08-16-2022 11:07 AM
Dear all
i hope it is okay that i come up with a topic which i found in similar posts in this forum. But i did not find the solution which could help me.
We are a non profit-organisation in Munich (Germany): The 'shaere' - www.shaere.net. We are building a culture & education centre for everyone who is intersted. i setup the IT and the Wifi area as follows:
ISP via Service Provider 'colt' working good.
Unifi Security Gateway USG-Pro-4
Cisco ME-3600-24FS-M bringing internet to 8 different technical rooms and there on a switch (STP only activated on the Cisco ME - disabled on the 8 switches)
From the 8 Switches we have patched to 17 Access Points (Ubiquiti WIFI 6 Lite & Long Range) - all sharing the same SSID
Now we are getting these errors:
Aug 16 11:58:02.781: %SW_MATM-4-MACFLAP_NOTIF: Host 1ed0.1954.1f4f in vlan 1 is flapping between port Gi0/3 and port Gi0/12
Aug 16 18:19:27.712: %SW_MATM-4-MACFLAP_NOTIF: Host 1ed0.1954.1f4f in vlan 1 is flapping between port Gi0/12 and port Gi0/6
Aug 16 18:23:56.744: %SW_MATM-4-MACFLAP_NOTIF: Host 1ed0.1954.1f4f in vlan 1 is flapping between port Gi0/4 and port Gi0/5
In this example it was the same mac adress but every day it is a different one and sometimes even more then one. Every flapping causes that the internet is down while the WIFI signal stays strong.
Do you have an idea what could cause this and how we could handle this? Is there any setting we can change on either the
Main Cisco Switch
The 8 switches in the technical rooms
The 17 APs we are using?
Thanks in advance. I appreciate your support!
Cheers
Jakob
08-16-2022 03:40 PM
Is the mac address 1ed0.1954.1f4f a client device or something else?
I have seen this before if the wireless is running in flexconnect as the clients roam between switches it doesnt normally cause issues.
Does internet go down for wired and wireless clients?
08-17-2022 05:17 AM
Yes, its a client device - in this case a smartphone. It happens also for PCs or Laptops.
And yes: Internet does go down for wird & wireless Clients, so it seems to happen on the switch(es), not on the APs, as we also do not get any error in the Unifi Controller/ the USG Pro 4.
08-16-2022 04:09 PM
are you using default VLAN1 for wifi network ? i would advise to change to different VLAN (other than 1)
if this MAC belongs to end device, and you have roaming, that is normal most cases and you see the logs like you mentioned.
08-17-2022 05:18 AM
Yes we are using default VLAN 1 for the whole network. What does it change using another VLAN?
Can i somehow get a setting to "accept net flapping"?
08-17-2022 09:14 AM
If that is Wireless roaming client, you should accept.
changing from VLAN 1 to new VLAN, you need to create new VLAN and Interface and move all ports to VLAN X (new created)
08-17-2022 09:55 AM
thank you for the fast feedback.
could you explain, why that is going to solve the issue?
08-18-2022 01:07 AM
This should stabilize the VLAN 1 a little bit. Although those messages (the roaming) should cause no issues, as the message is just informative.
I'm a bit surprised that even the wired clients have problems, do you maybe have a drawing how everything is connected?
08-20-2022 12:28 AM
08-21-2022 01:20 PM
Yeah in this scenario the port flap messages are quite normal. If client A moves from the second left AP to the third and back (for roaming reasons), you have a port flap message. This is nothing to worry about in this scenario and normal.
Now wired clients on the other hand should definitively not cause any such message while they are plugged in. Are you sure you have such errors?
08-22-2022 12:04 PM
Hi patoberli
thanks for your feedback. it sounds calming that this scenario is normal. the question now is, why the internet connection gets lost - for wifi and wired... yes i am sure that i have such errors...
Cheers,
Jakob
08-23-2022 11:06 PM
This is very weird and should not happen. I assume the Cisco doesn't log any other errors when this happens?
Do you have any features enabled on the ports, like BPDU Guard or Root Guard?
How long is the connection gone if it is gone?
08-23-2022 11:39 PM
No there are no other errors coming up.
BPDU/Root Guard are not enabled. Should i try this?
The connection is gone for about 20 seconds.
Cheers,
Jakob
08-27-2022 12:51 PM
You can, but I doubt it will help a lot. I think you might need to do a capture on the switch uplink port and try to recreate this issue. If you ping from client a to client b, both wired, will they also have an outage?
Also ping the gateway while you try this.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide