6xCat375V2 Stack - Losing packets.

[ALEXANDR LYKOV]

Hi!

We'd had 5 switches in the stack and everything was ok. but then we connected a new 6th switch so I can see periodical short loosing packets. It happened on many hosts connected to these stack at different time. Cpu load is ok and no any errors in logs. Any others debugs?

(From my host)

ping -t 10.8.8.xx

[...]

Reply from 10.8.8.xx: bytes=32 time<1ms TTL=127

Request timeout

Request timeout

Reply from 10.8.8.xx: bytes=32 time<1ms TTL=127

[...]

From this switch (vlan int)

SW01#ping 10.8.8.45 repeat 100000

[...]

!!!!!!!!!!!!!!!!!!!!....!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

[...]

Switch Ports Model              SW Version            SW Image                

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

     1 52    WS-C3750V2-48TS    12.2(58)SE2           C3750-IPSERVICESK9-M    

*    2 52    WS-C3750V2-48TS    12.2(58)SE2           C3750-IPSERVICESK9-M    

     3 52    WS-C3750V2-48TS    12.2(58)SE2           C3750-IPSERVICESK9-M    

     4 52    WS-C3750V2-48PS    12.2(58)SE2           C3750-IPSERVICESK9-M    

     5 52    WS-C3750V2-48TS    12.2(58)SE2           C3750-IPSERVICESK9-M    

     6 52    WS-C3750V2-48TS    12.2(58)SE2           C3750-IPSERVICESK9-M    

SW01#sh switch stack-ports

  Switch #    Port 1       Port 2

  --------    ------       ------

    1           Ok           Ok  

    2           Ok           Ok  

    3           Ok           Ok  

    4           Ok           Ok  

    5           Ok           Ok  

    6           Ok           Ok  

SW01#sh switch stack-ring speed

Stack Ring Speed        : 32G

Stack Ring Configuration: Full

Stack Ring Protocol     : StackWise

[ALEXANDR LYKOV]

only L2 switching. no routing.

[Vitaliy Zinatov]

Were taken into account the recommendations of this article?

http://www.cisco.com/en/US/products/hw/switches/ps5023/products_configuration_example09186a00807811ad.shtml

So can you print here Traceroute for 10.8.8.xx

The problem arises in the entire switch or specifically in hosts sitting on the 6th switch?

[ALEXANDR LYKOV]

Solved. CPU load was not OK on some switches in stack.

There was high Port-security load.

I had this config:

interface FastEthernet x/x/x

switchport access vlan 128

switchport mode access

switchport port-security

switchport port-security aging time 2

switchport port-security violation restrict

srr-queue bandwidth share 10 10 60 20

priority-queue out

mls qos trust device cisco-phone

mls qos trust cos

auto qos voip cisco-phone

macro description cisco-phone | cisco-phone | cisco-phone

spanning-tree portfast

spanning-tree bpduguard enable

service-policy input AutoQoS-Police-CiscoPhone

end

now I've changed it adding this:

switchport port-security aging time 5

switchport port-security aging type inactivity