IPS Working Group M. Rajagopal, R. Bhagwat, LightSand Communications INTERNET-DRAFT E. Rodriguez, Lucent Technologies <draft-ietf-ips-fcovertcpip-00.txt> V. Chau, Gadzoox Networks (Expires March 31, 2001) S. Berman, Vixel S. Wilson, Brocade Communications M. O'Donnell, McDATA C. Carlson, QLogic Fibre Channel Over TCP/IP (FCIP) Status of this Memo This document is an Internet-Draft and is in full conformance with all provisions of Section 10 of RFC 2026 [1]. 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 monthsx and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as Reference material or to cite them other than as ``work in progress''. The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/lid-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html 1. Abstract Fibre Channel (FC) is a dominant technology used in Storage Area Networks (SAN). The purpose of this draft is to specify a standard standard way of encapsulating FC frames over TCP/IP and to describe mechanisms that allow islands of FC SANs to be interconnected over IP-based networks. FC over TCP/IP relies on IP-based network services to provide the connectivity between the SAN islands over LANs, MANs, or WANs. The FC over TCP/IP specification relies upon TCP for congestion control and management and upon both TCP and FC for data error and data loss recovery. FC over TCP/IP treats all classes of FC frames the same -- as datagrams. 2. Conventions used in this document The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [2]. Rajagopal, et al. [Page 1]
Internet-Draft Fibre Channel over TCP/IP October 2000 3. Motivation and Objectives Fibre Channel (FC) is a gigabit speed networking technology primarily used for Storage Area Networking (SAN). Fibre Channel is standardized under American National Standard for Information Systems of the National Committee for Information Technology Standards (ANSI-NCITS) and has specified a number of documents describing its protocols, operations, and services [13]. The motivation behind connecting remote sites include disk or tape backup and live mirroring, or simply distance extension between two FC devices or FC Switch clusters (SAN islands). A fundamental assumption made in this specification is that the Fibre Channel traffic is carried over the IP network in such a manner that the Fibre Channel fabric and all Fibre Channel devices on the fabric are unaware of the fact. This means that the FC datagrams must be delivered in such time as to comply with existing Fibre Channel specifications. The FC traffic may span LANs, MANs and WANs, so long as this fundamental assumption is adhered to. Fibre Channel operates at Gigabit speeds. This specification is written such that Fibre Channel traffic may be transported over an IP backbone that has been engineered to have equivalent or better bit-error-rate (BER) and line speed as the Fibre Channel environments being bridged. While the tunneling of Fibre Channel traffic over other IP networks not so engineered is not precluded, the above environment is an important one, and this specification has been written so that to optimize for such traffic, while not over-burdening other configured IP networks. The main objectives of this document are to: 1) specify the TCP encapsulation and mapping of FC frames 2) apply the mechanism described in (1) to an FC network using IP as a backbone, or more generally, between any two FC devices. 3) address any FC concerns arising from tunneling FC traffic over an IP network, including security, data integrity (loss), congestion, and performance. This will be accomplished, where appropriate, by utilizing the existing IETF-specified suite of protocols. 4) be backwards compatible with existing FC specifications. While new work may be undertaken in T11[13] to optimize and enhance the bridging of FC networks and fabrics, this specification will not require adherence to such future works. Rajagopal, et al. [Page 2]
Internet-Draft Fibre Channel over TCP/IP October 2000 4. FCIP Protocol 4.1 FCIP Device In this specification, the term FCIP device generally refers to any device that encapsulates FC frames into TCP segments and re- assembles TCP segments to regenerate FC frames. Note: In an actual implementation, the FCIP device may be a stand- alone box or integrated with an FC device such as an FC back- bone switch (BBW) or integrated with any TCP/IP device such as an IP switch or an IP router. The FCIP device is a transparent translation point. The IP network is not aware of the FC payload that it is carrying. Likewise, the FC fabric and the FC end nodes are unaware of the IP-based transport. 4.2 Protocol The FCIP protocol specifies the TCP/IP encapsulation, mapping and routing of FC frames and applies these mechanisms to an FC network utilizing IP for its backbone, or more generally, between any two FC devices. The FCIP protocol is summarized below: 1. All FCIP protocol devices are peers and communicate using TCP/IP. Each FCIP device behaves like a TCP end-node from the perspective of the IP-based network. That is, these devices do not perform IP routing or IP switching but simply forward FC frames. 