Cisco C2960X - clients getting DHCP errors

[it inx]

Hello, i am having issues with connections on our 4 stack switches C2960X, Windows clients getting errors on startup or on reconnecting to the network and are not receiving a DHCP address of our Windows DHCP server. They are getting an automatic ipaddress 169.x.x.x, when doing ipconfig /release and renew, then they get an ipaddress of the DHCP server.

The error in the clients System - Eventlog is id:4199. It says that there is an ipaddress conflict with an other device, and the macaddress of that "other device" is in fact the lan-interface of the switch where the computer is connected. 

The system has detected an address conflict for the IP address 10.0.13.2, where the system has the network hardware address 00-41-D2-DA-54-11. This may disrupt the operation of the network on this system.

So why is the lan-interface of the switch holding the ipaddress?

When testing on a different VLAN and with another DHCP we are getting the same errors, so i don't think the problem is the DHCP server.

We have many more other Cisco C2960X switches, clients on those switches are getting without any problem DHCP addresses from the same DHCP-server.

Anybody know what the issue could be?

Kind regards

[Mark Elsen]

 - @it inx  Not accepted : assigning an ip address based on the mac address of a device is the essence of what a DHCP server does Assigning any kind of address is always possible.

 M.

[pieterh]

if the port connects to a single host and only uses vlan1 then you need to configure the port as edge-port/access-port
with the current configuration, the switch first tries to negotiate if this port needs to be a (dot1q)trunk or access port.
during this negotiation, the connected host allready started DHCP discovery, which times out -> APIPA address (169.x.x.x)
hence the errors, which disappear after a while when the host starts a new DHCP request cycle
and the switch now has determined this port is not connected to another switch
and dynamically configured this port as an access-port, the DHCP-request can succeed

for configuring using the CLI you can read:  https://www.cisco.com/c/en/us/td/docs/switches/lan/catalyst2960x/software/15-2_2_e/int_hw_components/configuration_guide/b_int_1522e_2960x_cg/b_int_152ex_2960-x_cg_chapter_010.html#con_1107751 
(or search for similar document for other software version running on your switch
https://www.cisco.com/c/en/us/support/switches/catalyst-2960-x-series-switches/series.html 

Configure switch ports by using the switchport interface configuration commands.

• Access Ports
• Trunk Ports

[it inx]

Hi @pieterh. In the 4th screenshot the port is configured as an edge-port already. Or am i misinterpreting something?

[it inx]

What is also want to add, when we put another switch (with no managment capabilities - HP Officeconnect 1420) between the clients and the Cisco switch, then there is no problem. Clients receive directly the dhcp address. 

[marcinkirszniak]

Maybe that will help:

https://www.cisco.com/c/en/us/support/docs/ios-nx-os-software/8021x/116529-problemsolution-product-00.html