Internet-Draft Grenville Armitage Bellcore February 4th, 1995 Support for Multicast over UNI 3.1 based ATM Networks. Status of this Memo This document was submitted to the IETF IP over ATM WG. Publication of this document does not imply acceptance by the IP over ATM WG of any ideas expressed within. Comments should be submitted to the ip- atm@matmos.hpl.hp.com mailing list. Distribution of this memo is unlimited. This memo is an internet draft. Internet Drafts are working documents of the Internet Engineering Task Force (IETF), its Areas, and its Working Groups. Note that other groups may also distribute working documents as Internet Drafts. Internet Drafts are draft documents valid for a maximum of six months. Internet Drafts may be updated, replaced, or obsoleted by other documents at any time. It is not appropriate to use Internet Drafts as reference material or to cite them other than as a "working draft" or "work in progress". Please check the lid-abstracts.txt listing contained in the internet-drafts shadow directories on nic.ddn.mil, nnsc.nsf.net, nic.nordu.net, ftp.nisc.src.com, or munnari.oz.au to learn the current status of any Internet Draft. Abstract This memo describes a Multicast Address Resolution Server (MARS) architecture that allows ATM based IP hosts to support RFC 1112 style Level 2 IP multicast using the ATM Forum's UNI 3.1 point to multipoint connection service. It also describes how this architecture can be generalized to support other protocols wishing to multicast over UNI 3.1 based ATM service. [Editorial note: The differences between this version and 03.txt are substantial in the area of multicast server support. This impacts on Chapter 8, and anything referring to MARS_MSERV. Two control VCs have been identified and named, two sequence numbers are now used, and three major appendices have been added discussing issues that cannot at this time be standardized. The MARS_JOIN/LEAVE message format has been extended by 32 bits, and Armitage Expires August 4th, 1995 [Page 1] Internet Draft February 4th, 1995 modified to support multiple address groups. Scattered editorial/clarificatory changes have been made to the rest of the document. Editorial notes will be removed.] 1. Introduction. Multicast support allows a source host or protocol entity to send a packet to multiple destinations simultaneously using a single, local 'transmit' operation. This facility is utilized by network layer protocols such IP. Most models, like the one described in RFC 1112 [1] for IP multicasting, assume sources may send their packets to an abstract 'multicast group addresses'. Link layer support for such an abstraction is assumed to exist, and is provided by technologies such as Ethernet. ATM is being utilized as a new link layer technology to support a variety of protocols, including IP. With RFC 1483 [2] the IETF defined a multiprotocol mechanism for encapsulating and transmitting packets using AAL5 over ATM Virtual Channels (VCs). However, the ATM Forum's currently published signalling specification (UNI 3.0 [4], with additions for UNI 3.1 released in late 1994) does not provide the multicast address abstraction. Unicast connections are supported by point to point, bidirectional VCs. Multicasting is supported through point to multipoint VCs. The key limitation is that the sender must have prior knowledge of each intended recipient, and explicitly establish a VC with itself as the root node and the recipients as the leaf nodes. The main goal of this document is to define an address registration and distribution mechanism that allows UNI 3.1 based networks to support the multicast service of protocols such as IP. The second goal is to define specific endpoint behaviour and management of point to multipoint VCs. As the IETF is currently in the forefront of using wide area multicasting this document's descriptions will focus on IP version 4 (IPv4). A final chapter will note the more general application of the architecture. The Multicast Address Resolution Server (MARS), a distant relative of the ATM ARP Server introduced in RFC 1577 [3], acts as a registry of multicast group membership. MARS messages, based on the ATM ARP format, support the distribution of multicast group membership information between MARS and hosts or endpoints. Endpoint address resolution entities query the MARS when a multicast group address needs to be resolved. The actual mechanism for multicasting data packets may be through meshes of point to multipoint VCs, or the use of Multicast Servers. To provide for asynchronous notification of group membership changes the MARS manages two point to multipoint VCs Armitage Expires August 4th, 1995 [Page 2] Internet Draft February 4th, 1995 - one out to all endpoints desiring multicast support, and the other to all multicast servers registered as providing support to any multicast groups. The choice of mesh or multicast server is configurable on a group by group basis. The numerical size of link layer multicast groups will be constrained by practical concerns such as limited VC support within endpoint ATM interfaces. Each MARS manages a 'cluster' of ATM-attached endpoints. A cluster is defined as a set of endpoints willing to be grouped together as link layer members of multicast groups. It is assumed that specially configured routers are used to pass multicast traffic between clusters. This document explicitly avoids specifying the nature of inter-cluster multicast routing protocols. The mapping of clusters to other constrained sets of endpoints (such as Logical IP Subnets) is left to network administrators. A simple approach in overlaid IP environments would be for each LIS to be served by a separate MARS, with the cluster being built from the LIS members. IP multicast routers would interconnect each LIS as they do with conventional subnets. However, there is no requirement that a cluster be limited to a single LIS. Section 2 provides an overview of IP multicast and what RFC 1112 required from Ethernet. Section 3 outlines the set of generic functions that should be available to clients of a local host's UNI 3.1 signalling service. Section 4 specifies the encapsulation to be used for MARS messages and multicast packet traffic. The basic behaviour for the sending side of an interface is described in section 5, with section 6 covering the mechanism whereby a host joins and leaves multicast groups. Sections 7 covers the way in which hosts respond to dynamic group membership changes. Configuring the use of Multicast Servers is covered in section 8. Support for multicast routers is described in section 9, and section 10 explains the features included to improve the reliability of the membership management mechanisms. Section 11 discusses the application of this document beyond IP. Section 12 is a summary of the documents key points. The appendices provide discussion on issues that arise out the implementation of this memo. Appendix A discusses MARS and endpoint algorithms for parsing MARS messages. Appendix B describes the particular problems introduced by the current IGMP paradigms, and possible interim work-arounds. Finally, Appendix C covers the various designs that are possible for multicast server support within clusters. This document assumes an understanding of concepts explained in greater detail in RFC 1112, RFC 1577, UNI 3.1, and . 2. Review of RFC 1112 and IP Multicast over Ethernet. Under IP version 4 (IPv4) ddresses in the range of 224.0.0.0 and 239.255.255.255 are termed 'Class D' or 'multicast group' addresses. In RFC 1112 the behaviour of the transmit and receive sides are quite independent, making the concept of being a 'member' of an IP multicast group imprecise at the link layer interface. The interface must support the transmission of IP packets to an IP multicast group address, whether or not the node considers itself a 'member' of that group. Consequently, group membership is effectively irrelevant to the transmit side of the link layer interfaces. No address resolution is required to transmit packets - an algorithmic mapping from IP multicast address to Ethernet multicast address is performed locally before the packet is sent out the local interface in the same 'send and forget' manner as a unicast IP packet. Joining and Leaving an IP multicast group is more explicit on the receive side - with the primitives JoinLocalGroup and LeaveLocalGroup affecting what groups the local link layer interface should accept packets from. When the IP layer wants to receive packets from a group, it issues JoinLocalGroup. When it no longer wants to receive packets, it issues LeaveLocalGroup. A key point to note is that changing state is a local issue, it has no affect on other hosts attached to the Ethernet. IGMP is defined in RFC 1112 to support IP multicast routers attached to a given subnet. Hosts issue IGMP Report messages when they perform a JoinLocalGroup, or in response to an IP multicast router sending an IGMP Query. By periodically transmitting queries IP multicast routers are able to identify what IP multicast groups have non-zero membership on a given subnet. A specific IP multicast address, 224.0.0.1, is allocated for the transmission of IGMP Query messages. All IP multicast hosts must issue JoinLocalGroup for 224.0.0.1 during their initialisation. Each host keeps a list of IP multicast