Case studies

Case studies

Earlier in Introduction to case studies we introduced the Alice and Bob case studies where Alice is deploying a private government cloud and Bob is deploying a public cloud each with different security requirements. Here we discuss how Alice and Bob would address monitoring and logging in the public vs a private cloud. In both instances, time synchronization and a centralized store of logs become extremely important for performing proper assessments and troubleshooting of anomalies. Just collecting logs is not very useful, a robust monitoring system must be built to generate actionable events.

Alice’s private cloud

As Alice is building out a new cloud within an existing organization, she is able to leverage several existing tools inside the existing business unit to help monitor and log her environment. She assigns a resource to the Security Operations Center (SOC) to monitor and respond to alerts coming from the new infrastructure. She uses a currently existing Security Event and Incident Management (SEIM) solution, and configures secure logging to the SEIM event collector. Alice and the SOC analyst build the SEIM views so that logs are correlated by type, and trigger alerts on unexpected or “interesting” events, such as a successful login by a user immediately after a string of failed login attempts within a given timeframe. The SOC analyst is also given escalation protocols and contact information so that when an specific event occurs, the analyst will escalate to the proper resource and coordinate communications about the event.

Alice ensures that only her resource in the SOC has access to SEIM feeds from her systems as well as the database back-end. She also gives the services teams read-only access to separate views SEIM feeds so that they can perform compliance-related duties such as regular log review or create tickets for tracking and other engagement. Finally, based on the default reporting that is currently built out, Alice may choose to hire a consultant to ensure specific events are captured for compliance purposes, or that specific events trigger the proper SEIM workflows.

Bob’s public cloud

Bob is also building out a new cloud group, and outlines both a Security Operations Center (SOC), as well as the processes and procedures for responding to security events. As Bob’s cloud will be able to be used by “anyone with a credit card” he wants to have a well-defined policy in place for issues that may arise. He outlines how SOC analysts can respond to malicious traffic detection, instances being compromised, external requests, and more. To aid the SOC he configures an Elasticsearch, Logstash, and Kibana (ELK) cluster to collect, correlate and identify security events. Bob also builds a Django based web application for his customers to perform self-service log retrieval from the Logstash systems. He does not allow user requests and ensures tenant validation so that tenants only have access to their own log events. Finally, Bob configures an NTP server to ensure all events maintain a consistent timeline.

Creative Commons Attribution 3.0 License

Except where otherwise noted, this document is licensed under Creative Commons Attribution 3.0 License. See all OpenStack Legal Documents.