How to Schedule RDS Instances with an AWS Lambda function?

Why Use an AWS Lambda Function?

Creating an AWS scheduler to control the start and stop time of your database achieves a reduction of the cost that comes with holding up active resources. You don’t need all your databases running 24/7, and you manually choosing when to stop the DB is not an ideal long-term solution. Fortunately, you can schedule the activity of your database without manual interference.

Unlike the very simple option of stopping your DB through the AWS console, this process would be a bit more detailed. The added advantage of a flexible AWS scheduler and not having to worry about the DB starting automatically a week later makes the small process more worth it.

How to Write an AWS RDS Stop Instance Function on Lambda?

Create a policy

First, we must create a policy for the IAM role to be attached to.

• Go to the AWS console
• Choose IAM service
• Click on Policies
• Click on Create Policy
• Click on the JSON tab and paste the JSON shown below to gain access to certain RDS actions and Cloud Watch Log Events

AWS has a visual editor for creating the policy but you can just copy-paste the JSON code below for your convenience.

{
“Version”: “2012–10–17”,
“Statement”: [
{
“Effect”: “Allow”,
“Action”: [
“rds:DescribeDBInstances”,
“rds:StopDBInstance”,
“rds:StartDBInstance”
],
“Resource”: “*”
},
{
“Effect”: “Allow”,
“Action”: [
“logs:CreateLogGroup”,
“logs:CreateLogStream”,
“logs:PutLogEvents”
],
“Resource”: “*”
}
]
}

• Give the policy a name, manual_RDS_Schedule and add a description. Click on ‘Create Policy’

Create a role

The IAM role will allow making service requests which are necessary for scheduling the AWS RDS stop instance, it also grants access to CloudWatch logs.

• Go to the IAM console
• Go to Roles
• Click on Create Roles
• Choose Lambda in the AWS Service section
• Search for Policy_ on the search bar and the policy you constructed earlier will pop up.
• Give the role a name, ManualScheduleRDS and click Create Role.

• Attach the policy, manual_RDS_Schedule to the role

Create the AWS Lambda function

Before creating a Lambda function, make sure your region is the same as the region where you created the DB instance.

Since you can control instances in multiple regions, choosing the right region is necessary for the function to work. You could use the same function later on for your instances hosted on multiple regions.

You’ll be able to receive the AWS RDS stop instance name and the Availability Zone from the administrator.

• Go to services
• Click on Lambda
• Click on Create Function
• Choose Author from Scratch

Enter the following information on the window:

• Name: StopRDSInstances
• Runtime: Python 3.8
• Role: Choose an existing role
• Role Name: ManualScheduleRDS
• Click on Create Function

You will find an ARN resource created for the Lambda function in the top right corner. The resource allows the function to access the GetFunctionConfiguration Lambda API and the AWS lambda environment variable.

Now, return to the ManualScheduleRDS role and click on Add Inline Policy.

Now, paste the following JSON in the editor.

{
“Version”: “2012–10–17”,
“Statement”: [
{
“Effect”: “Allow”,
“Action”: “lambda:GetFunctionConfiguration”,
“Resource”: “arn:aws:lambda:<region>:<Account_id>:function:StopRDSInstances”
}
]
}

Assign the ARN on the JSON code and then click Save.

AWS Lambda will now appear as a Resource in the function.

Update the Code

Add the following Python code to the function.

import sys
import botocore
import boto3
from botocore.exceptions import ClientError
def lambda_handler(event, context):
rds = boto3.client(‘rds’)
lambdaFunc = boto3.client(‘lambda’)
print(‘Trying to get Environment variable’)
try:
funcResponse = lambdaFunc.get_function_configuration(
FunctionName=’StopRDSInstances’
)
DBinstance = funcResponse[‘Environment’][‘Variables’][‘DBInstanceName’]
print(‘Stopping RDS service for DBInstance : ‘ + DBinstance)
except ClientError as e:
print(e)
try:
response = rds.stop_db_instance(
DBInstanceIdentifier=DBinstance
)
print(‘Success :: ‘)
return response
except ClientError as e:
print(e)
return
{
‘message’ : “Script execution completed. See Cloudwatch logs for complete output”
}

The line of code responsible for stopping the RDS is rds.stop_db_instance.

The AWS Lambda environment variable, DBInstanceName is held by the variable DBInstance. This environment variable is the key-value pair that will let you use this Lambda function in other environments like Dev, QA or Prod.

Configure the function

• Enter the Key-Value pair, DBInstanceName and <userDB instance name>.
• Set the execution role to ManualScheduleRDS
• Save the function

Test the function

• Select Test and then Configure Test from above
• Select Create New Test Events
• Choose the Hello World event template

Save and you’re done! Now you have a Lambda function to stop your RDS instance. If you receive a message saying “Instance<> is not in an available state”, then your DB is not started.

Moving onto Start Functions

The process for setting up the Start function is identical to the Stop function, with most of the changes going to the names and some lines of code.

• Name your function StartRDSInstances
• Update the Inline policy with the following JSON

{
“Version”: “2012–10–17”,
“Statement”: [
{
“Effect”: “Allow”,
“Action”: “lambda:GetFunctionConfiguration”,
“Resource”: “arn:aws:lambda:<region>:<Account_id>:function:StartRDSInstancest”
}
]
}

• Save the policy as policy_rds_start
• Update the function code

import sys
import botocore
import boto3
from botocore.exceptions import ClientError
def lambda_handler(event, context):
# TODO implement
rds = boto3.client(‘rds’)
lambdaFunc = boto3.client(‘lambda’)
print(‘Trying to get Environment variable’)
try:
funcResponse = lambdaFunc.get_function_configuration(
FunctionName=’StartRDSInstances’
)
#print (funcResponse)
DBinstance = funcResponse[‘Environment’][‘Variables’][‘DBInstanceName’]
print(‘Starting RDS service for DBInstance : ‘ + DBinstance)
except ClientError as e:
print(e)
try:
response = rds.start_db_instance(
DBInstanceIdentifier=DBinstance
)
print(‘Success :: ‘)
return response
except ClientError as e:
print(e)
return
{
‘message’ : “Script execution completed. See Cloudwatch logs for complete output”
}

• Follow the rest of the steps as mentioned earlier.

How to Set the Schedule?

Now that you have the two Lambda functions at your beck and call, you just need to determine when to invoke them. For this, you will have to set a rule on AWS CloudWatch.

• Go to CloudWatch
• Click on Rules
• Choose ‘Schedule’ and then add the Cron expression to choose your time of the day
• Add either the start or stop Lambda functions created earlier as the ‘Target’ to invoke

Your scheduled RDS Instance is now ready to go, you need not concern yourself with stopping and restarting your instances anymore. For an easier alternative, you could automate this entire process with simple workflows.

Check out the tutorial video here.