Network Working Group T. Johannsen Request for Comments: 1608 Dresden University Category: Experimental G. Mansfield AIC Systems Laboratory M. Kosters Network Solutions,Inc. S. Sataluri AT&T Bell Laboratories March 1994
Representing IP Information in the X.500 Directory
Status of this Memo
This memo defines an Experimental Protocol for the Internet community. This memo does not specify an Internet standard of any kind. Discussion and suggestions for improvement are requested. Distribution of this memo is unlimited.
Abstract
This document describes the objects necessary to include information about IP networks and IP numbers in the X.500 Directory. It extends the work "Charting networks in the X.500 Directory" [1] where a general framework is presented for representing networks in the Directory by applying it to IP networks. This application of the Directory is intended to support the work of IP network assigning authorities, NICs, as well as other applications looking for a mapping of IP numbers to data of related networks. Furthermore, Autonomous Systems and related routing policy information can be represented in the Directory along with their relationship to networks and organizations.
Johannsen, Mansfield, Kosters & Sataluri [Page 1]
RFC 1608 IP Information in the X.500 Directory March 1994
Table of Contents
1. Introduction 2 2. IP images of networks 3 2.1 IP network image 3 2.2 IP node image 5 2.3 IP network interface image 6 2.4 Autonomous Systems 7 3. Number assignment information 7 3.1 Delegated Block object 8 3.2 IP Group object 9 3.3 IP Reference object 10 3.4 AS Block object 10 3.5 AS Reference object 10 4. Directory tree 11 4.1 IP image objects 11 4.2 AS objects 11 4.3 Namespace objects 11 4.4 Relationship to organizational entries 13 5. Security Considerations 14 6. Authors' Addresses 15 References 16 Appendix: OID tables 17
Information related to the Internet Network Infrastructure is created and stored by a number of different organizations, such as the Internet Assigned Numbers Authority (IANA), Internet Registry (IR), Network Information Centers (NICs), and the NSFNET Network Operations Center (NOC). This information is generally "mastered" (stored and maintained) by these organizations on a centralized basis, i.e., there is a single place to look for a definitive list of entries for these categories. This has worked well in the past but given the tremendous growth of the Internet and its number of users and networks, it is essential that a distributed schema be used.
The X.500 Directory offers the appropriate technology for implementing this distributed method of managing network infrastructure information.
The following goals are addressed in this document:
o Provision of IP specific images of network elements o Mapping from Network Number to network, owner, provider etc. o Support of delegation of IP address blocks o Storage of high-level routing policies and AS information o Support of "classless" network address formats
Johannsen, Mansfield, Kosters & Sataluri [Page 2]
RFC 1608 IP Information in the X.500 Directory March 1994
o Provision of several organizational images of a network
It may be noted that the document basically consists of two parts. In the first part, an IP specific extension of the general framework "Charting networks in the Directory" [1] is given. Objects defined by the application of this framework are related to IP numbers. An IP namespace is defined in the second part of this paper, referring to IP network elements defined at the beginning.
As defined in [1], networks are modeled as a set of subnetworks and nodes, called network elements in the remainder. To obtain a particular view of a network element, images were introduced. Below, images of network elements for an IP specific view are defined. Please note that images contain references to underlying physical information about elements.
In the remainder, ipStringSyntax is used as attribute type for all attributes that are to hold an IP number. Currently, there are two definitions for a syntax like this:
o IpAddress as of [5] o ip as of [6]
It is suggested to use CaseIgnoreStringSyntax for implementations for the time being with the convention to use the ordinary IP syntax. Nevertheless, an OID has been reserved for ipStringSyntax (see Appendix).
IP network image is one application of network images and therefore inherits the networkImageClass. This class is given below for reference only, its definition can be found in [1].
networkImage OBJECT CLASS SUBCLASS of ImageCommunicationObject MAY CONTAIN externalGateway :: distinguishedNameSyntax, /* points to one or more nodes that act as gateway for the protocol application this image refers to */
An IP network combines all network elements that have a common IP network number. Presently, IP network numbers fall into one of the classes A, B, or C. However, sub- and supernetworking is already done broadly, and classless networks numbers are expected to be assigned
Johannsen, Mansfield, Kosters & Sataluri [Page 3]
RFC 1608 IP Information in the X.500 Directory March 1994
soon. [2] proposes to assign bitwise contiguous blocks of class-C- addresses to all networks with more than 255 nodes. The definition of IP network, therefore, is always related to common network number and network mask.
