The following analytic identifies an AWS IAM account with concurrent sessions coming from more than one unique IP address within the span of 5 minutes. This behavior could represent a session hijacking attack whereby an adversary has extracted cookies from a victims browser and is using them from a different location to access corporate online resources. When a user navigates the AWS Console after authentication, the API call with the event name
DescribeEventAggregates is registered in the AWS CloudTrail logs. The Splunk Threat Research team leveraged this event name to identify 2 concurrent sessions. The presence of this event occurring from two different IP addresses is highly unlikely. As users may behave differently across organizations, security teams should test and customize this detection to fit their environments.
- Type: TTP
Product: Splunk Enterprise, Splunk Enterprise Security, Splunk Cloud
- Last Updated: 2023-02-01
- Author: Bhavin Patel, Splunk
- ID: 51c04fdb-2746-465a-b86e-b413a09c9085
Kill Chain Phase
- CIS 10
1 2 3 4 5 `cloudtrail` eventName = DescribeEventAggregates src_ip!="AWS Internal" | bin span=5m _time | stats values(userAgent) values(eventName) values(src_ip) as src_ip dc(src_ip) as distinct_ip_count by _time user_arn | where distinct_ip_count > 1 | `aws_concurrent_sessions_from_different_ips_filter`
The SPL above uses the following Macros:
aws_concurrent_sessions_from_different_ips_filter is a empty macro by default. It allows the user to filter out any results (false positives) without editing the SPL.
List of fields required to use this analytic.
How To Implement
You must install Splunk AWS Add on and Splunk App for AWS. This search works with AWS CloudTrail logs.
Known False Positives
A user with concurrent sessions from different Ips may also represent the legitimate use of more than one device. Filter as needed and/or customize the threshold to fit your environment.
Associated Analytic Story
|42.0||70||60||User $user_arn$ has concurrent sessions from more than one unique IP address $src_ip$ in the span of 5 minutes.|
The Risk Score is calculated by the following formula: Risk Score = (Impact * Confidence/100). Initial Confidence and Impact is set by the analytic author.
source | version: 1