Telefonaktiebolaget Lm Ericsson (publ)

Faseela K

Vyshakh Krishnan C H

Ashutosh Bisht

Vinayak Joshi

Muthukumaran Kothandaraman

Joel Halpern

Nobin Mathew

Deepthi V V

Vishal Thapar

Riyazahmed D Talikoti

Manu B

Siva Kumar V V K A Perumalla

D Arunprakash

A method by a first controller in a software defined networking (SDN) network for programming a switch in the SDN network to use a controller port as a watch port. The method includes generating an instruction for the switch to create a first group entry for a first group in a packet processing pipeline of the switch, where the first group entry includes a first bucket that specifies a first controller port as a watch port and an action for the switch to forward packets to the first controller via the first controller port, where the first controller port being specified as the watch port in the first bucket indicates that execution of the action specified by the first bucket is to be contingent upon a liveness of the first controller port and sending the instruction to the switch to cause the switch to create the first group entry.

Controller watch port for robust software defined networking (SDN) system operation

A method implemented by a first controller in a software defined networking (SDN) network to monitor a health of a second controller, where the first controller and the second controller are part of a cluster of controllers, where controllers in the cluster communicate with switches over an in-band network, and where controllers in the cluster communicate with each other over an out-of-band network. The method includes sending a first control message to a switch over the in-band network in response to a determination that the second controller cannot be reached over the out-of-band network, where the first control message includes a health check request message and an instruction for the switch to send the health check request message to the second controller, and where the switch is a designated switch that has a control channel connection to the first controller and the second controller over the in-band network.

Robust node failure detection mechanism for SDN controller cluster

A method by a software defined networking (SDN) controller to configure a switch to perform translation module bypass in a container orchestration system. The method includes receiving a translation rule for a flow from a load balancer, sending translation module bypass instructions to a switch in response to receiving the translation rule for the flow, where the translation module bypass instructions include instructions for the switch to stop sending packets belonging to the flow to the translation module and to apply a network address translation specified by the translation rule for the flow to the packets belonging to the flow, and send an indication to the load balancer that the packets belonging to the flow are to bypass the translation module to cause the load balancer to disable timeout processing for the flow in the translation module.

Service handling in software defined networking based container orchestration systems

A method is implemented by a switch in a software defined networking (SDN) network managed by a controller to achieve hitless resynchronization during a controller upgrade. The method includes installing an upgraded set of flow entries so that a packet processing pipeline of the switch includes both a non-upgraded set of flow entries and the upgraded set of flow entries, processing non-tunneled packets using the non-upgraded set of flow entries, processing tunneled packets that have a tunnel upgrade status indicator set in a tunnel header using the non-upgraded set of flow entries, while processing tunneled packets that do not have a tunnel upgrade status indicator set in a tunnel header using the upgraded set of flow entries, and processing non-tunneled packets using the upgraded set of flow entries after all switches managed by the controller have installed upgraded flow entries.

Mechanism for hitless resynchronization during SDN controller upgrades between incompatible versions

A method is implemented by a network device to reduce bandwidth and compute resources needed by the controller of a software defined networking (SDN) controller by distributing monitoring of virtual network functions (VNFs) to data plane nodes (DPNs) in a SDN network. The method includes receiving a monitoring request from the SDN controller to monitor a VNF, receiving a byte sequence from the SDN controller for the VNF, and generating flow table entries for monitoring the VNF by matching the byte sequence with monitoring response messages from the VNF.

Scalable application level monitoring for SDN networks

Methods for tracing a packet in a pipeline comprising a set of tables, in an SDN using OpenFlow. The packet comprises a trace bit, and is provided with a TTL value. A first network node indicates to a second network node to have each flow table decrement the TTL value by 1, based upon the trace bit. The first network node initially sets the TTL to be 1, and then sends the packet to a first table in the second network node. The first network node receives the packet, from another table in the set, and for every remaining table, and one table at a time: a) increments the TTL value by 1, and b) repeats the sending, and the receiving, until a last table is reached or a criterion is met. The first network node then indicates a route followed by the packet.

First network node, second network node, and methods performed thereby for tracing a packet in a pipeline

Exemplary embodiments include methods and/or procedures for handling messages from one or more data-plane nodes (DPNs) to a plurality of controller nodes configured in a cluster, comprising: receiving a message from a DPN; tagging the message with an incarnation identifier (IID) associated with the DPN; and performing HD handling of the tagged message, which includes comparing a value of the IID of the tagged message with a value of a global IID associated with the DPN, discarding the tagged message if the value of the IID of the tagged message is less than the value of the global IID, and processing the tagged message if the value of the IID of the tagged message is not less than the value of the global IID. Exemplary embodiments also include controller nodes, controller clusters, and computer-readable media embodying operations of the exemplary methods and/or procedures.

