Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements
Webcast-Catalyst9k
New Member

Cisco 2960x Stack flapping

I ran into a problem with a Cisco 2960 stack.

 

To me it looks like there is a bug in the 15.2.2E firmware (c2960x-universalk9-mz.152-2.E.bin) regarding stacking.

 

The stack contain two switches 2960x with IP-Lite license.

The cabling was done according to the best practice as redundant cabling as in the document Figure 4. http://www.cisco.com/c/en/us/products/collateral/switches/catalyst-2960-x-series-switches/white_paper_c11-728327.html

 

 

After about 24 to 48 hours the whole network goes down and is not responsive anymore. The switches do not respond to any ipcm pings.

 

The Log contains:

%SW_MATM-4-MACFLAP_NOTIF: Host 4437.e667.34b9 in vlan 1 is flapping between port Te2/0/1 and port Te2/0/2
%PLATFORM_IPC-3-COMMON: Unknown IPC message type 65535 size 60
-Traceback= 539544z 27E3D4Cz 271C3B4z 271EA78z 27049D8z 26FBF5Cz 26FBF04z 2704C5Cz 270543Cz 26FC00Cz 23A65D8z
%SW_MATM-4-MACFLAP_NOTIF: Host 4437.e658.4564 in vlan 1 is flapping between port Te2/0/1 and port Te2/0/2
%SW_MATM-4-MACFLAP_NOTIF: Host 4437.e65e.2345 in vlan 1 is flapping between port Te2/0/2 and port Te2/0/1
%SW_MATM-4-MACFLAP_NOTIF: Host 70f3.9503.5234 in vlan 1 is flapping between port Te2/0/2 and port Te2/0/1
%SW_MATM-4-MACFLAP_NOTIF: Host 0090.3302.6665 in vlan 1 is flapping between port Te2/0/2 and port Te2/0/1
%SW_MATM-4-MACFLAP_NOTIF: Host 70f3.9503.2345 in vlan 1 is flapping between port Te2/0/1 and port Te2/0/2
%SW_MATM-4-MACFLAP_NOTIF: Host 4437.e666.9796 in vlan 1 is flapping between port Te2/0/2 and port Te2/0/1
%SW_MATM-4-MACFLAP_NOTIF: Host 4437.e65e.3246 in vlan 1 is flapping between port Te2/0/2 and port Te2/0/1
%SW_MATM-4-MACFLAP_NOTIF: Host fc4d.d44a.5643 in vlan 1 is flapping between port Te2/0/1 and port Te2/0/2

 

After a reboot of the stack slave, the switch hangs in the redundant configuration at the Stack-master selection. If you disable one of the two redundant stack ports, the stack starts to work again as it should be.

%STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 1 has changed to state DOWN
%STACKMGR-4-SWITCH_ADDED: Switch 2 has been ADDED to the stack
%STACKMGR-5-SWITCH_READY: Switch 2 is READY

 

The operation is complete normal, with a non redundant configuration.

If you try to enable the redundant configuration again, the problem immediately escalates again.

%STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 1 has changed to state UP
%IPC-2-INVALIDZONE: Invalid IPC Zone 0xA0000000.
-Traceback= 539544z 10040D4z 27E3CC0z 271C3B4z 271EA78z 27049D8z 26FBF5Cz 26FBF04z 2704C5Cz 270543Cz 26FC00Cz 15BA6CCz 15BA690z 15BABDCz 15B7DFCz 1F5D004z
%IPC-2-INVALIDZONE: Invalid IPC Zone 0xA0000000.
-Traceback= 539544z 10040D4z 27E3CC0z 271C3B4z 271EA78z 27049D8z 26FBF5Cz 26FBF04z 2704C5Cz 270543Cz 26FC00Cz 2DE3E7Cz
 

 

 

 

My temporary solution to that is for the moment is the run the stack cabling as non-redundant as shown in Figure 6 with the 15.2.2.E firmware.

 

The next step is to downgrade the firmware to c2960x-universalk9-mz.150-2.EX5.bin

 

Update: The downgrade to 15.0.2.EX5 did solve the flapping problem.

 

 

Everyone's tags (1)
1 ACCEPTED SOLUTION

Accepted Solutions
New Member

you can deactivate a stack

you can deactivate a stack port by this command:

 

 

switch 1 stack port 2 disable

 

 

 

and enable the port again with:

 

switch 1 stack port 2 enable 

 

2 REPLIES
New Member

Seems that the switch hangs

Seems that the switch hangs because it gets overloaded by a traffic loop.

Does the switch stack with redundancy works ok on a different software version?

New Member

you can deactivate a stack

you can deactivate a stack port by this command:

 

 

switch 1 stack port 2 disable

 

 

 

and enable the port again with:

 

switch 1 stack port 2 enable 

 

962
Views
5
Helpful
2
Replies
CreatePlease to create content