3750 Stack connected to ESA Hosts management ports Flapping

[cscheper1]

Gi1/0/1 - 1/0/6 and Gi2/0/1 - 2/0/6 are access interfaces and connected to 6 ESA hosts' management interfaces set to active/active HA.  1/0/2 and 2/0/1 are constantly flapping. 

STP shows all 12 interfaces Desg FWD Cost:4 as expected.

The ESA's default gateway is the VLAN Interface for all 12 access ports in the stack, and randomly some ESA hosts will lose connectivity to the gateway.

I also get:

Jan 8 08:01:15: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 1 has changed to state UP
Jan 8 08:01:19: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 1 Switch 2 has changed to state DOWN
Jan 8 08:01:20: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 1 Switch 2 has changed to state UP
Jan 8 08:01:27: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 1 Switch 2 has changed to state DOWN
Jan 8 08:01:28: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 1 Switch 2 has changed to state UP
Jan 8 08:01:28: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 1 has changed to state DOWN
Jan 8 08:01:29: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 1 has changed to state UP
Jan 8 08:01:50: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 1 has changed to state DOWN
Jan 8 08:01:51: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 1 has changed to state UP
Jan 8 08:02:07: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 1 Switch 2 has changed to state DOWN
Jan 8 08:02:08: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 1 Switch 2 has changed to state UP
Jan 8 08:02:11: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 1 Switch 2 has changed to state DOWN
Jan 8 08:02:12: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 1 Switch 2 has changed to state UP
Jan 8 08:02:15: %STACKMGR-4-STACK_LINK_CHANGE: Stack Port 2 Switch 1 has changed to state DOWN

I cannot figure out why these two interfaces are flapping.  Thoughts please?

[Mark Malone · Accepted solution]

You could have 2 separate issues there , the stack cable could be hardware they are known to fail , if you have a spare stack cable you could replace it to see , also there are known software bugs for stackports on 3750s so it could be software too but you could rule that out with and upgrade , if you run t6he show tech through cli analyser it may identify if a software problem on your specific version

Cisco Bug: CSCsj80738 - Catalyst 3750 stack experiences stack port flapping issue

, the issue with the 2 ports flapping , what exactly is happening , is it having stp issues at layer 2 flapping ? are the interfaces clean of errors ? are they setup any differently than the orher MGMT access ports ?

This will show you if there flapping at l2
SH SPANNing-tree DEtail | I IEEE|occur|from|is exec

[cscheper1]

Cisco Bug: CSCsj80738 - Catalyst 3750 stack experiences stack port flapping issue

Our 3750 switch if operated in a high temperature environment

, the issue with the 2 ports flapping , what exactly is happening , is it having stp issues at layer 2 flapping ? are the interfaces clean of errors ? 

Not flapping on the other end of the link.  Not STP.  The interfaces are clean: 

3 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 1600 multicast, 0 pause input
0 input packets with dribble condition detected
350871676 packets output, 50742358913 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets

are they setup any differently than the orher MGMT access ports ?

No, simple access ports.

This will show you if there flapping at l2 
SH SPANNing-tree DEtail | I IEEE|occur|from|is exec

VLAN0XXX is executing the rstp compatible Spanning Tree protocol
Number of topology changes 1546 last change occurred 02:34:14 ago
from StackPort1

But isn't that expected with the interfaces flapping?

[Mark Malone]

stp notifications for stackport can be normal as the port syncs stp between the switches through the stackport ,but I'm not an expert on how exactly its done just what ive read on it , when I check my stacks though I'm not seeing any stackports listed as stp changes at all

if you break into another switch on the stack and run the show span command again does it show anything different , you can do that by just issuing command #session x where x is the switch number in the stack

how are you seeing its flapping , is it just in logs on Cisco side  , does the MGMT port g1/0/1 or 2/0/1 have there own individual ip behind it you could run a ping to , to see how unstable it is

that bug id was just an example I would still run the show tech through the analyzer to rule out the software to be sure , its a free tool and quite powerful  

[cscheper1]

I am Digging!  Thank You.

[cscheper1]

To Clarify: ESAs, and vCenter Hosts.  Sorry.