Fabric switchover for systems with control plane and fabric plane on same board

• US 9,866,428 B2
• Filed: 05/29/2015
• Issued: 01/09/2018
• Est. Priority Date: 03/18/2011
• Status: Active Grant

• Current Assignee: Juniper Networks Incorporated
• Original Assignee: Juniper Networks Incorporated
• Inventors: Sharma, Usha
• Primary Examiner(s): Sloms, Nicholas
• Application Number:
  US14/725,493
• Publication Number:
  US 20150263884A1
• Time in Patent Office: 956 Days
• Field of Search: None
• CPC Class Codes: G06F 11/16 Error detection or correction of the data by redundancy in hardware
• CPC Class Codes: G06F 11/2005 using redundant communication controllers
• CPC Class Codes: H04L 41/0654 using network fault recovery ring fault isolation or reconfiguration in loop networks without recovery actions by a network management system H04L12/437
• CPC Class Codes: H04L 49/25 Routing or path finding in a switch fabric
• CPC Class Codes: H04L 49/557 Error correction, e.g. fault recovery or fault tolerance
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

A network device includes a first and a second control board, with each control board having a separate control plane and fabric plane. The network device is configured to establish an active control plane on the first control board and an active fabric plane on the second control board. The network device is configured to establish a standby fabric plane on the first control board and a standby control plane on the second control board. The network device is configured to detect a failure on the first control board or the second control board and switch the status of the control plane on the second control board from standby to active, when the failure is on the first control board, and switches the status of the fabric plane on the first control board from standby to active when the failure is on the second control board.

Claims

1. A method comprising:

verifying links between a first fabric plane and input/output units of a network device that includes a first control card and a second control card;

detecting, by a first control plane of the first control card and after verifying the links between the first fabric plane and the input/output units, a first failure of a second control plane, of the second control card, while the first control plane is in standby,

the first control card including the first control plane and the first fabric plane, and

the second control card including the second control plane and a second fabric plane;

switching, by the first control plane and based on the first control plane detecting the first failure of the second control plane, the first control plane from being in standby to being active;

detecting, by the first control plane, a second failure of the second fabric plane; and

initiating, by the first control plane and based on detecting the second failure of the second fabric plane, a switchover to the first fabric plane without again verifying the links with the input/output units after detecting the second failure of the second fabric plane and before initiating the switchover to the first fabric plane.

2. The method of claim 1, further comprising:

determining that the first control card includes the first control plane and the first fabric plane;

determining that the second control card includes the second control plane and the second fabric plane; and

detecting, before detecting the first failure of the second control plane, a redundant control board configuration based on determining that the first control card includes the first control plane and the first fabric plane and based on determining that the second control card includes the second control plane and the second fabric plane.

3. The method of claim 1, further comprising:

establishing, before detecting the first failure of the second control plane, a partially-redundant configuration using the first control card and the second control card.

4. The method of claim 3, where establishing the partially-redundant configuration includes:

establishing the first fabric plane to be active,

establishing the second fabric plane to be in standby, and

establishing the first control plane to be in standby.

5. The method of claim 1, further comprising:

forwarding traffic through the first fabric plane after verifying the links.

6. The method of claim 1, where switching the first control plane from being in standby to being active comprises:

switching the first control plane from being in standby to being active without blocking traffic through the first fabric plane.

7. The method of claim 1, further comprising:

maintaining, by the first control plane and while switching the first control plane from being in standby to being active, a first link with a first input/output unit and a second link with a second input/output unit,

the links including the first link and the second link, and

the input/output units including the first input/output unit and the second input/output unit.

19. The method of claim 1, further comprising:

detecting a repair of the second control card after switching the first control plane from being in standby to being active, and

initiating a return of the network device to a partially redundant configuration after detecting the repair of the second control card.

20. The method of claim 1, where verifying the links comprises:

signaling each of the input/output units; and

receiving a respective verification regarding a respective link, of the links, from each of the input/output units.

