Link Failure Detection and Traffic Redirection in an Openflow Network

• US 20110286324A1
• Filed: 05/19/2011
• Published: 11/24/2011
• Est. Priority Date: 05/19/2010
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Skoldstrom, Pontus, Bellagamba, Elisa, Kempf, James
• Granted Patent: US 8,665,699 B2
• Time in Patent Office: Days
• US Class Current: 370/219
• CPC Class Codes: H04L 41/0659 by isolating or reconfiguring faulty entities
• CPC Class Codes: H04L 41/0677 Localisation of faults
• CPC Class Codes: H04L 41/40 using virtualisation of network functions or resources, e.g. SDN or NFV entities
• CPC Class Codes: H04L 43/0805 by checking availability
• CPC Class Codes: H04L 43/20 the monitoring system or the monitored elements being virtualised, abstracted or software-defined entities, e.g. SDN or NFV
• CPC Class Codes: H04L 45/22 Alternate routing
• CPC Class Codes: H04L 45/28 using route fault recovery
• CPC Class Codes: H04L 45/50 using label swapping, e.g. multi-protocol label switch [MPLS]

Abstract

Failure detection and traffic redirection are implemented in an OpenFlow switch. Link failure detection packets, such as Bidirectional Forwarding Detection (BFD) packets, are periodically sent out on links to peer OpenFlow switches, such as via the Multi-protocol Label Switching (MPLS) Transport Profile (MPLS-TP). Link failure detection packets are received from the peer OpenFlow switches on the links, and monitored. A link failure is detected if no incoming link failure detection packets are received on a link for a predetermined interval. In the event of a link failure, traffic is redirected from the failed link to a backup link by altering entries on a flow table of the OpenFlow switch.

Claims

1. An apparatus configured to operate as an OpenFlow switch, comprising:

a transmitter operative to periodically send outgoing link failure detection packets on one or more links, each link being used to carry traffic to a peer OpenFlow switch;

a receiver operative to receive incoming link failure detection packets, from each peer OpenFlow switch, over the one or more links;

a link failure detection unit operative to monitor the reception of incoming link failure detection packets and further operative to determine that a link has failed if incoming link failure detection packets have not been received on that link for a predetermined interval; and

a link failover unit operative to redirect traffic intended for a failed link to a backup link in response to a determination of a link failure.

2. The apparatus of claim 1 further comprising one or more packet generating virtual ports operative to generate outgoing link failure detection packets, wherein each packet generating virtual port is operative to generate outgoing link failure detection packets with a predetermined period.

3. The apparatus of claim 1 further comprising:

a flow table configured with a plurality of multicast entries, wherein each multicast entry matches outgoing link failure detection packets with a predetermined period, and operative to identify a multicast group unit that the matching outgoing link failure detection packets should be sent to; and

one or more multicast group units each associated with a multicast group, each multicast group comprising one or more links on which outgoing link failure detection packets are to be sent with the predetermined period.

4. The apparatus of claim 3 wherein each multicast group unit is configured to replicate outgoing link failure detection packets generated by a packet generating virtual port, such that there is an outgoing link failure detection packet for each of one or more links in the multicast group associated with the multicast group unit.

5. The apparatus of claim 1 further comprising a field modifying virtual port for each of the one or more links, wherein the field modifying port is operative to insert link-specific data into one or more fields of the outgoing link failure detection packets.

6. The apparatus of claim 1 further comprising a link information database operative to store the link-specific information relating to each of the one or more links.

7. The apparatus of claim 6 wherein the information relating to each of the one or more links comprises information selected from the group consisting of an identifier for the OpenFlow switch, an identifier for a peer OpenFlow switch, one or more identifiers for a link failure detection session between the OpenFlow switch and a peer OpenFlow switch, the predefined interval for the link failure detection session, and information identifying the backup link.

8. The apparatus of claim 1 further comprising a packet processing unit operative to identify incoming link failure detection packets and further operative to forward identified incoming link failure detection packets to the link failure detection unit.

9. The apparatus of claim 1 wherein link failover unit is configured to add or modify one or more entries in a flow table in order to redirect traffic intended for a failed link on to a backup link.

10. The apparatus of claim 1 wherein the link failure detection packets comprise Bidirectional Forwarding Detection (BFD) packets.