IP networks have a very close relationship to the Domain Name System. Several attributes are introduced to take care of these relations. Though we do not intend to abolish the existing DNS service, the schema defined here is able to provide the mapping IP number to domain name and vice versa.
An IP network image object as defined below is intended to provide technical and administrative data for one IP network. Attributes hold information that a NIC-WHOIS server would reveal for the network, and more.
ipNetworkImage OBJECT CLASS SUBCLASS of NetworkImage MUST CONTAIN ipNetworkImageName :: CaseIgnoreString, /* common name */ ipNwNumber :: IPStringSyntax, /* the IP network number for this (sub)network */ ipNwMask :: IPStringSyntax /* mask that applies to ipNwNumber in order to define classless network number; also used for routing */
MAY CONTAIN /* DNS related info ; e.g. - */ associatedDomain :: CaseIgnoreStringSyntax, /* the domain name associated to this IP network; As there is not always a 1:1 mapping between traditional IP numbers and domain names, the name given here should contain the "closest match". 1) (sub)network does not have a domain name of its own, but is part of a bigger domain: take name of that domain 2) network is divided into several domains, therefore having more than one domain name: list all of them. Note: a reverse mapping of domain names to networks/nodes can be achieved by a modified implementation of RFC1279 */ inAddrServer :: DistinguishedNameSyntax, /* refers to the ipNodeImageObject of the inaddr Server that holds information about
Johannsen, Mansfield, Kosters & Sataluri [Page 4]
RFC 1608 IP Information in the X.500 Directory March 1994
this network */ /* routing policy; e.g. - */ asNumber :: integerStringSyntax, /* number of Autonomous System this network belongs to */ acceptedUsagePolicy :: caseIgnoreStringSyntax, /* semantics to be defined */ /* Any other - */ provider :: DistinguishedNameSyntax, /* points to network provider */ onlineDate :: uTCTimeSyntax /* date when network got connected to the Internet */
If a node in the network is running the IP protocol, an ipNodeImageObject should be created for this node. This image is a subclass of nodeImageClass and holds IP specific information. The nodeImageClass is given below for reference only, its definition can be found in [1].
nodeImage OBJECT CLASS SUBCLASS of ImageCommunicationObject /* no attributes common for all nodeImages have been defined yet */
An ipNodeImage is used to obtain technical and administrative information on a host. The object is defined as follows:
ipNodeImage OBJECT CLASS SUBCLASS of NodeImage MUST CONTAIN ipNodeName :: CaseIgnoreString /* common name, it is advised to use the hostname for this purpose */ MAY CONTAIN protocol :: CaseIgnoreString, /* name and version of IP protocol running */ domainName :: CaseIgnoreString, /* the complete domain name of this node; CNAMEs can be stored additionally to the DNS A record name; further relationships, like MX record entries, should be taken care of by the domain name tree according to RFC 1279 */
Johannsen, Mansfield, Kosters & Sataluri [Page 5]
RFC 1608 IP Information in the X.500 Directory March 1994
The most important IP related information of a node (its IP addresses) is registered with ipNetworkInterfaceImageObjects. This picture is accurate as a node can have several IP addresses, but at most one per interface. Furthermore, it shows clearly the relationship between interface and neighboring IP network.
IpNetworkInterfaceImage is a subclass of networkInterfaceImageClass. The networkInterfaceImageClass is given below for reference only, its definition can be found in [1]. Note that for ipNetworkInterfaceImage all references are drawn in the context of IP, i.e., networkInterfaceAddress becomes an IP address and connectedNetwork points to an ipNetworkImageObject.
