CAT 3500 Switch issue

Unanswered Question
Sep 10th, 2009

HI,

In our network we have ont CAT 6513 as the core switch and from the core switch uplinks connected two access switches CAT 3500 A and B

Switch CAT 3500 A with the vlan 5

Switch CAT 3500 B with the vlan 6

and we interconnected between the two CAT 3500 with the uplink

servers with the two NIC cards will be connected on the Switch CAT 3500 A and Switch CAT 3500 B and we will do teaming the Both server NIC card so redunacncy will be there

and reason also our internal VLAN traffic will be within the CAT 3500 not go through the core switch

Obousily STP is enabled and blocked one port

everything working fine but,

now later 6 months we have got issue that servers connected to the both the switches CAT 3500 are getting packet loss

and we try to login to the switches its hanged and not through console port also

might be i thought both the switches are gone to High cpu utlization

can u suggest us is there any issue with the network design and why its happened for both switches

any broadcast stroam

thanks

I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
MUHAMMAD SHAHEEN Thu, 09/10/2009 - 03:35

Hi,

as far as i know, you cannot do cross switch teaming on these standalone switches. Check the server if it is set to active/active in teaming configuration then you do have network loop/ spanning tree issue.

try changing to your server team as active/standby and that may help.

You may need to disconnect one of the teamed nic from the switch as your switches are seeing same mac address on 2 different switches.

For teaming to work as active/active, You have to use Chassis and plug each server nic into different blade slots, or you can use 3750 switches if they are stacked.

Hope this helps....

Shaheen

vinoth.kumar Fri, 09/11/2009 - 07:00

HI,

As i seen in my both the switches it peaks up to 100% CPU utlization on the both switches

I have attached the

sh version - might be an IOS issue if that case can anyone suggest us the IOS code to fix up the issue

sh process CPU - were i found the below are some using the CPU utlization and what service is this?

35 14969376 11182157 1338 0.39% 0.39% 0.39% 0 Vegas Statistics

40 3048 47528503 0 0.00% 0.00% 0.00% 0 VegasPM

41 252 1871203 0 0.00% 0.00% 0.00% 0 VUR_MGR bg proce

sh proc cpu history - which shows it increase the high cpu utilization when ever the traffic increase and getting packet loss

Cisco IOS Software, C3550 Software (C3550-IPBASE-M), Version 12.2(25)SED, RELEASE SOFTWARE (fc1)

Copyright (c) 1986-2005 by Cisco Systems, Inc.

Compiled Sat 22-Oct-05 09:27 by yenanh

Image text-base: 0x00003000, data-base: 0x00A02660

System image file is "flash:c3550-ipbase-mz.122-25.SED/c3550-ipbase-mz.122-25.SED.bin"

