After testing this in the lab I can say that the book is incorrect. I tested
with 2 5505s and 2 6509s. In both cases even with the config rev. zero on
both switches one of the VTP server switches will overwrite the others
database when the trunk between them is brought up. The same switch did not
win each time so I conclude that there is no determining factor past the
revision number. The switch that wins seems to be the one which sends out
the VTP update first.
In theory the book is correct. I have seen it written in several places
(including CCO) that a switch will ignore a VTP advertisment with a rev.
number which is the same, however this is not the case.
Rich
----- Original Message -----
From: "Nelluri Reddy"
To:
Sent: Sunday, July 08, 2001 11:26 PM
Subject: Re: Cisco 5505 switch puzzle [7:11335]
> Stefan
>
> That is what I thought too. After reading the same book and
> documentation on CCO, I thought that I was safe as long as the VTP
> revision numbers were zero on both. I wonder if the IOS version numbers
> have any effect!
>
> Stefan Dozier wrote:
> >
> > Gareth Hinton wrote:
> > >
> > > Yeah that was sort of what I was getting at.
> > > It would be nice if you could put the switches into a learning
> > > mode, where
> > > all server mode switches learn from all other server mode
> > > switches
> > > initially, but that's not the case, so one of them will decide
> > > it's got the
> > > highest VTP version and all others VLAN info will be wiped out.
> > > I can't remember what the negotiation process consists of when
> > > both versions
> > > are zero, or if I ever knew. It's not an ideal way of doing it
> > > anyway.
> > > Probably easier to cut/paste vlan info on to one switch and
> > > make sure it's
> > > got the highest version.
> > >
> > > I've got a few sites with multiple 5500's in one VTP domain and
> > > they're all
> > > set to Server mode. Never caused any problems I can think of.
> > > Off to get another book out to find out which switch wins when
> > > both are
> > > version zero.
> > >
> > > Gaz
> >
> > Quoting from the Cisco Press BCMSC book , chap 3 page 112....
> >
> > Two databases with the same configuration number will not update each
> other,
> > because they assume that they both have the same information.
> >
> > Stefan
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=11480&t=11335
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
