Cant find this problem in the labs which is a shame but have today seen a really weird issue on a home Lab and wondered if anyone had seen it before.
I have a switch, a 2960 which has 24 ports. A windows based server looks after dhcp for each of the subnets apart from vl1. The DHCP server has option 242 configured to push a tftp server address 10.10.2.1 to the phone to get its config. Option 242 MCIPADD=10.10.2.1,MCPORT=1719,HTTPSRVR=10.10.2.1 Vlan 1 10.10.1.1/24 vlan 10 10.10.10.1/24 vlan 20 10.10.20.1/24 the config on ports fa0/1 thru 10 look like this: swit port mode acc swit acc vlan 10 swit voice vlan 20 The remaining ports are un-configured as were not needed so i thought if something was plugged into them at worst will auto negotiate, So I go down to my basement and plug a phone into ports 1 thru 10, with a pc on the back the phone gets a dhcp lease from vl10, the pc from vl20 as expected. So far text book setup (there is no QOS on this yet) but here is where it gets weird. If I plug a pc or phone into an unconfigured port (as i did by accident) ie 11 thru 24 all hell breaks loose. I see duplicate address errors, EIGRP flapping, what appears to be a broadcast storm. Granted with duplicate addresses or a broadcast storm eigrp is going to flap ;-) I thought if a switchport was unconfigured it would try and auto negotiate, now with a phone with a built in switch I assume a trunk will form, but if you plug a pc, I would expect an access port to be negotiated, and then only Vlan1 Am I missing something here? Thanks for the thoughts this is bugging me. tfn Gary _______________________________________________ For more information regarding industry leading CCIE Lab training, please visit www.ipexpert.com
