cat 6500 problem on 10b port

julxu · ‎10-24-2007

I have upgraded catalyst 6500 IOS to new version (s72033-ipservicesk9-mz.122-33.SXH.bin) and than, all the module SFM-capable 48-port 10/100 Mbps RJ45 have some strange problem.

The ports on the module will have many input errors, and CRC; the server behind the ports also will lose real connection to the network. However, if you do â€œshow int statusâ€ on the cat6500, it still connected, and if you log into the server (which linked to the port) on console, it still recons the connection available.

I have also found I can easily fix the problem by shut/noshut on the port. But, the problem will occur later. And also, there is always some CPU intensive sometime before with no reason.

Could anyone give me some tip what problem it is? And what I can do with the problem?

Any comments will be appreciated

Thanks in advance

lamav · ‎10-24-2007

Hi.

It sounds like you have a duplex mismatch problem as a result of autonegotiation.

Do one of two things:

1. Hardcode both the server NIC and the switch port for speed and duplex (example: 100/full or 1000/full).

2. Set both the server NIC and the switchport to "auto". Sometimes, though, the server NIC's autonegotiation protocol may have added proprietary features that the Cisco switch does not support -- and that may cause both sides to lose sync when the periodic renegotiation takes place. The industry has had a problem standardizing the Autonegotiation protocol.

HTH

Victor

lamav · ‎10-24-2007

Oh yeah, never set one side of the connection to "auto" while hardcoding the other side. It only works if you are dealing with a Gigabitethernet, otherwise you will have a duplex mismatch.

julxu · ‎10-25-2007

I have no good luck.

the server is as5200, configure as:

interface Ethernet0

ip address 10.1.1.1 255.255.255.0

no ip redirects

no ip directed-broadcast

no ip route-cache

no ip mroute-cache

the port on cat6500:

interface FastEthernet9/4

description as5200

switchport

switchport access vlan 196

switchport mode access

wrr-queue bandwidth 20 100 200

wrr-queue random-detect max-threshold 3 90

wrr-queue cos-map 1 1 0

wrr-queue cos-map 2 1 1 2

wrr-queue cos-map 3 1 3 4 6 7

mls qos vlan-based

spanning-tree portfast

before the cat6500 is hardcoded, and I have changed to auto 4 hours before, now, the problem came back.

Any advice?

Julie

lamav · ‎10-25-2007

can you post the speed and duplex seetings for the server and switch?

julxu · ‎10-25-2007

There is no actually configuration for speed and duplex for the ports. By default, it is auto. so, the setting on both are auto, am I right? Also, there is no speed/duplex related configure in globel.

cat6500 sh int

++++++++++++++

FastEthernet9/4 is up, line protocol is up (connected)

Hardware is C6k 100Mb 802.3, address is 0013.1a04.8903 (bia 0013.1a04.8903)

Description: caracalla

MTU 1500 bytes, BW 10000 Kbit, DLY 100 usec,

reliability 255/255, txload 1/255, rxload 1/255

Encapsulation ARPA, loopback not set

Keepalive set (10 sec)

Half-duplex, 10Mb/s, media type is 10/100BaseTX

input flow-control is off, output flow-control is unsupported

ARP type: ARPA, ARP Timeout 04:00:00

Last input 00:00:32, output 00:00:43, output hang never

Last clearing of "show interface" counters 1w0d

Input queue: 0/2000/0/0 (size/max/drops/flushes); Total output drops: 0

Queueing strategy: fifo

Output queue: 0/40 (size/max)

5 minute input rate 0 bits/sec, 0 packets/sec

5 minute output rate 0 bits/sec, 0 packets/sec

142018 packets input, 105758743 bytes, 0 no buffer

Received 3200 broadcasts (3169 multicasts)

0 runts, 0 giants, 0 throttles

72481 input errors, 64295 CRC, 0 frame, 0 overrun, 0 ignored

0 watchdog, 0 multicast, 0 pause input

0 input packets with dribble condition detected

144305 packets output, 45555226 bytes, 0 underruns

0 output errors, 9 collisions, 4 interface resets

0 babbles, 0 late collision, 0 deferred

0 lost carrier, 0 no carrier, 0 PAUSE output

0 output buffer failures, 0 output buffers swapped out

as5200 show int

+++++++++++++++

Ethernet0 is up, line protocol is up

Hardware is Lance, address is 00e0.1e88.7769 (bia 00e0.1e88.7769)

Internet address is 10.2.2.2/24

MTU 1500 bytes, BW 10000 Kbit, DLY 1000 usec, rely 255/255, load 1/255

Encapsulation ARPA, loopback not set, keepalive set (10 sec)

ARP type: ARPA, ARP Timeout 04:00:00

Last input 00:00:00, output 00:00:00, output hang never

Last clearing of "show interface" counters never

Input queue: 1/75/125/0 (size/max/drops/flushes); Total output drops: 0

Queueing strategy: fifo

Output queue 0/40 (size/max)

5 minute input rate 1000 bits/sec, 2 packets/sec

5 minute output rate 1000 bits/sec, 1 packets/sec

393597 packets input, 251894293 bytes, 70 no buffer

Received 64658 broadcasts, 0 runts, 0 giants, 125 throttles

15 input errors, 0 CRC, 0 frame, 0 overrun, 15 ignored, 0 abort

0 input packets with dribble condition detected

623309 packets output, 223481043 bytes, 0 underruns

9 output errors, 259 collisions, 256 interface resets

0 babbles, 0 late collision, 500 deferred

9 lost carrier, 0 no carrier

0 output buffer failures, 0 output buffers swapped out

#sh int statu m 9

Port Name Status Vlan Duplex Speed Type

Fa9/4 caracalla connected 196 a-half a-10 10/100BaseTX

Many Regards

julxu · ‎10-25-2007

Just add something.

And also, it seems cat6500 lost MAC address of the server first.