How do S series switches implement unplanned OSPF GR and planned OSPF GR

5

In unplanned GR mode, a device directly performs a active/standby switchover without sending a grace LSA to its peer. As a result, the peer does not enter the Helper state. After the standby card switches to the active card and goes Up, the device sends a grace LSA. Therefore, the smooth time cannot exceed the dead interval.
The timer in graceful period timer of IETF GR needs to be negotiated based on the timer value configured on the Restarter. The default timer value is 120s.

Other related questions:
How long is the default GR period of the OSPF routing protocol used by S series switches
For S series switches supporting OSPF, the default graceful restart (GR) period of the OSPF routing protocol is 120 seconds. If an OSPF neighbor relationship does not enter the Full state within 120 seconds, the GR fails. For example, when there are a large number of OSPF link-state advertisements (LSAs), LSA synchronization may not be complete within the default GR period, causing the GR to fail. In this case, run the graceful-restart period period command in the OSPF view to change the GR period on the Restarter.

Do CE series switches support the GR and NSR functions
CE series switches support the non-stop routing (NSR) function. When they perform an active/standby switchover, routing protocols do not flap. Support for graceful restart (GR) of CE series switches varies according to the routing protocols: - RIP does not support GR. - OSPF, IS-IS, and BGP do not support the GR restarter function (NSR is used instead), but support the GR helper function.

Whether S series switches support OSPF
For details on whether S series switches in V200R008 and earlier versions support or do not support OSPF, see the following: �?S7700/S9700/S12700: Support �?S9300: Support �?S6720EI: Support �?S6700EI: Support �?S5720HI: Support �?S5710HI: Support �?S5700HI: Support �?S5720EI: Support �?S5710EI: Support �?S5700EI: Support �?S5720SI/S5720S-SI: Support �?S5700SI: Not support �?S5710LI: Not support �?S5700LI: Not support �?S5700S-LI: Not support �?S3700HI: Support �?S3700EI: Support �?S3700SI: Not support �?S2750EI: Not support �?S2720EI: Not support �?S2700EI: Not support �?S2710SI: Not support �?S2700SI: Not support �?S1720: Not support For details on whether S series switches in V200R008 and earlier versions support or do not support OSPFv3, see the following: �?S7700/S9700/S12700: Support �?S9300: Support �?S6720EI: Support �?S6700EI: Support �?S5720HI: Support �?S5710HI: Support �?S5700HI: Support �?S5720EI: Support �?S5710EI: Support �?S5700EI: Support �?S5720SI/S5720S-SI: Support �?S5700SI: Not support �?S5710LI: Not support �?S5700LI: Not support �?S5700S-LI: Not support �?S3700HI: Support �?S3700EI: Support �?S3700SI: Not support �?S2750EI: Not support �?S2720EI: Not support �?S2700EI: Not support �?S2710SI: Not support �?S2700SI: Not support �?S1720: Not support

OSPF cost of S series switches
The ospf cost command can be executed on an S series switch that supports OSPF to configure the cost required for running the OSPF protocol on an interface. Application scenario: The interface cost can be automatically calculated by OSPF based on the interface bandwidth or manually configured by using the ospf cost command. If the costs and destination IP addresses of multiple routes discovered by one routing protocol are the same, load balancing can be implemented among the routes. Modify interface costs to implement load balancing as required. Configuration impact: If no interface cost is configured by using the ospf cost command, OSPF automatically calculates the interface cost based on the interface bandwidth. The calculation formula is as follows: Cost of the interface = Bandwidth reference value/Interface bandwidth. The integer of the calculated result is the cost of the interface. If the calculated result is smaller than 1, the cost value is 1. Changing the bandwidth reference value can change the cost of an interface. By default, the bandwidth reference value is 100 Mbit/s. Therefore, the default cost value of an Ethernet (100 Mbit/s) interface is 1 (100,000,000/the interface bandwidth). Example: Set the cost required by VLANIF 100 for running OSPF to 65. [HUAWEI] interface vlanif 100 [HUAWEI-Vlanif100] ospf cost 65 Set the cost required by GE0/0/1 for running OSPF to 65. [HUAWEI] interface gigabitethernet 0/0/1 [HUAWEI-GigabitEthernet0/0/1] undo portswitch [HUAWEI-GigabitEthernet0/0/1] ospf cost 65 Note: The cost calculated by OSPF is the cost of the route to the outbound interface of the destination network.

Definition of ABR in OSPF on S series switches
Question: What is the definition of ABR? Is a device with more than two areas configured an ABR? Answer: According to RFC 2328, an OSPF backbone area contains the ABRs of all areas. That is, an ABR must belong to a backbone area. If Area 2 and Area 3 are configured on a device with no interface in a backbone area, the device is not an ABR. If two or more areas are configured on an S series switch supporting OSPF and one of these areas is a backbone area, the switch is an ABR. Not all these areas must have interfaces in the Up state.

If you have more questions, you can seek help from following ways:
To iKnow To Live Chat
Scroll to top