I have a couple of Meraki MS switches that have links to "dumb" workgroup switches. My question is whether I should set the ports those are plugged into as trunks or access? These switches have devices that are on different vlans by virtue of having statically assigned IPs.
No matter how I set my ports I get constant vlan mismatch errors, those vary depending which vlan I set as native or access, but never go away.
The devices do seem to be working, nobody is reporting that they're not. Unfortunately I don't have access to these endpoints, nobody at this site seems to know what or where they are. The uplink cable disappears into the ceiling...
Of the two modes "trunk" is safest. It starts life with the expectation their might be spanning tree packets, while "access"mode starts life wit the assumption their wont be.
So I think I would tend to lean towards trunk mode in this case.
Next the VLAN mismatch errors. This suggest to me the MS is receiving spanning tree or CDP or LLDP packets from something to be able to detect a VLAN mismatch.
I would do a packet capture of one type of packet at a time and then see what you can discover about the over end of the cable. This might also reveal what the VLAN configuration should be.
The MS will tell you what clients are on its ports. Perhaps you could determine where one of these clients are, and then trace back from that client to find out what that plugs into.
Thanks for the response, most appreciated!
I changed the ports to trunk with no vlan tagging. That seems to have cleared up the vlan mismatch errors, but I see alot of devices on vlan 0 now. What are the implications of this, if any? For instance, would this keep devices that are on the same IP subnet, but have been tagged a vlan from seeing these devices?
I don't think I mentioned that this network is mostly building infrastructure, so the devices in question are things like security cameras, card readers, and alarms....