[118321] in North American Network Operators' Group
RE: Cisco VSS-1440 migration query
daemon@ATHENA.MIT.EDU (Jason Giles)
Mon Oct 19 13:09:00 2009
From: Jason Giles <jgiles@e-dialog.com>
To: "leland@taranta.discpro.org" <leland@taranta.discpro.org>,
"nanog@nanog.org" <nanog@nanog.org>
Date: Mon, 19 Oct 2009 13:06:16 -0400
In-Reply-To: <1255964840.6372.19.camel@leland-gandi>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
>From my test, all physical interfaces configs on switch 2 are factory defau=
lted and SVI interfaces deleted on switch 2 upon running the conversion com=
mands.
Switch 1 is intact with the new naming convention as well as SVI's.=20
I cannot comment on glbp as I do not run it. =20
/Jason
-----Original Message-----
From: Leland Vandervort [mailto:leland@taranta.discpro.org]=20
Sent: Monday, October 19, 2009 11:07 AM
To: nanog@nanog.org
Subject: Cisco VSS-1440 migration query
Hi All,=20
Trying to find an answer to a single technical point concerning a
migration of a fleet of Catalyst 6500's to VSS-1440.
I've had a scan through the documentation on CCO (whitepapers, config
guides, migration guides, etc.) but cannot find anything dealing with
this one specific point.
Background:
Assume that on both stand-alone chassis, you have a specific vlan
interface with L3 configuration, such as:
switch1:
vlan 10
name testing
!
interface vlan10
ip address 10.100.100.1 255.255.255.0
glbp 1 ip 10.100.100.254=20
glbp 1 preempt
glbp 1 load-balancing round-robin
!
switch2:
vlan 10
name testing
!
interface vlan10
ip address 10.100.100.2 255.255.255.0
glbp 1 ip 10.100.100.254
glbp 1 preempt
glbp 1 load-balancing round-robin
!
Once switch 1 is converted to VSS mode and is brought up as
ACTIVE/MASTER, it will keep it's layer3 configuration on the VLAN
interface.=20
The question is what happens to the L3 configuration on switch2 when it
comes up in STANDBY/SLAVE?
I can foresee one of two possibilities, but not sure which to expect
before trying to do this in a live enviroment:
either:=20
1. switch2 will boot up straight into NSF/SSO, inheriting its [new L3]
configuration from the active master, overwriting the pre-VSS L3
configuration. (would be preferable).=20
or
2. switch2 will boot up and enter RPR mode with incompatible
configuration for the interface, forcing me to have to manually remove
the L3 configuration from the interface before I can proceed with the
migration. (not really ideal with hundreds of vlans on these things...)
Any idea which one of these would be the case ?
(the glbp will be retained, as it will be running between two VSS pairs
with 10G trunks between them).
Thanks in advance
Leland
