Explain native AWS sample in detail; w/ test events

Updated with info from convo with OlegZ. Thx!
This commit is contained in:
Tim Colson
2023-11-13 10:31:37 -07:00
committed by GitHub
parent 27d7b1dca4
commit a8747a7080

View File

@@ -1,44 +1,95 @@
# Introduction
In this sample, you'll build a native GraalVM image with `spring-cloud-function` and set it up to run in AWS Lambda.
This example shows GraaalVM native spring-cloud-function application. The application itself is very simple and contains two functions - `uppercase` & `lowercase`.
Unless specific value is specified as `handler`, the application will fall back on `RoutingFunction` where you can pass the routing instruction
via `spring.cloud.function.definition` Message header. If using API Gateway you can pass such header as HTTP header.
# To run
## If you are on OSX Apple M1 Pro (arch64)
You first need to build a Docker image where you will actually build project.
To do that execute the following command form [project directory
The sample contains two functions - `uppercase` and `reverse` - so you can see how to route requests. A provided `RoutingFunction` will send messages to a handler function specified in a header named: `spring.cloud.function.definition` (demonstrated in the test section). The routing value can also be passed as an environment variable. If using API Gateway, you can pass this value as an HTTP header.
**Example function definition**
```
docker build -t "al2-graalvm19:native-uppercase" .
```
Start the container
```
docker run -dit -v `pwd`:`pwd` -w `pwd` -v ~/.m2:/root/.m2 al2-graalvm19:native-uppercase
```
Now navigate to the image terminal. Your working directory is alredy set for the root of the project. You can verify it by executing `ls`.
Build the project:
```
./mvnw clean -Pnative native:compile -DskipTests
@Bean
public Function<String, String> uppercase() {
return v -> {
System.out.println("Uppercasing " + v);
return v.toUpperCase();
};
}
```
Once the build finishes, you can deploy it.
> Note: If your function takes a Spring Message as an input parameter (e.g., Function<Message, ..>), the Lambda Context object will be available in the message header `aws-context`. See [AWSLambdaUtils.java](https://github.com/spring-cloud/spring-cloud-function/blob/main/spring-cloud-function-adapters/spring-cloud-function-adapter-aws/src/main/java/org/springframework/cloud/function/adapter/aws/AWSLambdaUtils.java#L67C44-L67C55) for details.
## Deploying to AWS LAmbda
Start *AWS Dashboard* and navigate to **AWS Lambda** Services
## To run the sample on macOS (Apple silicon arch64)
Click on `Create Function`. Enter `uppercase` for *function name*. For the runtime select `Provide your own bootstrap on Amazon Linux 2`.
Make sure you select the proper architecture (`x86_64` or `arm64`).
You first need to build the function, then you will deploy it to AWS Lambda.
Click on `Create Function` again.
### Step 1 - Build the native image
Next you need to upload your project, so click on `Upload From` and point to the ZIP file that was created by the build process (in the `target` directory).
Before starting the build, you must clone or download the code in **function-sample-aws-native**.
Once the file is uploaded navigate to the `Test` tab. You can change the input data or use the default. Basically you need to pas a String in a JSON format such as `"hello"` and you should see the output `"HELLO"`.
1. Change into the project directory: `spring-cloud-function-samples/function-sample-aws-native`
2. Run the following to build a Docker containter image which will be used to create the Lambda function zip file.
```
docker build -t "al2-graalvm19:native-uppercase" .
```
3. Start the container
```
docker run -dit -v `pwd`:`pwd` -w `pwd` -v ~/.m2:/root/.m2 al2-graalvm19:native-uppercase
```
4. In Docker, open the image terminal.
> Your working directory should default to the project root. Verify by running `ls` to view the files.
6. From inside the container, build the Lambda function:
```
./mvnw clean -Pnative native:compile -DskipTests
```
After the build finishes, you need to deploy the function.
### Step 2 - Deploy your function
You will first create the function, and then you will upload the zipped native image from the build process.
**Create the function**
1. Login to the **Amazon Web Services console**.
2. Navigate to the **Lambda service**.
3. Choose `Create Function`.
4. For **function name**, enter `uppercase`.
5. For runtime, select `Provide your own bootstrap on Amazon Linux 2`.
6. For architecture, select `arm64`.
7. Choose `Create Function` again.
**Upload the zip image**
1. Choose `Upload From` and choose the zip file created by the build, located in the `target` directory.
2. Wait for the image to upload.
### Step 3 - Test your function
Your test event will provide the information needed to select the `uppercase` or `reverse` handler functions.
1. From the Lambda console, navigate to the `Test` tab.
2. For test data, enter the following JSON:
```JSON
{
"payload": "hello",
"headers": {
"spring.cloud.function.definition": "uppercase"
}
}
```
3. Choose **Test**.
You should see uppercased output for the payload value: "HELLO"
4. Change the test data to the following JSON:
```JSON
{
"payload": "hello",
"headers": {
"spring.cloud.function.definition": "reverse"
}
}
```
5. Choose **Test**.
You should see reversed output for the payload value: "OLLEH"
**Congratulations!** You have built and deployed a Graal native image to AWS Lambda.