groups it has been JoinLocalGroup'd to. When a router issues an IGMP Query on 224.0.0.1 each host begins to send IGMP Reports for each group it is a member of. IGMP Reports are sent to the group address, not 224.0.0.1, "so that other members of the same group on the same network can overhear the Report" and not bother sending one of their own. IP multicast routers conclude that a group has no members on the subnet when IGMP Queries no longer elict associated replies. Armitage Expires August 4th, 1995 [Page 4] Internet Draft February 4th, 1995 3. Multicast support under UNI 3.1. This document will describe its operation in terms of 'generic' functions that should be available to clients of a UNI 3.1 signalling entity in a given ATM endpoint. The ATM model broadly describes 'AAL Users' as any entity that establishes and manages VCs and underlying AAL service to exchange data. An IP over ATM interface is a form of 'AAL User' (either directly, when VC multiplexing is used, or indirectly, when LLC/SNAP encpasulation is used). The most fundamental limitations of UNI 3.1's multicast support are: Only point to multipoint, unidirectional VCs may be established. Only the root node of a given VC may add or remove leaf nodes. Within these constraints, multicast group members can communicate by the use of multicast meshes, or multicast servers. With a mesh each transmitting host is the Root of a point to multipoint VC that has every other host in the group as a Leaf. The Multicast Server model has every group member send their packets directly to a 'server' entity somewhere on the ATM cloud, which then retransmits copies to all other members. This document defines the MARS-Endpoint signalling required to support both mechanisms. Issues relating to the architecture, operation, and management of multicast servers are discussed in Appendix C. The following generic signalling functions are presumed to be available to local AAL Users: L_CALL_RQ - Establish a unicast VC to a specific endpoint. L_MULTI_RQ - Establish multicast VC to a specific endpoint. L_MULTI_ADD - Add new leaf node to previously established VC. L_MULTI_DROP - Remove specific leaf node from established VC. L_RELEASE - Release unicast VC, or all Leaves of a multicast VC. The signalling exchanges and local information passed between AAL User and UNI 3.1 signalling entity with these functions is currently beyond the scope of this document. The following indications are assumed to be available to AAL Users, generated by by the local UNI 3.1 signalling entity: L_ACK - Succesful completion of a request to signalling entity. L_REMOTE_CALL - A new VC has been established to the AAL User. Armitage Expires August 4th, 1995 [Page 5] Internet Draft February 4th, 1995 ERR_L_RQFAILED - A remote ATM endpoint rejected an L_CALL_RQ, L_MULTI_RQ, or L_MULTI_ADD. ERR_L_RELEASE - A remote ATM endpoint has elected to terminate a pre-existing VC. The signalling exchanges and local information passed between AAL User and UNI 3.1 signalling entity with these functions is currently beyond the scope of this document. UNI 3.1 defines two ATM address formats - E.164 and ISO NSAP. In UNI 3.1 an 'ATM Number' is the primary identification of an ATM endpoint, and it may use either format. Under some circumstances an ATM endpoint must be identified by both an E.164 address (identifying the attachment point of a private network to a public network), and an ISO NSAP address ('ATM Subaddress') identifying the final endpoint within the private network. For the rest of this document the term 'ATM Address' will be used to mean either a single 'ATM Number' or an 'ATM Number' combined with an 'ATM Subaddress'. 4. Overview of the Multicast Address Resolution Server. The MARS may reside within any ATM endpoint that is directly addressable by the endpoints it is serving. Endpoints wishing to join a multicast cluster must be configured with the ATM address of the node on which the cluster's MARS resides. This is the cluster's Primary MARS. If a cluster is to be served by a backup MARS, endpoints are configured with the ATM address of a Secondary MARS. Section 10 will discuss the relationship between the Primary MARS and Secondary MARS during failure conditions. Although a Secondary MARS is optional, endpoint implementations must be capable of utilizing them as described in section 10. References to 'the MARS' in following sections will be assumed to mean the acting MARS for the cluster. Architecturally the MARS is similar to the RFC 1577 ARP Server, although there is little overlap between the information they manage. Whilst the ARP Server keeps a table of {IP,ATM} address pairs for all IP endpoints in the LIS, the MARS keeps extended tables of {multicast address, ATM.1, ATM.2, ..... ATM.n} mappings. It can either be configured with certain mappings, or dynamically 'learn' mappings. The MARS distributes group membership information to cluster members over a point to multipoint VC known as the ClusterControlVC. When supporting multicast servers within a cluster, the MARS also establishes a separate point to multipoint VC known as the ServerControlVC. All cluster members are leaf nodes of ClusterControlVC. All registered multicast servers are leaf nodes of ServerControlVC (Section 8 will discuss the use of ServerControlVC). Armitage Expires August 4th, 1995 [Page 6] Internet Draft February 4th, 1995 The MARS message format is an extension of the ATM ARP message format. By default all MARS messages MUST be LLC/SNAP encapsulated in accordance with RFC 1483, using the same encapsulation as ATM ARP: LLC = 0xAA-AA-03 OUI = 0x00-00-00 PID = 0x08-06 The default for data traffic carried on point to multipoint VCs is LLC/SNAP encapsulation with a header appropriate to the protocol being carried. For IP traffic this is defined in RFC 1483 as: LLC = 0xAA-AA-03 OUI = 0x00-00-00 PID = 0x08-00 The choice of common encapsulation and message format means that MARS and ARP Server functionality may be implemented within a common entity if a network designer so chooses. 5. Transmitting to Multicast groups. [Editorial note: This section has discarded the MARS_MSERV function of version ipmc-03.txt. MARS_MSERV is now used in an entirely different fashion. Endpoint VC management is now entirely independent of whether the group is mesh or mc server supported.] The following description will be in terms of an IP/ATM interface that is capable of transmitting packets to a Class D address at any time, without prior warning. When a packet arrives for transmission, and there is no outgoing VC already marked as serving the packet's multicast destination address, the MARS is queried for the set of ATM endpoints currently making up the multicast group. The query is executed by issuing a MARS_REQUEST. The MARS_REQUEST message is formatted as an ATM ARP_REQUEST with type code of 11 (decimal). The reply from the MARS may take one of two forms: MARS_MULTI - Sequence of MARS_MULTI messages return the set of endpoints in the group. MARS_NAK - No mapping found, group is empty. The request/response traffic MUST occur on a point to point VC established by the host to the MARS. Where the MARS and ARP Server Armitage Expires August 4th, 1995 [Page 7] Internet Draft February 4th, 1995 are co-resident, this VC may be shared between ATM ARP traffic and MARS traffic. 5.1 Retrieving Group Membership from the MARS. If the MARS had no mapping for the desired Class D address a MARS_NAK will be returned. In this case the IP packet MUST be discarded silently. If a match is found in the MARS's tables it proceeds to return addresses ATM.1 through ATM.n in a sequence of one or more MARS_MULTIs. A simple mechanism is used to detect and recover from loss of MARS_MULTI messages. Each MARS_MULTI carries a new boolean field x, and a 15 bit integer field y - expressed as MARS_MULTI(x,y). Field y acts as a sequence number, starting at 1 and incrementing for each MARS_MULTI sent. Field x acts as an 'end of reply' marker. When x == 1 the MARS response is considered complete. In addition, each MARS_MULTI may carry multiple ATM addresses from the set {ATM.1, ATM.2, .... ATM.n}. A MARS MUST minimise the number of MARS_MULTIs transmitted by placing as many group member's addresses in a single MARS_MULTI as possible. The limit on MARS_MULTI message length MUST be the MTU of the underlying VC. Assume n ATM addresses must be returned, each MARS_MULTI is limited to only p ATM addresses, and p << n. This would require a sequence of k MARS_MULTI messages (where k = (n/p)+1, using integer arithmetic), transmitted as follows: MARS_MULTI(0,1) carries back {ATM.1 ... ATM.p} MARS_MULTI(0,2) carries back {ATM.(p+1) ... ATM.