So, everything seems to be fine, the network is working, but looking into the Cisco logs we find frequently repeating messages like:
*Mar 22 16:49:59.821: %CDP-4-NATIVE_VLAN_MISMATCH: Native VLAN mismatch discovered on GigabitEthernet0/24 (1), with CENTRAL-CISCO GigabitEthernet0/7 (14).
What does this message mean?
Why?
First of all, this is not an error message. Quite often CDP-4-NATIVE_VLAN_MISMATCH is just informational for the administrator.
Second, as the message itself says (Native VLAN mismatch discovered): Cisco has detected that the port on the other end has a different Native VLAN. That is, for example, on one side untagged traffic goes out/in on VLAN=A, and on the other side - on VLAN=B. Since the packets are untagged, i.e. they pass through an ACCESS port or through the Native VLAN of a Trunk port, this situation is possible and, moreover, can actually be used to connect differently configured networks.
As an example: on one of the core switches we see traffic from a remote office on VLAN=20, while on the office side the LAN VLAN is, say, VLAN 50. How do we connect them? The simplest way is to switch the ports to Access mode and set one side to Access=VLAN_20 and the other to Access=VLAN_50.
If other VLANs also need to be carried across, then instead of Access we configure the ports as Trunk with different Native VLANs.
And what do we end up with? Everything works, but the logs get flooded with "%CDP-4-NATIVE_VLAN_MISMATCH" and "Native VLAN mismatch discovered" messages.
Why? Because Cisco devices use VTP and check configurations against each other so as to immediately notify the administrator of possible problems. Among other things, the devices check whether the Native VLAN is the same on both sides of the port, even if one of the sides has the port configured as Access.
From Cisco's point of view, configurations must be identical within the same VTP domain. If the situation described above occurs (i.e. networks with different VLANs are indeed being connected via an Access link), then the domains should be different.
Solution.
If everything works as it should and, in your view, the message is just an unnecessary (unneeded) warning - let's get rid of it. Why clutter the logs with unnecessary information (especially since the log length is limited and a genuinely important message could get overwritten because of these "%CDP-4-NATIVE_VLAN_MISMATCH" entries).
The solution is to move one of the devices into a different VTP domain. Yes, it's enough for at least one of the Ciscos to be in a different VTP domain - and these messages will disappear.
So, let's log in to one of the Ciscos and move it to a different domain:
CISCO> enable
CISCO# config t
CISCO(config)# vtp mode transparent
CISCO(config)# vtp domain MY_NEW_VTP_NAME
where instead of MY_NEW_VTP_NAME you specify a new VTP domain name (no particular naming requirement).
And that's it, the message will go away.
Comments