2. There is no requirement for an FCIP device to establish a login with a peer before communication begins. However, FCIP devices may authenticate the IP packet before accepting it using the IPSec protocols. Each IP datagram is treated independently and an FCIP device receiver simply listens to the appropriate socket value contained in the TCP header. 3. Each FCIP device may be statically or dynamically configured with a list of IP addresses corresponding to all the participating FCIP devices. Dynamic discovery of participating FCIP devices may be performed using Internet protocols such as LDAP, DHCP or other discovery protocols. It is outside the scope of this specification to describe any static or dynamic scheme for participating FCIP device IP address discovery. 4. Discovery of FC addresses (accessible via the FCIP device) is provided by techniques and protocols within the FC architecture. These techniques and protocols are described in Fibre Channel ANSI standards ([3], [7], [15]). The FCIP device does not participate in the discovery of FC addresses. Routing in the IP plane and the FC plane are largely independent. Rajagopal, et al. [Page 3]
Internet-Draft Fibre Channel over TCP/IP October 2000 5. The exact path (route) taken by an FC over TCP/IP encapsulated packet follows the normal procedures of routing any IP packet. From the perspective of the FCIP devices this communication is between only two FCIP devices for any given packet. 6. An FCIP device may send FC encapsulated TCP/IP packets to more than one FCIP device. However, these encapsulated packets are treated as separate instances and are not correlated in any way by the FCIP protocol devices. The source FCIP device routes its packets based on the 3-byte FC destination Address Identifier (D_ID) contained in each FC frame. 7. An IP packet may make use of the IPSec protocol to provide secure communications across the IP-based network. 8. Any re-ordering of data link frames due to MTU fragmentation will be recovered in accordance with a normal TCP reliable delivery behavior. Any re-ordering of FC frames due to IP packet re-ordering will be resolved via the standard TCP reliable delivery behavior. 9. FCIP relies on both TCP error recovery mechanism and normal FC recovery mechanisms to detect and recover from data loss due to any loss of IP packets. 10. FC over TCP/IP encapsulated IP packets shall indicate the use of the Premium Service in the DSCP bits in the IP header. 11. The TCP layer in the sending FCIP device shall package each FC frame handed down by the FC layer into a TCP segment and set the PSH control flag in the TCP header to ensure that the entire FC frame is sent in one TCP segment. If the FC frame cannot be packaged in one TCP segment (e.g. the FC frame size is greater than TCP MSS), the last part of the FC frame must occupy one TCP segment and the PSH of that segment must be set. 5. FCIP Encapsulation 5.1 FC Frame Format All FC frames have a standard format much like LAN's 802.x protocols. However, the exact size of each frame varies depending on the sizes of the variable fields. The sizes of the variable fields range from 0 to 2112-bytes as shown in the FC Frame Format in Fig. 1 resulting in the minimum size FC Frame of 36 bytes and the maximum size FC frame of 2148 bytes. Rajagopal, et al. [Page 4]
Internet-Draft Fibre Channel over TCP/IP October 2000 +------+--------+-----------+----//-------+------+------+ | SOF |Frame |Optional | Frame | CRC | EOF | | (4B) |Header |Header | Payload | (4B) | (4B) | | |(24B) |<----------------------->| | | | | | Data Field = (0-2112B) | | | +------+--------+-----------+----//-------+------+------+ Fig. 1 FC Frame Format Start Of Frame (SOF) and End Of Frame (EOF) Delimiters: On an FC link, Start-of-Frame (SOF) and End-Of-Frame (EOF) are called Ordered Sets and are sent as special words constructed from the 10-bit comma character (K28.5) followed by three additional 10-bit data characters. On a non-Fibre Channel link, the SOF and EOF delimiters are both byte-encoded and 4-bytes long. On an FC link the SOF delimiter serves to identify the beginning of a frame and prepares the receiver for frame reception. The SOF contains information about the frame's Class of Service, position within a sequence, and in some cases, connection status. The EOF delimiter identifies the end of the frame and the final frame of a sequence. In addition, it serves to force the running disparity to negative. The EOF is used to end the connection in connection-oriented classes of service. It is therefore important to preserve the information conveyed by the delimiters across the IP-based network, so that the receiving FCIP device can correctly reconstruct the FC frame in its original SOF and EOF format before forwarding it to its ultimate FC destination on the FC link. Start of Frame (SOF) and End of Frame (EOF) byte-encodings are defined in Annex A. Although the SOF and EOF codes are 32-bits, the format makes use of a single-byte to represent each FC Ordered Set. The remaining three bytes associated with these delimiters in the TCP encapsulated FC frames do not necessarily carry any meaningful information. Frame Header: The Frame Header is 24-bytes long and has several fields that are associated with the identification and control of the payload. Current FC Standards allow up to three Optional Header fields [4],[5]: - Network_Header (16-bytes) - Association_Header (32-bytes) - Device_Header (up to 64-bytes). Rajagopal, et al. [Page 5]
