dslreports logo
 
    All Forums Hot Topics Gallery
spc
Search similar:


uniqs
4570
meta
join:2004-12-27
00000

meta

Member

L2TPv3 MTU problem/question.

Good evening everyone (and happy halloween)

Im having a bit of confusion regarding some L2TPv3 configuration.

I have 2 directly connected routers (via SRP5/0) are tunneling their fastethernet1/0 interfaces via L2TPv3.

R1:
...
pseudowire-class asdf
 encapsulation l2tpv3
 ip local interface SRP5/0
 ip pmtu
!
interface FastEthernet1/0
 no ip address
 duplex full
 xconnect 1.1.1.2 41 encapsulation l2tpv3 pw-class asdf
!
interface SRP5/0
 ip address 1.1.1.1 255.255.255.248
!
...
 

R2:
...
pseudowire-class asdf
 encapsulation l2tpv3
 ip local interface SRP5/0
 ip pmtu
!
interface FastEthernet1/0
 no ip address
 duplex full
 xconnect 1.1.1.1 41 encapsulation l2tpv3 pw-class asdf
!
interface SRP5/0
 ip address 1.1.1.2 255.255.255.248
!
...
 

The interfaces have the appropriate MTU:
Router#     show int SRP5/0 | I MTU
  MTU 4470 bytes, BW 622000 Kbit, DLY 100 usec, 
Router#     show int fast1/0 | I MTU
  MTU 1500 bytes, BW 100000 Kbit, DLY 100 usec, 
 

When i ping from the switches on either end of fa1/0:
Switch#ping 2.2.2.2 size 1480
Sending 5, 1480-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 4/4/4 ms
 
Switch#ping 2.2.2.2 size 1481
Sending 5, 1481-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds:
.....
Success rate is 0 percent (0/5)
Switch#
 

It doesnt look like any fragmentation should be necessary. There should be a 1500 byte (or larger) path between the two switches even WITH l2tpv3 overhead.
Can anybody help me figure out whats going wrong with this setup?

Thanks.

TomS_
Git-r-done
MVM
join:2002-07-19
London, UK

TomS_

MVM

Can you ping the IP of the opposing tunnel end points with > 1500 bytes?

Maybe it takes the MTU command on the SRP interface, but doesnt actually do anything with it...?
meta
join:2004-12-27
00000

2 edits

meta

Member

Good thought, tried that:
Router#ping 1.1.1.2 size 4400 df-bit
Sending 5, 4400-byte ICMP Echos to 1.1.1.2, timeout is 2 seconds:
Packet sent with the DF bit set
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/4 ms
 
Router#ping 1.1.1.1 size 4400 df-bit
Sending 5, 4400-byte ICMP Echos to 1.1.1.1, timeout is 2 seconds:
Packet sent with the DF bit set
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/3/4 ms
Router#
 

I have also tried with and without "ip pmtu" in the pseudowire class.

In other news, when i do this on higher end equipment (7500s, 7600s, etc) with encapsulation mpls instead of l2tpv3 i have no MTU problem (providing the MPLS transit inbetween the endpoints can pass the full data without fragmenting it)

Is there any debugging or show commands i could use to detect packet drops due to size / unfragmentability? Or figure out what the L2TPv3 tunnel is trying to do? Theres no VTI with an MTU or anything so far as i could tell, i looked through the show l2tun commands and didnt see anything particularly interesting.