torchserve 0.11.0 tests update (#271)

Signed-off-by: sharvil10 <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

pytorch/serving/README.md

@@ -49,6 +49,12 @@ docker container stop server
 
 As demonstrated in the above example, models must be registered before they can be used for predictions. The best way to ensure models are pre-registered with ideal settings is to modify the included [config file](./config.properties) for the torchserve server.
 
+> [!NOTE]
+> Since torchserve 0.11.1 torchserve asks for token authentication for security. We've disabled it in the config.properties by setting `disable_token_authorization=true`. If you want to enable the authentication you can find more details in the [documentation](https://github.com/pytorch/serve/blob/master/docs/token_authorization_api.md).
+
+> [!NOTE]
+> Since torchserve 0.11.1 the model API has been disabled by default. We enable the model API by setting `enable_model_api=true` in provided config.properties.
+
 1. Add your model to the config file
 
     ```properties

pytorch/serving/config.properties

@@ -10,3 +10,5 @@ allowed_urls=https://s3.amazonaws.com/.*,https://torchserve.pytorch.org/.*
 ipex_enable=true
 cpu_launcher_enable=true
 cpu_launcher_args=--use_logical_core
+disable_token_authorization=true
+enable_model_api=true

pytorch/serving/model-store/grpc-test.sh

@@ -25,6 +25,7 @@
 #
 #       For reference:
 #           https://docs.docker.com/develop/develop-images/build_enhancements/
+set -e
 
 apt-get -y update
 apt-get -y install curl

pytorch/serving/model-store/rest-test.sh

@@ -26,6 +26,8 @@
 #       For reference:
 #           https://docs.docker.com/develop/develop-images/build_enhancements/
 
+set -e
+
 apt-get -y update
 apt-get -y install curl
 # curl -O https://torchserve.pytorch.org/mar_files/squeezenet1_1.mar

pytorch/serving/tests.yaml

@@ -54,6 +54,8 @@ ipex-serving-rest-inference:
   volumes:
   - src: $PWD/pytorch/serving/model-store
     dst: /home/model-server/model-store
+  - src: $PWD/pytorch/serving/wf-store
+    dst: /home/model-server/wf-store
   env:
     ENABLE_TORCH_PROFILER: 'true'
   shm_size: 1g
@@ -65,6 +67,8 @@ ipex-serving-grpc-inference:
   volumes:
   - src: $PWD/pytorch/serving/model-store
     dst: /home/model-server/model-store
+  - src: $PWD/pytorch/serving/wf-store
+    dst: /home/model-server/wf-store
   env:
     ENABLE_TORCH_PROFILER: 'true'
   shm_size: 1g

pytorch/serving/wf-store/rest-test.sh

@@ -26,6 +26,8 @@
 #       For reference:
 #           https://docs.docker.com/develop/develop-images/build_enhancements/
 
+set -e
+
 apt-get -y update
 apt-get -y install curl
 
@@ -47,4 +49,4 @@ curl --fail -X POST http://127.0.0.1:8080/wfpredict/dog_breed_wf -T Dog1.jpg
 curl --fail -O https://raw.githubusercontent.com/pytorch/serve/master/examples/Workflows/dog_breed_classification/model_input/Dog2.jpg
 curl --fail -X POST http://127.0.0.1:8080/wfpredict/dog_breed_wf -T Dog2.jpg
 
-rm -rf ./*.war ./Cat.jpg ./Dog1.jpg ./Dog2.jpg ../model-store/*.mar
+rm -rf ./*.war ./Cat.jpg ./Dog1.jpg ./Dog2.jpg