KodeKloud Notes

In this guide, we'll configure cross-account access between a Production AWS account (owns an S3 log bucket) and a Development AWS account (hosts a Log Analysts group). This setup enables secure, temporary access to logs without sharing long-term credentials.

Scenario

AWS Account	Resource	Purpose
Production Account	S3 bucket (`log-bucket`)	Stores application log files
Development Account	IAM Group (`Log Analysts`)	Needs permission to list and read log files

Our objective is to let the Log Analysts group assume a role in the Production account to retrieve logs.

High-Level Architecture

Create an IAM Role in the Production Account
Attach an inline S3 policy to that role
Update the S3 Bucket Policy to trust the role
Assume the role from the Development Account and verify access

The image is a diagram showing a request to provide log access to a Log Analysts group, involving a production account with an S3 bucket and a dev account with a log access role.

Cross-Account Access Components

Component	Description
Trust Relationship	IAM Role trust policy in the Production account allowing Dev account to assume it
Role Assumption	`sts:AssumeRole` call from Dev account for temporary credentials
Permissions Boundary	Inline policy (or managed) on the role controls S3 access
Resource Policy	S3 Bucket policy grants the role `s3:ListBucket` and `s3:GetObject`

The image explains IAM Cross Account Access Capability, highlighting the implementation of cross-account access, the need for a trust relationship, role assumption by users, and the security benefits of resource isolation.

Note

Be explicit in your trust policy to avoid granting unintended access. Restrict Principal to specific IAM roles or account IDs.

Demo Walkthrough

Follow these steps to implement and test cross-account S3 access.

1. Create the IAM Role in Production

Create a trust policy (trust-policy.json):

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Principal": {
        "AWS": "arn:aws:iam::DEV_ACCOUNT_ID:root"
      },
      "Action": "sts:AssumeRole"
    }
  ]
}

Use the AWS CLI to create the role:

aws iam create-role \
  --role-name DevLogAccessRole \
  --assume-role-policy-document file://trust-policy.json \
  --description "Allows Dev account to access logs" \
  --profile prod-account

2. Attach an Inline S3 Access Policy

Define s3-access-policy.json:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "s3:ListBucket",
        "s3:GetObject"
      ],
      "Resource": [
        "arn:aws:s3:::log-bucket",
        "arn:aws:s3:::log-bucket/*"
      ]
    }
  ]
}

Attach it to the role:

aws iam put-role-policy \
  --role-name DevLogAccessRole \
  --policy-name S3LogAccess \
  --policy-document file://s3-access-policy.json \
  --profile prod-account

3. Update the S3 Bucket Policy

Create or edit your bucket policy (bucket-policy.json):

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Sid": "AllowDevRoleAccess",
      "Effect": "Allow",
      "Principal": {
        "AWS": "arn:aws:iam::PROD_ACCOUNT_ID:role/DevLogAccessRole"
      },
      "Action": [
        "s3:ListBucket",
        "s3:GetObject"
      ],
      "Resource": [
        "arn:aws:s3:::log-bucket",
        "arn:aws:s3:::log-bucket/*"
      ]
    }
  ]
}

Apply it:

aws s3api put-bucket-policy \
  --bucket log-bucket \
  --policy file://bucket-policy.json \
  --profile prod-account

Warning

Ensure the bucket policy’s Principal matches the exact ARN of the role. Using wildcards may expose your bucket to unintended access.

4. Assume the Role and Verify Access

From the Development account, assume the role:

aws sts assume-role \
  --role-arn arn:aws:iam::PROD_ACCOUNT_ID:role/DevLogAccessRole \
  --role-session-name LogAnalysisSession \
  --profile dev-account > assume-role-output.json

Export temporary credentials:

export AWS_ACCESS_KEY_ID=$(jq -r '.Credentials.AccessKeyId' assume-role-output.json)
export AWS_SECRET_ACCESS_KEY=$(jq -r '.Credentials.SecretAccessKey' assume-role-output.json)
export AWS_SESSION_TOKEN=$(jq -r '.Credentials.SessionToken' assume-role-output.json)

List and retrieve logs:

aws s3 ls s3://log-bucket
aws s3 cp s3://log-bucket/example.log .

References

Watch Video

Watch video content