I have seen recently that during firmware upgrades the virtual IP does not failover to the surviving fabric interconnect within the expected time of 30 seconds to 1 minute. I'm referring to the mgmt0 interfaces of the UCS FI's and the VIP and the management plane failover. I've done some monitoring and packet capture during the process and see it isn't until about 3.5 minutes that the failover of the VIP happens to the subordinate and am able to access UCSM again. By that time, the downed Primary is also responding to ping again by this time. Has anyone else seen this behavior? I have now seen it 3 times all on upgrades to 2.2.3. I'm opening a TAC case also, but wanted to see if the community has also seen this behavior.
Thanks in advance.
