使用Google Cloud PHP SDK调用Vertex AI遇503 UNAVAILABLE错误求助

问题详情 #

使用Google Cloud PHP SDK实现AI预测服务时，调用Vertex AI接口返回503 UNAVAILABLE错误（错误码14），但通过Curl发起完全相同的请求却能得到正常响应。模型标识符$model格式为projects/{project}/locations/{location}/publishers/google/models/*。代码三四天前运行完全正常，当前突然出现该错误。

相关代码 #

生成内容的核心函数 #

function generate_content_sample(string $model): void
{
    $predictionServiceClient = new PredictionServiceClient();

    $contentsParts = new Part();
    $contentsParts->setText("loremp ipsum"); 

    $contentsParts = [new Part()];
    $content = (new Content())
        ->setRole('user')
        ->setParts($contentsParts);

    $contents = [$content,];
    $request = (new GenerateContentRequest())
        ->setModel($model)
        ->setContents($contents);

    try {
        $response = $predictionServiceClient->generateContent($request);
        printf('Response data: %s' . PHP_EOL, $response->serializeToJsonString());
    } catch (ApiException $ex) {
        printf('Call failed with message: %s' . PHP_EOL, $ex->getMessage());
    }
}

控制器调用代码 #

public function generate(Request $request)
{
    try {
        $vertexAIService = new VertexAIService();

        $projectId = 'Lorem ipsum';
        $locationId = 'Lorem ipsum';
        $modelId = 'Lorem ipsum';

        $formattedEndpoint = sprintf(
            "projects/%s/locations/%s/publishers/google/models/%s",
            $projectId,
            $locationId,
            $modelId
        );

        // $content = WebScrapingController::scrape($request->url);
        // $prompt = $request->prompt . ': ' . $content;

        $response = $vertexAIService->generate_content_sample($formattedEndpoint, "Lorem ipsum");

        FacadesLog::debug('Response: ' . $response);

        return response()->json([
            'response' => $response
        ]);
    } catch (Exception $e) {
        return response()->json([
            'error' => $e->getMessage()
        ], 500);
    }
}

错误返回结果 #

{
"message": "<!DOCTYPE html>\n<html lang=en>\n  <meta charset=utf-8>\n  <meta name=viewport content=\"initial-scale=1, minimum-scale=1, width=device-width\">\n  <title>Error 503 (Server Error)!!1</title>\n  <style>\n    *{margin:0;padding:0}html,code{font:15px/22px arial,sans-serif}html{background:#fff;color:#222;padding:15px}body{margin:7% auto 0;max-width:390px;min-height:180px;padding:30px 0 15px}* > body{background:url(//www.google.com/images/errors/robot.png) 100% 5px no-repeat;padding-right:205px}p{margin:11px 0 22px;overflow:hidden}ins{color:#777;text-decoration:none}a img{border:0}@media screen and (max-width:772px){body{background:none;margin-top:0;max-width:none;padding-right:0}}#logo{background:url(//www.google.com/images/branding/googlelogo/1x/googlelogo_color_150x54dp.png) no-repeat;margin-left:-5px}@media only screen and (min-resolution:192dpi){#logo{background:url(//www.google.com/images/branding/googlelogo/2x/googlelogo_color_150x54dp.png) no-repeat 0% 0%/100% 100%;-moz-border-image:url(//www.google.com/images/branding/googlelogo/2x/googlelogo_color_150x54dp.png) 0}}@media only screen and (-webkit-min-device-pixel-ratio:2){#logo{background:url(//www.google.com/images/branding/googlelogo/2x/googlelogo_color_150x54dp.png) no-repeat;-webkit-background-size:100% 100%}}#logo{display:inline-block;height:54px;width:150px}\n  </style>\n  <a href=//www.google.com/><span id=logo aria-label=Google></span></a>\n  <p><b>503.</b> <ins>That’s an error.</ins>\n  <p>The service you requested is not available at this time.<p>Service error -27.  <ins>That’s all we know.</ins>\n",
"code": 14,
"status": "UNAVAILABLE",
"details": []
}

使用的依赖版本 #

• google/apiclient: ^2.18
• google/cloud-ai-platform: ^1.13

解决建议 #

1. 修复请求体逻辑错误 #

核心函数中存在明显的变量覆盖问题：先创建带文本的Part对象，随后又将其重新赋值为空Part数组，导致请求内容为空。修正代码：

function generate_content_sample(string $model): void
{
    $predictionServiceClient = new PredictionServiceClient();

    // 创建包含文本的Part并传入内容数组
    $contentPart = (new Part())->setText("loremp ipsum"); 
    $content = (new Content())
        ->setRole('user')
        ->setParts([$contentPart]);

    $request = (new GenerateContentRequest())
        ->setModel($model)
        ->setContents([$content]);

    try {
        $response = $predictionServiceClient->generateContent($request);
        printf('Response data: %s' . PHP_EOL, $response->serializeToJsonString());
    } catch (ApiException $ex) {
        printf('Call failed with message: %s' . PHP_EOL, $ex->getMessage());
    }
}

2. 指定正确的API端点 #

PHP SDK可能默认使用了错误的端点，手动指定与模型匹配的区域端点：

$predictionServiceClient = new PredictionServiceClient([
    'apiEndpoint' => 'us-central1-aiplatform.googleapis.com' // 替换为你的模型所在区域
]);

3. 更新依赖库 #

当前使用的google/cloud-ai-platform版本较旧，可能存在已知兼容性问题，执行更新：

composer update google/cloud-ai-platform google/apiclient

4. 验证服务账号权限 #

确认SDK使用的服务账号拥有aiplatform.predictions.generate权限，同时检查账号是否被限制访问Vertex AI服务。

5. 检查区域一致性 #

确保模型所在区域与客户端指定的API端点区域完全匹配，避免跨区域访问导致的服务不可用。

内容的提问来源于stack exchange，提问作者Jordi Tomás Molina