linshenkx/prompt-optimizer · error · APIError

Connection failed: ${error.message}

Error message

Connection failed: ${error.message}

What it means

Same network-failure branch as the CORS case, but detectCrossOriginError concluded it is a plain connectivity failure — the endpoint is unreachable, refused, or DNS fails, and it's not a cross-origin browser block. Message is prefixed 'Connection failed:'.

Source

Thrown at packages/core/src/services/llm/adapters/openai-adapter.ts:177

          throw new APIError('API returned empty model list')
        }

        return models
      }

      throw new APIError('Unexpected API response format')
    } catch (error: any) {
      console.error('[OpenAIAdapter] Failed to fetch models:', error)

      // 连接错误处理（包括跨域检测）
      if (error.message && (error.message.includes('Failed to fetch') ||
          error.message.includes('Connection error'))) {
        const isCrossOriginError = this.detectCrossOriginError(error, baseURL)

        if (isCrossOriginError) {
          throw new APIError(`Cross-origin connection failed: ${error.message}`)
        } else {
          throw new APIError(`Connection failed: ${error.message}`)
        }
      }

      // API返回的错误信息
      if (error.response?.data) {
        throw new APIError(`API error: ${JSON.stringify(error.response.data)}`)
      }

      // 其他错误,保持原始信息
      throw new APIError(error.message || 'Unknown error')
    }
  }

  // ===== 参数定义（用于buildDefaultModel） =====

  /**
   * 获取参数定义
   * 基于 OpenAI 官方文档: https://platform.openai.com/docs/api-reference/chat/create

View on GitHub (pinned to 3e677b1d9f)

Solutions

1. Confirm the server is running: curl $BASE_URL/models from the same environment
2. Fix scheme/port/typos in baseURL (http vs https, correct port)
3. If server-side, add reachability checks/retry with backoff for transient failures
4. Check DNS/firewall/VPN if the host should be reachable

Example fix

// before
const models = await adapter.getModelsAsync()

// after
if (!(await isReachable(baseURL))) throw new Error(`LLM endpoint ${baseURL} unreachable`)
const models = await adapter.getModelsAsync()

async function endpointReachable(baseURL: string): Promise<boolean> {
  try { const u = new URL(baseURL); await fetch(u.origin, { mode: 'no-cors' }); return true } catch { return false }
}

function isConnectionFailure(e: unknown): boolean {
  return e instanceof APIError && e.message.startsWith('Connection failed')
}

try { models = await adapter.getModelsAsync() }
catch (e) {
  if (isConnectionFailure(e)) return withBackoff(() => adapter.getModelsAsync(), 3)
  throw e
}

When it happens

Trigger: getModelsAsync() against a local server that isn't running (ECONNREFUSED), wrong port, DNS failure for the base URL host, TLS errors, or no network.

Common situations: Gateway process stopped, wrong port in baseURL, typo in hostname, VPN/offline machine, server restarting.

Related errors

• Cross-origin connection failed: ${error.message}
• formatExecutionErrorMessage(error)
• GENERATION_FAILED
• GENERATION_FAILED
• UNSUPPORTED_TEST_TYPE

AI-assisted analysis of linshenkx/prompt-optimizer@3e677b1d9f (2026-08-27). Data as JSON: /api/errors/a0dfccce6521a5bf. Report an issue: GitHub.