Notes
1 The word “incidents” is generally understood in the context of cybersecurity, wherein an attack or incidental compromise of systems affects the confidentiality, integrity, and availability of those systems and the data stored, hosted, or processed on them. However, there is a need to account for other, non-malicious triggers of failure, such as natural disasters, which can similarly affect cloud services.
2 This framework of incident response draws on the Cloud Security Alliance’s “Cloud Incident Response (CIR) Framework” and the National Institute of Standards and Technology’s “Computer Security Incident Handling Guide”. See: CSA, “Cloud Incident Response (CIR) Framework.” Cloud Security Alliance (CSA), May 4, 2021, https://cloudsecurityalliance.org/artifacts/cloud-incident-response-framework/ and NIST, “Computer Security Incident Handling Guide.” National Institute of Standards and Technology (NIST), U.S. Department of Commerce, Special Publication 800-61, Revision 2, August 2012, http://dx.doi.org/10.6028/NIST.SP.800-61r2.
3 See: Lydia Leong, “Multicloud failover is almost always a terrible idea,” Gartner, October 14, 2021, https://blogs.gartner.com/lydia_leong/2021/10/14/multicloud-failover-is-almost-always-a-terrible-idea/.
4 These may include information- and intelligence-sharing arrangements (including cross-border arrangements) between stakeholders which facilitate not just reporting but also detection (especially in the case of multinational corporations).
5 While the Cybersecurity and Infrastructure Security Agency (CISA) in the U.S. plays this role, not all governments maintain a similar coordinating body. See: CISA, “Cybersecurity and Infrastructure Security Agency,” Cybersecurity and Infrastructure Security Agency, n.d., https://www.cisa.gov/.
6 However, there is debate over whether the “cloud sector” can be considered a coherent and bounded sector for regulatory purposes.
7 See: CSA, “Cloud Incident Response (CIR) Framework,” Cloud Security Alliance (CSA), May 4, 2021, https://cloudsecurityalliance.org/artifacts/cloud-incident-response-framework/.
8 For example, as is the case with Government Intervention in Extremis, some may seek the authority to assume control over cloud providers or to direct providers to work with government agencies in cases of clear threats to national security.
9 Including information about the incident’s impact on the affected business’ mission and finances, as well as technical details, including the types of vulnerabilities exploited and indicators of similar incidents. See: NIST, “Computer Security Incident Handling Guide,” National Institute of Standards and Technology (NIST), U.S. Department of Commerce, Special Publication 800-61, Revision 2, August 2012, http://dx.doi.org/10.6028/NIST.SP.800-61r2.
10 While all customers benefit from insight into an incident and the response of concerned stakeholders, enterprise customers may particularly benefit from transparency and communication, given that contracts for enterprise cloud deployments often place additional security responsibilities on enterprise customers. Whereas consumer cloud offerings are almost entirely managed by the cloud provider.
11 Governments may wish to exercise significant control over a major cloud provider and direct its recovery efforts following a major cloud incident. For example, the Australian government has expressed that “In an emergency, we see a role for Government to use its enhanced threat picture and unique capabilities to take direct action to protect a critical infrastructure entity or system in the national interest. These powers would be exercised with appropriate immunities and limited by robust checks and balances. The primary purpose of these powers would be to allow Government to assist entities to take technical action to defend and protect their networks and systems” (Emphasis in original.). See: Australian Government, “Protecting Critical Infrastructure and Systems of National Significance,” Australian Government, Department of Home Affairs, August 2020, https://www.homeaffairs.gov.au/reports-and-pubs/files/protecting-critical-infrastructure-systems-consultation-paper.pdf and Australian Government, “Critical Infrastructure – Government assistance in practice,” (Diagram) Australian Government, Department of Home Affairs, n.d., https://www.homeaffairs.gov.au/reports-and-pubs/files/ci-government-assistance-in-practice.pdf.
12 Requirements on disclosure timelines to government agencies vary both within and among nations, with some governments calling for 72-hour timelines and others for as low as 12-hours. These requirements may also vary depending on the function and sectors (such as, critical infrastructure and government services) being served. See: Michael Kans, “Congress Debates Cyber Incident Reporting Deadlines in the NDAA,” Just Security, October 26, 2021, https://www.justsecurity.org/78745/congress-debates-cyber-incident-reporting-deadlines-in-the-ndaa/.