networkInterfaceImage OBJECT CLASS SUBCLASS of ImageCommunicationObject MAY CONTAIN networkInterfaceAddress :: caseIgnoreStringSyntax, /* this interface's address in the context the image refers to, e.g. IP number, NSAP */ connectedNetwork :: distinguishedNameSyntax /* pointer to networkImageObject that describes the network this interface is attached to in terms of the protocol or application the images indicates */
Additionally, ipNetworkInterfaceImageObject has the following properties:
ipNetworkInterfaceImage OBJECT CLASS SUBCLASS of NetworkInterfaceImage MUST CONTAIN ipNetworkInterfaceName :: CaseIgnoreStringSyntax /* It is suggested that the interface name is derived from the name of the logical device this interface represents for the operating system, e.g. le0, COM1 */ MAY CONTAIN ipNwMask :: IPStringSyntax /* mask that applies to networkInterfaceAddress for routing of packets to nodes on the connected (broadcast) network; Note: This is only a fraction of the routing table information for this interface, namely for one hop. */
Johannsen, Mansfield, Kosters & Sataluri [Page 6]
RFC 1608 IP Information in the X.500 Directory March 1994
Autonomous Systems (AS) are defined in asObject which is a subclass of imageCommunicationObject. It provides technical and administrative information of an AS. Furthermore, routing policies can be stored with the AS object. The definition of the AS object is aligned with the corresponding database object defined in [3].
as OBJECT CLASS SUBCLASS of ImageCommunicationObject MUST CONTAIN asNumber :: IntegerStringSyntax
MAY CONTAIN asName :: CaseIgnoreStringSyntax, /* if there is a name different from AS */ asIn :: CaseIgnoreStringSyntax, /* accepted routes from other AS */ /* for syntax and semantics refer [3] */ asOut :: CaseIgnoreStringSyntax, /* generated routes announced to others */ /* for syntax and semantics refer [3] */ asDefault ::CaseIgnoreStringSyntax, /* how default routing is handled */ /* for syntax and semantics refer [3] */ asGuardian :: DistinguishedNameSyntax, */ /* DN of guardian of this AS */ lastModifiedDate :: UTCtimeSyntax */ /* important as routes change frequently */
Note that routing policies for an Autonomous System are represented by the {asIn, asOut} attributes of asObject. Due to performance constraints of present X.500 technology, it will probably not be used directly by routers for dynamic routing. However, it certainly can be used in network management systems to determine the allowed paths [i.e., in accordance with published policies] between two networks. This will be useful in finding alternate paths, and evaluating the connectivity of networks.
In the following, Directory objects have been defined to represent IP and AS (Autonomous System) namespace in the Directory. Their purpose is to provide
o mapping from IP number to IP network element (network or node) o mapping from IP number to AS number and vice versa o assignment and delegation information
Johannsen, Mansfield, Kosters & Sataluri [Page 7]
RFC 1608 IP Information in the X.500 Directory March 1994
The need for a global, distributed database supporting the objectives arises mainly from distributed IP- and AS-number assignment.
Describing all IP numbers with one of the new objects delegatedBlock, ipGroup and ipReference leads to the desired information. AS number information is stored with the objects asBlock and asReference. Furthermore, all assigned numbers have some properties in common. Therefore, an objectclass assignedNumberClass is introduced. This class exports attributes to delegatedBlock, ipGroup, ipReference, asBlock, and asReference.
AssignedNumberClass is defined as follows ("number" always refers to IP number of delegatedBlock, network, host, and AS number, resp.):
assignedNumberClass OBJECT CLASS SUBCLASS of top MAY CONTAIN assBy :: DistinguishedNameSyntax, /* refers to an organization or organizationalRole that assigned the number to assTo (see below) */ assTo :: DistinguishedNameSyntax, /* refers to organization or organizationalRole that the number was assigned to. This does not imply that assTo "owns" this number now. */ assDate :: uTCTimeSyntax, /* date of assignment for this number */ nicHandle :: CaseIgnoreStringSyntax, /* gives the unique ID for a description related to this number. format: "handle : nic-domain-name" example: MAK21 : rs.internic.net */ relNwElement :: DistinguishedNameSyntax, /* the network element related to this number (network or node) */
This object provides information on a block of IP addresses delegated to some local-authority or service provider. Only contiguous blocks can be represented with the following schema. If an organization (say, a NIC) has been assigned several IP network numbers which do not form a contiguous block, it might want to use a different form of representing that fact (e.g., using imageNetworks). The delegatedBlock object holds lower and upper bounds of the block.
Note that the above does not make any assumption about the network masks being constrained by byte boundaries. We can thus represent subnetting within a "network (number)" that often happens within an
Johannsen, Mansfield, Kosters & Sataluri [Page 8]
RFC 1608 IP Information in the X.500 Directory March 1994
organization in the same framework.
