Deploying and Integrating write up (#5)
This commit is contained in:
jbirddog
GitHub
parent
ee39e83295
commit
00f4877a5f
|
|
@ -0,0 +1,103 @@
|
|||
# Deploying and Integrating With SpiffArena
|
||||
|
||||
This guide shows you how to deploy the demo `Connector Proxy` as an `AWS Lambda Function` as well as how to integrate it with [SpiffArena](https://www.spiffworkflow.org/pages/spiffarena/). The [Getting Started Guide](https://www.spiffworkflow.org/posts/articles/get_started/) will be used as the basis for integration but the steps should easily map to any custom installation.
|
||||
|
||||
There is an assumption that you have access to login to the AWS Console and can create/deploy Lambda functions.
|
||||
|
||||
## Building the zip
|
||||
|
||||
One option when deploying a Lambda function is to upload a zip file containing the source code or executable. In the root of this repository run:
|
||||
|
||||
```
|
||||
make zip
|
||||
```
|
||||
|
||||
This will create a zip file containing the [lambda entry point function](https://github.com/jbirddog/connector-proxy-lambda-demo/blob/main/connector_proxy_lambda_demo/lambda_function.py#L5) as well as all the dependencies needed to execute the connectors. For this example the libraries [spiffworkflow-proxy](https://github.com/sartography/spiffworkflow-proxy) is used for discovering connectors and [connector-http](https://github.com/sartography/connector-http) is an example connector that provides http get and post requests.
|
||||
|
||||
Once `make zip` completes `connector_proxy_lambda_demo.zip` will be available in the repository root.
|
||||
|
||||
## Creating the Lambda Function
|
||||
|
||||
Log in to the AWS Console and navigate to the Lambda section.
|
||||
|
||||
|
||||
|
||||
From there choose `Create function`.
|
||||
|
||||
|
||||
|
||||
Choose to `Author from scratch` and select the most recent Python runtime.
|
||||
|
||||
|
||||
|
||||
Under `Advanced Settings` check `Enable function URL`. For this demo we will use the `NONE` auth type to keep things simple.
|
||||
|
||||
|
||||
|
||||
After hitting the `Create function` button you will be taken to your new Lambda function:
|
||||
|
||||
|
||||
|
||||
In the bottom right of the first section is a link to your Lamba's function URL. Click it for a hello world response.
|
||||
|
||||
|
||||
|
||||
## Deploying the Lambda Function
|
||||
|
||||
If you scroll down you will see a section with the example code created with your Lambda function. We are going to replace this with the contents of our zip file. Choose `Upload from` and select `.zip file`.
|
||||
|
||||
|
||||
|
||||
After a confirmation dialog you will see your lambda has been updated:
|
||||
|
||||
|
||||
|
||||
Click your function URL again to see a hello from our deployed Connector Proxy.
|
||||
|
||||
## Integrating With SpiffArena
|
||||
|
||||
Congrats, your Connector Proxy has been deployed as a Lambda function. To integrate it with SpiffArena we simply need to update an environment variable and restart the backend. If using the [Getting Started Guide](https://www.spiffworkflow.org/posts/articles/get_started/) open the docker-compose.yml file, else edit the environment variable in the way appropriate for your deployment. The variable we need to change is called `SPIFFWORKFLOW_BACKEND_CONNECTOR_PROXY_URL`.
|
||||
|
||||
Example diff using the function URL from the above images:
|
||||
|
||||
```
|
||||
diff --git a/docker-compose.yml b/docker-compose.yml
|
||||
index 95b87b39..7d55c492 100644
|
||||
--- a/docker-compose.yml
|
||||
+++ b/docker-compose.yml
|
||||
@@ -26,7 +26,7 @@ services:
|
||||
SPIFFWORKFLOW_BACKEND_URL: "http://localhost:${SPIFF_BACKEND_PORT:-8000}"
|
||||
|
||||
SPIFFWORKFLOW_BACKEND_BPMN_SPEC_ABSOLUTE_DIR: "/app/process_models"
|
||||
- SPIFFWORKFLOW_BACKEND_CONNECTOR_PROXY_URL: "http://spiffworkflow-connector:8004"
|
||||
+ SPIFFWORKFLOW_BACKEND_CONNECTOR_PROXY_URL: "https://crbxgaopinfxqscntkqixjbl4e0gigpm.lambda-url.us-east-1.on.aws"
|
||||
SPIFFWORKFLOW_BACKEND_DATABASE_URI: "mysql+mysqlconnector://root:${SPIFF_MYSQL_PASS:-my-secret-pw}@spiffworkflow-db:${SPIFF_MYSQL_PORT:-8003}/spiffworkflow_backend_development"
|
||||
SPIFFWORKFLOW_BACKEND_LOAD_FIXTURE_DATA: "false"
|
||||
SPIFFWORKFLOW_BACKEND_OPEN_ID_CLIENT_ID: "spiffworkflow-backend"
|
||||
|
||||
```
|
||||
|
||||
Restart:
|
||||
|
||||
```
|
||||
docker compose down
|
||||
docker compose up -d
|
||||
```
|
||||
|
||||
## Testing
|
||||
|
||||
Create a new process model as described in the [Getting Started Guide](https://www.spiffworkflow.org/posts/articles/get_started/). Add a `Service Task` and in its properties panel you will see a drop down to pick which connector in your `Connector Proxy` to call. In this demo we deployed http get and post connectors:
|
||||
|
||||
|
||||
|
||||
Choose the `http/GetRequest` operator ID and enter the [dog fact api](https://dog-api.kinduff.com/api/facts) URL. Remember to quote it since parameters are evaluated as Python expressions.
|
||||
|
||||
|
||||
|
||||
Run the process and once complete you can see the response in the workflow:
|
||||
|
||||
|
||||
|
||||
## Congrats!
|
||||
|
||||
You have successfully deployed your own `Connector Proxy` as an `AWS Lambda function` and integrated it with `SpiffArena`. You made a call from a workflow to get a dog fact. Now imagine that call was to talk to an external system relevant to your business processes.
|
||||
Loading…
Reference in New Issue