LAN, Switching and Routing 

Packet Loss - Every 1 in 15 Packets 


Feb 18, 2005, 1:54pm PST 
We have a switch gig6/2 directly connected to a router gig 3/2. When
we ping across this connection, we drop every 15th packet. We've
researched and can find nothing wrong.
Any ideas? 
Is it fair to assume this is a config issue (not hardware)? 
Is the mtu setting on the interface the issue? (mtu is set like this
on other ports working fine)

switch3#ping 
Protocol [ip]: 
Target IP address: 172.18.xx.yy1 
Repeat count [5]: 100 
Datagram size [100]: 1500 
Timeout in seconds [2]: 
Extended commands [n]: y 
Source address or interface: 172.18.xx.yy2 
Packet sent with a source address of 172.18.xx.yy2 
!!!!!!!!!!!!!!.!!!!!!!!!!!!!!.!!!!!!!!!!!!!!.!!!!!!!!!!!!!!.!!!!!!!!!!!!!!.!!!!!!!!!!!!!!.!!!!!!!!!!
Success rate is 94 percent (94/100), round-trip min/avg/max = 1/1/4 ms 

On Switch 
GigabitEthernet6/2 is up, line protocol is up (connected) 
Hardware is C6k 1000Mb 802.3, address is 000f.24ee.1180 (bia 000f.24ee.1180) 
Description: GigE connection to CR2FITCWI Gig3/2 
Internet address is 172.18.xx.yy2/30 
MTU 1526 bytes, BW 1000000 Kbit, DLY 10 usec, 
reliability 255/255, txload 1/255, rxload 1/255 
Encapsulation ARPA, loopback not set 
Keepalive set (10 sec) 
Full-duplex, 1000Mb/s, media type is SX 
Media-type configured as SFP connector 
input flow-control is off, output flow-control is on 
Clock mode is auto 
ARP type: ARPA, ARP Timeout 04:00:00 
Last input 00:00:02, output 00:00:00, output hang never 
Last clearing of "show interface" counters 37w5d 
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 3000 bits/sec, 1 packets/sec 
5 minute output rate 0 bits/sec, 0 packets/sec 
L2 Switched: ucast: 612 pkt, 39168 bytes - mcast: 5849176 pkt, 1439036257 bytes 
L3 in Switched: ucast: 265227088 pkt, 68413357525 bytes - mcast: 0
pkt, 0 bytes mcast
L3 out Switched: ucast: 439526622 pkt, 101313141953 bytes mcast: 0 pkt, 0 bytes 
382115122 packets input, 84882675436 bytes, 0 no buffer 
Received 5686 broadcasts, 0 runts, 11817162 giants, 
0 input packets with dribble condition detected 
476732950 packets output, 109537179878 bytes, 

interface GigabitEthernet6/2 
description GigE connection to CR2FITCWI Gig3/2 
mtu 1526 
ip address 172.18.xx.yy2 255.255.255.252 
ip router isis 49 
speed nonegotiate 
tag-switching ip 
isis metric 30 


On Router 
sh int gig 3/2 
GigabitEthernet3/2 is up, line protocol is up 
Hardware is GigMac 3 Port GigabitEthernet, address is 0008.e23d.18c2
(bia 0008.e23d.18c2)
Description: GigE connection to Tel7609 S6p2 
Internet address is 172.18.xx.yy1/30 
MTU 1526 bytes, BW 1000000 Kbit, DLY 10 usec, rely 255/255, load 1/255 
Encapsulation ARPA, loopback not set 
Keepalive set (10 sec) 
Full-duplex mode, link type is force-up, media type is SX 
output flow-control is unsupported, input flow-control is off 
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 
Queueing strategy: fifo 
Output queue 0/40, 0 drops; input queue 0/75, 0 drops 
5 minute input rate 5000 bits/sec, 0 packets/sec 
5 minute output rate 0 bits/sec, 0 packets/sec 
477606053 packets input, 107711899149 bytes, 0 no buffer 
Received 1995932 broadcasts, 3 runts, 140912 giants, 0 throttles 
6 input errors, 
0 watchdog, 7259704 multicast, 0 pause input 
382988101 packets output, 83406662197 bytes, 5 interface reset 
6 lost carrier, 0 no carrier, 0 pause output 


interface GigabitEthernet3/2 
description GigE connection to Tel7609 S6p2 
mtu 1526 
ip address 172.18.xx.yy1 255.255.255.252 
no ip directed-broadcast 
ip router isis 49 
no negotiation auto 
tag-switching ip 
isis metric 30 level-1

solid_foundry-ga,

I reviewed the interface configs on the devices and noticed the MTU on
both is set to 1526. This 1526 increases the payload of the frame
beyond the "normal" payload. I recommend changing the MTU back to the
default 1500 on both sides of the connection. Both interfaces are
recording giants due to the MTU 1526. The tag-switching adds to the
already over sized Ethernet frame and can cause some unexpected
results. Tag-switching increases a normal 1500 frame is to 1518 frame.
The reason it does not work to the router is hard to say without
knowing more about the router. It could be a bug or a hardware problem
with the interface, but I was start by changing the MTU to 1500 on
both sides of the connection. Best network design, implementation,
installation practices included using default setting whenever
possible, this includes MTU settings. RFC 1191 address MTU.

Also, the speed between the router and switch should be hard coded to
Gig, and not left to negotiation auto on the GigE6/2. I have seen Auto
Neg cause problems with network connectivity.

I would also try sending various size PING packets across the link and
monitor the rate of sucess/loss. Record the results and use it to
determine if this corrects the problem. If all else fail, contact
Foundry. I have worked with Foundry TAC. Make sure you have a current
contract with Foundry before calling TAC.

Good luck.