Internet-Draft Fibre Channel over TCP/IP October 2000 Frame Payload: The FC Frame Payload is transparent to the FCIP device. An FC application level payload is called an Information Unit at the FC-4 Level. This is mapped into the Frame Payload of the FC Frame. A large Information Unit is segmented using a structure consisting of FC Sequences. Typically, a Sequence consists of more than one FC frames. FCIP does not maintain any state information regarding the relationship of frames within an FC Sequence. CRC: The CRC is 4-bytes long and uses the same 32-bit polynomial used in FDDI and is specified in ANSI X3.139 Fiber Distributed Data Interface. Note: When FC frames are encapsulated into TCP segments, the CRC is untouched. 5.2 FC Frame Mapping to TCP/IP packets Fig.2 shows the mapping of the FC frame into an TCP/IPv4 Packet. The FC to TCP/IP mapping (and the reverse) mapping is one-to-one since the maximum size of the encapsulated FC Frame along with the header fields does not exceed 2148 bytes. The minimum size FC Frame is 36 bytes resulting in a maximally minimum IP MTU size of 156 bytes. (The maximally minimum MTU size is the IP packet with the minimum size payload and the maximum size IP headers and TCP headers). The maximum size FC frame is 2148 bytes resulting in a nominal IP packet size of 2188 bytes. Fig. 2 shows the format of the IPv4 packet with the standard 20-byte IP fixed header, a 40-byte optional IP header, the standard 20-byte TCP fixed header and TCP options. For the case of the maximum size payload of 2148 bytes, the maximum TCP/IPv4 packet size is 2268 bytes. The maximum size FC frame can cause the IP packet to be fragmented when the data link cannot support this MTU size. Furthermore, TCP may also segment the data if the maximum segment size (MSS) is less than the size of the frame. When an IP packet is fragmented, the required parts of the header must be copied by all fragments and the option field may or may not be copied. When a TCP packet is segmented, normal TCP operations are responsible for receiving and re-sequencing the data prior to passing it on to the FC processing portion of the device. Rajagopal, et al. [Page 6]
Internet-Draft Fibre Channel over TCP/IP October 2000 +-----------+--------------+----------+-----------+-----------+ | IP Header |IP Opt. Header|TCP Header|TCP Options| FC Frame | |(20 bytes) | (40 bytes |(20 bytes)| (40 bytes |(2148 bytes| | | Max) | | Max) | Max) | +-----------+--------------+----------+-----------+-----------+ Fig. 2 Format of an IPv4 Payload carrying FC <Placeholder for IPv6 format> If IPSec is used for security it introduces its own headers and the IP packet size increase depends on the exact mode of IPSec usage (AH versus ESP, Tunnel versus Transport). However, this additional increase in the IP packet size due to IPSec headers is relatively small (see [8], [9], [10]), and the maximum size IP packet will remain within its maximum size of 65535 bytes. Adding IPSec header may, in some cases, lead to fragmentation if it exceeds the data link MTU. IP Header Field Setting: DSCP (6 bits): The Differentiated Service Code Points (DSCP) [6] shall be set to correspond to the Premium Service. This service provides "Expedited Forwarding" at each IP hop (Per Hop Behavior (PHB)). Protocol (8 bits): This 8-bit field defines the higher level protocol that uses the service of the IP layer. In this case, this is set to the TCP Protocol Value 6 as defined in [12]. Source IP Address (32 bits): This is the IP address of the ingress FCIP device that is transmitting the IP packet containing the FC frame. Destination IP Address (32 bits): This is the IP address of the egress FCIP device that is receiving the IP packet containing the FC frame. The FCIP specification treats all classes of FC frames the same - as datagrams. Since FC Primitives and Primitive Sequences are not exchanged between FCIP devices there may be times when a frame is lost in the IP network. When this event occurs it is the responsibility of the communicating Fibre Channel devices to detect and correct the errors. The FCIP devices shall not generate Fibre Channel F_BSY and/or F_RJT frames or otherwise participate in FC frame recovery. FCIP may not be suitable for the transport of Class 1 traffic since these frames are treated the same way as any Class 2 or 3 frame by the FCIP devices. Rajagopal, et al. [Page 7]
