Hi All,
We have a Dell M1000e Chassis containing four Cisco 3130 switches in a stack. On two occasions now when we have rebooted a blade server, the server has lost full or partial connectivity with the rest of the network, as if the Port channel / vlan trunk is failing to renegotiate. The logs on the Cisco switches show nothing of note, no errors. However when I try to issue the command show etherchannel 'n' the VTY session hangs. I can log back onto the switch via a new session, but I am unable to clear the hung session, either by clear Line or clear line vty. show tcp brief doesn't show the hung session, but a show tcp vty n does show the following:
tty1, virtual tty from host 10.10.0.101
Connection state is CLOSED, I/O status: 10, unread input bytes: 0
Mininum incoming TTL 0, Outgoing TTL 255
Local host: 172.31.200.195, Local port: 23
Foreign host: 10.10.0.101, Foreign port: 52602
If I continue to issue the show interface port-channel or show etherchannel eventually I deplete the number of VTY lines and I am locked out of the switch
On the previous occasion this happened we had to reboot the whole chassis to fix both the hung vty sessions and also the blade server not seeing the network.
This has happened twice in the last 6 months now, and it is obviously a fault, but unsure if it's hardware with the chassis/switch or an IOS bug.
The current IOS on the switches is 12.2(40)EX1 - CBS31X0-UNIVERSAL-M
Any ideas?