Robust message processing for a software-defined networking (SDN) controller cluster

A method is implemented by a network device where the method performs a trace flow process in a packet processing pipeline of the network device. The packet processing pipeline includes a trace table and a forward table. The method encompasses forwarding a trace packet to the trace table, forwarding a copy of the trace packet to a trace analyzer in response to determining that a trace bit is set in the trace packet, forwarding the trace packet to the forward table, and determining a next stage of the packet processing pipeline based on trace packet meta data and a value stored in a trace register.

Method and system for tracing packets in software defined networks

A method is implemented by a network device for enabling destination network address translation in a cloud network. The method includes determining that packets having a first public address as a source address and a second public address as a destination address are to be forwarded to a first host that is assigned a first private address and sending a first advertisement message to a gateway indicating that packets having the first public address as a source address and the second public address as a destination address are to be forwarded to a first switch connected to the first host, where the first switch is configured to translate the destination address of those packets from the second public address to the first private address assigned to the first host.

Efficient network address translation (NAT) in cloud networks

A system for tracing packets in a Software Defined Networking (SDN) network. The system includes a switch and an SDN controller communicatively coupled to the switch. The SDN controller is configured to program a packet processing pipeline of the switch to cause the switch to determine, at a first flow table of the packet processing pipeline, whether a packet is a trace packet that was recirculated based on content of a first field associated with the packet and in response to a determination that the packet is a trace packet that was recirculated, set a first field associated with the packet to indicate that tracing is disabled for the packet, set a second field associated with the packet to indicate that tracing is enabled for the packet, determine a flow table of the packet processing pipeline from which the packet was recirculated, and direct the packet to that flow table.

Optimized datapath troubleshooting with trace policy engine

A method is implemented by a switch in a software defined networking (SDN) network to trace packets in a packet processing pipeline of the switch. The method includes creating a copy of a received packet to function as a trace packet. The method further includes, at each of one or more subsequent flow tables that the trace packet traverses, appending an identifier of that flow table to a recorded route of the trace packet and resubmitting the trace packet to that flow table to be processed without packet tracing. The method further includes sending, at an egress table, the trace packet to a controller along with the recorded route of the trace packet.

Apparatus and method to trace packets in a packet processing pipeline of a software defined networking switch

A method and system for automatic redirection of control channel messages in response to a control channel failure with a target switch in a software defined networking (SDN) network, the automatic redirection occurring before a topology of the SDN network is updated in response to the control channel failure. The method includes determining whether a control channel message is to be sent to an SDN controller, checking availability of the control channel, selecting a backup port to reach the SDN controller in response to the control channel being unavailable, and forwarding the control channel message to the SDN controller via the backup port.

Mechanism for control message redirection for SDN control channel failures

A method implemented by a first switch in a software defined networking (SDN) network to monitor a tunnel between the first switch and a second switch in the SDN network. The method includes generating a first flow entry that matches packets received over the tunnel, generating a second flow entry that matches packet received over the tunnel, where the second flow entry has a priority that is lower than a priority of the first flow entry, removing the first flow entry and transmitting a flow removed message to an SDN controller in response to a determination that the first flow entry has timed out, maintaining a statistic associated with the second flow entry, and transmitting a statistics trigger event message to the SDN controller in response to a determination that the statistic associated with the second flow entry exceeds a threshold value.

Optimizing tunnel monitoring in SDN

A method implemented by a switch in a software defined networking (SDN) network to monitor a service node communicatively coupled to the switch. The method includes generating a first flow entry that matches packets received from the service node, generating a second flow entry that matches packet received from the service node, wherein the second flow entry has a priority that is lower than a priority of the first flow entry, removing the first flow entry and transmitting a flow removed message to an SDN controller in response to a determination that the first flow entry has timed out, maintaining a statistic associated with the second flow entry, and transmitting a statistics trigger event message to the SDN controller in response to a determination that the statistic associated with the second flow entry exceeds a threshold value.

Optimizing service node monitoring in SDN

A method and system provide for inline packet response generation implemented by a network device functioning as a switch in a software defined networking (SDN) network. The method configures the flow control pipeline to enable the inline response generation without use of the control channel and SDN controller after configuration. The method includes connecting with the SDN controller, receiving a packet out data packet from SDN controller with a template message and a buffer identifier (ID) for the template message, and identifying the received packet out data packet as containing the template message. The method further includes installing the template message into a buffer with corresponding buffer ID, receiving a first data packet from the SDN controller identifying matching criteria and the buffer ID, and updating a flow control pipeline to match on the matching criteria and to point to the buffer with the buffer ID.

Growing community of inventors

Bengaluru, India

Faseela K