Internet-Draft Fibre Channel over TCP/IP October 2000 5.3 FC frame mapping to TCP Segment The FC-to-TCP mapping (and reverse mapping) may not necessarily be one-to-one as the FC frame size may exceed the TCP Maximum Segment Size (MSS). In this case, the TCP layer in the sending FCIP device may segment the FC frame into multiple TCP segments. The sending device must ensure that the last TCP segment contains only the last part of the encapsulated FC frame and that last TCP segment does not contain the beginning of another FC frame. The last TCP segment must also have the PSH flag set so that the receiving FCIP device knows to send the frame to the FC layer above it. The fields in the TCP header are explained in the following paragraphs: Source Port: The port number selected by the sending FCIP device. Destination Port: The well-known port number assigned for FC encapsulation. This number is assigned by ... <to be added> Sequence Number: The sequence number of the SOF byte-code of the FC frame or the first byte of the current TCP segment payload. Acknowledge Number: If the ACK flag is set, this field contains the sequence number of the next byte that the receiving TCP expects to receive. Data Offset: The number of 32-bit words in the TCP header. The use of TCP options are not precluded for FCIP devices. Control (flag) bits: URG: Urgent Pointer field significant - must be 0; the use of urgent data is not allowed for segments containing encapsulated FC frames. ACK: Acknowledgment field significant - FCIP devices use this bit to indicate that the Acknowledge number is valid. PSH: Push Function - The sending FCIP device shall set this bit to 1 if the entire FC frame fits inside a TCP Rajagopal, et al. [Page 8]
Internet-Draft Fibre Channel over TCP/IP October 2000 segment or the last portion of an FC frame (containing the EOF) is sent in its own TCP segment. RST: Reset the connection - the FCIP device shall set this flag according to the state of the TCP connection and in compliance to the TCP specifications. SYN: Synchronize sequence numbers - the FCIP device shall set this flag according to the state of the TCP connection and in compliance to the TCP specifications. FIN: No more data from sender - the FCIP device shall set this flag according to the state of the TCP connection and in compliance to the TCP specifications. Window: The FCIP device shall operate this field according to the state of the TCP connection and in compliance to the TCP specifications. Checksum: The FCIP device shall calculate the checksum over the entire FC frame or the portion of the FC frame encapsulated inside the current segment. The checksum shall also include the TCP header and the pseudo IP header as specified in [16]. Urgent Pointer: The FCIP devices shall not use this field as the urgent function is not allowed in tunneling FC frames over the IP network. Options and Padding: FCIP devices are allowed to use TCP options and padding as specified by the TCP specifications [16]. 5.4 Fibre Channel Bit and Byte Ordering Fibre Channel's FC-1 Level defines the method used to encode data prior to transmission and subsequently decode the data upon reception. The method encodes 8-bit bytes into 10-bit characters to improve the transmission characteristics of the serial data stream. In Fibre Channel, data fields are aligned on word boundaries. A word in FC is defined as 4 bytes or 32 bits. The resulting transmission word after the 8-bit to 10-bit encoding consists of 40 bits. Data words or Ordered Sets (special FC-2 Level control words) from the FC-2 Level map to the FC-1 Level with no change in order Rajagopal, et al. [Page 9]
Internet-Draft Fibre Channel over TCP/IP October 2000 and the bytes in the word are transmitted in the Most Significant Byte first to Least Significant Byte order. The transmission order of bits in each byte is from the Least Significant Bit to the Most Significant Bit. 6. FCIP Network 6.1 FC Backbone Switches Fibre Channel Standards [3] describe the operation of and interactio betweenn FC Switches. Two distinct levels of switch interconnections are specified. Autonomous Regions (AR) are defined to allow clusters of FC Switches to be connected across a backbone network called an FSPF-backbone. An Autonomous Region is administratively defined with each AR encompassing one or more FC Domains. The FSPF-backbone network is formed from one or more Backbone Switches (BSW) that run the FSPF-backbone routing protocol. The FSPF-backbone routing protocol is based on OSPF and the FSPF backbone may consist of an arbitrary mesh network. A BSW may communicate with multiple neighbors. As specified in [3], native FC frames traverse the backbone between BSW neighbors. The FSPF-backbone Routing Protocol messages are exchanged between BSWs on the FSPF backbone. An example