Inserting, Processing And Deleting IPv6 Extension Headers

In IPv6 optional internet-layer information is encoded in extension headers. As specified by , "extension headers (except for the Hop-by-Hop Options header) are not processed, inserted, or deleted by any node along a packet's delivery path, until the packet reaches the node (or each of the set of nodes, in the case of multicast) identified in the Destination Address field of the IPv6 header". The statement quoted above identifies nodes upon which extension headers are not processed, inserted or deleted. It does not imply that extension headers can be processed, inserted or deleted on any other node along a packet's delivery path. This document provides guidance regarding the processing, insertion and deletion of IPv6 extension headers. It clarifies the statement quoted above and updates .

The following terms are used in this document: Source node - An IPv6 source node accepts data from an upper-layer protocol, encapsulates it in an IPv6 header, and sends the resulting IPv6 packet to a destination node. Destination node - An IPv6 destination node receives an IPv6 packet and delivers its payload to an upper-layer protocol. Delivery path - A packet's delivery path is a series of nodes that a packet traverses on route to its destination. The delivery path includes the destination node. Segment - A segment is a series of links and nodes in a packet's delivery path. The IPv6 Routing header steers packets from segment to segment along the delivery path. If a packet contains a Routing header, its delivery path can contain multiple segments. If a packet does not contain a Routing header, its delivery path contains only one segment. Segment egress node - A segment egress node terminates a segment. When a packet arrives at a segment egress node, its IPv6 Destination Address identifies a resource that belongs to the node. All destination nodes are also segment egress nodes.

The terms defined in of this document should be added to Section 2 of . of this document quotes text from . That text should be replaced with the text contained by of this document.

"Extension headers (except for the Hop-by-Hop Options header) are not processed, inserted, or deleted by any node along a packet's delivery path, until the packet reaches the node (or each of the set of nodes, in the case of multicast) identified in the Destination Address field of the IPv6 header. The Hop-by-Hop Options header is not inserted or deleted, but may be examined or processed by any node along a packet's delivery path, until the packet reaches the node (or each of the set of nodes, in the case of multicast) identified in the Destination Address field of the IPv6 header. The Hop-by-Hop Options header, when present, must immediately follow the IPv6 header. Its presence is indicated by the value zero in the Next Header field of the IPv6 header."

Source nodes can send packets that include extension headers. Extension headers are not inserted by subsequent nodes along a packet's delivery path. The Hop-by-Hop Options header can be processed by any node in a packet’s delivery path. The Destination Options header and Routing header can be processed by any segment egress node, including the destination node. The Fragment header can be processed by the destination node only. If a packet includes a Fragment header, the Authentication header and Encapsulating Security Payload header can be processed by the destination node only. However, if the packet does not include a Fragment header, the Authentication header and Encapsulating Security Payload header can be processed by any segment egress node, including the destination node. Except for the following fields, extension headers are not modified by nodes along a packet's delivery path: The Segments Left field in the Routing header. Type-specific data in the Routing header. Option Data in the Destination Options header. Extension headers are not deleted by any node along a packet's delivery path, until the packet reaches the destination node (or each of the set of destination nodes, in the case of multicast).

The following are reasons why extension headers are not inserted by nodes along a packet's delivery path: MTU black holing - Many IPv6 nodes dynamically discover Path MTU (PMTU). Having discovered the PMTU, they send packets whose size approaches, but does not exceed the PMTU. Adding an extension header to such a packet can cause MTU black holing. Incompatibility with the IPv6 Authentication Header - IPv6 Authentication header processing relies on the immutability of the Payload Length field in the IPv6 header. When a node along a packet's delivery path inserts an extension header, it updates the Payload Length field in the IPv6 header. This causes Authentication header processing to fail. Semantic corruption - Insertion of an extension header can change the meaning of a packet. The following are reasons why extension headers are not deleted by any node along a packet's delivery path, until the packet reaches the destination node: Incompatibility with the IPv6 Authentication Header - IPv6 Authentication header processing relies on the immutability of the Payload Length field in the IPv6 header. When a node along a packet's delivery path deletes an extension header, it updates the Payload Length field in the IPv6 header. This causes Authentication header processing to fail. Semantic corruption - Deletion of an extension header can change the meaning of a packet.

This document does not introduce any new security considerations.

This document does not request any IANA actions.

Thanks to Bob Hinden, Brian Carpenter, and Jinmei Tatuya for their comments and review.