we have this issue with HP Proliant DL380 G7 and Esxi 6.0U3 HP custom image: we configured snmp on the servers to send the trap on our collector, all works well with 3 of the 4 nics; we configured the port 1 and 3 (vmnic0 and vmnic2) in teaming for a vswitch (with management) and the port 2 and 4 (vmnic1 and vmnic3) in teaming with another vswitch; when we disconnect a cable we receive the 18006 trap (for link down), while the link is up we have the 18005; on one nic, instead, we have only the trap when we disconnect the cable but we haven't the follow trap when we reconnect it.
We noticed that one of this port is one of the 2 inside the management network nic; I also tried to dedicate another 2 nics for only management network, but we have the same behaviour.
The teaming are configured, for load balancing, with Route based on originating port ID; the 2 nics are in active/active state and we have 2 cisco switches that have one cable for nic..so port 1 is on switch A and port 3 is on switch B; the switches are configured with HSRP.
Do you have any idea why we don't receive the trap 18005 for only one nic? We have this issue on 2 servers and also tried the same configuration but with the cables linked on one switch..same issue.
issue resolved : with wireshark I've seen that when I reconnected the cable I have ping failure even if there was another interface in the teaming; with this behaviour the snmp trap was lost; I have done this command in the switches on the related vmware host port:
With this the network interfaces are up almost instantly so no package and no snmp trap is lost.