network consisting of 4 ARs and an FSPF backbone consisting of 3 links is given in Fig. 3. There is no restriction in adding other links to this network as needed. The connection between BSWs below may in fact form a fully connected mesh. _______ _______ | | | | | AR #1 |_____ _____| AR #4 | |_______| | | |_______| ___|___ ___|___ | BSW 1 |---------------------| BSW 4 | |_______| |_______| ___|___ _______ | BSW 2 |---------------------| BSW 3 | |_______| |_______| ___ ___ | | _______ | | | | | | | AR #2 |----- -----| AR #3 | |_______| |_______| Fig. 3 Example Network showing FSPF-Backbone Switching Architecture Note: BSW 1 knows it is connected to BSWs 2 and 4; BSW 2 knows it is connected to BSWs 1 and 3; BSW 4 knows it is connected to BSWs 1. Rajagopal, et al. [Page 10]
Internet-Draft Fibre Channel over TCP/IP October 2000 An FCIP device provides a single, logical interface to the FSPF- backbone protocol connecting multiple BSW neighbors on the IP- network. From the FSPF-backbone routing's point of view, the connection to each neighbor on the IP-network is treated as a separate logical FC link. In FCIP, the native FC frames are first encapsulated in TCP segments which then traverse the IP-based network. The IP network provides a new transport path for each emulated FSPF-backbone link. The IP network itself may consist of any number of hops between two FCIP devices. Also, the route taken by the IP packet between any two FCIP devices is dictated by normal IP routing. A functional and logical diagram of an IP-based FSPF-backbone for the example network given in Fig. 3 is shown in Fig. 4. In this figure, each BSW is logically connected to other BSWs. _______ _______ | | | | | AR #1 |--- | AR #4 | |_______| | ______ ________ ______ |_______| __|_ __ | | | | | | ___|___ | BSW 1 |---| FCIP |--| IP |--| FCIP |--| BSW 4 | |_______| |______| | Network| |______| |_______| | | -------- ______ | | ______ ______ | | | | | | _______ | BSW 2|---| FCIP |-----| |---| FCIP |---| BSW 3 | |______| |______| |______| |_______| ________ | ___|___ | | | | | | AR #2 |__| | AR #3 | |________| |_______| Fig. 4 Example Network showing an IP-based FC Backbone Switching Architecture The IP-based network has transformed the FSPF-backbone into a fully connected network. From the perspective of each BSW all remote BSWs therefore appear to be neighbors. The FSPF-backbone routing protocol computations would make the IP-based network topology appear as a fully connected mesh. The FSPF-backbone routing protocol exchanges between BSWs occur transparently to the FCIP devices. Encapsulated FC frames are routed on the IP network according to the normal IP routing Rajagopal, et al. [Page 11]
Internet-Draft Fibre Channel over TCP/IP October 2000 procedures. In this mode, the FSPF-backbone routing protocol lays over the IP network and has no knowledge of the underlying IP protocol and IP routing or the underlying technology that carries the IP datagram. This concept is shown in Fig.5 ________ _______ | AR #1 | | AR #2 | | |-- | | |________| | ______ ________ _____ |_______| __|___ | | | | | | ____|__ | BSW 1|---| FCIP |--| IP |--|FCIP |--| BSW 2 | | | | | | Network| | | | | |______| |______| |________| |_____| |_______| <--------------> IP Routing <----------------------------------> FSPF-backbone Routing Plane Fig. 5 FC packet routing over IP based backbone network Note: IP Network routing may consist of multiple paths 6.2 FC Device The protocol encapsulation and mapping of the FC frame described in earlier sections applies equally to any pair of FC devices (e.g. switch to switch or host to storage subsystem) wishing to tunnel FC frames across an IP-based network. Any FC routing protocol exchanges may still occur transparent to the FCIP devices. It should be noted that Fibre Channel Primitive Sequences and Primitives are not exchanged between FCIP devices. 7. Security Considerations Using a wide-area, general purpose network such as an IP internet in a position normally occupied by physical cabling introduces some security problems not normally encountered in Fibre Channel storage networks. Normal media are typically protected physically from outside access; IP internets typically invite outside accesses. The general effect is that the security of the entire Fibre Channel internetwork is only as good as the security of the entire IP internet through which it tunnels. The following broad classes of attacks are possible: Rajagopal, et al. [Page 12]
