Best Practice 9.1 – Understand your
strategy for SAP application and database security event analysis
Without keeping security logs at the appropriate levels of granularity, vital data required for incident response, forensic security analysis, and threat modeling can be lost. SAP security staff must be able to evaluate potential security incidents affecting SAP systems in alignment with your business security requirements. For SAP workloads running on AWS, the AWS services described in the Well-Architected Framework Security Pillar are a helpful starting point in conjunction with the following suggestions.
-
Well-Architected Framework [Security]: Detection – Configure
Suggestion 9.1.1 – Determine which logs are required to detect security events
For individual SAP software and supported databases refer to the SAP NetWeaver Guide
Finder as well as the SAP NetWeaver Security Guide for what logs might be applicable (for
example, read access logging
-
SAP Documentation: SAP NetWeaver Guide Finder
-
SAP Documentation: ABAP Platform Security Guide
-
SAP Documentation: Security Logging
Suggestion 9.1.2 – Develop mechanisms for storing and analyzing logs
Having relevant data regarding potential security events is necessary for any secure
SAP installation, but it is equally important to store that data securely and have the
necessary tools for searching and analyzing the data in an efficient and timely manner. One
option within AWS includes using the CloudWatch Agent to store instance logs and SAP application logs relevant to
security in an Amazon CloudWatch log group. Such logs could also be exported to
Amazon S3 for holistic log analysis and for integration with third-party log
analytics solutions
Refer to the following for help with assembling, combining, and analyzing your SAP on AWS security logs:
-
SAP Lens [Security]: Suggestion 7.4.4 - Consolidate user and authorization events in a Security Information and Event Management (SIEM) system for analysis
-
SAP on AWS Blog: Set up observability for SAP HANA databases with Amazon CloudWatch Application Insights
-
SAP on AWS Blog: SAP HANA monitoring: A serverless approach using Amazon CloudWatch
-
SAP on AWS Blog: SAP Monitoring: A serverless approach using Amazon CloudWatch
Suggestion 9.1.3 – Use machine learning to analyse and determine events of importance
Consider applying pattern recognition, anomaly detection, or both to security logs to
assist in determining potential threats and events of importance to your SAP workload. AWS
managed services, such as AWS Security Hub
-
AWS Video: An Overview of AWS Security Hub
-
AWS Documentation: Getting started with GuardDuty