Swicth stack issue - lost connections

[acerda]

Hi All,

We have 4 switches connected under stack. The issue is basically that we are lossing the connections (flapping) we got a CPU history of 90% but without memory issue, It happens when the business time start and we have around 90% of the interfaces in used per users. I would like to get your suggest in order to do a workaround

The following is the current IOS for the switches

Switch Ports Model SW Version SW Image

------ ----- ----- ---------- ----------

* 1 52 WS-C3750-48P 12.2(25)SEE3 C3750-IPBASEK9-M

2 52 WS-C3750-48P 12.2(25)SEE3 C3750-IPBASEK9-M

3 52 WS-C3750-48P 12.2(25)SEE3 C3750-IPBASEK9-M

4 52 WS-C3750-48P 12.2(25)SEE3 C3750-IPBASEK9-M

And the show switch

SgoL3IntA4#sh switch

Current

Switch# Role Mac Address Priority State

--------------------------------------------------------

*1 Master 0019.300e.7c00 15 Ready

2 Member 0019.3034.fa00 14 Ready

3 Member 0019.3035.2e80 13 Ready

4 Member 0019.3035.2e00 12 Ready

5 Member 0000.0000.0000 0 Provisioned

Please advise anything

Thanks

Andres

[Joseph W. Doherty]

Have you reviewed possible issues for high CPU on 3750s?

e.g.

http://www.cisco.com/en/US/products/hw/switches/ps5023/products_tech_note09186a00807213f5.shtml

http://www.cisco.com/en/US/docs/switches/lan/catalyst3750/software/troubleshooting/cpu_util.html