| MILE | C. Inacio |
| Internet-Draft | CMU |
| Intended status: Informational | D. Miyamoto |
| Expires: May 17, 2017 | UTokyo |
| November 13, 2016 |
MILE Implementation Report
draft-ietf-mile-implementreport-10
This document is a collection of implementation reports from vendors, consortiums, and researchers who have implemented one or more of the standards published from the IETF INCident Handling (INCH) and Management Incident Lightweight Exchange (MILE) working groups.
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."
This Internet-Draft will expire on May 17, 2017.
Copyright (c) 2016 IETF Trust and the persons identified as the document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.
This draft is a collection of information about Security Incident reporting protocols, and the implementation of systems that use them to share such information. It is simply a collection of information, it makes no attempt to compare the various standards or implementations. As such, it will be of interest to Network Operators who wish to collect and share such data.
Operationally, Operators would need to decide which incident data collection group they want to be part of, that choice will strongly influence their choice of reporting protocol and applications to gather and distribute the data.
This document is a collection of implementation reports from vendors and researchers who have implemented one or more of the standards published from the INCH and MILE working groups. The standards include:
The implementation reports included in this document have been provided by the team or product responsible for the implementations of the mentioned RFCs. Additional submissions are welcome and should be sent to the draft editor. A more complete list of implementations, including open source efforts and vendor products, can also be found at the following location:
The Anti-Phishing Working Group (APWG) is one of the biggest coalitions against cybercrime, especially phishing. In order to collect threat information in a structured format, APWG provides a phishing and cybercrime reporting tool which sends threat information to APWG by tailoring information with IODEF format, based on RFC5070 and RFC5901.
The Advanced Cyber Defense Centre (ACDC), is an European wide activity to fight against botnets. ACDC provides solutions to mitigate on-going attacks, as well as consolidating information provided by various stakeholders into a pool of knowledge. Within ACDC, IODEF is one of the supported schema for exchanging the information.
Research and Education Networking Information Sharing and Analysis Center (REN-ISAC) is a private community of the research and higher education members for sharing threat information, and employs IODEF formatted-messages to exchange information.
REN-ISAC also recommends using an IODEF attachment provided with a notification email for processing rather than relying on parsing of the email body text. The tools provided by REN-ISAC is designed to handle such email.
The EMC/RSA RID agent is an open source implementation of the IETF standards for the exchange of incident and indicator data. The code has been released under a MIT license and development will continue with the open source community at the Github site for RSA Intelligence Sharing:
The code implements the RFC6545, Real-time Inter-network Defense (RID) and RFC6546, Transport of RID over HTTP/TLS protocol. The code supports the evolving RFC5070-bis Incident Object Description Exchange Format (IODEF) data model from the work in the IETF working group Managed Incident Lightweight Exchange (MILE).
Japan's National Institute of Information and Communications Technology (NICT) Network Security Research Institute implemented open source tools for exchanging, accumulating, and locating IODEF-SCI (RFC7203, [RFC7203]documents.
Three tools are available from GitHub. These tools assist the exchange of IODEF-SCI documents between parties. IODEF-SCI is RFC7203 that extends IODEF so that IODEF document can embed structured cybersecurity information (SCI). For instance, it can embed MMDEF, CEE, MAEC in XML and CVE identifiers.
The three tools are generator, exchanger, and parser. The generator generates IODEF-SCI documents or appends an XML to an existing IODEF document. The exchanger sends the IODEF document to a specified correspondent node. The parser receives, parses, and stores the IODEF-SCI document. The parser also creates an interface that enables users to locate IODEF-SCI documents which have previously been received. The code has been released under a MIT license and development will continue on GitHub.
Note that users can enjoy using this software at their own risk.
Available Online:
n6 is a platform for processing security-related information, developed by NASK (Poland Research and Academic Computer Network), Computer Emergency Response Team (CERT) Polska. The n6 API provides a common and unified way of representing data across the different sources that participate in knowledge management.
n6 exposes a REST-ful API over HTTPS with mandatory authentication via TLS client certificates, to ensure confidential and trustworthy communications. Moreover, it uses an event-based data model for representation of all types of security information.
Each event is represented as a JSON object with a set of mandatory and optional attributes. n6 also supports alternative output data formats for keeping compatibility with existing systems - IODEF and CSV - although these formats lack some of the attributes that may be present in the native JSON format.
Available Online:
Deep-Secure Guards are built to protect a trusted domain from:
Deep-Secure Guards support HTTPS and XMPP (optimised server to server protocol) transports. The Deep-Secure Guards support transfer of XML based business content by creating a schema to translate the known good content to and from the intermediate format. This means that the Deep-Secure Guards can be used to protect:
Deep-Secure Guards also support the SMTP transport and perform deep content inspection of content including XML attachments. The Mail Guard supports S/MIME and Deep Secure is working on support for the upcoming PLASMA standard which enables an information centric policy enforcement of data use.
