Multilink (MPLS) intfc taking LOTS of input errors while T1s clean..?

mikesuter · ‎01-13-2009

I have a Multilink interface that is taking a bunch of input errors (accumulating rapidly) while the underlying T1s are running clean. Don't recall seeing this before, and despite my efforts searching can't seem to find anything relevant.

interface Multilink123

bandwidth 3088

ip address x.x.x.x 255.255.255.252

no peer neighbor-route

ppp chap hostname xxx

ppp multilink

ppp multilink fragment disable

ppp multilink group 123

service-policy output voice

interface Serial0/0/0:0

no ip address

encapsulation ppp

ppp chap hostname xxxxx

ppp multilink

ppp multilink group 123

!

interface Serial0/0/1:0

no ip address

encapsulation ppp

ppp chap hostname xxx

ppp multilink

ppp multilink group 123

#sh int multi 123 | incl errors

2991771 input errors, 0 CRC, 2991771 frame, 0 overrun, 0 ignored, 0 abort

0 output errors, 0 collisions, 0 interface resets

[ errors actively increasing.. ]

#sh int multi 123 | incl errors

2992394 input errors, 0 CRC, 2992394 frame, 0 overrun, 0 ignored, 0 abort

0 output errors, 0 collisions, 0 interface resets

[ underlying t1s are clean! ]

#sh int ser 0/0/0:0 | incl errors

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

0 output errors, 0 collisions, 0 interface resets

#sh int ser 0/0/1:0 | incl errors

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

0 output errors, 0 collisions, 0 interface resets

Has anyone seen this before?

bagvanliju · ‎01-14-2009

Hi

input errors are combination of all errors,If 2992394 input errors, 0 CRC, 2992394 frame,

try to reconfigure your Multilink,check individual T1's are error free by checking the loop.

ujil

2go2marketing · ‎01-15-2009

What is your MTU set to?

Giuseppe Larosa · ‎01-15-2009

Hello Mike,

try to use

show ppp multilink to see if there are error counters also here.

You have input errors all of them of type Frame (meaning CRC errors or a non integer number of octets)

Number of packets received incorrectly having a CRC error and a noninteger number of octets. On a serial line, this is usually the result of noise or other transmission problems.

see

http://www.cisco.com/en/US/docs/ios/12_0/interface/command/reference/irshowin.html#wp1020497

but the multilink is not a physical serial interface so

sh ppp multilink may help

Hope to help

Giuseppe

msuter · ‎01-15-2009

sh ppp multilink doesn't appear to show anything too serious:

RTR#show ppp multilink

Multilink123, bundle name is host1056

Endpoint discriminator is host1056

Bundle up for 21:46:17, total bandwidth 3072, load 12/255

Receive buffer limit 24000 bytes, frag timeout 1000 ms

0/0 fragments/bytes in reassembly list

6 lost fragments, 79602 reordered

0/0 discarded fragments/bytes, 0 lost received

0x15DA55 received sequence, 0x2B8600 sent sequence

Member links: 2 active, 2 inactive (max not set, min not set)

Se0/0/1:0, since 21:46:17

Se0/0/0:0, since 21:31:24

Se0/1/0:0 (inactive)

Se0/1/1:0 (inactive)

No inactive multilink interfaces

MTU is 1500

Thanks!

Giuseppe Larosa · ‎01-15-2009

Hello Mike,

I see something I don't like:

Se0/1/0:0 (inactive)

Se0/1/1:0 (inactive)

No inactive multilink interfaces

what if you remove ppp multilink on these unused interfaces ?

another note: I always used low numbers for multilink what if you configure again a ppp bundle but with a low index like 5 ?

(all right this is just a note)

Hope to help

Giuseppe