Hi all,
I'm having an issue with a couple of ESXi hosts connected to a 4507 switch.
The hosts are connected with 4 GB links to to the switch, and the port channel is configured as trunk, with 5 vlans allowed, with on
Interfaces:
Gi1/25 ESX01 connected trunk a-full a-1000 10/100/1000-TX
Gi1/26 ESX01 connected trunk a-full a-1000 10/100/1000-TX
Gi2/25 ESX01 connected trunk a-full a-1000 10/100/1000-TX
Gi2/26 ESX01 connected trunk a-full a-1000 10/100/1000-TX
Po12 ESX01 connected trunk a-full a-1000
PO:
12 Po12(SU) - Gi1/25(P) Gi1/26(P) Gi2/25(P)
Gi2/26(P)
One of the interface config:
interface GigabitEthernet1/25
description ESX01
switchport trunk allowed vlan 2,20,22,900
switchport mode trunk
channel-group 12 mode on
spanning-tree portfast trunk
Po Config:
interface Port-channel12
description ESX01
switchport
switchport trunk allowed vlan 2,20,22,900
switchport mode trunk
The issue is that every once in a while a port in the Po dropes out, and joins back in without the physical link failing or anyone changing something, and the host stays down untill a series of sh/no sh are issued on interfaces:
001063: Nov 13 13:46:21: %EC-5-UNBUNDLE: Interface GigabitEthernet1/26 left the port-channel Port-channel12
001064: 000146: Nov 13 13:46:21: %EC-5-UNBUNDLE: STANDBY:Interface GigabitEthernet1/26 left the port-channel Port-channel12
001065: Nov 13 13:46:26: %EC-5-BUNDLE: Interface GigabitEthernet1/26 joined port-channel Port-channel12
001066: 000147: Nov 13 13:46:27: %EC-5-BUNDLE: STANDBY:Interface GigabitEthernet1/26 joined port-channel Port-channel12
- there are no erros on the interface;
- the cables have been changed;
- spanning tree looks to be stable (not confirmed)
- this happens the same way on the other host.
- hardware tests (diagnostics) look okay
Did anyone see this kind of problems before?
