[92450] in North American Network Operators' Group
RE: Lucent GBE (4 x VC4) clues needed
daemon@ATHENA.MIT.EDU (David Temkin)
Thu Sep 21 09:33:33 2006
Date: Thu, 21 Sep 2006 06:32:03 -0700
In-Reply-To: <20060921131217.GA32715@mx.ytti.net>
From: "David Temkin" <dave@rightmedia.com>
To: "Saku Ytti" <saku+nanog@ytti.fi>, <nanog@merit.edu>
Errors-To: owner-nanog@merit.edu
> -----Original Message-----
> From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu] On=20
> Behalf Of Saku Ytti
> Sent: Thursday, September 21, 2006 9:12 AM
> To: nanog@merit.edu
> Subject: Lucent GBE (4 x VC4) clues needed
>=20
>=20
> (oops technical question in nanog, wearing my asbestos suit)
>=20
> Consider this topology
>=20
> GSR - 3750 --(GE over 4xVC4) - NSE100 - NSE100 --(GE over=20
> 4xVC4) -- 3550 - GSR
>=20
> All other fibres are dark fibres, except marked.
>=20
> When we ping either NSE100 <-> GSR leg, when there is no=20
> background traffic there is no packet loss. If there is even=20
> few Mbps, lets say 10Mbps of background traffic we get 1-5%=20
> packet loss on 1500 bytes, and bit less packet loss on small=20
> packets. As background traffic increases packet loss quickly=20
> increases.
>=20
> We tried to replace (GSR-3750) with 7600, but same issue persisted.
>=20
> We've measured both Lucent GBE legs with having loop in other=20
> end and pushing tests from EXFO and Smartbits gear through=20
> the loop, no errors can be detected in RFC tests.
>=20
> There isn't very much that can be configured in the Lucent,=20
> and we've tried pretty much every setting. We've tried to set=20
> autonego on and off in every gear in the path, without any=20
> changes to observed behaviour. We've also tried to use use=20
> 1xVC4, without any changes to the behaviour. All VC4's in=20
> given leg are using same path.
> Even though we test the packet loss pinging from router link=20
> to router link, same packet loss is experienced for transit=20
> traffic also. We've tried to turn PXF off in NSE100. Packets=20
> between NSE100 <-> NSE100 over dark fibre are not lost.
>=20
> We're pretty much utterly without clues. All I can think off=20
> is some obscure IFG issue, that is, NSE100 would have less=20
> than perfect timing for IFG which would confuse Lucent=20
> regarding what is part of which frame. Does stuff like this=20
> really happen?
>=20
> NSE100 drops bad IP packets in PXF and there is only shared=20
> counter, so I can't tell if I get CRC for IP, I just loose=20
> the packets. But IS-IS is not handled in PXF, and I get=20
> %CLNS-4-LSPCKSUM and %CLNS-3-BADPACKET messages over both=20
> Lucent legs, but not between the NSE100's.
> So I assume the packets are not dropped, but broken.
>=20
>=20
> I swear next time I'll complain about some political issue, thanks,
> --
> ++ytti
>=20
Silly question (considering that you stated that IS-IS is borked also,
which is not handled by PXF - but did you try disabling PXF?
There's a reason why Cisco discontinued every product that "features"
it. It's broken.=20