(2p)} [.......] MARS_MULTI(1,k) carries back { ... ATM.n} If k == 1 then only MARS_MULTI(1,1) is sent. Typical failure mode will be losing one or more of MARS_MULTI(0,1) through MARS_MULTI(0,k-1). This is detected when y jumps by more than one between consecutive MARS_MULTI's. An alternative failure mode is losing MARS_MULTI(1,k). A timer MUST be implemented to flag the failure of the last MARS_MULTI to arrive. A default value of 10 seconds is suggested. If a 'sequence jump' is detected, the host MUST wait for the MARS_MULTI(1,k), discard all results, and repeat the MARS_REQUEST. If a timeout occurs, the host MUST discard all results, and repeat the MARS_REQUEST. Armitage Expires August 4th, 1995 [Page 8] Internet Draft February 4th, 1995 Corruption of cell contents will lead to loss of a MARS_MULTI through AAL5 CPCS_PDU reassembly failure, which will be detected through the mechanisms described above. If the MARS is managing a cluster of endpoints spread across different but directly accessible ATM networks it will not be able to return all the group members in a single MARS_MULTI. The MARS_MULTI message format allows for either E.164, ISO NSAP, or (E.164 + NSAP) to be returned as ATM addresses. However, each MARS_MULTI message may only return ATM addresses of the same type. The returned addresses MUST be grouped according to type (E.164, ISO NSAP, or both) and returned in a sequence of separate MARS_MULTI parts. 5.2 MARS_REQUEST, MARS_MULTI, MARS_MSERV, and MARS_NAK formats. MARS_REQUEST is based on an ATM ARP_REQUEST, but with an 'operation type value' of 11 (decimal). The multicast address being resolved is placed into the the target protocol address field (ar$tpa). The hardware type (ar$hrd) is set to 19 (decimal), and in IP environments the protocol type is 2048 (decimal). Section 6.6 of RFC 1577 should be consulted for specific details and coding of the ar$shtl, ar$sstl, ar$thtl, and ar$tstl fields. MARS_NAK is the MARS_REQUEST returned with operation type value of 16 (decimal). The MARS_MULTI message is identified by an 'operation type value' of 12 (decimal). The message format is: Data: ar$hrd 16 bits Hardware type ( 19 decimal, 0x13 hex) ar$pro 16 bits Protocol type ar$shtl 8 bits Type & length of source ATM number (q) ar$sstl 8 bits Type & length of source ATM subaddress (r) ar$op 16 bits Operation code (MARS_MULTI) ar$spln 8 bits Length of source protocol address (s) ar$thtl 8 bits Type & length of target ATM number (x) ar$tstl 8 bits Type & length of target ATM subaddress (y) ar$tpln 8 bits Length of target multicast group address (z) ar$tnum 16 bits Number of target ATM addresses returned (N). ar$seqxy 16 bits Boolean flag x and sequence number y. ar$msn 32 bits MARS Sequence Number. ar$sha qoctets source ATM number ar$ssa roctets source ATM subaddress ar$spa soctets source protocol address ar$tha.1 xoctets target ATM number 1 ar$tsa.1 yoctets target ATM subaddress 1 ar$tpa zoctets target multicast group address Armitage Expires August 4th, 1995 [Page 9] Internet Draft February 4th, 1995 ar$tha.2 xoctets target ATM number 2 ar$tsa.2 yoctets target ATM subaddress 2 [.......] ar$tha.N xoctets target ATM number N ar$tsa.N yoctets target ATM subaddress N ar$seqxy is coded with flag x in the leading bit, and sequence number y coded as an unsigned integer in the remaining 15 bits. 0 1 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |x| y | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ar$tnum is an unsigned integer indicating how many pairs of {ar$tha,ar$tsa} (i.e. how many group member's ATM addresses) are present in the message. ar$msn is an unsigned 32 bit number filled in by the MARS before transmitting each MARS_MULTI. Its use is described further in section 10. Section 6.6 of RFC 1577 should be consulted for specific details and coding of all other fields. As an example, assume we have a multicast cluster using 4 byte protocol addresses, 20 byte ATM numbers, and 0 byte ATM subaddresses. For n group members in a single MARS_MULTI we require a (44 + 20n) byte message. If we assume the default MTU of 9180 bytes, we can return a maximum of 456 group member's addresses in a single MARS_MULTI. 5.3 Establishing the Multicast VC. Following the completion of the MARS_MULTI reply the endpoint may establish a new point to multipoint VC, or reuse an existing one. If establishing a new VC, an L_MULTI_RQ is issued for ATM.n, followed by an L_MULTI_ADD for every member of the set {ATM.1, ....ATM.(n-1)} (assuming the set is non-null). The packet is then transmitted over the newly created VC just as it would be for a unicast VC. After transmitting the packet, the local interface holds the VC open and marks it as the active path out of the host for any subsequent IP packets being sent to that Class D address. When establishing a new multicast VC is is possible that one or more returned endpoints may reject an L_MULTI_RQ or L_MULTI_ADD. If this occurs then the endpoint's ATM address is dropped from the set {ATM.1, ATM.2, .... ATM.n} returned by the MARS, and the creation of the multipoint VC continues. Armitage Expires August 4th, 1995 [Page 10] Internet Draft February 4th, 1995 Multicast VCs have the potential to be expensive in their use of resources. Therefore each VC MUST have a configurable inactivity timer associated with it. If the timer expires, an L_RELEASE is issued for that VC, and the Class D address is no longer considered to have an active path out of the local host. The timer SHOULD be no less than 1 minute, and a default of 20 minutes is RECOMMENDED. Choice of specific timer periods is beyond the scope of this document. VC consumption may also be reduced by endpoints noting when a new group's set of {ATM.1, ....ATM.n} matches that of a pre-existing VC out to another group. With careful local management, and assuming the QoS of the existing VC is sufficient for both groups, a new pt to mpt VC may not be necessary. Algorithms for performing this type of optimization are not discussed here, and are not required for conformance with this memo. Section 7 describes the endpoint's response to group membership changes while the VC is open. Section 10 describes the mechanism for ensuring hosts remain up to date with changes that occur while the VC is open. 6. Joining and Leaving Multicast Groups. A cluster member is a 'group member' (in the sense that it receives packets directed at the group) when its ATM address appears in the MARS's table entry for the group's multicast address. A key requirement within each cluster is the distribution of group membership information between the MARS and cluster members. Two new messages are defined: MARS_JOIN and MARS_LEAVE. These are sent to the MARS by endpoints joining or leaving a multicast group. The MARS propagates these messages back out to the cluster over its ClusterControlVC, to ensure the knowledge is distributed in a timely fashion. ClusterControlVC is an outgoing, point to multipoint VC with each cluster member as a leaf node. RFC1112 expects that IP multicast routers are capable of behaving 'promiscuously'. This functionality may be emulated by allowing routers to request that the MARS returns them as 'wild card' members of all Class D addresses. However, a problem inherent in the current ATM model is that completely promiscuous behaviour may be wasteful of reassembly resources on the router's ATM interface. This document describes a generalisation to the notion of 'wild card' entries, enabling routers to limit themselves to 'blocks' of the Class D address space. The application of this facility is described in greater detail in Section 9. Armitage Expires August 4th, 1995 [Page 11] Internet Draft February 4th, 1995 A block can be as small as 1 (a single group) or as large as the entire Class D address space (default IPv4 'promiscuous' behaviour). A block is defined as all addresses between, and inclusive of, a address pair. The key extensions required to manage the MARS table entries are: Two new message types: MARS_JOIN carries one or more pairs (specifying one or more blocks of groups being joined) and a unicast ATM address (of the node joining). MARS_LEAVE carries one or more pairs (specifying one or more blocks of groups being left) and a unicast ATM address (of the node leaving). When a MARS_JOIN is received by the MARS it adds the specified ATM address to the table entry for the specified multicast group address(es). When a MARS_LEAVE is received by the MARS it removes the specified ATM address from the ARP entry for the specified multicast group address(es). MARS_JOIN and MARS_LEAVE messages arriving from individual hosts are processed locally by the MARS and retransmitted on ClusterControlVC (possibly after modification, as detailed in Section 8). All endpoints MUST ignore MARS_JOIN or MARS_LEAVE messages that simply confirm information already held. The MARS retransmits redundant messages, but otherwise takes no action. Section 7 describes how endpoints utilize retransmitted MARS_JOIN and MARS_LEAVE messages. Cluster members MUST only include a single pair in each JOIN/LEAVE message they issue. They MUST be able to process multiple pairs in JOIN/LEAVE messages received on ClusterControlVC from the MARS (the interpretation being that the join/leave operation applies to all addresses in range from to inclusive, for every pair). In IPv4 environments JoinLocalGroup now results in two messages being transmitted: MARS_JOIN, sent over a VC to the ARP Server. It identifies the single IP group being joined, and the host's unicast ATM address. Armitage Expires August 4th, 1995 [Page 12] Internet Draft February 4th, 1995 An IGMP Report, except for 224.0.0.1 (in accordance with RFC1112). In IPv4 environments LeaveLocalGroup now results in a MARS_LEAVE being sent over a VC to the MARS, identifying the IP group being left, and the host's unicast ATM address. Endpoints with special requirements (e.g. multicast routers) may directly issue MARS_JOINs and MARS_LEAVEs specifying blocks of multicast group addresses. No IGMP Report is issued for such operations in IP environments. An endpoint must register with a MARS in order to become a member of a cluster and be added as a leaf to ClusterControlVC. Registration is covered in section 6.2. 