBody))
upstreamMsg = sanitizeUpstreamErrorMessage(upstreamMsg)
upstreamDetail := ""
if s.cfg != nil && s.cfg.Gateway.LogUpstreamErrorBody {
maxBytes := s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes
if maxBytes <= 0 {
maxBytes = 2048
}
upstreamDetail = truncateString(string(evBody), maxBytes)
}
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: resp.StatusCode,
UpstreamRequestID: requestID,
Kind: "failover",
Message: upstreamMsg,
Detail: upstreamDetail,
})
return nil, &UpstreamFailoverError{StatusCode: resp.StatusCode}
}
respBody = unwrapIfNeeded(isOAuth, respBody)
upstreamMsg := strings.TrimSpace(extractUpstreamErrorMessage(respBody))
upstreamMsg = sanitizeUpstreamErrorMessage(upstreamMsg)
upstreamDetail := ""
if s.cfg != nil && s.cfg.Gateway.LogUpstreamErrorBody {
maxBytes := s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes
if maxBytes <= 0 {
maxBytes = 2048
}
upstreamDetail = truncateString(string(respBody), maxBytes)
log.Printf("[Gemini] native upstream error %d: %s", resp.StatusCode, truncateForLog(respBody, s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes))
}
setOpsUpstreamError(c, resp.StatusCode, upstreamMsg, upstreamDetail)
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: resp.StatusCode,
UpstreamRequestID: requestID,
Kind: "http_error",
Message: upstreamMsg,
Detail: upstreamDetail,
})
contentType := resp.Header.Get("Content-Type")
if contentType == "" {
contentType = "application/json"
}
c.Data(resp.StatusCode, contentType, respBody)
return nil, fmt.Errorf("gemini upstream error: %d", resp.StatusCode)
if upstreamMsg == "" {
return nil, fmt.Errorf("gemini upstream error: %d", resp.StatusCode)
}
return nil, fmt.Errorf("gemini upstream error: %d message=%s", resp.StatusCode, upstreamMsg)
}
var usage *ClaudeUsage
@@ -1076,7 +1281,32 @@ func sanitizeUpstreamErrorMessage(msg string) string {
return sensitiveQueryParamRegex.ReplaceAllString(msg, `$1***`)
}
func (s *GeminiMessagesCompatService) writeGeminiMappedError(c *gin.Context, upstreamStatus int, body []byte) error {
func (s *GeminiMessagesCompatService) writeGeminiMappedError(c *gin.Context, account *Account, upstreamStatus int, upstreamRequestID string, body []byte) error {
upstreamMsg := strings.TrimSpace(extractUpstreamErrorMessage(body))
upstreamMsg = sanitizeUpstreamErrorMessage(upstreamMsg)
upstreamDetail := ""
if s.cfg != nil && s.cfg.Gateway.LogUpstreamErrorBody {
maxBytes := s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes
if maxBytes <= 0 {
maxBytes = 2048
}
upstreamDetail = truncateString(string(body), maxBytes)
}
setOpsUpstreamError(c, upstreamStatus, upstreamMsg, upstreamDetail)
appendOpsUpstreamError(c, OpsUpstreamErrorEvent{
Platform: account.Platform,
AccountID: account.ID,
UpstreamStatusCode: upstreamStatus,
UpstreamRequestID: upstreamRequestID,
Kind: "http_error",
Message: upstreamMsg,
Detail: upstreamDetail,
})
if s.cfg != nil && s.cfg.Gateway.LogUpstreamErrorBody {
log.Printf("[Gemini] upstream error %d: %s", upstreamStatus, truncateForLog(body, s.cfg.Gateway.LogUpstreamErrorBodyMaxBytes))
}
var statusCode int
var errType, errMsg string
@@ -1184,7 +1414,10 @@ func (s *GeminiMessagesCompatService) writeGeminiMappedError(c *gin.Context, ups
"type": "error",
"error": gin.H{"type": errType, "message": errMsg},
})
return fmt.Errorf("upstream error: %d", upstreamStatus)
if upstreamMsg == "" {
return fmt.Errorf("upstream error: %d", upstreamStatus)
}
return fmt.Errorf("upstream error: %d message=%s", upstreamStatus, upstreamMsg)
}
type claudeErrorMapping struct {