11. The apparatus of claim 10 wherein the one or more links comprise Multi-protocol Label Switching (MPLS) tunnels, and wherein the BFD packets are carried using the MPLS Transport Profile (MPLS-TP).

12. The apparatus of claim 11 wherein the packet generating virtual ports generate BFD packets with a Generic Associated Channel Label (GAL) and each of the field modifying virtual ports is operative to insert an Associated Channel header (ACH), an ACH Type-Length-Value (TLV) header, and one or more ACH TLVs.

13. The apparatus of claim 11 further comprising, for each of the one or more MPLS tunnels, an MPLS tunnel ingress virtual port operative to assign an MPLS label on to the outgoing BFD packets.

14. The apparatus of claim 12 further comprising, for each of the one or more MPLS tunnels, an MPLS tunnel egress virtual port operative to remove an MPLS label from any incoming packets and further operative to forward the incoming packets to a packet processing unit if the incoming packets have a GAL.

15. The apparatus of claim 12 wherein the packet processing unit is configured to determine if an incoming packet is an incoming BFD packet.

16. The apparatus of claim 12 wherein the flow table is configured with a plurality of MPLS tunnel entries, each MPLS tunnel entry matching incoming packets received on a specific MPLS tunnel and identifying the MPLS tunnel egress virtual port to which the matching incoming packets should be sent.

17. A method of operating an OpenFlow switch, comprising the steps of:

periodically sending outgoing link failure detection packets on one or more links, each link being used to carry traffic to a peer OpenFlow switch;

receiving incoming link failure detection packets, from each peer OpenFlow switch, over the one or more links;

monitoring the reception of incoming link failure detection packets;

determining that a particular link has failed if incoming link failure detection packets have not been received on that link for a predetermined interval; and

redirecting traffic intended for the failed link to a backup link.

18. The method of claim 17 further comprising:

configuring the OpenFlow switch with one or more packet generating virtual ports operative to generate the outgoing link failure detection packets, each packet generating virtual port operative to generate outgoing link failure detection packets with a predetermined period.

19. The method of claim 17 further comprising configuring the OpenFlow switch with one or more multicast group units, each associated with a multicast group, each multicast group comprising one or more links on which outgoing link failure detection packets are sent with the predetermined period.

20. The method of claim 19 further comprising:

comparing outgoing link failure detection packets to a flow table, the flow table being configured with a plurality of multicast entries, wherein each multicast entry is operative to match outgoing link failure detection packets with a predetermined period and is further operative to identify a multicast group unit; and

sending the outgoing link failure detection packets to the multicast group unit identified in the matching multicast entry.

21. The method of claim 20 further comprising, at each multicast group unit, replicating outgoing link failure detection packets generated by a packet generating virtual port, such that there is an outgoing link failure detection packet for each of one or more links in the multicast group associated with the multicast group unit.

22. The method of claim 21 further comprising configuring the OpenFlow switch with a field modifying virtual port for each of the one or more links.

23. The method of claim 22 further comprising:

sending each of the replicated outgoing link failure detection packets to the appropriate field modifying port; and

inserting, at the field modifying port, link-specific data into one or more fields of the outgoing link failure detection packets.

24. A nontransient machine-readable medium including one or more computer programs operative to cause a computing apparatus operatively connected to a communication network to operate as an OpenFlow switch, the computer programs operative to cause a computing apparatus to perform the steps of:

periodically sending outgoing link failure detection packets on one or more links, each link being used to carry traffic to a peer OpenFlow switch;

receiving incoming link failure detection packets, from each peer OpenFlow switch, over the one or more links;

monitoring the reception of incoming link failure detection packets;

determining that a particular link has failed if incoming link failure detection packets have not been received on that link for a predetermined interval; and

redirecting traffic intended for the failed link to a backup link.

25. A computer system operatively connected to a communication network and configured to operate as an OpenFlow switch, the computer system comprising:

a transmitter operative to periodically send outgoing link failure detection packets on one or more links connected to the communication network, each link being used to carry traffic to a peer OpenFlow switch;

a receiver operative to receive incoming link failure detection packets over the one or more links from each peer OpenFlow switch; and

a processor operative to monitor the reception of incoming link failure detection packets, further operative to determine that a link has failed if incoming link failure detection packets have not been received on the link for a predetermined interval, and still further operative to redirect traffic intended for a failed link to a backup link.