The Incident Object Description Exchange Format, documented in the RFC5070, defines a data representation that provides a framework for sharing information commonly exchanged by Computer Security Incident Response Teams (CSIRTs) about computer security incidents. IncMan Suite implements the IODEF standard for exchanging details about incidents, either for exporting or importing activities. This has been introduced to enhance the capabilities of the various Computer Security Incident Response Teams (CSIRT), to facilitate collaboration and sharing of useful experiences, sharing awareness on specific cases.
The IODEF implementation is specified as an XML schema, therefore all data are stored in an xml file; in this file all the data of an incident are organized in a hierarchical structure to describe the various objects and their relationships.
The IncMan Suite relies on IODEF as a transport format, composed by various classes for describing the entities which are part of the incident description. For instance the various relevant timestamps (detection time, start time, end time, and report time), the techniques used by the intruders to perpetrate the incident, the impact of the incident, technical and non-technical (time and monetary) and obviously all systems involved in the incident.
Each incident defined in the IncMan Suite can be exported via a User Interface feature and it will create a xml document. Due to the nature of the data processed, the IODEF extraction might be considered privacy sensitive by the parties exchanging the information or by those described by it. For this reason, specific care needs to be taken in ensuring the distribution to an appropriate audience or third party, either during the document exchange and subsequent processing.
The xml document generated will include description and details of the incident along with all the systems involved and the related information. At this stage it can be distributed for import into a remote system.
The IncMan Suite provides the functionality to import incidents stored in files and transported via IODEF-compliant xml documents. The importing process comprises of two steps: first, the file is inspected to validate if it is well formed, then all data are uploaded inside the system.
If the incident already exists in the system with the same incident id, the new one being imported will be created under a new id. This approach prevents accidentally overwriting existing info or merging inconsistent data.
The IncMan Suite also includes a feature to upload incidents from emails.
The incident, described in xml format, can be stored directly into the body of the email message or transported as an attachment of the email. At regular intervals, customizable by the user, the IncMan Suite monitors for incoming emails, filtered by a configurable white-list and black-list mechanism on the sender's email account, then a parser processes the received email and a new incident is created automatically, after having validated the email body or the attachment to ensure it is well formed format.
XMPP is enhanced and extended through the XMPP Extension Protocols (or XEPs). XEP-0268 [XEP-0268] describes incident management (using IODEF) of the XMPP network itself, effectively supporting self-healing the XMPP network. In order to more generically cover the incident management of a network over the same network, XEP-0268 requires some updates. We are working on these changes together with a new XEP that supports "social networking" over XMPP, enhancing the publish-and-subscribe XEP (XEP-0060 [XEP-0060]). This now allows nodes to publish and subscribe to any type of content and therefore receive the content. XEP-0060 will be used to describe IODEF content. We now have an alpha version of the server-side software and client-side software required to demonstrate the "social networking" capability and are currently enhancing this to support Cyber Incident management in real-time.
MANTIS provides an example implementation of a framework for managing cyber threat intelligence expressed in standards such as STIX, CybOX, IODEF, etc. The aims of providing such an example implementation are:
HP has developed HP Threat Central, a security intelligence platform that enables automated, real-time collaboration between organizations to combat today's increasingly sophisticated cyber attacks. One way automated sharing of threat indicators is achieved is through close integration with the HP ArcSight SIEM for automated upload and consumption of information from the Threat Central Server. In addition HP Threat Central supports open standards for sharing threat information so that participants who do not use HP Security Products can participate in the sharing ecosystem. It is planned that future versions also support IODEF for the automated upload and download of threat information.
DAEDALUS is a real-time alert system based on a large-scale darknet monitoring facility that has been deployed as a part of the nicter system of NICT, Japan. DAEDALUS consists of an analysis center (i.e., nicter) and several cooperative organizations. Each organization installs a darknet sensor and establishes a secure channel between it and the analysis center, and continuously forwards darknet traffic toward the center. In addition, each organization registers the IP address range of its livenet at the center in advance. When these distributed darknet sensors observe malware activities from the IP address of a cooperate organization, then the analysis center sends an alert to the organization. The future version of DAEDALUS will support IODEF for sending alert messages to the users.
Collaborative Incident Management System (CIMS) is a proof-of-concept system for collaborative incident handling and for the sharing of cyber defence situational awareness information between the participants, developed for the Cyber Coalition 2013 (CC13) exercise organized by NATO. CIMS was implemented based on Request Tracker (RT), an open source software widely used for handling incident response by many CERTs and CSIRTs.