13 While this occurs in the U.S. under the auspices of the Cybersecurity and Infrastructure Security Agency (CISA) and in other individual countries, not all governments maintain a centralized cybersecurity agency to perform this function. See: CISA, “Cybersecurity and Infrastructure Security Agency.” Cybersecurity and Infrastructure Security Agency (CISA), n.d., https://www.cisa.gov/.
14 Cloud providers and their customers should develop and integrate incident notification matrices in their SLAs and/or contracts, laying out each party’s responsibilities in crisis communications. For instance, the Cloud Security Alliance sets out the following division of responsibilities for various cloud incident response scenarios: (1) For a security incident occurring in the platform or service layer for a PaaS or SaaS application, the response should be driven by the cloud provider; (2) if a security incident is occurring in the application layer for a PaaS application, the customer should be driving the response; and (3) in the case of a security incident occurring in the platform layer for an IaaS infrastructure cloud, the response should be driven jointly by the customer and the cloud provider to determine if it originated in the customer’s environment or the cloud provider’s environment. (See: CSA, “Cloud Incident Response (CIR) Framework,” Cloud Security Alliance (CSA), May 4, 2021, https://cloudsecurityalliance.org/artifacts/cloud-incident-response-framework/.) Additionally, the retention of digital forensic evidence should be seen as a shared responsibility. (See: Ben Martini and Kim-Kwang Raymond Choo, “An integrated conceptual digital forensic framework for cloud computing,” ScienceDirect, Digital Investigation, vol. 9, issue 2, November 2012, pages 71-80, https://www.sciencedirect.com/science/article/abs/pii/S174228761200059X.)
15 See: Michael Kans, “Congress Debates Cyber Incident Reporting Deadlines in the NDAA,” Just Security, October, 26 2021, https://www.justsecurity.org/78745/congress-debates-cyber-incident-reporting-deadlines-in-the-ndaa/.
16 This can be achieved through an Information Sharing Analysis Center (ISAC). For example, the Cloud Security Alliance runs the Cloud Cyber Incident Sharing Center (CloudCISC) which facilitates incident data sharing between participating cloud providers. (See: CSA, “CloudCISC,” Cloud Security Alliance (CSA), n.d., https://cloudsecurityalliance.org/research/working-groups/cloudcisc/.) In order to be successful, members must participate equally and actively, which requires that they have the willingness and ability to discuss security incidents that have affected their organizations. Participation in these programs by cloud service providers may not be uniform, with some members possibly contributing more actively than others.
17 Cloud providers and their customers should develop and integrate incident notification matrices in their SLAs and/or contracts, laying out each party’s responsibilities in crisis communications. For instance, the Cloud Security Alliance sets out the following division of responsibilities for various cloud incident response scenarios: (1) For a security incident occurring in the platform or service layer for a PaaS or SaaS application, the response should be driven by the cloud provider; (2) if a security incident is occurring in the application layer for a PaaS application, the customer should be driving the response; and (3) in the case of a security incident occurring in the platform layer for an IaaS infrastructure cloud, the response should be driven jointly by the customer and the cloud provider to determine if it originated in the customer’s environment or the cloud provider’s environment. (See: CSA, “Cloud Incident Response (CIR) Framework,” Cloud Security Alliance (CSA), May 4, 2021, https://cloudsecurityalliance.org/artifacts/cloud-incident-response-framework/.) Additionally, the retention of digital forensic evidence should be seen as a shared responsibility. (See: Ben Martini and Kim-Kwang Raymond Choo, “An integrated conceptual digital forensic framework for cloud computing,” ScienceDirect, Digital Investigation, vol. 9, issue 2, November 2012, pages 71-80, https://www.sciencedirect.com/science/article/abs/pii/S174228761200059X.)
18 See: Michael Kans, “Congress Debates Cyber Incident Reporting Deadlines in the NDAA,” Just Security, October, 26 2021, https://www.justsecurity.org/78745/congress-debates-cyber-incident-reporting-deadlines-in-the-ndaa/.
19 See: Michael Kans, “Congress Debates Cyber Incident Reporting Deadlines in the NDAA,” Just Security, October, 26 2021, https://www.justsecurity.org/78745/congress-debates-cyber-incident-reporting-deadlines-in-the-ndaa/.