This schema does lead to some granularity in the otherwise flat IP- number space. Further, the granularity is significant as it may be used to identify the administrator of the block - a service provider or a domain manager. E.g., it fits well into the schema of aggregating networks for routing purposes as has been proposed in [4].
The object delegatedBlock is of the form:
delegatedBlock OBJECT CLASS SUBCLASS of AssignedNumberClass MUST CONTAIN delegatedBlockName :: caseIgnoreStringSyntax, lowerBound :: IPStringSyntax, /* smallest IP address belonging to the block, e.g. 195.100.0.0 */ upperBound :: IPStringSyntax /* highest IP address belonging to the block, e.g. 195.103.255.255 */
The attribute relNwElement (inherited from AssignedNumberClass) can point to a networkImage covering all networks within the block if this makes sense.
This object provides information for an IP network number. Its purpose is basically only to
o show that the number has been assigned, and o provide a reference to the descriptive ipNetworkObject for this network.
Regardless of the actual value of x, IP group objects may exist for IP numbers x.0.0.0, x.y.0.0 and x.y.z.0. This approach includes "classical" class-A, -B and -C network addresses as well as any kind of super- and subnetworking.
The IP group object is a subclass of assignedNumberClass. The attribute relNwElement points to an ipNetworkImage as defined above.
ipGroup OBJECT CLASS SUBCLASS of AssignedNumberClass MUST CONTAIN ipGroupName :: IPStringSyntax, /* IP number; x.0.0.0 or x.y.0.0 or x.y.z.0
Johannsen, Mansfield, Kosters & Sataluri [Page 9]
RFC 1608 IP Information in the X.500 Directory March 1994
where x, y, z in 1..255 */ ipNwMask :: IPStringSyntax /* mask that applies to all numbers within the group; used to define classless networking; */
An AS reference object is used to show that an Autonomous System number has been assigned (and thus can not be given to somebody else). Similar to ipGroup, asReference does not contain technical details about an autonomous system itself but rather points (with relNwElement) to a descriptive asObject.
According to [1], IP image entries will be stored underneath the organization / organizationalUnit entry of the entity responsible for that network. The case that such an entry does not yet exist in the white-pages pilot is discussed in 4.4 below.
The technical and administrative description of an AS is basically maintained by NICs, network providers, or other special organizations. It is suggested that these organizations build a subtree for information on AS which they are responsible for.
The new IP namespace objects build a single tree in the Directory. It is suggested that this tree will have a root of type organizationalUnit within @o=Internet@ou=Network Information.
objectClass= organizationalUnit organizationalUnitName= IP networks description= root of IP number tree
RFC 1608 IP Information in the X.500 Directory March 1994
The tree is built under an administrative and an implementational view. Nowadays, network numbers usually are assigned to organizations by (national) Network Information Centers (NIC) which themselves have got a block of IP network numbers assigned from another authority (e.g., IR at top level). This concept of delegated blocks falling apart in smaller delegated blocks and IP network numbers is used to model the Directory tree. Thus, an ipGroup object is always subordinate of a delegated block object (namely the delegated block including this IP number). Network numbers that were directly assigned by a top-level authority, i.e., have not been object of a delegation to a local assigning authority, will all be at one level in the Directory. Already today, however, we find many delegations within the traditional class A-, B- and C-addresses. Such a delegation is represented by a delegated block object, having the assigned IP network numbers as subordinates. Also, part of the block can be further delegated to another authority, leading to another delegated block object within the parent delegated block's tree. Usually, subordinates of ipGroup objects are ipReferences, i.e., single IP addresses as assigned to nodes. To support subnetworking, it is also allowed to divide ipGroups into several subnetwork ipGroups, each representing an IP subnetwork. In such cases, subnetwork numbers are given as subordinates to the assigned IP network number. Network masks clarify what the subnetwork addresses are.
RFC 1608 IP Information in the X.500 Directory March 1994
disadvantages in the look-up procedure. In that case one might think of combining both object classes with the aim to provide one object describing administrative and technical data for an IP network.
