Twilio-Lambda is a result of toying with Twilio API and trying to come up with something useful to use in everyday coding life. What I decided to build was a very generic lambda function which could be used in the cloud to alert developers when something serious has occurred with a service and immediate attention is needed therefore an sms text would be a nice way to immediately alert someone there is an issue. This is where the project started. After I developed this, I thought it might be really nice to have some data around what kind of messages are being sent, and even potentially run some analytics to determine if I am receiving too many alarms etc.
I tried to use a wide range of tech stack and cloud technologies for this project. I am using serverless architecture for the API portion, using a public API Gateway to front two lambda functions. This is secured by an API key for access. The Lambda functions are written in Node and use the twilio-node npm package to interact with the Twilio API. Since we are dealing with sensitive account information, I am also anonymizing sensitive data as well. As for the Web portion, I took advantage of React with Typescript, Styled Components, and Ag-Grid for my table. I decided to deploy the web portion in a Docker container on Heroku. The container also contains an express webserver that acts as an intermediary between the API Gateway and the web application.
- React
- Typescript
- Ag-Grid
- Docker
- Heroku
- API Gateway
- Lambda
- Jenkins
Devs can be notified directly via SMS
You can use the GUI to look at all the possible messages, errors, and warnings which are being sent out.
You can filter Rows by quick text, in addition to robust sorting and filtering which comes with Ag-Grid
I used Jenkins to orchestrate all deployment for this project. I wrote individual Jenkins jobs for deploying the Lambda functions, and a deployment pipeline for the docker container on Heroku.
This was more a fun experimenting with the API project, but I do think there are some really cool things you could do with this project. Like imagine when you get alerted via sms you could respond with different responses that would cause a particular action to occur on the resource that triggered the alert. There are lots of really interesting paths you can go down.