Internet-Draft Fibre Channel over TCP/IP October 2000 1. Unauthorized Fibre Channel controllers can gain access to resources through normal Fibre Channel processes. 2. Unauthorized agents can monitor and manipulate Fibre Channel traffic flowing over physical media used by the IP internet and under control of the agent. To a large extent, these security risks are typical of the risks facing any other application using an IP internet. They are mentioned here only because Fibre Channel storage networks are not normally suspicious of the media. Fibre Channel storage network administrators will need to be aware of these additional security risks. Security protocols and procedures used in other IP applications may also be used for FC over TCP/IP. For Virtual Private Networks, both authentication and encryption are generally desired, because it is important both to (1) assure that unauthorized users do not penetrate the virtual private net- network and (2) assure that eavesdroppers on the network cannot read messages sent over the network. IPSec provides 3 main facilities: an authentication-only function referred to as Authentication Header (AH) , a combined authentication/encryption function called Encapsulating Security Payload (ESP), and a key exchange function. Because both features are generally desirable, ESP may be more suitable than AH. The key exchange function allows for manual exchange of keys as well as an automated scheme. The IPSec specifications described in [8], [9], [10], and [11] covers these topics. It is beyond the scope of this document to discuss specific uses of the IPSec protocols. Note: Use of IPSec protocol is optional. 8. Data Integrity Considerations 8.1 Loss Recovery from data loss due to IP datagram loss is provided via the TCP reliable delivery mechanism. Note: Due to varying TCP timeouts, competing FC and TCP recovery schemes is a possibility. This issue is addressed in section 8.4. 8.2 Corruption Data corruption is detected at two different levels: TCP checksum and Fibre Channel CRC. Data corruption detected at the TCP level shall be recovered via TCP reliable data recovery mechanisms. Rajagopal, et al. [Page 13]
Internet-Draft Fibre Channel over TCP/IP October 2000 Data corruption detected at the Fibre Channel level shall be handled within the Fibre Channel end nodes. 8.3 Timeouts Fibre Channel has two important timeouts to be consider in FCIP. These are: ED_TOV, and R_A_TOV. ED_TOV determines the life of an individual Fibre Channel frame in any particular fabric element. The effects of ED_TOV on the fabric as a whole are typically cumulative since each fabric element contains it's own ED_TOV timers for any frame received. R_A_TOV determines the life of an individual Fibre Channel frame in the fabric as a whole. For a fabric, R_A_TOV implies that no particular frame will remain in (and thus be emitted from) the fabric after the timer expires. TCP has the TCP acknowledgement timeout. This is a variable timeout value. <TBD: Need to elaborate on TCP timeouts and define how Fibre Channel timeouts map to TCP timeouts.> 8.4 Recovery Mechanisms When an FC over TCP/IP encapsulated frame is transported over an IP network, there is a possibility of the frame getting dropped. This can happen if there is congestion along the path within the IP network, if there are no empty buffers available on one of the incoming ports, due to bit errors, etc. When this happens, the TCP acknowledgement will not be received by the source, and normal TCP retry mechanisms will be activated. An issue may arise during these recovery mechanisms, since TCP timeout is variable, and may exceed Fibre Channel FC ED_TOV/ R_A_TOV timeouts. <Placeholder: Need to add details on how to handle this here.> 8.5 IP Fragmentation: The Fibre Channel maximum transmittable unit (MTU) is 2148 bytes. A maximum of 60 bytes of TCP header increases the MTU of the IP payload to 2208 bytes. It is preferable that FCIP packets encompass the TCP + FC MTU to avoid fragmentation of FCIP packets. The resulting packet size exceeds the MTU of some IP physical layers (e.g. Ethernet MTU = 1518 bytes). FCIP devices should handle fragmentation and must handle re-assembly of FCIP packets. An FCIP device may use Path MTU Discovery (RFC 1191) or an equivalent mechanism to adjust FCIP packet size to avoid Rajagopal, et al. [Page 14]
Internet-Draft Fibre Channel over TCP/IP October 2000 fragmentation. Alternatively, the MTUs of all FC nodes may be manually set to match the path MTU of the IP network. 8.6 TCP Segmentation: 8.7 Ordering: In an IP network, packets are not guaranteed to be delivered in the same order in which they were transmitted. But, TCP does handle ordering. This means that frames will be delivered in the order that they were transmitted by the FCIP device. Note, this does not necessarily mean that the frames will be transmitted by the FCIP device in the order transmitted from the source end FC device, or received by the destination end FC device in the order received by the FCIP device. This is due to the fact that FC frames may be reordered in the FC network/fabric itself. 