18 7864 2651625 2 0.00% 0.00% 0.00% 0 Net Background

19 140 124864 1 0.00% 0.00% 0.00% 0 Logger

20 2812 3742200 0 0.00% 0.00% 0.00% 0 TTY Background

21 7336 3742214 1 0.00% 0.00% 0.00% 0 Per-Second Jobs

26 2188 3727386 0 0.00% 0.00% 0.00% 0 Vegas Storm Cont

27 39616 93549578 0 0.00% 0.00% 0.00% 0 Vegas LED Proces

28 0 1 0 0.00% 0.00% 0.00% 0 SCQ_PROCESS

29 6996 14909292 0 0.00% 0.00% 0.00% 0 RAM Access

30 50056 85250973 0 0.00% 0.00% 0.00% 0 SW Frame Ager

31 80 69 1159 0.00% 0.00% 0.00% 0 VLAN Info Update

32 0 1 0 0.00% 0.00% 0.00% 0 JADE Gbit lockup

33 828 7484340 0 0.00% 0.00% 0.00% 0 L2TM Process

34 2868 7484342 0 0.00% 0.00% 0.00% 0 L3TM

35 14969376 11182157 1338 0.39% 0.39% 0.39% 0 Vegas Statistics

36 64656 51973817 1 0.00% 0.00% 0.00% 0 HMATM Learn proc

37 636 3742206 0 0.00% 0.00% 0.00% 0 HMATM Age proces

38 56 106 528 0.00% 0.00% 0.00% 0 VL2MM

39 0 1 0 0.00% 0.00% 0.00% 0 Vegas Bridging

40 3048 47528503 0 0.00% 0.00% 0.00% 0 VegasPM

41 252 1871203 0 0.00% 0.00% 0.00% 0 VUR_MGR bg proce

42 0 2 0 0.00% 0.00% 0.00% 0 EAPoUDP Process

43 0 3 0 0.00% 0.00% 0.00% 0 CEF switching ba

44 5232 3742205 1 0.00% 0.00% 0.00% 0 PI MATM Aging Pr

45 0 2 0 0.00% 0.00% 0.00% 0 DTP Protocol

46 4 62374 0 0.00% 0.00% 0.00% 0 Port-Security

47 28 115 243 0.15% 0.03% 0.00% 0 Exec

50 20 374241 0 0.00% 0.00% 0.00% 0 Ethchnl

51 324 3742407 0 0.00% 0.00% 0.00% 0 UDLD

57 114648 1088242 105 0.00% 0.00% 0.00% 0 CDP Protocol

60 0 2 0 0.00% 0.00% 0.00% 0 AAA Dictionary R

61 24 31189 0 0.00% 0.00% 0.00% 0 DHCP Snooping

62 12 62374 0 0.00% 0.00% 0.00% 0 IP Input

63 0 1 0 0.00% 0.00% 0.00% 0 ICMP event handl

64 1647560 3789568 434 0.00% 0.00% 0.02% 0 Spanning Tree

65 1064 62416 17 0.00% 0.00% 0.00% 0 CEF background p

66 0 1 0

vinoth.kumar Fri, 09/11/2009 - 07:03

HI,

As i seen in my both the switches it peaks up to 100% CPU utlization on the both switches

I have attached the

sh version - might be an IOS issue if that case can anyone suggest us the IOS code to fix up the issue

sh process CPU - were i found the below are some using the CPU utlization and what service is this?

35 14969376 11182157 1338 0.39% 0.39% 0.39% 0 Vegas Statistics

40 3048 47528503 0 0.00% 0.00% 0.00% 0 VegasPM

41 252 1871203 0 0.00% 0.00% 0.00% 0 VUR_MGR bg proce

sh proc cpu history - which shows it increase the high cpu utilization when ever the traffic increase and getting packet loss

sh run -- running config

RACK19-SWA#sh version

Cisco IOS Software, C3550 Software (C3550-IPBASE-M), Version 12.2(25)SED, RELEASE SOFTWARE (fc1)

System image file is "flash:c3550-ipbase-mz.122-25.SED/c3550-ipbase-mz.122-25.SED.bin"

26 2188 3727386 0 0.00% 0.00% 0.00% 0 Vegas Storm Cont

27 39616 93549578 0 0.00% 0.00% 0.00% 0 Vegas LED Proces

35 14969376 11182157 1338 0.39% 0.39% 0.39% 0 Vegas Statistics

36 64656 51973817 1 0.00% 0.00% 0.00% 0 HMATM Learn proc

37 636 3742206 0 0.00% 0.00% 0.00% 0 HMATM Age proces

vinoth.kumar Fri, 09/11/2009 - 07:16

HI,

i forget a add a point that last 2 months back also we found the same issue as like the 3500 with the cisco 3750 switches also that getting imprpoer responce in terms of connectivity

its happening for the switches which is connected to sun solaris server

is there any issue with the sun OS and IOS

Leo Laohoo Sat, 09/12/2009 - 17:06

There are no issues with any clients running any OS when connected to any network appliance. If the you have observed that the switch is undergoing high CPU utilization, you should consider investigating where this issue is coming from. Do you have a loop issue? Have you tried moving the servers somewhere else? Are there any line errors on the switch ports?

The 3500XL's weakness is the lack of auto-negotiation capabilities. So speed/duplex can be a trial-and-error. Have you checked that end?

Actions

This Discussion