以 (StatusCode.UNAVAILABLE, Socket closed)> 终止的 RPC 集合点

Question

• tensorflow-gpu 1.10.0
• 张量流服务器 1.10.0

我已经部署了一个服务于多个模型的 tensorflow 服务器。 客户端代码就像client.py这个，我调用predict函数。

channel = implementations.insecure_channel(host, port)
stub = prediction_service_pb2.beta_create_PredictionService_stub(channel)
request = predict_pb2.PredictRequest()

def predict(data, shape, model_name, signature_name="predict"):
    request.model_spec.name = model_name
    request.model_spec.signature_name = signature_name
    request.inputs['image'].CopyFrom(tf.contrib.util.make_tensor_proto(data, shape=shape))
    result = stub.Predict(request, 10.0)
    return result.outputs['prediction'].float_val[0]

我有大约 100 个具有相同配置的客户端。 下面是调用predict函数的示例代码：

from client import predict
while True:
    print(predict(data, shape, model_name))
    # time.sleep some while

首先，当我运行客户端代码时，我可以正确收到响应。 但几个小时后，客户端因错误而崩溃

_Rendezvous of RPC that terminated with (StatusCode.UNAVAILABLE, Socket closed)

我尝试将我的客户端代码修改为

def predict(data, shape, model_name, signature_name="predict"):
    channel = implementations.insecure_channel(host, port)
    stub = prediction_service_pb2.beta_create_PredictionService_stub(channel)
    request = predict_pb2.PredictRequest()
    request.model_spec.name = model_name
    request.model_spec.signature_name = signature_name
    request.inputs['image'].CopyFrom(tf.contrib.util.make_tensor_proto(data, shape=shape))
    result = stub.Predict(request, 10.0)
    return result.outputs['prediction'].float_val[0]

这意味着每次调用predict 函数时，我都会尝试与 tfs 服务器建立连接。但是这段代码也像以前一样失败了。

那么我应该怎么处理这种情况呢？

Answer 1

最后我在return 之前添加了一个channel.close()，它工作正常。