Cisco Support Community
Community Member


we had a very wierd issue yesterday and stll waiting for Cisco to find out what happened. I needed four 10G interfaces for a project to connect our Shark appliance to our 6509 switches. The 6509 switches have four 10G blades. I took one from blade two and one from blade nine on each 6509. The interfaces were shut down, so I thought that I can configure the ports before hand as I did. I put the following configuration on interface ten2/7 on each 6509:

interface TenGigabitEthernet2/7


switchport capture

switchport capture allow vlan 110,115,117,118,160,161


mls qos trust dscp   <<<<<<<<This is on every interface on thr 6509s

And the following configuration on ten9/7 on each 6509:

interface TenGigabitEthernet9/7


switchport capture

switchport capture allowed vlan 120,125,499,580,999


mls qos trust dscp

Everything was fine until I put the configuration on ten2/7 on the second 6509. By then all other three configurations were in place. Suddenly we started to have wierd issues like not being able to connect to some routers or switches, the connection to the Internet slowed down, some servers and application had connectivity issues, etc. We have 4948 switches plugged into blade 1 and 2 and two 3750X stacks plugged in to blade 9. None of the 4948s had anythin in the log, but the two 3750X stacks had a ton of the following alerts:

Nov 15 11:26:44.956 EDT: %SW_MATM-4-MACFLAP_NOTIF: Host e839.3500.9edb in vlan 115 is flapping between port Te1/1/1 and port Gi1/0/17

Nov 15 11:26:45.459 EDT: %SW_MATM-4-MACFLAP_NOTIF: Host 5cf3.fc94.ed54 in vlan 117 is flapping between port Te1/1/1 and port Gi1/0/44

After removing the configuration from the last interface I configured everything went back to normal. Did any of you ever saw anything that an administratively down interface that has no cable in it causes issues?


Everyone's tags (1)
CreatePlease to create content