6.1 Format of the MARS_JOIN and MARS_LEAVE Messages. The MARS_JOIN message is indicated by an operation type value of 14 (decimal). MARS_LEAVE has the same format and operation type value of 15 (decimal). The message format is: Data: ar$hrd 16 bits Hardware type (19 decimal) ar$pro 16 bits Protocol type ar$shtl 8 bits Type & length of source ATM number (q) ar$sstl 8 bits Type & length of source ATM subaddress (r) ar$op 16 bits Operation code (MARS_JOIN or MARS_LEAVE) ar$spln 8 bits Length of source protocol address (s) ar$tpln 8 bits Length of multicast group address (z) ar$pnum 16 bits Number of multicast group address pairs (N) ar$resv 16 bits Reserved. ar$msn 32 bits MARS Sequence Number. ar$sha qoctets source ATM number (E.164 or ATM Forum NSAPA). ar$ssa roctets source ATM subaddress (ATM Forum NSAPA). ar$spa soctets source protocol address ar$min.1 zoctets Minimum multicast group address - pair.1 ar$max.1 zoctets Maximum multicast group mask - pair.1 [.......] ar$min.N zoctets Minimum multicast group address - pair.N ar$max.N zoctets Maximum multicast group mask - pair.N Refer to RFC 1577, section 6.6 for the coding of the ar$shtl and ar$sstl fields. For conventional IPv4 environments ar$spln and ar$tpln are both set to 4. Note that the message format differs from ATMARP_REPLY in the fields after ar$op. ar$msn is an unsigned 32 bit number filled in by the MARS before re-transmitting a MARS_JOIN or MARS_LEAVE. The originator SHOULD set it to zero, although it will be ignored by the MARS. Its use is described further in section 10. Armitage Expires August 4th, 1995 [Page 13] Internet Draft February 4th, 1995 A join/leave message carries a set {, , ...., }, with at least one pair. ar$pnum indicates how many pairs are included in the message. To simplify MARS and endhost interpretation, the following restrictions are imposed: Assume max(N) is the field from the Nth pair. Assume min(N) is the field from the Nth pair. Assume a join/leave message arrives with K pairs. The following must hold: max(N) < min(N+1) for 1 <= N < K max(N) >= min(N) for 1 <= N <= K In plain english, the set must specify an ascending sequence of address blocks. The definition of "greater" or "less than" may be protocol specific. In IP environments the addresses are treated as simple unsigned binary values. 6.1.1 Important IPv4 default values. The JoinLocalGroup and LeaveLocalGroup operations are only valid for a single group. For any arbitrary group address X the associated MARS_JOIN or MARS_LEAVE MUST specify a single pair . A router choosing to behave strictly in accordance with RFC1112 MUST specify the entire Class D space. The associated MARS_JOIN or MARS_LEAVE MUST specify a single pair <224.0.0.0, 239.255.255.255>. The use of alternative values is discussed in Section 9. 6.2 Registering with the MARS. Two separate signalling paths exist between cluster members and their associated MARS. The first is a transient point to point VC that cluster members establish to the MARS when they need to issue MARS_REQUESTs, MARS_JOINs, or MARS_LEAVEs. This VC is used by the MARS to return MARS_MULTI messages. It has an associated idle timer, and is dismantled if not used for a configurable period of time. The minimum suggested value for this time is 1 minute, and the RECOMMENDED default is 20 minutes. The second signalling path is ClusterControlVC. Every endpoint registered as a cluster member is added as a leaf node to this VC, which exists for the lifetime of the MARS. It is used to re- distribute MARS_JOIN and MARS_LEAVE messages received by the MARS from individual cluster members. Registration with the MARS as a cluster member occurs when an endpoint issues a MARS_JOIN for a protocol specific multicast group address. Once this occurs the endpoint is added as a leaf node to ClusterControlVC. Armitage Expires August 4th, 1995 [Page 14] Internet Draft February 4th, 1995 In IPv4 environments the 'all nodes' Class D address of 224.0.0.1 is used to register with the MARS. RFC 1112 requires that all hosts (including routers) that wish to participate in Level 2 IP multicasting must explicitly issue a JoinLocalGroup for group 224.0.0.1 when they initialise (Level 1 is not supported by this memo). The JoinLocalGroup to 224.0.0.1 will result in an MARS_JOIN being transmitted from the host to the MARS. If an IPv4 endpoint issues a LeaveLocalGroup for 224.0.0.1 it will also be considered to have ceased membership of all other groups for which it may have joined. The MARS MUST flush that endpoint's ATM address from any Class D address entries it appears in. Finally, the endpoint is released as a Leaf node from ClusterControlVC. If the MARS receives an ERR_L_RELEASE on ClusterControlVC indicating that a cluster member has died, that member's ATM address MUST be removed from all groups for which it may have joined. Registration of endpoints for other protocols is currently beyond the scope of this document. 7. Endpoint management of point to multipoint VCs. Once a cluster member has established a new VC to the members returned in a MARS_MULTI response it must: Monitor traffic on ClusterControlVC for updates to the group's membership. Revalidate a group's membership if a leaf node releases itself from the VC. 7.1 Monitoring updates on ClusterControlVC. When a cluster member joins or leaves a particular multicast group it is not sufficient to simply update the mapping table in the cluster's MARS. Endpoints that are already transmitting to the multicast group's members must be informed of the change so they may add or remove a leaf node as appropriate. Cluster members track MARS_JOIN and MARS_LEAVE messages retransmitted by the MARS to determine when another endpoint joins or leaves a group or block of groups. If a MARS_JOIN is seen that refers to (or encompasses) a group for which the transmit side already has a VC open, the new member's ATM address is extracted and an L_MULTI_ADD issued locally. This ensures that hosts already sending to a given group will immediately add the new member to their list of recipients. It also ensures that routers joining a 'block' of groups are added by all endpoints currently Armitage Expires August 4th, 1995 [Page 15] Internet Draft February 4th, 1995 sending to groups within the block. If a MARS_LEAVE is seen that refers to (or encompasses) a group for which the transmit side already has a VC open, the old member's ATM address is extracted and an L_MULTI_DROP issued locally. This ensures that hosts already sending to a given group will immediately drop the old member from their list of recipients. In an IPv4 environment any endpoint leaving 224.0.0.1 is assumed to be ceasing support for IP multicast operation. If a MARS_LEAVE is seen that refers to group 224.0.0.1 then the ATM address of the endpoint specified in the message MUST be removed from every multipoint VC on which it is listed as a leaf node. The transmit side of the interface MUST NOT shut down an active VC to a group for which the receive side has just executed a LeaveLocalGroup. This behaviour is consistent with the model of hosts transmitting to groups regardless of their own membership status. If a MARS_JOIN or MARS_LEAVE arrives with ar$pnum == 0 it carries no pairs, and is only used for validation as described in section 10. 7.2 Revalidating when leaf nodes drop themselves. During the life of a multipoint VC an ERR_L_RELEASE may be received indicating that a leaf node has terminated its participation at the ATM level. The ATM endpoint associated with the ERR_L_RELEASE MUST be removed from the locally held set {ATM.1, ATM.2, .... ATM.n} associated with the VC. After a random period of time between 1 and 10 seconds the endpoint MUST revalidate the associated group's membership by re-issuing a MARS_REQEUEST. The returned set of members {NewATM.1, NewATM.2, .... NewATM.n} is compared with the set already held locally. L_MULTI_DROPs are issued on the group's VC for each node that appears in the original set of members but not in the revalidated set of members. L_MULTI_ADDs are issued on the group's VC for each node that appears in the revalidated set of members but not in the original set of members. 