OpenC2-org / openc2-working-group

The Open Command and Control Forum promotes the global development and adoption of the OpenC2 language and reference material.
Other
1 stars 0 forks source link

[Security Model] Need to define an optional extension to assure message integrity #3

Closed romanojd closed 8 years ago

romanojd commented 8 years ago

PROBLEM

When used as a command, an actuator must be able to detect possible changes to OpenC2 messages.


POTENTIAL SOLUTION

In many cases, OpenC2 commands can be wrapped in a container that will, among other things, address message integrity. In some cases, it will not be practical to use an external message integrity mechanism. In those instances, we should define an optional extension of the OpenC2 language to address the need to assure the integrity of commands.

jmbrule commented 8 years ago

There will be no argument wrt the necessity of message integrity, however there is little value in adding this to the openC2 syntax. ALL of the layers within the OSI stack have integrity mechanisms available, thus openC2 implementors can simpley reference one or more pre-exiting RFC's. Adding this complexity is contrary to openC2 design principles (lightweight, leverage pre-existing standards etc)

rsharo commented 8 years ago

Recommend identifying criteria for transport within the language spec, while explicitly leaving implementation out of the spec.

We should explicitly state that the OpenC2 contains no provisions for anti-tamper and that external provisions are required if comms are to be secure/reliable.

jmbrule commented 8 years ago

I would ike to close this action. Please provide additonal comments by COB 5/23/2016. I will take an action to produce a construct

romanojd commented 8 years ago

On behalf of Joe Brule, this issue has been resolved. The resolution details are described in this construct.