Systems and methods for improved fault tolerance in solicited information handling systems
An information handling system is provided that uses an overlay network controller to improve fault tolerance and minimize downtime in a solicited system. The network controller may determine standby paths for the solicited communications path and reroute communications to one of the standby paths when a fault is detected in the communications path.
1. An information handling system, comprising:
a first endpoint coupled to a second endpoint through one or more network devices of a network, the first endpoint comprising a fibre channel over Ethernet (FCoE) initialization protocol (FIP) snooping bridge for establishing a session with the second endpoint by encapsulating a FCoE frame with fibre channel information, the session including an expiration time; and
a network controller in communication with the first endpoint, the second endpoint, and the one or more network devices, the network controller configured to:
receive status information from the first endpoint, the second endpoint, and the one or more network devices, the status information comprising information regarding the one or more network devices and network links along a first path from the first endpoint through the one or more network devices to the second endpoint;
determine one or more standby paths between the first endpoint and second endpoint through the one or more network devices based on the status information;
determine whether there is a link failure in the first path preventing communication between the first endpoint and the second endpoint using the first path for the session based on the status information;
when the expiration time has been reached, forward instructions to the first endpoint to establish a new session with the second endpoint by encapsulating a second FCoE frame with fibre channel information, the new session including a second expiration time; and
in response to the link failure, when the expiration time has not been reached, select one of the one or more standby paths to be a current path, and provide instructions to the first endpoint to reroute communication between the first and second endpoints using the selected standby path.
2. The information handling system of claim 1 , further comprising:
an initiator coupled to the first endpoint; and
a target coupled to the second endpoint.
3. The information handling system of claim 2 , wherein the target comprises a storage area network (SAN).
4. The information handling system of claim 1 , wherein the second endpoint comprises a fibre channel forwarder.
5. The information handling system of claim 4 , wherein the session is used to communicate FCoE information.
6. The information handling system of claim 1 , wherein the session is used to communicate remote direct memory access (RDMA) over Converged Ethernet (RoCE) information.
7. The information handling system of claim 1 , wherein a time to reroute communication from the first path to the selected standby path is less than the expiration time, the expiration time being a keep-alive signal expiration time associated with the information handling system.
8. The information handling system of claim 1 , wherein each of the one or more network devices comprises a respective agent, each respective agent being configured to collect the status information and send the collected status information to the network controller.
9. The information handling system of claim 1 , wherein the network controller comprises a memory storing additional information describing the first path and the one or more standby paths.
10. The information handling system of claim 9 , wherein the network controller is further configured to update a table stored in the memory to remove any standby path of the one or more standby paths that relies upon one or more links associated with the link failure.
11. The information handling system of claim 1 , wherein the status information includes one or more selected from the group consisting of synchronization information, keep-alive signal information, fabric or communicated information type information, link failure information, and latency between one or more of the one or more network devices.
12. A network controller for an information handling system, the network controller comprising:
a memory;
a network interface component coupling the network controller to a network comprising a first endpoint, a second endpoint and one or more network devices, the first endpoint comprising a fibre channel over Ethernet (FCoE) initialization protocol (FIP) snooping bridge; and
one or more processors coupled to the memory and the network interface component, the one or more processors being configured to:
receive status information from the first endpoint, the second endpoint, and the one or more network devices, the status information comprising status information associated with the one or more network devices and network links along a first path from the first endpoint through the one or more network devices to the second endpoint established between the first endpoint and the second endpoint through the network upon establishment of a session between the first endpoint and the second endpoint, the session being established by the FIP snooping bridge by encapsulating a FCoE frame with fibre channel information, the session including an expiration time;
determine one or more standby paths between the first endpoint and the second endpoint based on the status information;
detect a link failure in the first path preventing communication between the first endpoint and the second endpoint using the first path for the session based on the received status information;
when the expiration time has been reached, forward instructions to the first endpoint to establish a new session with the second endpoint by encapsulating a second FCoE frame with fibre channel information, the new session including a second expiration time; and
in response to the link failure, when the expiration time has not been reached, select one of the one or more standby paths to be a current path and provide instructions to the first endpoint to reroute communication between the first and second endpoints using the selected standby path.
13. The network controller of claim 12 , wherein the session is used to communicate fibre channel over Ethernet (FCoE) information.
14. The network controller of claim 12 , wherein the session is used to communicate remote direct memory access (RDMA) over Converged Ethernet (RoCE) information.
15. The network controller of claim 12 , wherein the status information is collected by and received from a respective agent installed in each of the one or more network devices.
16. A method for rerouting information in an information handling system, comprising:
receiving, at a network controller, status information from a first endpoint, a second endpoint, and one or more network devices in a network, the status information comprising status of the one or more network devices and network links along a first path from the first endpoint through the one or more network devices to the second endpoint during a session, the session being established by the first endpoint by encapsulating a FCoE frame with fibre channel information, the session including an expiration time;
determining, by one or more processors of the network controller, one or more standby paths between the first endpoint and the second endpoint;
detecting, by the one or more processors, a link failure in the first path preventing communication between the first endpoint and the second endpoint using the first path based on the received status information;
when the expiration time has been reached, forward instructions to the first endpoint to establish a new session with the second endpoint by encapsulating a second FCoE frame with fibre channel information, the new session including a second expiration time; and
in response to the link failure, when the expiration time has not been reached, selecting one of the one or more standby paths to be a current path and providing instructions to the first endpoint to reroute communication between the first and second endpoints using the selected standby path.
17. The method of claim 16 , wherein the status information is received from a respective agent installed in each of the one or more network devices.
18. The method of claim 16 , wherein the session is used to communicate fibre channel over Ethernet (FCoE) information.
19. The method of claim 18 , wherein the first endpoint comprises a FCoE initialization protocol (FIP) snooping bridge and the second endpoint comprises a fibre channel forwarder.
20. The method of claim 19 , wherein the FIP snooping bridge is coupled to an initiator and the fibre channel forwarder is coupled to a target.