Can't connect to server

Question

Can't connect to server

Opened this issue 6 years ago · 4 comments

Hi,

I am trying to run the example on the ReadMe. After updating the yaml to my image file, and sending a request from the client, I got the following error
$ python resnet_client.py --server localhost --port 9000 cat_sample.jpg Traceback (most recent call last): File "resnet_client.py", line 155, in <module> main() File "resnet_client.py", line 94, in main args.server, args.port, args.model, jpeg_batch) File "resnet_client.py", line 149, in predict_and_profile result = stub.Predict(request, 60.0) # 60 second timeout File "/Users/kahmun/tf-serving-k8s-tutorial/client/client/lib/python2.7/site-packages/grpc/beta/_client_adaptations.py", line 310, in __call__ self._request_serializer, self._response_deserializer) File "/Users/kahmun/tf-serving-k8s-tutorial/client/client/lib/python2.7/site-packages/grpc/beta/_client_adaptations.py", line 196, in _blocking_unary_unary raise _abortion_error(rpc_error_call) grpc.framework.interfaces.face.face.AbortionError: AbortionError(code=StatusCode.UNAVAILABLE, details="Connect Failed")

The external-IP for the LoadBalancer is

$ kubectl get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 1h resnet-service LoadBalancer 10.101.188.76 <pending> 9000:31072/TCP 1h

I read that when running in minikube, i should use NodePort instead, but switching to NodePort returns the same error. Is there anything that I am still doing wrong?

Also I was using the Dockerfile which was deleted in a previous commit, I don't think that is the issue could it?

Answer 1 · 2018-09-05T05:18:15.000Z

Please switch to the branch march-2018 for the tutorial. The master branch is currently broken since we are in the process of updating it for another tutorial. :)

…

On Tue, Sep 4, 2018 at 7:57 AM Kah Mun ***@***.***> wrote: Hi, I am trying to run the example on the ReadMe. After updating the yaml to my image file, and sending a request from the client, I got the following error $ python resnet_client.py --server localhost --port 9000 cat_sample.jpg Traceback (most recent call last): File "resnet_client.py", line 155, in <module> main() File "resnet_client.py", line 94, in main args.server, args.port, args.model, jpeg_batch) File "resnet_client.py", line 149, in predict_and_profile result = stub.Predict(request, 60.0) # 60 second timeout File "/Users/kahmun/tf-serving-k8s-tutorial/client/client/lib/python2.7/site-packages/grpc/beta/_client_adaptations.py", line 310, in __call__ self._request_serializer, self._response_deserializer) File "/Users/kahmun/tf-serving-k8s-tutorial/client/client/lib/python2.7/site-packages/grpc/beta/_client_adaptations.py", line 196, in _blocking_unary_unary raise _abortion_error(rpc_error_call) grpc.framework.interfaces.face.face.AbortionError: AbortionError(code=StatusCode.UNAVAILABLE, details="Connect Failed") The external-IP for the LoadBalancer is $ kubectl get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 1h resnet-service LoadBalancer 10.101.188.76 <pending> 9000:31072/TCP 1h I read that when running in minikube, i should use NodePort instead, but switching to NodePort returns the same error. Is there anything that I am still doing wrong? Also I was using the Dockerfile which was deleted in a previous commit, I don't think that is the issue could it? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#2>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADj0k91VQCO5yj0ayFEklOjrRqKbofXCks5uXpTWgaJpZM4WZKb_> .

Answer 2 · 2018-09-05T08:23:49.000Z

Hi Brian,

Thanks for the reply, but I am working on the upgrade-to-tf-1.8 branch if that is the march-2018 branch that you are referring to. It still returns me the same error. Is the new tutorial going to be updated soon?

Answer 3 · 2018-09-05T23:53:22.000Z

Hi Kah, We will have a new tutorial out by early next week. I will be working on the github documentation for a couple of days afterwards, so I will let you know once the project is fully up to date. I may have some time later next week to look into the minikube issue. May I ask what your computer's native OS is? I've only tested this on my macbook which might have had some local configurations that I forgot to log. Best, Brian

…

On Wed, Sep 5, 2018 at 1:24 AM Kah Mun ***@***.***> wrote: Hi Brian, Thanks for the reply, but I am working on the upgrade-to-tf-1.8 branch if that is the march-2018 branch that you are referring to. It still returns me the same error. Is the new tutorial going to be updated soon? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#2 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADj0k048ySJhBwSgGanDC74XLN5TSxyFks5uX4ofgaJpZM4WZKb_> .

Answer 4 · 2018-09-10T02:29:28.000Z

Hi Brian,

I have tested both on MacOS and Ubuntu 18.04. Both have the same results.
Thanks for the help.