Cisco 4948 switches excessive drops

balla-zoltan · ‎01-26-2012

Hi,

We have some issues with our 4948 switches namely there are a lot of drops on interfaces. Here is an example:

GigabitEthernet1/19 is up, line protocol is up (connected)

Hardware is Gigabit Ethernet Port, address is 001b.d417.a792 (bia 001b.d417.a792)

Description: USPHL0ESX18 RSA

MTU 1500 bytes, BW 100000 Kbit, DLY 100 usec,

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

Encapsulation ARPA, loopback not set

Keepalive set (10 sec)

Full-duplex, 100Mb/s, link type is auto, media type is 10/100/1000-TX

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

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

Last input never, output never, output hang never

Last clearing of "show interface" counters 6d17h

Input queue: 0/2000/0/0 (size/max/drops/flushes); Total output drops: 17577272

Queueing strategy: fifo

Output queue: 0/40 (size/max)

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

5 minute output rate 23000 bits/sec, 33 packets/sec

3083 packets input, 202707 bytes, 0 no buffer

Received 299 broadcasts (0 multicasts)

0 runts, 0 giants, 0 throttles

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

0 input packets with dribble condition detected

86584193 packets output, 52059988888 bytes, 0 underruns

0 output errors, 0 collisions, 0 interface resets

0 babbles, 0 late collision, 0 deferred

0 lost carrier, 0 no carrier

0 output buffer failures, 0 output buffers swapped out

Here is the interface configuration:

interface GigabitEthernet1/19

description USPHL0ESX18 RSA

switchport access vlan 120

speed 100

duplex full

spanning-tree portfast

The drops happen in a burst. The case was opened with TAC, but it looks like we have a run around.

Thanks

Fabienne Stephanoff · ‎02-01-2012

Hello,

Did you capture a sniffer trace of this interface while the drops were occurring? This looks like oversubscription. A capture would tell you for sure using the wireshark I/O graph feature.

Is qos enabled?

Cheers,

Fabienne

LAN-Switching TAC

balla-zoltan · ‎02-01-2012

Thank you for your esponse. We have no empty ports o use wireshark. The interesting part is that there is only one device plugged into the port. It is the RSA card for that server is only for management, so, in theory, should not have a lot of traffic. QoS is not enabled on the port.

glen.grant · ‎02-01-2012

How is the far end port configured? Seeing you hardcoded the 4948 port the far end would have to be hardcoded also otherwise the far end will default to 100/half duplex . Output drops usuaully means the far end cannot handle the traffic fast enough .

balla-zoltan · ‎02-01-2012

Thank you for your response. The interface on the server is set to 100/full, just like the switch port. The interface does not have a lot of traffic because the RSA card is only for server management.

jkukorica · ‎10-01-2012

Szia!

Jutottál valamire az esetedben? Nekünk is van 4948-as és növekvő output drop, amit semmi sem indokol.

gigabitethernet 1/41

Power consumed by backplane : 0 Watts

Mod Ports Card Type Model Serial No.

---+-----+--------------------------------------+------------------+-----------

1 48 1000BaseX (SFP) Supervisor WS-C4948

M MAC addresses Hw Fw Sw Status

--+--------------------------------+---+------------+----------------+---------

1 001f.ca75.72c0 to 001f.ca75.72ef 1.5 12.2(31r)SGA 12.2(54)SG1 Ok

Rövid idővel nullázás után !

* GigabitEthernet1/41 0 0 0 457632 67000 4 92921000 14218 0

* GigabitEthernet1/42 0 0 0 0 90135000 9246 13000 7 0

* GigabitEthernet1/43 0 0 0 0 1568000 786 53448000 8173 0

* GigabitEthernet1/44 0 0 0 0 1186000 315 60776000 9244 0

* GigabitEthernet1/45 0 0 0 0 4807000 1667 62027000 8756 0

* GigabitEthernet1/46 0 0 0 0 4144000 1733 58523000 8508 0

* GigabitEthernet1/47 0 0 0 0 0 0 0 0 0

Köszi!

Joseph W. Doherty · ‎10-01-2012

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

The drops happen in a burst. The case was opened with TAC, but it looks like we have a run around.

If drops happen in a burst, then you might be experiencing traffic bandwidth bursts. If so, drops can happen.

Can't run port at gig? If not, next alternate would be increase buffer resources.