模擬 OpenAI API 的錯誤

當您在應用程式中使用 OpenAI API 時，您應該測試應用程式如何處理 API 錯誤。 Dev Proxy 可讓您使用 GenericRandomErrorPlugin，模擬任何 OpenAI API 上的錯誤。

提示

在命令提示字元中執行 devproxy config get openai-throttling以下載此預設。

在 [開發 Proxy 安裝] 資料夾中，找出 [config] 資料夾。 在 [config] 資料夾中，建立名為 openai-errors.json的新檔案。 在程式代碼編輯器中開啟檔案。

在參考 plugins的 GenericRandomErrorPlugin 陣列中建立新的物件。 定義要監看外掛程式的OpenAI API URL，並新增外掛程式組態的參考。

{
  "$schema": "https://raw.githubusercontent.com/dotnet/dev-proxy/main/schemas/v1.0.0/rc.schema.json",
  "plugins": [    
    {
      "name": "GenericRandomErrorPlugin",
      "enabled": true,
      "pluginPath": "~appFolder/plugins/DevProxy.Plugins.dll",
      "configSection": "openAIAPI",
      "urlsToWatch": [
        "https://api.openai.com/*"
      ]
    }
  ]
}

建立外掛程式的設定物件，以便向外掛程式提供錯誤回應的位置。

{
  "$schema": "https://raw.githubusercontent.com/dotnet/dev-proxy/main/schemas/v1.0.0/rc.schema.json",
  "plugins": [    
    {
      "name": "GenericRandomErrorPlugin",
      "enabled": true,
      "pluginPath": "~appFolder/plugins/DevProxy.Plugins.dll",
      "configSection": "openAIAPI",
      "urlsToWatch": [
        "https://api.openai.com/*"
      ]
    }
  ],
  "openAIAPI": {
    "errorsFile": "errors-openai.json"
  }
}

在相同的資料夾中，建立 errors-openai.json 檔案。 此檔案包含外掛程式傳送錯誤回應時可傳回的可能錯誤回應。

{
  "$schema": "https://raw.githubusercontent.com/dotnet/dev-proxy/main/schemas/v1.0.0/genericrandomerrorplugin.schema.json",
  "errors": [
    {
      "request": {
        "url": "https://api.openai.com/*"
      },
      "responses": [
        {
          "statusCode": 429,
          "headers": [
            {
              "name": "content-type",
              "value": "application/json; charset=utf-8"
            }
          ],
          "body": {
            "error": {
              "message": "Rate limit reached for default-text-davinci-003 in organization org-K7hT684bLccDbBRnySOoK9f2 on tokens per min. Limit: 150000.000000 / min. Current: 160000.000000 / min. Contact support@openai.com if you continue to have issues. Please add a payment method to your account to increase your rate limit. Visit https://beta.openai.com/account/billing to add a payment method.",
              "type": "tokens",
              "param": null,
              "code": null
            }
          }
        },
        {
          "statusCode": 429,
          "headers": [
            {
              "name": "content-type",
              "value": "application/json; charset=utf-8"
            }
          ],
          "body": {
            "error": {
              "message": "Rate limit reached for default-text-davinci-003 in organization org-K7hT684bLccDbBRnySOoK9f2 on requests per min. Limit: 60.000000 / min. Current: 70.000000 / min. Contact support@openai.com if you continue to have issues. Please add a payment method to your account to increase your rate limit. Visit https://beta.openai.com/account/billing to add a payment method.",
              "type": "requests",
              "param": null,
              "code": null
            }
          }
        },
        {
          "statusCode": 429,
          "addDynamicRetryAfter": true,
          "headers": [
            {
              "name": "content-type",
              "value": "application/json; charset=utf-8"
            }
          ],
          "body": {
            "error": {
              "message": "The engine is currently overloaded, please try again later.",
              "type": "requests",
              "param": null,
              "code": null
            }
          }
        }
      ]
    }
  ]
}

使用組態檔啟動 Dev Proxy：

devproxy --config-file "~appFolder/config/openai-errors.json"

當您使用呼叫 OpenAI API 的應用程式時，Dev Proxy 會隨機傳回您在 errors-openai.json 檔案中定義的其中一個錯誤回應。

深入瞭解 GenericRandomErrorPlugin。

GenericRandomErrorPlugin