[j-nsp] MTU problems over VPLS

[Eduardo Barrios]

We use RSVP LSPs and set the Adspec attribute to autodiscover the LSP MTU hop by hop during the path setup. We also set the MTU on the PE-CE facing interface. Finally set the VPLS mtu in the routing-instance --- It has to match on all PEs that we run the VPLS instance, we do this to interop with Alcatel.

HTH,
Eduardo


-----Original Message-----
From: juniper-nsp-bounces at puck.nether.net [mailto:juniper-nsp-bounces at puck.nether.net] On Behalf Of Luca Salvatore
Sent: Tuesday, February 12, 2013 12:58 AM
To: juniper-nsp at puck.nether.net
Subject: [j-nsp] MTU problems over VPLS

I have a few sites connected via a VPLS core.  The core devices are all MX 10 routers connected via 10Gb fibre.
I'm having problems doing file copies (SCP between two Centos VMs).

The issue is that the file copy never gets anywhere, on the Centos CLI it sits at 0% then says 'stalled'
To fix this issue I have just set the MTU on the Centos machines to be 1400 - when this is in place the copy works and I get nice speeds.

I don't believe I should have to modify the MTU though, shouldn't path MTU discovery take care of this?

For example - I have done some TCPdumps, I can see the sender is sending traffic with the DF bit set, however I don't see any 'ICMP fragmentation needed' coming back from the MX saying the MTU is too big, I assume this should be the case.

I haven't modified any of the MTU's on the MX, everything is just the default.

I also have normal layer 3 running over the fibre between the routers and when I use that I don't see any issues, so it must be something to do with VPLS.

Any thoughts would be greatly appreciated.

Thanks
Luca.

_______________________________________________
juniper-nsp mailing list juniper-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp