[74500] in North American Network Operators' Group
RE: 10GE access switch router
daemon@ATHENA.MIT.EDU (Temkin, David)
Wed Sep 29 11:54:01 2004
Date: Wed, 29 Sep 2004 11:51:42 -0400
From: "Temkin, David" <temkin@sig.com>
To: "Bill Woodcock" <woody@pch.net>, "Deepak Jain" <deepak@ai.net>
Cc: "Frederic NGUYEN" <fnguyen@t-online.fr>, <nanog@merit.edu>
Errors-To: owner-nanog-outgoing@merit.edu
Bill,
With the right amount of prep work and understanding of how the
stacking works, you can control everything you complained about.
I complained about the same stuff until I read the document that
explains how to:
1) Renumber a switch in the cluster (and all of it's interfaces with it)
2) Hot swap a new switch into the cluster
3) - and this one's sweet - upgrade the s/w on the entire cluster in one
shot, even if they're different models
4) Control which switch is the master so that adding a new switch to the
stack doesn't chance screwing up your configs.=20
5) Permanently remove all stacking config from the switch
The actual backplane has lived up performance wise in the testing I've
done, but I haven't come anywhere near testing it to 32gbps.
Just the same as thousands of people have wiped out every VLAN on their
network by putting in a switch with a higher VTP revision number with no
VLANs defined, it takes a learning curve to work well with these
suckers.
Granted - the software has been somewhat buggy - but those aren't the
merits I'm debating.
http://www.cisco.com/en/US/customer/products/hw/switches/ps5023/products
_configuration_guide_chapter09186a00801a6558.html
> -----Original Message-----
> From: owner-nanog@merit.edu [mailto:owner-nanog@merit.edu] On=20
> Behalf Of Bill Woodcock
> Sent: Wednesday, September 29, 2004 11:33 AM
> To: Deepak Jain
> Cc: Frederic NGUYEN; nanog@merit.edu
> Subject: Re: 10GE access switch router
>=20
>=20
> On Tue, 28 Sep 2004, Deepak Jain wrote:
> > Just a note, if you want redundant 10GE uplinks you=20
> need to get two of
> > these and stack them. The stacking interface does not=20
> reduce the amount
> > of switching bandwidth to the front ports IIRC.
>=20
> ...and the stacking interface is actually pretty lousy, from=20
> our testing.
> We were anticipating really liking it, but we haven't touched=20
> it again, since our lab work. Obviously it precludes=20
> hot-swappability, but beyond that, using it wipes any=20
> preexisting configuration on all but the first box (and out=20
> of two, I don't know how to predict which it will decide is=20
> first, in advance), and it leaves the port-numbering screwed=20
> up on any boxes that have used it, in perpetuity.
>=20
> -Bill
>=20
>=20
>=20
IMPORTANT: The information contained in this email and/or its =
attachments is confidential. If you are not the intended recipient, =
please notify the sender immediately by reply and immediately delete =
this message and all its attachments. Any review, use, reproduction, =
disclosure or dissemination of this message or any attachment by an =
unintended recipient is strictly prohibited. Neither this message nor =
any attachment is intended as or should be construed as an offer, =
solicitation or recommendation to buy or sell any security or other =
financial instrument. Neither the sender, his or her employer nor any =
of their respective affiliates makes any warranties as to the =
completeness or accuracy of any of the information contained herein or =
that this message or any of its attachments is free of viruses.