9. Performance Considerations Mapping the IP header DSCP bits to correspond to Premium Service provides a preferred service at each IP Router Per Hop Behavior (PHB) [6]. The FCIP protocol does not necessitate the "cracking" of the FC Frame (except for attaching the correct byte-encoded SOF and EOF) nor does it call for any FC payload processing. This allows any FC traffic to be tunneled across at high throughput rates. In the case where there is no data link fragmentation, each FC frame has a one-to-one mapping to a TCP segment; the TCP segment has a one-to-one mapping to an IP dategram; and the IP datagram has a one-to-one mapping to a data link frame. This has the tendency to further improve the throughput performance. Note: Class 1 FC traffic expects dedicated bandwidth. This specification does not address this requirement. The Flow Control Protocol (discussed in the next section) provides the ability to stream gigabit FC data when using a large window size. 10. Flow Control and Congestion Management <Text to be added> 11. References: [1] Bradner, S., "The Internet Standards Process -- Revision 3", BCP 9, RFC 2026, October 1996. Rajagopal, et al. [Page 15]
Internet-Draft Fibre Channel over TCP/IP October 2000 [2] Bradner, S., "Key words for use in RFCs to Indicate RequirementLevels", BCP 14, RFC 2119, March 1997 [3] NCITS 321-200x (ANSI) T11/Project 1305-D/Rev 4.7 "Fibre Channel Switch-Fabric-2", (FC-SW-2) September 29, 2000 (www.t11.org) [4] Fibre Channel Physical and Signaling Interface-3 (FC-PH-3), Rev. 9.4, ANSI X3.303-1998 [5] The Fibre Channel Consultant: A Comprehensive Introduction, "Robert W. Kembel", Northwest Learning Associates, 1998 [6] Nichols, K., Blake, S., Baker, F. and D. Black, " Definition of the Differentiated Services Field (DS Field) in the IPv4 and Ipv6 Headers", RFC 2474, December 1998. [7] NCITS T11/Project 1238-D/Rev4.7 "Fibre Channel Backbone", (FC- BB) June 8, 2000 (www.t11.org) [8] Kent, S. and Atkinson, R., "Security Architecture for the Internet Protocol", RFC 2401, Nov 1998 [9] Kent, S. and Atkinson, R., "IP Authentication Header", RFC 2402, Nov 1998 [10] Kent, S. and Atkinson, R., "IP Encapsulating Security Payload (ESP)", RFC 2406, Nov 1998 [11] Maughan, D. et all, "Internet Security Association and Key Management Protocol (ISAKMP)", RFC 2408, Nov 1998 [12] http://www.isi.edu/in-notes/iana/assignments/protocol-numbers [13] http://www.t11.org [14] Fibre Channel Physical and Signaling Interphace (FC-PH), Rev 4.3, ANSI X3.230-1994. [15] Fibre Channel NCITS 321-200x (ANSI) T11/Project 1356- D/Rev4.3 " Fibre Channel - Generic Services 3", June 2000 (www.t11.org)). [16] ISI, "Transmission Control Protocol", RFC 793, Sep 1981 12. Acknowledgments Rajagopal, et al. [Page 16]
Internet-Draft Fibre Channel over TCP/IP October 2000 13. Authors' Addresses Murali Rajagopal LightSand Communications, Inc. 375 Los Coches St. Milpitas, CA 95035 Phone: +1 408 404 3164 Email: muralir@lightsand.com Raj Bhagwat LightSand Communications, Inc. 375 Los Coches St. Milpitas, CA 95035 Phone: +1 408 404 3194 Fax: Email: rajb@lightsand.com Elizabeth G. Rodriguez Lucent Technologies 1202 Richardson Drive, Suite 210 Richardson, TX 75080 Phone: +1 972 231 0672 Fax: +1 972 671 5476 Email: egrodriguez@lucent.com Vi Chau Gadzoox Networks, Inc. 16241 Laguna Canyon Road, Suite 100 Irvine, CA 92618 Phone: +1 949 789 4639 Fax: +1 949 453 1271 Email: vchau@gadzoox.com Stuart Berman Vixel Corporation 15245 Alton Parkway Suite 100 Irvine, CA 92618 Phone: +1 949 450 6100 Fax: +1 949 753 9500 Email: sberman@vixel.com Steve Wilson Brocade Communications Systems, Inc. 1745 Technology Drive San Jose, CA. 95110 Rajagopal, et al. [Page 17]
Internet-Draft Fibre Channel over TCP/IP October 2000 Phone: 408-487-8128 Fax: 408-487-8101 email: swilson@brocade.com Craig W. Carlson QLogic Corporation 6321 Bury Drive Eden Prairie, MN 55346 Phone: +1 952 932 4064 Email: craig.carlson@qlogic.com Michael E. O'Donnell McDATA Corporation 310 Interlocken Parkway Broomfield, Co. 80021 Phone: +1 303 460 4142 Fax: +1 303 465 4996 Email: modonnell@mcdata.com ANNEX A: Fibre Channel EOF and SOF Encodings A.1 Ordered Sets On a Fibre Channel link, Ordered Sets (OS) are sent as special out-of-band words constructed from a combination of the 10-bit comma character (K28.5) followed by 3 additional 10-bit data characters. The Ordered Sets defined by Fibre Channel include the Frame Delimiters, Start of Frame (SOF) and End of Frame (EOF), other Primitive Signals (IDLE, R_RDY, ARB, OPN, CLS, MRK), and Primitive Sequences (OLS,NOS, LR, LRR, LIP, LPB, LPE). When Fibre Chanel frames are encapsulated in an IP packet, the Byte-encoded frame format is used. The Byte-encoded frame format uses 32-bit OS Code Words to represent valid Fibre channel frame delimiters. This format uses a single-byte OS Code to represent each FC Ordered Set. FC Over IP makes use of the OS Codes defined in Annex A of [7] for the frame delimiters. SOF and EOF codes defined in the figures (see below) in this Annex are inserted into the FC frame. Primitive Signals and Primitive Sequences are stripped at the FCIP boundary and not sent between FCIP devices over the IP network. Rajagopal, et al. [Page 18]
