Implement an Uptime Monitor Using AWS Lambda, EventBridge, SNS, and SQS.
You can use AWS EventBridge rules to schedule a Lambda function to run at a specific time or on a set interval. Combining these two services unlocks a vast array of use cases and workloads. Anything from regular web scraping jobs to automated daily reminders can be implemented by scheduling Lambda functions with EventBridge rules.
In this article, we will be scheduling a Lambda function to run every 5 minutes to check the health of a website. The Lambda function will send a simple HTTP request to the website server and
check the response code. If the response code is not 200, the Lambda function will send an alert message to an SNS topic that can be subscribed to. This is how I implemented uptime monitoring for my monitoring and alerting SaaS, Komonitor.
The only prerequisites for this article is having an AWS account and Node environment set up.
Setting up AWS resources
This section will cover setting up the required AWS resources via the AWS console.
Lambda Function
First, we need to create a Lambda function that will be used to check the health of our website.
Navigate to the AWS Lambda console and create a new Lambda function and give it a name.
After creating it, it should be visible in the console like this:
Right now, the function only has template code that prints a hello-world message. Later in this article we will implement the
uptime checking logic in the Lambda function.
EventBridge Rule
In order to schedule the Lambda function to run every 5 minutes, we need to create an EventBridge rule.
Navigate to the AWS EventBridge console and click the Create Rule button.
In the Define rule detail step, give the rule a name and set the Rule type to Schedule and proceed to the next step.
On the next step, define the schedule pattern and set the Rate expression to 5 minutes.
Now, select the targets for this event rule. Select AWS Service and then Lambda function. In the dropdown, the name Lambda function created earlier should be an option, select it.
That is all the configuration needed to set up the event rule, proceed through the following steps and create the rule.
To confirm that the rule was created and configured to target the Lambda function, navigate to the Lambda console and take a look
at the function overview. The rule should be visible in the function overview and configuration tab like this:
SNS Topic and Lambda Permissions
Our uptime monitor will be capturing alerts via an SNS topic that over parties can subscribe to. Go to the SNS console and create a new FIFO topic.
In order for the Lambda function to publish messages to the SNS topic, we need to edit the Lambda function’s IAM role and give it permission to publish to the topic.
Go back to the Lambda console for the function, select the Configuration tab, then select Permissions. You will see the
Execution role listed. Click on it to be navigated to the IAM console.
Click on Add permission and then Create inline policy.
In the inline policy visual editor, set the service to SNS and under actions, filter for and select Publish.
For Resources, enter the SNS topic ARN. Save the inline policy and make sure it is attached to the Lambda execution role.
SQS queue and subscribing to SNS
We want to be able to see our alerts whenever our Lambda publishes a message to the SNS topic, so we will create an SWS queue that will subscribe to messages from the SNS Topic and allow us to view messages in the AWS console.
Go to the SQS console and create a new FIFO queue like so:
After creating the queue, you will see an overview of the queue.
Click on Subscribe to Amazon SNS topic and select the SNS topic you created earlier and create a subscription.
Now whenever we publish alert messages to the SNS topic, those messages will appear in the SQS queue.
All of the required AWS resources are now set up. In the next section, we will implement the Lambda function code.
Lambda code
Now, let’s write the code that will be executed by the Lambda function to perform uptime checking.
We will write the Lambda function in JavaScript and use the Node.js runtime with NPM.
Create a new directory on your system and run:
Then, install the AWS SDK and fetch packages:
npm install aws-sdk node-fetchTo use ESM modules, open package.json and add "type": "module".
Open index.js and write the following code (Be sure to fill in your own website URL and SNS topic ARN.):
Note that in this simple implementation, we hardcoded the website URL and SNS topic ARN.
Typically, we would want to read these values from the event that triggered the Lambda function or from environment variables.
To upload the code and dependencies to the Lambda function, we need to zip the files.
On Linux, this can be done using:
zip -r package.zip . *This command will create a package.zip file in the current directory, and it will contain everything needed for Lambda.
On the Lambda function overview, go to the Code tab and upload the zip file containing the code.
Making sure everything works
We can run the Lambda manually in the Test tab.
If the Lambda receives a 200 OK status code from the website it is monitoring, it will not publish a message to the SNS topic
and will instead log that the website is up like so:
If we want to see the alerts in action, change the website URL to https://httpstat.us/400 so that we always get a 400 status code.
Reupload a new zip file to the Lambda function, then test again. The Lambda logs will indicate that the website is down that it published a message to the SNS topic.
To see the message, go to the SQS console for the queue and click on Send and receive messages. Click on Poll for messages, and you should be able to see messages in the queue. Click on it to see the message details, which should look like this JSON object:
{"Type": "Notification","MessageId": "a44c27f8-9c68-507e-8ba9-7d3b45a76115","SequenceNumber": "10000000000000011000","TopicArn": "arn:aws:sns:us-east-1:060780781184:Website-Alert-Topic.fifo","Subject": "Website Down Alert","Message": "Received 400 status code from https://httpstat.us/400","Timestamp": "2022-04-03T17:21:19.940Z","UnsubscribeURL": "<URL>"}
If you see a message like above in the queue, your uptime monitor is working! Try extending this simple implementation to cover multiple URL’s, regions, and more.
Cleanup
To avoid getting random bills from AWS, we need to clean up the resources we created.
Make sure to delete the Lambda function, EventBridge rule, SNS topic, and SQS queue when they are no longer needed.
Originally published at https://komonitor.com on April 4, 2022.