One of the functionality implemented in CIMS was the ability to import and export IODEF messages in the body of emails. The intent was to verify the suitability of IODEF to achieve the objective of collaborative incident handling. The customized version of RT could be configured to send an email message containing an IODEF message whenever an incident ticket was created, modified or deleted. These IODEF messages would then be imported into other incident handling systems in order to allow participating CSIRTs to use their usual means for incident handling, while still interacting with those using the proof-of-concept CIMS. Having an IODEF message generated for every change made to the incident information in RT (and for the system to allow incoming IODEF email messages to be associated to an existing incident) would in some way allow all participating CSIRTs to actually work on a "common incident ticket", at least at the conceptual level. Of particular importance was the ability for users to exchange information between each other concerning actions taken in the handling of a particular incident, thus creating a sort of common action log, as well as requesting/tasking others to provide information or perform specified action and correlating received responses to the original request or tasking. As well, a specific "profile" was developed to identify a subset of the IODEF classes that would be used during the exercise, in an attempt to channel all users into a common usage pattern of the otherwise flexible IODEF standard.
AirCERT was implemented by CERT/CC of Carnegie Mellon's Software Engineering Institute CERT division. AirCERT was designed to be an Internet-scalable distributed system for sharing security event data. The AirCERT system was designed to be an automated collector of flow and IDS alerts. AirCERT would collect that information into a relational database and be able to share reporting using IODEF and Intrusion Detection Message Exchange Format (RFC4765, [RFC4765]). AirCERT additionally used SNML [SNML] to exchange information about the network. AirCERT was implemented in a combination of C and Perl modules and included periodic graphing capabilities leveraging RRDTool.
AirCERT was intended for large scale distributed deployment and eventually the ability to sanitize data to be shared across administrative domains. The architecture was designed to allow collection of data at a per site basis and to allow each site to create data sharing based on its own particular trust relationships.
The CyberFed system was implemented and deployed by Argonne National Laboratory to automate the detection and response of attack activity against Department of Energy (DoE) computer networks. CyberFed automates the collection of network alerting activity from various perimeter network defenses and logs those events into its database. CyberFed then automatically converts that information into blocking information transmitted to all participants. The original implementation used IODEF messages wrapped in an XML extension to manage a large array of indicators. The CyberFed system was not designed to describe a particular incident as much as to describe a set of current network blocking indicators that can be generated and deployed machine-to-machine.
CyberFed is primarily implemented in Perl. Included as part of the CyberFed system are scripts which interact with a large number of firewalls, IDS/IPS devices, DNS systems, and proxies which operate to implement both the automated collection of events as well as the automated deployment of black listing.
Currently CyberFed supports multiple exchange formats including IODEF and STIX. OpenIOC is also a potential exchange format that US DoE is considering.
The section aims at sharing the tips for development of IODEF-capable systems.
For implementing IODEF-capable systems, it is feasible to employ code generators for XML Schema Document (XSD). The generators are used to save development costs since they automatically create useful libraries for accessing XML attributes, composing messages, and/or validating XML objects. The IODEF XSD was defined in section 8 of RFC5070, and is availabe at http://www.iana.org/assignments/xml-registry/schema/iodef-1.0.xsd.
However, there still remains some issues. Due to the complexity of IODEF XSD, some code generators could not generate code from the XSD file. The tested code generators were as follows.
For instance, we have tried to use XML::Pastor, but it could not properly understand its schema due to the complexity of IODEF XSD. The same applies to RXSD and JAXB. Only PyXB, CodeSynthesis XSD and Xsd.exe were able to understand the complex schema.
Unfortunately, there is no recommended workaround. A possible workaround is a double conversion of XSD file. This entails the XSD being serialized into XML, and afterwards the resulting XML is converted back into an XSD. The resultant XSD was successfully processed by the all tools above.
It should be noted that IODEF uses '-' (hyphen) symbols in its classes or attributes, listed as follows.
According to the language specification, many programing language prohibit having '-' symbols in the name of class. The code generators must replace or remove the '-' when building the librarlies. They should have the name space restore the '-' when outputting the XML along with IODEF XSD.
iodeflib is an open source implementation written in Python. This provides simple but powerful APIs to create, parse and edit IODEF documents. It was designed in order to keep its interface as simple as possible, whereas generated libraries tend to inherit the complexity of IODEF XSD. In addition, the iodeflib interface includes functions to hide some unnecessarily nested structures of the IODEF schema, and adding more convenient shortcuts.
This tool is available through the following link:
IODEF.pm is an open source implementation written in Perl. This also provides a simple interface for creating and parsing IODEF documents, in order to facilitate the translation of the a key-value based format to the IODEF representation. The module contains a generic XML DTD parser and includes a simplified node based representation of the IODEF DTD. It can hence easily be upgraded or extended to support new XML nodes or other DTDs.
This tool is available through the following link:
Here notes some tips to avoid problems.
The MILE Implementation report has been compiled through the submissions of implementers of INCH and MILE working group standards. A special note of thanks to the following contributors:
This memo includes no request to IANA.
This draft provides a summary of implementation reports from researchers and vendors who have implemented RFCs and drafts from the MILE and INCH working groups. There are no security considerations added in this draft because of the nature of the document.