Internet-Draft Fibre Channel over TCP/IP October 2000 The frame delimiters are identified by their position. An encapsulated Byte-encoded frame must use the corresponding 32-bit OS Code Word as the first and last words in the encapsulated PDU. Fibre Channel frame delimiters shall be encoded in the format shown in Table below. Table 1. Frame Delimiter Format +---+----------------+----------------+----------------+-------------+ |Wrd| <31:24> | <23:16> | <15:08> | <07:00> | +---+----------------+----------------+----------------+-------------+ |0 | OS Code | Reserved | +---+----------------+----------------+----------------+-------------+ A.2 Encoded FC Frame Delimiters The SOF OS-codes are a single byte encoding of the SOF Ordered Set. The first word in an encapsulated Byte-encoded FC frame shall map the SOF Ordered Set to the corresponding 32-bit OS Code Word. The EOF OS-codes are a single byte encoding of the EOF Ordered Set. The last word in an encapsulated Byte-encoded FC frame shall map the the EOF Ordered Set to the corresponding 32-bit OS Code Word. +-----------------+----------------+ | OS-Code | Delimiter Name | | (hex) | | +-----------------+----------------+ | 0x28 | SOFf | +-----------------+----------------+ | 0x3F | SOFc1 | +-----------------+----------------+ | 0x2F | SOFi1 | +-----------------+----------------+ | 0x37 | SOFn1 | +-----------------+----------------+ | 0x3D | SOFc2 | +-----------------+----------------+ | 0x2D | SOFi2 | +-----------------+----------------+ | 0x35 | SOFn2 | +-----------------+----------------+ | 0x3E | SOFc3 | +-----------------+----------------+ | 0x2E | SOFi3 | +-----------------+----------------+ | 0x36 | SOFn3 | +-----------------+----------------+ | 0x39 | SOFc4 | +-----------------+----------------+ | 0x29 | SOFi4 | Rajagopal, et al. [Page 19]
Internet-Draft Fibre Channel over TCP/IP October 2000 +-----------------+----------------+ | 0x31 | SOFn4 | +-----------------+----------------+ | 0x38 | SOFcf | +-----------------+----------------+ | 0x30 | SOFnf | +-----------------+----------------+ | 0x41 | EOFn | +-----------------+----------------+ | 0x42 | EOFt | +-----------------+----------------+ | 0x46 | EOFdt | +-----------------+----------------+ | 0x44 | EOFrt | +-----------------+----------------+ | 0x49 | EOFni | +-----------------+----------------+ | 0x4E | EOFdti | +-----------------+----------------+ | 0x4F | EOFrti | +-----------------+----------------+ | 0x50 | EOFa | +-----------------+----------------+ ANNEX B: Relationship between FC over IP (FCIP, FCoverIP) and IP over FC (IPFC) IPFC describes the encapsulation of IP packets in FC frames. It is intended to facilitate IP communication over an FC network. FC over IP describes the encapsulation of FC frames in TCP segments which in turn are encapsulated inside IP packets for transporting over an IP network. It gives no consideration to the type of FC frame that is being encapsulated. Therefore, the FC frame may actually contain an IP packet as described in the IP over FC specification (RFC 2625). In such a case, the encapsulated IP packet would have: IP Header FC Header IP Header TCP header Note: The two IP headers would not be identical to each other. One would have information pertaining to the final destination while the other would have information pertaining to the FCIP device. The two documents focus on different objectives. As mentioned above, implementation of FC over IP will lead to IP encapsulation Rajagopal, et al. [Page 20]
Internet-Draft Fibre Channel over TCP/IP October 2000 within IP. While perhaps inefficient, this should not lead to issues with IP communication. One caveat: if a Fibre Channel device is encapsulating IP packets in an FC frame (e.g. an IPFC device), and that device is communicating with a device running IP over a non-FC medium, a second IPFC device will need to act as a gateway between the two networks. This scenario is not specifically addressed by FC over IP. There is nothing in either specification to prevent a single device from implementing both FC-over-IP and IP-over-FC, but this is implementation specific, and is beyond the scope of this document. Full Copyright Statement Copyright (C) The Internet Society (1999). All Rights Reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to the Internet Society or other Internet organizations, except as needed for the purpose of developing Internet standards in which case the procedures for copyrights defined in the Internet Standards process must be followed, or as required to translate it into languages other than English. The limited permissions granted above are perpetual and will not be revoked by the Internet Society or its successors or assigns. This document and the information contained herein is provided on an "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Acknowledgement Funding for the RFC Editor function is currently provided by the Internet Society. [draft-ietf-ips-fcovertcpip-00.txt] [This INTERNET DRAFT expires on March 31, 2001] Rajagopal, et al. [Page 21]