8. Configuring for Multicast Servers or Multicast Meshes. Endpoint's assume that all groups are supported by meshes of point to multipoint VCs. Under certain circumstances the consumption of VCs and AAL resources around the cluster can make meshes unattractive, despite their performance advantages. The MARS protocol provides a Armitage Expires August 4th, 1995 [Page 16] Internet Draft February 4th, 1995 mechanism for introducing multicast servers on a per-multicast group basis, and in a manner that is completely transparent to cluster members. The multicast server has two key roles: Providing one (or a limited number of) leaf nodes for outgoing VCs from cluster members. Constructing a single point to multipoint VC, with each group memember as a leaf. This reduces the AAL consumption to one per group, rather than one per sender per group. The MARS must keep two sets of mappings for each multicast group address supported by multicast servers. The original {multicast address, ATM.1, ATM.2, ... ATM.n} mapping (the 'host map', although it includes routers) is augmented by a parallel {multicast address, server.1, server.2, .... server.K} mapping (the 'server map'). It is assumed that no ATM addresses appear in both the server and host maps for the same multicast group. Typically K will be 1, but it will be larger when multiple multicast servers are configured to share the data load of a given group. When the MARS receives a MARS_REQUEST for a multicast address that has both host and server maps it generates a response based on the identity of the request's source. If the requestor is a member of the server map for the requested group then the MARS returns the contents of the host map in a sequence of one or more MARS_MULTIs. Otherwise the MARS returns the contents of the server map in a sequence of one or more MARS_MULTIs. Servers use the host map to establish a basic distribution VC for the group. Cluster members will establish outgoing multipoint VCs to members of the group's server map, without being aware that their packets will not be going directly the multicast group's members. The MARS also maintains a point to multipoint VC out to any multicast servers it is aware of, called ServerControlVC. This serves an analogous role to ClusterControlVC, allowing the MARS to update the servers with group membership changes as they occur. A set of four MARS messages cover the current requirements: MARS_MSERV Register as multicast server for one or more groups. MARS_UNSERV Deregister as multicast server for one or more groups. MARS_SJOIN A JOIN message on ServerControlVC. MARS_SLEAVE A LEAVE message on ServerControlVC. Armitage Expires August 4th, 1995 [Page 17] Internet Draft February 4th, 1995 MARS_SJOIN/SLEAVE are identical in format to MARS_JOIN/LEAVE, but have different operation codes so that a node acting as both a cluster member and multicast server may distinguish between updates arriving on ServerControlVC and ClusterControlVC. 8.1 Registering and deregistering multicast servers. MARS_MSERV and MARS_UNSERV are identical to the MARS_JOIN message. MARS_MSERV uses the set {, , ...., } to specify one or more sets of multicast groups that a multicast server is willing to support. MARS_UNSERV indicates the set of groups that the multicast server is no longer willing to support. The operation code for MARS_MSERV is 11 (decimal), and MARS_UNSERV is 17 (decimal). When a node registers with MARS_MSERV the MARS adds the new ATM address to the server maps for each specified group, possibly constructing a new server map if this is the first multicast server for the group. If the multicast server is not already a leaf node of ServerControlVC it is added. When a node deregisters with MARS_UNSERV the MARS removes its ATM address from the server maps for each specified group, deleting the server map if this was the only server for the group. Both of these messages are sent to the MARS over a point to point VC, and echoed on ServerControlVC by the MARS (section 10 covers the use of this behaviour). The operation code is then changed to MARS_JOIN or MARS_LEAVE respectively, and a copy of the original message is transmitted on ClusterControlVC. The MARS retransmits but otherwise ignores redundant MARS_MSERV and MARS_UNSERV messages. It is assumed that at least one server will have registered to support a group before the first cluster member joins it. If a MARS_MSERV arrives for a group that has a non-null host map but no server map the default response of the MARS will be to drop the MARS_MSERV without any further action. The originating multicast server will eventually flag an error when repeated attempts to register fail. The opposite situation is where the last or only multicast server for a group deregisters itself while the group still has members. The default solution is for multicast servers to sever all VCs to which they are attached as leaf nodes when they deregister, forcing any active senders to the group to revalidate (as described in section 7). Since the MARS will have deleted the server map, the revalidation will result in the host map being return, and the group Armitage Expires August 4th, 1995 [Page 18] Internet Draft February 4th, 1995 reverts to being a mesh. This shall be the default mechanism until future work develops a more elegant approach. Appendix C discusses possible extensions to allow dynamic transitions between mesh and multicast server support while a group is active. However, these are not required for conformance with this memo. 8.2 Handling group membership changes. The existence of multicast servers supporting some groups but not others requires the MARS to intervene in the distribution of single and block join/leave updates to cluster members. The MARS_SJOIN and MARS_SLEAVE messages are identical to MARS_JOIN, with operation codes 18 and 19 (decimal) respectively. They exist to allow a node combining cluster member and multicast server to distinguish between information arriving on ClusterControlVC and ServerControlVC. When a cluster member issues MARS_JOIN or MARS_LEAVE for a single group, the MARS checks to see if the group has an associated server map. If the specified group does not have a server map the MARS_JOIN or MARS_LEAVE is retransmitted on ClusterControlVC. If it does have a server map two transmissions occur: A copy is made with type MARS_SJOIN or MARS_SLEAVE as appropriate and transmitted on ServerControlVC. This allows the server(s) supporting the group to note the new member and add it as a leaf node. The original message's ar$pnum field is set to 0, and it is transmitted back using the VC it arrived on (rather than ClusterControlVC). (Section 10 requires cluster members have a mechanism to confirm the reception of their message by the MARS. For mesh supported groups, using ClusterControlVC serves dual purpose of providing this confirmation and distributing group update information. When using multicast servers there is no reason for having all cluster members process and discard null join/leave messages on ClusterControlVC). Receipt of a block join/leave (e.g. from a router coming on-line) requires a more complex response. Cluster members must be directly informed of which mesh supported groups the block covers. Multicast servers must also be informed in case they support one of the groups covered by the block being joined. Armitage Expires August 4th, 1995 [Page 19] Internet Draft February 4th, 1995 The solution is for the MARS to 'punch holes' in the block of addresses supplied in the join/leave message, creating a set of pairs that excludes those addresses/groups supported by the multicast servers. This hole-punched set is then sent out on ClusterControlVC, ensuring the router is immediately noted by senders to any mesh supported groups in the block. The original MARS_JOIN/LEAVE is then converted to a MARS_SJOIN/SLEAVE and transmitted on ServerControlVC. Appendix A discusses some algorithms for 'hole punching'. If punching holes in the originally specified block leaves a null set, the ar$pnum field is set to zero before sending the modified MARS_JOIN/LEAVE on ClusterControlVC. 8.3 Multicast server architectures. Specification of multicast server architectures, and the synchronisation of multiple multicast servers supporting single multicast groups, is beyond the scope of this document and is expected to be the subject of further work. Appendix C discusses some possible approaches. 9. Utilizing blocks for for multicast routers. Multicast routers are required for the propagation of multicast traffic beyond the constraints of a single cluster. There is a sense in which they are multicast servers acting at the next higher layer, with clusters rather than individual endpoints as their abstract sources and destinations. Multicast routers typically participate in higher layer multicast routing algorithms and policies that are beyond the scope of this memo (e.g. DVMRP [5] in the IPv4 environment). It is assumed that the multicast routers will be implemented over the same sort of IP/ATM interface that a multicast host would use. They will use the basic services described in the preceeding sections to join and leave multicast groups as necessary, and will register with the MARS as a cluster member. The rest of this section will assume a simple IPv4 scenario where the scope of a cluster has been limited to a particular LIS that is part of an overlaid IP network. Not all members of the LIS are necessarily registered cluster members. Armitage Expires August 4th, 1995 [Page 20] Internet Draft February 4th, 1995 9.1 Sending to a Group. If the multicast router needs to transmit a packet to a group within the cluster it opens a VC in the same manner as a normal host would. Once a VC is open, the router watches for MARS_JOIN and MARS_LEAVE messages and responds to them as a normal host would. The multicast router's transmit side MUST implement inactivity timers to shut down idle outgoing VCs, as for normal hosts. As with normal host, the multicast router does not need to be a member of a group it is sending to. 9.2 Promiscuously Joining Groups. Once registered and initialised, the simplest model of IPv4 multicast router operation is for it to issue a MARS_JOIN encompassing the entire Class D address space. In effect it becomes 'promiscuous', as it will be a leaf node to all present and future multipoint VCs established to IPv4 groups on the cluster. How a router chooses which groups to propagate outside the cluster is beyond the scope of this memo. Consistent with RFC 1112, IP multicast routers may retain the use of IGMP Query and IGMP Report messages to ascertain group membership. 