What need I check when LSP flapping occurs

9

When LSP flapping occurs, check the following information:
Whether the peer end can be pinged. Run the display mpls ldp peer command in the system view to check the peer address (Transport-Address field). Run the ping command to check for traffic loss.
Whether route flapping occurs. Run the display ip routing-table command in the system view to check all the current routing information. Set the destination address to check all the current routing information mapping the destination address. Frequently check routing information in a period to check for flapping.
Whether LDP session flapping occurs. Run the display mpls ldp session command in the system view to check information about all LDP sessions. Set the peer address to check information about the LDP session with the specified peer. Frequently check LDP session information in a period to check for flapping.
Whether lsp-trigger all is configured.
Whether exceptions occur by enabling the debugging function.

Other related questions:
When will an LSP flap and how do I know that an LSP flap occurred
If routing flapping or LSP session flapping occurs, LSP flapping occurs. LSPs may also flap when loop detection is configured or address overlapping, multiple links, or ring topologies exist. To determine whether LSP flapping occurs, run the display mpls lsp command to check the number of LSPs. If the number of LSPs frequently changes, LSP flapping occurs. You can also run the snmp-agent trap enable feature-name mpls_lspm command to enable trap for the MPLS LSPM module. If the LSP alarms are frequently generated, LSP flapping occurs. In this case, run the undo snmp-agent trap enable command to restore the default trap settings. The following alarm information is displayed: #Jul 25 22:49:18 2020 HUAWEI LSPM/5/TRAP:1.3.6.1.2.1.10.166.2.0.2 LSP 210958 Changes to Down #Jul 25 22:49:19 2020 HUAWEI LSPM/5/TRAP:1.3.6.1.2.1.10.166.2.0.1 LSP 210958 Changes to Up

How do I check whether LSP flapping occurs on an AR
You can run the display mpls lsp command to check the number of LSPs. If the number of LSPs changes frequently, LSP flapping occurs. In this case, run the snmp-agent trap enable feature-name mpls_lspm command. If the alarm about LSPs is frequently generated, LSP flapping occurs. Run the undo snmp-agent trap enable command. For example: #Jul 25 22:49:18 2020 HUAWEI LSPM/5/TRAP:1.3.6.1.2.1.10.166.2.0.2 LSP 210958 Changes to Down #Jul 25 22:49:19 2020 HUAWEI LSPM/5/TRAP:1.3.6.1.2.1.10.166.2.0.1 LSP 210958 Changes to Up The preceding information shows that the LSP alternates between Up and Down.

Do I need to specify the next hop or egress when configuring a static LSP on the ingress node
When configuring a static LSP on the ingress node, ensure that configuration items match routing information, including the destination address, mask, next hop, and outbound interface;otherwise,the LSP cannot become Up. The system calculates the next hop if you specify the outbound interface. The calculated next hop is the IP address of the local outbound interface. The system calculates the outbound interface if you specify the next hop (the inbound interface of the peer end). The calculated outbound interface is still the local outbound interface. Therefore, errors may occur during the static LSP configuration on the ingress node. Pay attention to the following points: If the outbound interface is specified for a static route, you must specify the outbound interface when configuring the static LSP on the ingress node. In other situations, you need to specify the next hop rather than only the outbound interface. The fundamental rule is that the static ingress LSP route fully matches routing information.

How Can I Check the Exception That Occurs When OperationCenter Interconnects with All-In-One FusionManager
【Product�? ManageOne OperationCenter 【Version�? 3.0, 3.0.5, 3.0.7, 3.0.8, and 3.0.9 【Question�? How can I check the exception that occurs when OperationCenter interconnects with All-In-One FusionManager? 【Answer�? Step 1 Check the current OperationCenter version by following the instructions provided in How Do I Check the OperationCenter Version? and check All-In-One FusionManager that interconnects with OperationCenter. The OperationCenter version must be V100R002C30SPC108 or later. OperationCenter earlier than V100R002C30SPC108 cannot interconnect with All-In-One FusionManager. Step 2 Check whether the interfaces of All-In-One FusionManager are compatible with OperationCenter. ----End

Problem and solution when disk isolation occurs
You can perform the following operations when disk isolation occurs: The following causes may result in disk isolation: Bit error Reinserting disks repeatedly Disk power connection problem 1. Bit error Check the bit error of back-end SAS disks. Search keywords err inc and disable disk phy in the SES log. Note: phy:9 phymon***disable disk phy in the log shows that disk phy 9 is isolated. That is, the disk in slot 9 is isolated (phy0 to phy23 corresponds to disk 0 to 23). Troubleshooting 1. Before removing a faulty disk, collect S.M.A.R.T. information. 2. If conditions permit, insert the isolated disk to other slots to check whether isolation is caused by the disk or the slot. If isolation is caused by the disk, apply for disk replacement. If isolation is caused by the slot, check whether the slot has any foreign objects. Check the bit error on Fibre Channel disks. Search keyword lcv that is Fibre Channel bit errors in the SES log. If HD 0 and lcv ffff are displayed, the information indicates that a large quantity of bit errors are produced in slot 0 and cause disk isolation. The back-end Fibre Channel bit errors can spread from the port to the disk. If a Fibre Channel disk is isolated, check whether bit errors occur on the port by using the following methods: Check on the ISM. Enter fc allinfo in MML mode. Note: If any information displayed is not 0, bit errors exist. If bit errors are detected on the port, verify whether bit error are generated in the link. For details about how to verify, see the troubleshooting cases for a single link failure of the Fibre Channel enclosure disk caused by bit errors. Troubleshooting: If only one disk fails, verify the failure by using the above method. If a link fails, replace the optical module and optical cables and verify the failure. If a link does not fail, use the same method as one carried out on the SAS disk. If multiple disks are faulty, refer to the troubleshooting cases for a single link failure of the Fibre Channel enclosure disk caused by bit errors. 2. Reinserting disks repeatedly Note: A drive can isolate the disk from other ones if intermittent disconnections occur on the disk. Reinserting disks repeatedly may lead to disk isolation. Verify whether the disk is reinserted many times within a short period. If such a case exists, reinserting disk may result in disk isolation. Troubleshooting: Reinsert the disk. 3. Disk power connection problem Note: If the disk enclosure is affected by violent shaking, disk power may be insecurely connected and the disk is isolated. Troubleshooting: Contact R&D engineers for further analysis.

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