Re: [cti-stix] Current thoughts on Event Object

Hi Sarah,

Below is my take on the different objects.

Object 2&3 can be combined but it might be better to leave them separate since the consumers that work with them are different today.

Objects	Purpose	Producer	Consumer	CTI Requirements
Object1	Captures data from sensors and is sent to a data analytics platform for further processing	Sensor - Produces events/alerts based on configured threshold rules Produces sightings to indicators	Data analytics platform that does further processing on the data	Event Trigger timestamp - time when the event was seen by the sensor Event sighted timestamp - time when the sighting was made Tool that made the sighting/sent the alert It could refer to a sighting or an alert or be an independent blob of data (for STIX unsupported notifications) Relationship to observed data or sighting
Object2	Captures details about an event/alert that needs further enrichment. Object2 might contain: Collection of Object1 Sightings Observed data Blob of non STIX data - artifact observables Object2 evolves to contain enrichment information, relationships to related objects and verdict/judgement IT Ticket/Case creation - Information in this object will be exported to case management tools in case it is a true positive Could do some preventative actions	Data Analytics platform output in form of Behavior analytics TI fusing with internal data Saved searches Alerts from Network behavior tools Alerts from Endpoint behavior tools User initiated ticket	SOC tools /analysts connect to different enrichment providers to gather additional context such as DNS, IP reputation, WhoIs info etc. Try to identify if the event is related to a historical event If new indicators are discovered/created, share them with threat intel products	Ability to store enrichment information against each artifact of the event. Example: relate IP->Domain->URL, Domain->registrant, IP->ISP->geographical area, IP, file hash->malware, campaign, threat actor etc. 2. Ability to relate to other historical records such as logs and flows - from forensic tools. 3. Ability to relate to older events/incidents 4. Ability to add judgement/verdict - true/false positive 5. Assignee - who is working on this event 6. Assignee notes - notes captured by the analyst/investigator 7. Property/Relationship to COAs for investigation & or preventative measures
Object3	IR team gets this object from User ticket/case Subset of object2 (with the bad artifacts only) and will do the following: Preventative actions - block/honeypot bad actors Triage - identify affected internal hosts Mitigative actions - Apply Segmentation policies for affected resources Remediative actions - program access policies, network policies, remove malware, reimage etc. Create/Update IT ticket/case	User ticket/case Subset of object2	IR team/tools to COAs to block/honeypot bad actors Triage - identify affected internal hosts COAs for segmentation Execute scripts to program access policies, network policies etc.	Relationship to Object2, if derived from it Property for Affected assets, hosts, servers Property for priority/severity Property/Relationship to COA with results of the COA execution Store timestamps for each of the actions
Object4	Subset of Object2 and 3 with relevant details			Report or Bundle?

Thanks,

Jyoti Verma

Technical Leader,

CTO office Security Business Group,

Cisco Systems Inc.

From: <cti-stix@lists.oasis-open.org> on behalf of Sarah Kelley <Sarah.Kelley@cisecurity.org>
Date: Monday, September 11, 2017 at 6:40 AM
To: "cti-stix@lists.oasis-open.org" <cti-stix@lists.oasis-open.org>
Subject: [cti-stix] Current thoughts on Event Object

CTI-TC,

I have made an attempt to summarize the discussion to date surrounding the Event object. I have tried to use the diagrams that we’ve been referencing that show several workflows and where I see the various SDOs we have been discussing.

I think there are five separate things that we’ve been talking about with regards to “Event”:

Object 1) Some sort of front end device sensing something. You could call this an alert, an event, a log line, etc.

Object 2) A somewhat more mature version of the thing above. It may be one to one, but it may not be. There may be many of Object #1 that make up one Object #2.

Object 3) A way to document an incident/investigation that is either ongoing or concluded. This may (or may not) have come from an evolution out of an Object 1 or 2. Likely, most IR tools will not use Object 3 directly, so it should probably be a sort of summary object that can allow the results of an IR investigation to be linked back to all the related data (like Objects 1 and 2, but also Actors, malware, etc)

Object 4) Some sort of automated method of reporting, for cases such as mandatory reporting (like to US CERT).

Object 5) A way to group a bunch of related data together.

I see these as five distinct SDOs. Objects 1 and 2 are very similar, and could be represented by one SDO, but if that was the case, then the SDO would need to have a relationship type of “this arglebargle is one of many arglebargles that make up this larger arglebargle”.

Object 1 has a lot of logistical overlap with a Sighting. There would need to be some sort of deconfliction between what properties were trying to be represented by Object 1 to make sure it wasn’t just a duplicate of Sighting (which, by definition, is something that was ‘seen’, just as Object 1 was described to be). This could also just be an Observed Data object, if the it wasn’t actually “sighted”.

Object 2 is something that would likely be looked at by a SOC analyst. It isn’t really something that involves IR, but rather “is this thing my sensor generated good or bad?”

Object 3 is for your IR people or your CERT.

Object 4 is likely distinct from the Report object as we have it now, as it’s not likely to be in the same vein as a ‘published report’, but rather, “Here are the series of questions that US CERT makes me answer every time we have an ‘incident’”.

Object 5 is basically what MISP has been asking for.

I believe Objects 1-3 roughly correlate to the first three objects from the FireEye proposal of “Event”, “Alert”, and “Investigation”. Object 5 is closer to their “Grouping” object.

All that being said, many of these tasks are not currently done in STIX (if not most of them). In the diagram, we have the TIP (in green) as being a separate object that lives alongside the workflow, but isn’t really IN the workflow. That being said, if we added objects 1, 2, 4, and 5, I think it could allow for easier data flow into/out of a TIP. (I don’t think the IR object will ever be used directly by the types of tools that produce that data, but maybe I’m wrong.)

Personally, I think the Event object as it currently stands is somewhat of a combination of Object 2 and Object 3. If people agree that these are really separate objects, I think we could scope out a few properties and turn the current Event object into Object 2 or Object 3 fairly easily (or easily split it into two objects). I think Object 1 is out of scope for 2.1 (unless it’s already covered by sighting/Observed Data). I think Object 4 is out of scope for 2.1. I think Object 5 will be covered by the “collection vs. report” debate/object, for which we should soon have a proposal.

Thoughts are appreciated.

Sarah Kelley

Senior Cyber Threat Analyst

Multi-State Information Sharing and Analysis Center (MS-ISAC)

31 Tech Valley Drive

East Greenbush, NY 12061

sarah.kelley@cisecurity.org

518-266-3493

24x7 Security Operations Center

SOC@cisecurity.org - 1-866-787-4722

This message and attachments may contain confidential information. If it appears that this message was sent to you by mistake, any retention, dissemination, distribution or copying of this message and attachments is strictly prohibited. Please notify the sender immediately and permanently delete the message and any attachments.

. . . . .
.....

cti-stix message