8. A network device comprising:

a first control card including;

a first control plane, and

a first fabric plane;

a second control card including;

a second control plane, and

a second fabric plane; and

input/output units,

the first control plane being to;

verify links between the first fabric plane and the input/output units,

detect, after verifying the links between the first fabric plan and the input/output units, a first failure of the second control plane while the first control plane is in standby,

switch, based on detecting the first failure of the second control plane, the first control plane from being in standby to being active;

detect a second failure of the second fabric plane; and

initiate, based on detecting the second failure of the second fabric plane, a switchover to the first fabric plane without again verifying the links with the input/output units after detecting the second failure of the second fabric plane and before initiating the switchover to the first fabric plane.

9. The network device of claim 8, where the first control plane is further to:

detect a repair of the second control card after switching the first control plane from being in standby to being active, and

initiate a return of the network device to a partially redundant configuration after detecting the repair of the second control card.

10. The network device of claim 8, where the first control plane is further to:

designate, before detecting the first failure of the second control plane, the first fabric plane to be active,

designate, before detecting the first failure of the second control plane, the second fabric plane to be in standby, and

designate, before detecting the first failure of the second control plane, the first control plane to be in standby.

11. The network device of claim 8, where the network device is to:

identify a failure of the first fabric plane, and

switch, based on the failure of the first fabric plane, the second fabric plane from standby to active without affecting communication between the first control plane and the input/output units of the network device.

12. The network device of claim 8,

where the input/output units include physical components, and

where the second control plane is to:

provide, before the first failure of the second control plane, instructions to the input/output units to forward traffic through the first fabric plane.

13. The network device of claim 8,

where the input/output units include physical components, and

where, when switching the first control plane from being in standby to being active, the first control plane is to:

switch the first control plane from being in standby to being active while maintaining active links between the first control plane and the input/output units to continue to allow the input/output units to process traffic.

14. A non-transitory computer-readable medium storing instructions, the instructions comprising:

one or more instructions that, when executed by a first control plane, cause the first control plane to;

verify links between a first fabric plane and input/output units of a network device that includes a first control card and a second control card;

detect a first failure of a second control plane while the first control plane is in standby,

the first control card including the first control plane and a first fabric plane, and

the second control card including the second control plane and a second fabric plane;

switch, based on the first control plane detecting the first failure of the second control plane, the first control plane from being in standby to being active;

detect a second failure of the second fabric plane; and

initiate, based on detecting the second failure of the second fabric plane, a switchover to the first fabric plane without again verifying the links with the input/output units after detecting the second failure of the second fabric plane and before initiating the switchover to the first fabric plane.

15. The non-transitory computer-readable medium of claim 14, where the instructions further comprise:

one or more instructions that, when executed by the first control plane, cause the first control plane to;

forward traffic through the first fabric plane after verifying the links.

16. The non-transitory computer-readable medium of claim 14, where the one or more instructions to switch the first control plane from being in standby to being active comprise:

one or more instructions that, when executed by the first control plane, cause the first control plane to;

switch the first control plane from being in standby to being active without blocking traffic through the first fabric plane.

17. The non-transitory computer-readable medium of claim 14,

where the instructions further comprise:

one or more instructions that, when executed by the first control plane, cause the first control plane to;

maintain, when switching the first control plane from being in standby to being active, a first link with a first input/output unit and a second link with a second input/output unit while a third link with the first input/output unit and a fourth link with the second input/output unit remain active,

where the links include the first link, the second link, the third link, and the fourth link, and

where the input/output units include the first input/output unit and the second input/output unit.

18. The non-transitory computer-readable medium of claim 14, where the instructions further comprise:

one or more instructions that, when executed by the first control plane, cause the first control plane to;

detect a repair of the second control card after switching the first control plane from being in standby to being active, and

initiate a return to a partially redundant configuration after detecting the repair of the second control card.