9.3 Forward Multicast Traffic Across the cluster. Under some circumstances the cluster may simply be another hop between IP subnets that have participants in a multicast group. [LAN.1] ----- IPmcR.1 -- [LIS] -- IPmcR.2 ----- [LAN.2] LAN.1 and LAN.2 are subnets (such as Ethernet) with attached hosts that are members of group X. IPmcR.1 and IPmcR.2 are multicast routers with interfaces to the LIS. A traditional solution would be to treat the LIS as a unicast subnet, and use tunneling routers. However, this would not allow hosts on the LIS to participate in the cross-LIS traffic. Assume IPmcR.1 is receiving packets promiscuously on its LAN.1 interface. Assume further it is configured to propagate multicast traffic to all attached interfaces. In this case that means the LIS. When a packet for group X arrives on its LAN.1 interface, IPmcR.1 Armitage Expires August 4th, 1995 [Page 21] Internet Draft February 4th, 1995 simply sends the packet to group X on the LIS interface as a normal host would (Issuing MARS_REQUEST for group X, creating the VC, sending the packet). Assuming IPmcR.2 initialised itself with the MARS as a member of the entire Class D space, it will have been returned as a member of X even if no other nodes on the LIS were members. All packets for group X received on IPmcR.2's LIS interface may be retransmitted on LAN.2. If IPmcR.1 is similarly initialised the reverse process will apply for multicast traffic from LAN.2 to LAN.1, for any multicast group. The benefit of this scenario is that cluster members within the LIS may also join and leave group X at anytime. 9.4 Restricted 'promiscous' Operation. Both unicast and multicast IP routers have a common problem - limitations on the number of AAL contexts available at their ATM interfaces. Being 'promiscuous' in the RFC 1112 sense means that for every M hosts sending to N groups, a multicast router's ATM interface will have M*N incoming reassembly engines tied up. It is not hard to envisage situations where a number of multicast groups are active within the LIS but are not required to be propagated beyond the LIS itself. An example might be a distributed simulation system specifically designed to use the high speed IP/ATM environment. There may be no practical way its traffic could be utilised on 'the other side' of the multicast router, yet under the conventional scheme the router would have to be a leaf to each participating host anyway. As this problem occurs at the link layer, it is worth noting that 'scoping' mechanisms at the IP multicast routing level do not provide a solution. In this situation the network administrator might configure their multicast routers to exclude sections of the Class D address space when issuing MARS_JOIN(s). Multicast groups that will never be propagated beyond the cluster will not have the router listed as a member by the MARS, and the router will never have to receive and ignore traffic from those groups. Another scenario involves the product M*N exceeding the capacity of a single router's interface (especially if the same interface must also support a unicast IP router service). A network administrator may choose to add a second node, to function as a parallel IP multicast router. Each router would be configured to Armitage Expires August 4th, 1995 [Page 22] Internet Draft February 4th, 1995 be 'promiscuous' over separate parts of the Class D address space, thus exposing themselves to only part of the VC load. This sharing would be completely transparent to IP hosts within the LIS. Restricted promiscuous mode does not break RFC 1112's use of IGMP Report messages. If the router is configured to serve a given block of Class D addresses, it will receive the IGMP Report. If the router is not configured to support a given block, then the existence of an IGMP Report for a group in that block is irrelevant to the router. All routers are able to track membership changes through the MARS_JOIN and MARS_LEAVE traffic anyway. Mechanisms for establishing these modes of operation are beyond the scope of this memo. 10. Robustness of interaction with the MARS. Transient problems may result in the loss of messages between the MARS, cluster members, and multicast servers. More serious problems may result in the failure of the MARS itself. There are two problem scenarios that are addressed. The first is the inability of a cluster member to send messages to the MARS itself, either through cell loss on the VC to the MARS, or the cluster member's inability to establish a VC to the MARS. The second is with the MARS_JOIN/SJOIN/LEAVE/SLEAVE messages re- transmitted from the MARS. If a cluster member or multicast server currently sending to a group misses an join update, the newly joined member misses out on some traffic to the group. If a cluster member or multicast server currently sending to a group misses a leave update, the cluster member that left will continue to receive packets unecessarily. 10.1 Ensuring the MARS hears you. A simple algorithm solves the first problem. Cluster members retransmit MARS_JOIN and MARS_LEAVE messages at regular intervals until they receive a copy back again, either on ClusterControlVC or the VC on which they are sending the messages. At this point the local endpoint can be certain that at least the MARS received it. Multicast servers retransmit MARS_MSERV and MARS_UNSERV messages at regular intervals until they receive a copy back on ServerControlVC. The interval should be no shorter than 5 seconds, and a default value of 10 seconds is recommended. After 5 retransmissions the attempt should be flagged locally as a failure. This should be considered as a MARS failure, and handled as described in section 10.2. Armitage Expires August 4th, 1995 [Page 23] Internet Draft February 4th, 1995 A 'copy' is defined as seeing a message of the same operation code containing the local host's identity in the source address fields. The pair set is not checked, and does not have to be the same (this is required so that cluster members may verify a MARS_JOIN they've sent even if the MARS's hole-punching creates a totally different set of pairs). 10.2 Temporary failure of the MARS. Two failure modes indicate problems with the MARS itself: If an ERR_L_RELEASE occurs for the cluster member's attachment to ClusterControlVC it may be assumed some problem exists with the MARS. If the cluster member receives ERR_L_RQFAILED when it attempts to establish a point to point VC to the MARS in order to send MARS messages. The cluster member should wait a random period of time between 1 and 10 seconds before attempting to re-register with the MARS. If the registration MARS_JOIN is successful (in accordance with section 10.1) then: The cluster member MUST then proceed to rejoin every group that its local higher layer protocol(s) have joined. It is recommended that a random delay between 1 and 10 seconds be inserted before the transmission of each MARS_JOIN. Finally, using the mechanism described in section 7, the cluster member MUST begin revalidating every multicast group it was sending to. The rejoin and revalidation procedure must not disrupt the cluster member's use of multipoint VCs that were already open at the time of the MARS failure. If the re-registration with the Primary MARS fails, and there is no configured Secondary MARS, the cluster member MUST wait for at least 1 minute before repeating the re-registration procedure. It is RECOMMENDED that the cluster member signals an error condition in some locally significant fashion. If the re-registration with the Primary MARS fails, and a Secondary MARS has been configured, the Secondary and Primary MARS addresses are swapped and the cluster member immediately repeats the re- registration procedure. If this is succesful the cluster member will resume normal operation using the Secondary MARS. It is RECOMMENDED Armitage Expires August 4th, 1995 [Page 24] Internet Draft February 4th, 1995 that the cluster member signals a warning of this condition in some locally significant fashion. If the attempt at re-registration with the Secondary MARS fails, the cluster member MUST wait for at least 1 minute before reverting back to the Primary MARS and starting the whole re-registration process over again. In the worst case scenario this will result in cluster members looping between registration attempts with the Primary MARS and Secondary MARS until network administrators manually intervene. Multicast servers shall behave in a similar manner to cluster members on this issue. 