Document Threat Vectors/Profiles

[magowan]

Trust Model needs to be advanced first -> Issue #117
This then provides the terminology and concepts to describe various threats and how we seek to address them. Will link to wider documentation if that exists to identify the solution.

We can however start to accumulate a list of threats to describe

• Protecting guest memory from host (relates to use of TEE)
• Orchestration apis (e.g. kubectl exec) (allow list of kata-agent apis)
• Protecting container images requests from host tampering (relates to container pull)
• Identifying the guest is as intended before starting workload (attestation combined with delivery of a secret required for access to what needs kept confidential (data or container encryption)
• Protecting guest boot/root volume from host
• Protecting container bundles from host interference (relates to storage [WIP/RFC]: Add documentation for confidential storage architecture documentation#23)
• Protecting persistent storage ( solution relates to storage [WIP/RFC]: Add documentation for confidential storage architecture documentation#23)

Future Topics

• Protecting secrets (kubernetes secrets)

[ariel-adam]

@magowan is this issue still relevant or can be closed?
If it's still relevant to what release do you think we should map it to (mid-November, end-December, mid-February etc...)?