Negotiated failover is something I'd rather not invoke: we have a network team here that I'm not a part of and my experience with EVERY network admin has been "the problem is not the network, it must be you" even with all the proof in the world of the network issue - including telling them how to fix their own problems. It's not quite that bad here, but we've had issues where ports have dropped unexpectedly and I really don't want a cluster failover to have to deal with as part of the overall issue. That said, it's a great option, but it doesn't work for my environment due to the split nature of the teams (and even if it wasn't I'd still be hesitant).

-----Original Message----- From: Kevin Graham [mailto:kgraham@industrial-marshmallow.com] Sent: Wednesday, August 27, 2008 3:29 PM To: Glenn Walker; James Beal; toasters@mathworks.com Subject: Re: issues with 10G Ethernet.

We're using the dual-port 10GbE TOE cards with no issues - that said, TOE is not enabled because we're using VIF (Filer turns TOE off with

VIF

• not desired by us, but no choice right now as we cannot have SPOF).

We've addressed this instead by using negotiated failover, both for 10GbE's and to convert failover 2x1gb's to single 4x1gb's; losing the switch the filer's uplinked to is rare enough of an event that (IMHO) a failover is

acceptable.

Perhaps disabling TOE would get you more reliability?

I'm still waiting for jumbos on the TOE to be stable...