10.3 The MARS Sequence Number. There is an unsigned 32 bit sequence number identified as ar$msn in most MARS messages. The following extensions govern its use: The MARS keeps two independent counters, Cluster Sequence Number (CSN) and Server Sequence Number (SSN). They are incremented every time a message is sent out ClusterControlVC or ServerControlVC respectively. [Editorial note - in ipmc-03.txt the counter was incremented only when a change occurred in the mapping tables. this is a simplification.] The current CSN is copied into the ar$msn field of MARS messages being sent to cluster members (either out ClusterControlVC or on an individual VC). The current SSN is copied into the ar$msn field of MARS messages being sent to multicast servers (either out ServerControlVC or on an individual VC). Cluster members and multicast servers track the increments of CSN or SSN to determine if they have missed any update messages. Calculations on the sequence numbers MUST be performed as unsigned 32 bit arithmetic, to ensure no glitches when the counters roll over. Every cluster member keeps its own 32 bit Host Sequence Number (HSN) to track the MARS's sequence number. Whenever a MARS_MULTI, MARS_JOIN, or MARS_LEAVE is received the following check is then performed on the ar$msn field of the new message: Seq.diff = ar$msn - HSN Armitage Expires August 4th, 1995 [Page 25] Internet Draft February 4th, 1995 ar$msn -> HSN {...process MARS message as appropriate...} if ((Seq.diff != 1) && (Seq.diff != 0)) then {...revalidate group membership information...} The basic result is that the cluster member attempts to keep locked in step with membership changes noted by the MARS. If it ever detects that a membership change occurred (in any group) without it noticing, it re-validates the membership of all groups it currently has multicast VCs open to. Revalidation involves treating each VC as though an ERR_L_RELEASE was received from a leaf node, and executing the procedure described in section 7. The ar$msn field of consecutive MARS_MULTIs sent in response to a MARS_REQUEST must be constant. If the ar$msn field changes then all the messages MUST be discarded at the completion of the response, and the MARS_REQUEST re-issued. One implication of this mechanism is that the MARS should serialize its processing of 'simultaneous' MARS_REQUEST, MARS_JOIN and MARS_LEAVE messages. Join and Leave operations should be queued within the MARS along with MARS_REQUESTS, and not processed until all the reply packets of a preceeding MARS_REQUEST have been transmitted. The MARS is free to choose a value of CSN and SSN. When a new cluster member starts up it should initialise HSN to zero. When the cluster member sends the MARS_JOIN to register, the HSN will be correctly set when it receives a copy of its MARS_JOIN from the MARS. If Seq.diff > 1 when the MARS_JOIN returns no action will be taken anyway, as the host will not have any multicast related VCs established at this stage. If a sequence number jump occurs when establishing a new group's VC the cluster member should not revalidate the membership of the group it just established. The membership returned in the MARS_MULTIs that carried the new ar$msn field should be considered already validated. A MARS should be carefully designed to minimise the possibility of CSN or SSN jumping unecessarily. Under normal operation only hosts that are affected by transient link problems will miss ar$msn updates and be forced to revalidate. If the MARS itself glitches it will be innundated with requests for a period as every cluster member attempts to revalidate. Multicast servers should utilize the ar$msn fields in exactly the same manner as cluster members. This will enable them to track the SSN, and recover from missing any MARS_SJOIN/SLEAVE traffic. Armitage Expires August 4th, 1995 [Page 26] Internet Draft February 4th, 1995 10.4 Why a Gobal sequence number? The CSN and SSN are global within the context of a given protocol (e.g. IP). They count ClusterControlVC and ServerControlVC activity without reference to the multicast group(s) involved. This may be perceived as a limitation, because there is no way for cluster members or multicast servers to isolate exactly which multicast group they may have missed an update for. An alternative was to try and provide a per-group sequence number. Unfortunately per-group sequence numbers are not practical. The current mechanism allows sequence information to be piggy-backed onto MARS messages already in transit for other reasons. The ability to specify blocks of multicast addresses with a single MARS_JOIN or MARS_LEAVE means that a single message can refer to membership change for multiple groups simultaneously. A single ar$msn field cannot provide meaningful information about each group's sequence. Multiple ar$msn fields would have been unwieldy. Any MARS or cluster member that supports different protocols MUST keep separate mapping tables and sequence numbers for each protocol. 10.5 Synchronizing the Primary and Secondary MARS. If a Secondary MARS exists for a given cluster then some mechanism is needed to ensure reasonable consistency between its mapping tables and those of the Primary MARS, especially as cluster members will only ever register with one MARS. The inter-server protocol also needs to cope with post-failure situations where some cluster members end up registered with the Primary and others with the Secondary. The definition of an inter-server protocol is beyond the current scope of this document, and is expected to be the subject of further work in the area. 11. Using the MARS in non-IP environments. An deliberate attempt has been made to describe the MARS and associated mechanisms in a manner independent of a specific higher layer protocol being run over the ATM cloud. The immediate application of this document will be in an IPv4 environment, and this is reflected by the focus of key examples. However, the coding of each MARS message means that any higher layer protocol identifiable by a two byte Ethernet Type code can be supported by a MARS. The 16 bit 'Protocol type' at the start of each MARS message, taken from the set of Ethernet Type codes. Every MARS MUST implement entirely separate logical mapping tables and support. Every cluster Armitage Expires August 4th, 1995 [Page 27] Internet Draft February 4th, 1995 member must interpret messages from the MARS in the context of the protocol type that the MARS message refers to. The LLC/SNAP encapsulation specified in section 4 should not be considered a hinderance in non-IP environments. Experimenters deploying IPX or AppleTalk over ATM are encouraged to use the architecture described in this document to support possible multicast needs. 12. Key Decisions and open issues. The key decisions this memo proposes: A Multicast Address Resolution Server (MARS) is proposed to co- ordinate and distribute mappings of ATM endpoint addresses to arbitrary higher layer 'multicast group addresses'. The specific case of IP version 4 multicast is used as the example. Individual multicast groups may be supported by multicast meshes between group members, or by multicast servers. The concept of 'clusters' is introduced to define the scope of a MARS's responsibility, and the set of ATM endpoints willing to participate in link level multicasting. MARS message formats and encapsulation allow co-resident MARS and ATM ARP Server implementations. New message types: MARS_JOIN, MARS_LEAVE, MARS_REQUEST. Allow endpoints to join, leave, and request the current membership list of multicast groups. New message type: MARS_MULTI. Allows multiple ATM addresses to be returned by the MARS in response to a MARS_REQUEST. New message types: MARS_MSERV, MARS_UNSERV. Allow multicast servers to register and deregister themselves with the MARS. New message types: MARS_SJOIN, MARS_SLEAVE. Allow MARS to pass on group membership changes to multicast servers. 'wild card' MARS mapping table entries possible, where a single ATM address is simultaneously associated with blocks of multicast group addresses. Some issues have not been addressed, although they may be in future revisions. MARS has no mechanism for realising cluster members have silently Armitage Expires August 4th, 1995 [Page 28] Internet Draft February 4th, 1995 died. The future development of ATM Group Addresses and Leaf Initiated Join to ATM Forum's UNI specification has not been addressed. The problems identified in this memo with respect to VC scarcity and impact on AAL contexts will not be fixed by such developments in the signalling protocol. Security Consideration Security consideration are not addressed in this memo. Acknowledgments The discussions within the IP over ATM Working Group have helped clarify the ideas expressed in this document. John Moy of Cascade Communications Corp. initially suggested the idea of wild-card entries in the ARP Server. Drew Perkins of Fore Systems provided rigorous and useful critique of early proposed mechanisms for distributing and validating group membership information. Susan Symington (and co-workers at MITRE Corp., Don Chirieleison, Rich Verjinski, and Bill Barns) clearly articulated the need for multicast server support, proposed a solution, and challenged earlier block join/leave mechanisms. Author's Address Grenville Armitage MRE 2P340, 445 South Street Morristown, NJ, 07960-6438 USA Email: gja@thumper.bellcore.com References [1] S. Deering, "Host Extensions for IP Multicasting", RFC 1112, Standford University, August 1989. [2] Heinanen, J., "Multiprotocol Encapsulation over ATM Adaption Layer 5", RFC 1483, USC/Information Science Institute, July 1993. [3] Laubach, M., "Classical IP and ARP over ATM", RFC1577, Hewlett- Packard Laboratories, December 1993 [4] ATM Forum, "ATM User-Network