I have a few switches that show "No connectivity" for only a minute or two multiple times throughout the day. The times these occur don't match among the switches and we haven't noticed or had any users report any internet connectivity issue during those times.
All switches are running 10.35 firmware and are not all the same model. They're a mixture of MS350, 225, 320 and 220.
Just wondering if anyone has any ideas on some possible causes.
I've something similar if there are VLAN inconsistencies. For example a switch using the default VLAN 1 settings, but the MX it is behind only has VLAN 10 and not an actual VLAN 1. Also how does your cabling look? Is it possible the cabling is worn, or longer than spec?
Thanks for the reply. Cabling looks all good. Most switches are connected via fiber. The SFPs are aftermarket but not experiencing any issues with traffic flow on the LAN side as far as I can see.
One thing we do have that we are planning on correcting is that all of the access switches connected to the core are set up with a native vlan mismatch. I.e port on core set as trunk with native vlan 3 going to access switch with native vlan 1 and the access ports on the switch are in vlan 1 for data.
I know this is bad, but to give the background on why is that we just replaced our HP core switch with a Meraki MS425-32) and given the small outage window we had, we decided to mimic the existing config/setup from the HP on the Meraki. This way we didn't have to touch the access switch configs during cut-over.
The plan is to now correct the native vlan mismatch and change the access ports on the access switches to the correct vlan site-by-site.
I'm not sure if this would have any effect on connectivity though, as everything seems to be working as far as LAN and internet connectivity for PCs and phones on the network.
That is very likely the issue. A network config I inherited from a previous IT admin had a VLAN mismatch like you are describing and it would cause a lot of little red lines in the connectivity of the switch. By creating a new template to properly bring the VLANs in sync resolved that issue.
Are these switches all running in layer 2 mode (aka, don't have layer 3 interfaces configured and doing routing)?
Cool. Thanks MacuserJim. I'll be re-mediating the first site soon so hopefully this resolves the issue!
The core has L3 interfaces and the access switches are all layer 2.