As Autonomous Systems are an additional namespace to the existing IP number space, they should go into a separate subtree. It is suggested that this is an organizationalUnit within @o=Internet@ou=Network Information.
objectClass= organizationalUnit organizationalUnitName= AS numbers description= root of Autonomous System number space
Similar to blocks of IP network numbers, blocks of AS numbers are sometimes delegated to another registry. This is expressed by asBlock objects. These objects come below the root of the AS number space. All AS numbers falling into such a block are stored as subordinates of the block. An AS block may have smaller AS blocks underneath if delegation is extended.
Organizational information (i.e., white-pages-like information about an organization, its departments and employees) occurs at several places in the network DIT - [org of IP-Number, org of AS-Number, org of Admin- contact, However, it will be basically mastered [administered, maintained] by the organization itself in the Directory Management Domain (DMD) over which the organization has the authority. This gives rise to some tricky problems - a typical example is that of a NIC which holds the AS, DNS, IP, ... subtrees of the DIT.
A good strategy would avoid explicit duplication of information. By explicit duplication of information we understand information being duplicated outside the directory framework, e.g., by having several master entries for one and the same piece of information. The only way to avoid duplication would be to have relevant entries point to the pertinent organizational entry for organizational information. But since
o most organizations do not, as yet, have an entry in the DIT and o the reliability of the access to an organizations DIT when stored in a remote DSA cannot be taken for granted,
the following framework is adopted to accommodate the conflicting requirements /conditions.
RFC 1608 IP Information in the X.500 Directory March 1994
o A copy of all the necessary organization-info is retained at the NICs DSA. Since only the necessary info will be kept the NIC will not be burdened to act as the repository of the organizations DIT. These objects may be kept in a separate subtree of affiliated-organizations [organizations affiliated to the NIC]. Though the affiliated organizations node does not really represent a locality, it is suggested to define the node as objectClass locality. This does not break the Directory schema when entries of organizations shall become subordinate to the NICs organization's entry.
o The problem of information duplication/consistency will arise when organizational DITs/DSAs do come into existence. At that stage a shadowing mechanism which will attempt to maintain the data consistency may be resorted to. The X.500/ISO 9594(1993) implementations are expected to provide appropriate shadowing mechanisms along X.525.
It may be noted that what is suggested is not a duplication of an entire white-pages-like structure at the NIC. It suggests an "affiliated organizations node". The entries under this node will be organization objects with a limited number of attributes, i.e., the attributes to hold the organization info necessary for the NIC: nothing more, nothing less. Operationally, and content wise the NIC DSA will hold exactly the amount of info that is desired by the NIC. When an organization sets up its DSA and when the organization informs the NIC about it, the NIC will set up the shadowing arrangement to obtain info on changes of interest [and forget about it].
It may be emphasized that the entries under affiliated organizations are physical entities [replicated and refined from the Master entries, if and when they exist...] rather than alias entries. If a NIC dislikes the idea of users poring over the entries in the affiliated organizations - appropriate access control can be applied. Though duplication is unavoidable, the proposal attempts to make it transparent, by delegating the responsibility of maintaining the integrity to the Directory.
This issue is discussed in greater detail in a separate document [7].
RFC 1608 IP Information in the X.500 Directory March 1994
References
[1] Mansfield, G., Johannsen, T., and M. Knopper, "Charting Networks in the X.500 Directory", RFC 1609, AIC Systems Laboratory, Dresden University, Merit Networks,Inc., March 1994.
[2] Gerich, E., "Guidelines for Management of IP Address Space", RFC 1466, Merit, May 1993.
[3] Bates, T., Jouanigot, J.-M., Karrenberg, D., Lothberg, P., and M. Terpstra, "Representation of IP Routing Policies in the RIPE Database", Document ripe-81, RIPE, February 1993.
[4] Fuller, V., Li, T., Yu, J., and K. Varadhan, "Supernetting: An Address Assignment and Aggregation Strategy", RFC 1338, BARRNet, cisco, MERIT, OARnet, June 1992.
[5] Rose, M., and K. McCloghrie, "Structure and Identification of Management Information for TCP/IP-based internets", STD 16, RFC 1155, Performance Systems International, Hughes LAN Systems, May 1990.
[6] Barker, P., and S. Kille, "The COSINE and Internet X.500 Schema", RFC 1274, University College London, November 1991.
[7] Mansfield, G., Johannsen, T., and J. Murai, J., "Deployment Strategy for the Directory in the Internet", AIC Systems Laboratory, Dresden University, Keio University, Work in Progress, July 1993.