Interface Specification Version 3.0", Englewood Cliffs, NJ: Prentice Hall, September 1993 Armitage Expires August 4th, 1995 [Page 29] Internet Draft February 4th, 1995 [5] D. Waitzman, C. Partridge, S. Deering, "Distance Vector Multicast Routing Protocol", RFC 1075, November 1988. [6] M. Perez, F. Liaw, D. Grossman, A. Mankin, E. Hoffman, A. Malis, "ATM Signaling Support for IP over ATM", Internet Draft, IP over ATM Working Group, draft-ietf-ipatm-sig-02.txt, November, 1994. Armitage Expires August 4th, 1995 [Page 30] Internet Draft February 4th, 1995 Appendix A. Parsing MARS messages. Implementations are entirely free to comply with the body of this memo in any way they see fit. This appendix is purely for clarification. A smart MARS implementation will pre-construct a set of pairs (P) that reflects the entire Class D space, excluding any addresses currently supported by multicast servers. The field of the first pair MUST be 224.0.0.0, and the field of the last pair MUST be 239.255.255.255. The first and last pair may be the same. This set is updated whenever a multicast server registers or deregisters. When the MARS must perform 'hole punching' it might consider the following algorithm: Assume the MARS_JOIN/LEAVE received by the MARS from the cluster member specied the block . Assume Pmin(N) and Pmax(N) are the and fields from the Nth pair in the MARS's current set P. Assume set P has K pairs. Pmin(1) MUST equal 224.0.0.0, and Pmax(M) MUST equal 239.255.255.255. (If K == 1 then no hole punching is required). Execute pseudo-code: create copy of set P, call it set C. index1 = 1; while (Pmax(index1) <= Emin) index1++; index2 = K; while (Pmin(index2) >= Emax) index2--; if (Pmin(index1) < Emin) Cmin(index1) = Emin; if (Pmax(index2) > Emax) Cmax(index2) = Emax; Set C is the required 'hole punched' set of address blocks. The resulting set C retains all the MARS's pre-constructed 'holes' Armitage Expires August 4th, 1995 [Page 31] Internet Draft February 4th, 1995 covering the multicast servers, but will have been pruned to cover the section of the Class D space specified by the originating host's values. The host end should keep a table, H, of open VCs in ascending order of Class D address. Assume H(x).addr is the Class address associated with VC.x. Assume H(x).addr < H(x+1).addr. The pseudo code for updating VCs based on an incoming JOIN/LEAVE might be: x = 1; N = 1; while (x < no.of VCs open) { while (H(x).addr > max(N)) { N++; if (N > no. of pairs in JOIN/LEAVE) return(0); } if ((H(x).addr <= max(N) && ((H(x).addr >= min(N)) perform_VC_update(); x++; } Armitage Expires August 4th, 1995 [Page 32] Internet Draft February 4th, 1995 Appendix B. Coping with IPv4 idiosyncracies. Implementing any part of this appendix is not required for conformance with this memo. It is provided solely to document issues that have been identified. The intent of section 5.3 is for cluster members to only have outgoing point to multipoint VCs when they are actually sending data to a particular multicast groups. However, in most IPv4 environments the multicast routers attached to a cluster will periodically issue IGMP Queries to ascertain if particular groups have members. The current IGMP specification attempts to avoid having every group member respond by insisting that each group member wait a random period, and responding if no other member has responded before them. The IGMP reply is sent to the multicast address of the group being queried. Unfortunately, as it stands the IGMP algorithm will be a nuisance for cluster members that are essentially passive receivers within a given multicast group. It is just as likely that a passive member, with no outgoing VC already established to the group, will decide to send an IGMP reply - causing a VC to be established were there was no need for one. This is not a fatal problem for small clusters, but will seriously impact on the ability of a cluster to scale. Various solutions exist, providing short and long term solutions to the problem. One long term solution would be to modify the IGMP algorithm, for example: If the group member has VC open to the group proceed as per RFC 1112 (picking a random reply delay between 0 and 10 seconds). If the group member does not have VC already open to the group, pick random reply delay between 10 and 20 seconds instead, and then proceed as per RFC 1112. If even one group member is sending to the group at the time the IGMP Query is issued then all the passive receivers will find the IGMP Reply has been transmitted before their delay expires, so no new VC is required. If all group members are passive at the time of the IGMP Query then a response will eventually arrive, but 10 seconds later than under conventional circumstances. The preceeding solution requires re-writing existing IGMP code, and implies the ability of the IGMP entity to ascertain the status of VCs on the underlying ATM interface. This is not likely to be available in the short term. Armitage Expires August 4th, 1995 [Page 33] Internet Draft February 4th, 1995 One short term solution is to provide something like the preceeding functionality with a 'hack' at the IP/ATM driver level within cluster members. Arrange for the IP/ATM driver to snoop inside IP packets looking for IGMP traffic. If an IGMP packet is accepted for transmission, the IP/ATM driver can buffer it locally if there is no VC already active to that group. A 10 second timer is started, and if an IGMP Reply for that group is received from elsewhere on the cluster the timer is reset. If the timer expires, the IP/ATM driver then establishes a VC to the group as it would for a normal IP multicast packet. Some network implementors may find it advantageous to configure a multicast server to support the group 224.0.0.1, rather than rely on a mesh. Given that IP multicast routers regularly send IGMP queries to this address, a mesh will mean that each router will permanently consume an AAL context within each cluster member. In clusters served by multiple routers the VC load within switches in the underlying ATM network will become a scaling problem. Finally, if a multicast server is used to support 224.0.0.1, another ATM driver level hack becomes a possible solution to IGMP Reply traffic. The ATM driver may choose to grab all outgoing IGMP packets and send them out on the VC established for sending to 224.0.0.1, regardless of the Class D address the IGMP message was actually for. Given that all hosts and routers must be members of 224.0.0.1, the intended recipients will still receive the IGMP Replies. The negative impact is that all cluster members will receive the IGMP Replies. Armitage Expires August 4th, 1995 [Page 34] Internet Draft February 4th, 1995 Appendix C. Issues relating to multicast servers. Implementing any part of this appendix is not required for conformance with this memo. It is provided to give some structure to further research and development on multicast server support within clusters. Various items are not addressed by this memo. They include: Automatic migration of cluster members from a mesh to a multicast server while a group is active. An elegant mechanism for migration of cluster members from multicast servers back to a mesh while the group is active. Additional intelligence in the MARS to perform load sharing between multicast servers if more than one registers for the same group. If one or more multicast servers attempt to register for a group that already has members, it would be nice to have current senders to the group migrate their outgoing VCs from the actual cluster members to the newly registered multicast server(s). One approach might be to have the MARS issue a sequence of fabricated MARS_JOINs for the multicast servers, followed by MARS_LEAVEs for each member of the group's current host map. What load this would place on the MARS, and its scalability, have not been considered. An elegant mechanism for the reverse migration might well be based around the reverse process. Issue MARS_JOINs for all entries in the host map, then issue MARS_LEAVEs for all remaining entries in the server map. In case of groups served by multiple multicast servers, the current expectation is that each server retrieves the entire group's membership with MARS_REQUESTs. This memo expects there to be an external mechanism for multiple multicast servers to synchronize the load sharing amongst themselves. Whether the MARS should to be extended to play a part is a subject for further work. An issue not immediately related to the MARS architecture is whether a multicast server retransmits using a point to multipoint VC out to group members, or a set of one VC per group member. The first approach makes better use of the underlying ATM fabric, but data sources that are also members of the group will receive copies of their own traffic back. The alternative avoids this problem, but at the expense of consuming more VCs and bandwidth on the path out of the multicast server itself. Armitage Expires August 4th, 1995 [Page 35] Internet Draft February 4th, 1995 Situations where either issue is a problem should simply revert to using a multicast mesh between participating endpoints, where the source never sees copies of its own packets, and the multicasting happens within the ATM switch fabrics. Armitage Expires August 4th, 1995 [Page 36]