Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. If you'd prefer to explore, try our test area to get started. And see here for current known issues.

Silver

Packet loss when routing between VLANs

I have a Catalyst 2950C switch running 12.1(13)EA1 code, connected via a dot1q trunk to a 2620-XM router, running 12.1(18). The router is routing between three different VLANs. From the switch, when I ping the default gateway (on the router) I will occasionally see large pings result in some packet loss (e.g. using extended pings with large packet sizes). I've tried the same test using a Cisco 2620 and a Catalyst 2900XL and the large pings work without any packet loss. The physical interfaces between the 2620-XM and 2950C switch are not reporting any errors (input/output errors, CRCs or late collisions), and both ends are hardwired to 100M/full-duplex. The only pointer is a large number of "deferred" frames (~5000) on the FastEthernet interface of the switch to which the router is connected. No such deferred frames have been identified on the corresponding switch port of 2620/Catalyst 2900XL (and in fact other similar 2620/2950 configurations) switch arrangements. Are these deferred frames significant?

(The only other difference between this and other configurations is a second dot1q trunk which links this 2950C switch to another 2950C switch via a 100BaseFX link. This trunk carries only one VLAN for the subnet in another building).

3 REPLIES
Silver

Re: Packet loss when routing between VLANs

Running Full Duplex you should not see any deferred on this interface. The deferred counter counts the number of times the interface has tried to send a frame, but found the carrier busy at the first attempt (Carrier Sense). This does not constitute a problem, and is part of normal Ethernet operation. Since Full Duplex operation disables Carrier Sense this is interesting. Can you verify that your duplex settings and speed is set correctly on both sides.

Regards,

Don

Silver

Re: Packet loss when routing between VLANs

Yes, both the router and its assocaited switch port are hardwired to 100Mbps and full-duplex..

Switch configuration:

interface FastEthernet0/24

switchport mode trunk

no ip address

duplex full

speed 100

!

switch#sh int fas0/24

FastEthernet0/24 is up, line protocol is up (connected)

Hardware is Fast Ethernet, address is 000b.be60.5958 (bia 000b.be60.5958)

MTU 1500 bytes, BW 100000 Kbit, DLY 1000 usec,

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

Encapsulation ARPA, loopback not set

Keepalive set (10 sec)

Full-duplex, 100Mb/s

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

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

Last input 00:00:39, output 00:00:01, output hang never

Last clearing of "show interface" counters never

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 ouxtput rate 0 bits/sec, 0 packets/sec

1525587 packets input, 698137821 bytes, 0 no buffer

Received 15676 broadcasts, 0 runts, 0 giants, 0 throttles

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

0 watchdog, 14888 multicast, 0 pause input

0 input packets with dribble condition detected

2643108 packets output, 788002276 bytes, 0 underruns

0 output errors, 0 collisions, 2 interface resets

0 babbles, 0 late collision, 7309 deferred

0 lost carrier, 0 no carrier, 0 PAUSE output

0 output buffer failures, 0 output buffers swapped out

There are no deferred frames on the FastEthernet0/0 interface of the router which connects to port fa0/24 of the switch. Switch port fa0/23 connects this site back to the core network over a 10Mbps wide-area (LAN-extension) link. This link is hardwired (at both ends) to 10Mbps full-duplex, but deferred frames are also been seen on this port.

switch#sh int fas0/23

FastEthernet0/23 is up, line protocol is up (connected)

Hardware is Fast Ethernet, address is 000b.be60.5957 (bia 000b.be60.5957)

Description: 10Mbps LES link

MTU 1500 bytes, BW 10000 Kbit, DLY 1000 usec,

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

Encapsulation ARPA, loopback not set

Keepalive set (10 sec)

Full-duplex, 10Mb/s

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

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

Last input 00:00:01, output 00:00:05, output hang never

Last clearing of "show interface" counters never

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 ouxtput rate 0 bits/sec, 0 packets/sec

1351759 packets input, 538246911 bytes, 0 no buffer

Received 428674 broadcasts, 0 runts, 0 giants, 0 throttles

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

0 watchdog, 428663 multicast, 0 pause input

0 input packets with dribble condition detected

889366 packets output, 263284692 bytes, 0 underruns

0 output errors, 0 collisions, 2 interface resets

0 babbles, 0 late collision, 26336 deferred

0 lost carrier, 0 no carrier, 0 PAUSE output

0 output buffer failures, 0 output buffers swapped out

I doubt whether it's a cabling issue (as it appears to be affecting more than one switch port configured for full duplex operation), and would expect to see other error port counters incrementing.

Silver

Re: Packet loss when routing between VLANs

I've had time to build this in the lab, using the same configs from the production environment. I believe I have managed to recreate the problem. I have downgraded the 2950C switch firmware to IOS 12.1(12c)EA1 and it appears to have cured the problem; I am no longer seeing "deferred" frames on full-duplex links, and I've ping'ed to and from the switch with large packets without a single packet drop (according to the extended ping output).

402
Views
0
Helpful
3
Replies