Core side flaps vs port channel member status and failover

[lcaruso]

Hi, I am looking for a better redundancy outcome, not to mention better understanding, given this scenario just experienced. A pair of SSO 9800-40s connected to a par of 9606s in SVL where a port channel member on the core side connecting to the active Primary starts flapping once per second due to a faulty transceiver with the result bringing down over one third of the APs across random vlans (clearly the load balancing hash algorithm sending traffic to the bit bucket). Should I not expect the 9800 side member interface to go down and failover to the standby, or said differently, is there a configuration to detect this link flapping such that the near end member signals a failover? Thanks in advance. 

[balaji.bandi · Accepted solution]

the failover scenarios are different - i think your situation its going up and down in seconds may not detected as failure i guess.(again what interface is that connected to WLC ?)

As i mentioned again depends how this connected each other  - RP connected back to back ?

check some failure scenarios :

https://www.ciscolive.com/c/dam/r/ciscolive/global-event/docs/2022/pdf/BRKEWN-2846.pdf

[marce1000 · Accepted solution]

        >... is there a configuration to detect this link flapping such that the near end member signals a failover? 
 - In general HA-SSO targets box failure and not link flapping  (the latter argument being 'enforced' by the fact that link flapping is 'outside' of the 9800 controller HA architecture)  ; you may look for improvements in the (HA-SSO) configuration by using the CLI command (on the primary controller) show tech wireless and feed the output into Wireless Config Analyzer

                                But the link should be repaired ASAP , 

 M.