NOTES: Dave W, Bill M JABBER: Stephen B. SACM WG @ IETF 105 17:40 - 19:10, Thursday July 25th, Van Horn ================== Intro / Agenda Bash - Chairs N/A WGLC wrap-up on ROLIE Software Descriptor Extension (Banghart) https://tools.ietf.org/wg/sacm/draft-ietf-sacm-rolie-softwaredescriptor/ Updates have been made in accordance with comments -07 Extended last call comments into -08 One "major" change is just a typo (application/coswid+cbor --> application/swid+cbor) Will move the draft forward post-WGLC WGLC wrap-up on Concise Software Identifiers (COSWID) (Waltermire) https://tools.ietf.org/wg/sacm/draft-ietf-sacm-coswid/ CBOR-encoded SWID; the "younger brother" of SWID tags 9 or so reviews during WGLC 3 reviewers with detailed comments - addressed by -12 revision 1 additional typo still in the -12 Mostly clarification stuff - editorial issues 1 normative change - UUID = 16-byte bstr New IANA registrations - "swid", "swidpath" and a registry for using CoSWID with SWIMA all open issues addressed, changes made based on all provided comments Ira on jabber - notes on RFC5198; to do with LF/CRLFs - Henk: its just "baggage from telnet"; dont really have to really worry about it right now - Ira: Agrees with keeping the 5198 reference - Kathleen would like to read it 2-week "double check" post-WGLC WGLC wrap-up on Endpoint Posture Collection (ECP) (Haynes/Fitzgerald-McKay) https://tools.ietf.org/wg/sacm/draft-ietf-sacm-ecp/ Lots of reviews Jess incorporating changes eventually; plan for an update in September Henk sent in comments today No show stoppers - Henk might not make it past his 97 issues Kathleen: CARIS Workshop - Adoption and (lack of) adoption on protocols For those protocols/data formats -- The work done didnt match what was wanted Using IODEF as storage format, but not necessarily for exchange Are things on the right track towards adoption Adam: Members want tools to interoperate quickly and easily - Tripwire blog: Cooperative ecosystem based on open and available tools, using standards Brett (Vendor Hat): The more complex the standard is, the less likely it will be implemented - Want simple, easy to integrate - Boil things down to bite-sized pieces that people can understand Adam: The standards-based stuff isnt the secret sauce, not the competitive advantage. Would having open tools/libraries help Brett: Would depend on the integrations and liability of the open source code and support level Standards are read/understood when the business requires it Henk: Look at the hackathons Dave: Vendors/NIST came initially to work on the next-gen of SCAP/OVAL/XCCDF SCAP2.0 effort is an attempt to rebuild the community - meeting outside of the IETF, new authoring formats around XCCDF/OVAL Hum: Should the WG devote some time in a virtual interim for more discussion on this? Summary: Strong support for, no objections. SACM Hackathon Progress Report (Munyan) Bill reviewed slides to share this hackathon's results. Used a modified OVAL by decoupling collection from the typically monolithic OVAL document model. System characteristics were collected when requested to do so and results were provided to a MAP of CBOR data. XML could be translated back from the MAP to XML. Slide: Thanks Stephen Banghart: Thanks for all of your (and Carl's) work. It would be great if you could brief the SCAP endpoint collection group. Henk: This is getting at what caused us to get stuck. This shows a good proof of concept that shows the operations needed to get clarity moving forward. This is a good basis to create a data model. More interation will be needed to complete this, and we might not want to rely on the hackathon alone. Bill: Is there a more generic way to represent something like the OVAL data that doesn't constanly require modifications for adding new data. Henk: A general purpose CDDL-based parser is needed for this. This work is in the queue. SACM Architecture (Montville) https://tools.ietf.org/wg/sacm/draft-ietf-sacm-arch/ (No updated draft - discuss next steps) Pending changes to the draft Adam/Bill working on now Draft clarifications Capability vs. Interface vs. Operation ----------------------- | v Interaction Sub-Architecture - Various collection systems; OVAL collection engine, EPCP implementations, YANG PUSH, etc Cooperative Tooling Architecture allows multiple approaches to work together (e.g., EPCP, OVAL) We will propose specific components and interactions for configuration management activities We will propose these things in the draft and move them out if/when necessary - If something crosses the line to solutions, we'll make the decision to move it out Henk: I am interested in helping with the operations in September. Arnaud: 3 points. 1) The diagram is not very friendly to those color blind (especially the right side). 2) Is there a formal language for capabilities/interfaces/operations? 3) There is maybe a link between this work and other endpooint work in the ITU, can ITU create a statement of interest to send to the IETF?. ROMAN: Yes. Describe the messaging infrastructure that allows transportation of a payload. Describe the instructions components can give to each other to operate on that payload - "Tell": Do a collection - "Ask": Watch this set of information for changes WGLC Feb 2020 ? - Information Model (Inacio) (No draft - discuss next steps) Chris posted a very rough -00 with only a handful of elements There's a composite type structure in there (Henk: Yes keep that) An "opaque" information field and a field of the opaque data type - If you can just have a blob of data, you dont have to standardize other elements - If the other party can understand the blob, then that's fine Actions: Some lists and enumerations - Will work with Bill/Adam to align with architecture Ira on Jabber: "Opaque" as a reference to the "x-name"(?) Dave Waltermire: This relates to the guidance in BCP178. Henk: This should be driven by running code. You should provide a category. Chris: I have "data use type". Chair: We can set the adoption milestone for this to September 2019. ? - Terminology (Birkholz/Montville) https://tools.ietf.org/wg/sacm/draft-ietf-sacm-terminology/ (No updated draft - discuss next steps) WG Status / Way Ahead - Chairs Roman: Milestone dates for those documents NOT in WGLC ROLIE SW Descriptor and CoSWID - Aug 2019 to IESG Architecture - Feb 2020 WGLC EPCP - October 2019 WGLC ROLIE Checklist Descriptor: WGLC March 2020 YANG PUSH January 2020 EPCP - Need to ask Jess about this one. Terminology - Architecture + 4 months (June 2020) VIRTUAL INTERIM Early Sept 9 or 16 Early to Mid Oct 14 AOB None.