Upgraded to a fiber trunk, much slower w/ errors

kent.barnes · ‎07-14-2008

I am seeing terrible network speeds after upgrading from a copper VLAN trunk to a fiber trunk. I also see alot of CRC errors on the mini-GBIC port.

The switches are Cisco Catalyst 3560's. Old config was 2 VLANS (we went through our landlords network to get between floors) one for voice, one for data. We did not want to have to piggy back on another network, so we had SM fiber installed. After switching the that fiber, network speed went to a crawl.

We have yet to test the fiber lines, that will be done soon and I do beleive bad media is a likely culprit.

Here is the config for our Core switch as well as the port the fiber was on...

Port g0/25 was the troublesome uplink.

!

interface GigabitEthernet0/25

description Uplink to SAS-SW-1K gb port 1

switchport trunk encapsulation dot1q

switchport mode trunk

speed nonegotiate

mls qos trust cos

ALso, here is the show interface for g0/25

SAS-SW-3L#show int g0/25

GigabitEthernet0/25 is down, line protocol is down (notconnect)

Hardware is Gigabit Ethernet, address is 0019.066d.af19 (bia 0019.066d.af19)

Description: Uplink to SAS-SW-1K gb port 1

MTU 1500 bytes, BW 1000000 Kbit, DLY 10 usec,

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

Encapsulation ARPA, loopback not set

Keepalive not set

Auto-duplex, Auto-speed, link type is force-up, media type is 1000BaseLX SFP

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

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

Last input 6d16h, output 6d16h, output hang never

Last clearing of "show interface" counters 3w3d

Input queue: 0/75/0/0 (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

94266365 packets input, 3342267418 bytes, 0 no buffer

Received 364398 broadcasts (0 multicast)

0 runts, 0 giants, 0 throttles

64612 input errors, 928 CRC, 0 frame, 0 overrun, 0 ignored

0 watchdog, 166576 multicast, 0 pause input

0 input packets with dribble condition detected

Here is the first floor switch, Port g0/1 was the fiber uplink...

interface GigabitEthernet0/1

description Uplink to SAS-SW-3L gb port 25

switchport trunk encapsulation dot1q

switchport mode trunk

speed nonegotiate

mls qos trust cos

Here is the show int for g0/1

GigabitEthernet0/1 is down, line protocol is down (notconnect)

Hardware is Gigabit Ethernet, address is 0013.c4ea.e301 (bia 0013.c4ea.e301)

Description: Uplink to SAS-SW-3L gb port 25

MTU 1500 bytes, BW 1000000 Kbit, DLY 10 usec,

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

Encapsulation ARPA, loopback not set

Keepalive not set

Auto-duplex, Auto-speed, link type is force-up, media type is 1000BaseLX SFP

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

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

Last input 6d16h, output 6d16h, output hang never

Last clearing of "show interface" counters 3w5d

Input queue: 0/75/0/0 (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

147601720 packets input, 3186375337 bytes, 0 no buffer

Received 9172008 broadcasts (0 multicast)

0 runts, 0 giants, 0 throttles

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

0 watchdog, 4123102 multicast, 0 pause input

0 input packets with dribble condition detected

Any idea as to what would cause all the errors on the core switch? the CRC and input errors? Could it be more than just bad media? We are still in the process of troubleshooting and testing, but any suggestions would be appreciated, thanks.

fherlan · ‎07-14-2008

Hi.

Did you try to lock the speed to 1000/FDX ?

Regards

Frank

s.marino · ‎07-14-2008

I would check the following:

Clean the fiber ends on the fiber patch cords and at the patch panels. If you don't have a tester then try to use a new fiber patch cords. See that helps, also - Have you rebooted the switches after changing the media from copper to fiber? If not, I would give that a try.

Sal

kent.barnes · ‎07-14-2008

I will give that a try, thank you. I the switches have been booted many times since the swap, and I do not have any spare patch cables.

tdrais · ‎07-14-2008

You appear to only see errors in one direction so it may be a single strand. A little hard to be sure since its all down/down

Try flipping the fiber just at the switch on both ends. Ie swap transmit and receive on both switches but do not change the end that is connected to the building fiber.

If the problem does not move then you have something other than a fiber problem.

Now swap the fiber at the connection to the building on both ends. If the problem does not move you have a issue with the building fiber if it does move one of your patch cables is bad.

Single mode is much more sensitive to alignment so make sure they did not do something like put in a multimode patch panel or that you have a multimode patch cable someplace in the path. If its all mulitmode it would work for "short" runs but mixing media tends to cause lots of issues.