From: Rik Guyler (Rik@xxxxxxxxxx)
Date: Fri Jul 26 2002 - 20:52:08 GMT-3
I agree, I have never seen that kind of down time in this setup.
I would also use the HSRP config here. Put the 2 sups with the MSFCs
together and the 2 without MSFCs in the other chassis. Config the 2 MSFCs
just like you would 2 separate routers. Remember - MSFCs don't synch like
the sup engine itself so you have to manually verify consistency in the
configs.
As for the trade-off, if you lose an entire core switch, meaning losing the
chassis with both MSFCs, you might as well be entirely down! ;-}
Rik
-----Original Message-----
From: Tim O'Brien [mailto:tobrien@cinci.rr.com]
Sent: Friday, July 26, 2002 6:21 PM
To: kym blair; cchurch@MAGNACOM.com; jezerski@broadcom.com;
anthonypace@fastmail.fm; Colin.Barber@telewest.co.uk; ccielab@groupstudy.com
Subject: RE: MSFC2 redundancy
Kym,
I believe that you have something misconfigured. HA is exactly that, High
Availability. If the first sup fails layer 2 should switch over in 1 to 2
seconds, and the MSFC should switch over as soon as HSRP detects the primary
down. I believe if I remember correctly it is like a max of 10 or 15
seconds. There is no way it should take a minute to fail over.
You are not running Manual-Mode MSFC Redundancy, are you?
Tim
CCIE 9015
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of kym
blair
Sent: Friday, July 26, 2002 4:30 PM
To: cchurch@MAGNACOM.com; jezerski@broadcom.com; anthonypace@fastmail.fm;
Colin.Barber@telewest.co.uk; ccielab@groupstudy.com
Subject: RE: MSFC2 redundancy
Chuck,
There are a couple drawbacks to using High Availability in MSFCs. Most
notably, you cannot also activate port security.
The advantage of HA is that if your active MSFC fails, you save several
minutes of downtime before the standby MSFC takes control of the switch (but
you'll still have about a minute of downtime, so it's not perfect). MSFCs
seldom fail, so if your environment can accept this risk, you might want to
opt for disabling HA so you can implement port security.
My 2 cents.
Kym
>From: Chuck Church <cchurch@MAGNACOM.com>
>Reply-To: Chuck Church <cchurch@MAGNACOM.com>
>To: "'Joseph Ezerski'" <jezerski@broadcom.com>, "'Anthony Pace'"
><anthonypace@fastmail.fm>, "'Colin Barber'"
><Colin.Barber@telewest.co.uk>, "'ccielab@groupstudy.com'"
><ccielab@groupstudy.com>
>Subject: RE: MSFC2 redundancy
>Date: Fri, 26 Jul 2002 08:39:34 -0400
>
>Thanks for the info. I was intending on using the HA on them , so I'm
>going with what's supported. Was what you saw while running IOS or
>CatOS on the switch?
>
>Chuck Church
>CCIE #8776, MCNE, MCSE
>Sr. Network Engineer
>Magnacom Technologies
>140 N. Rt. 303
>Valley Cottage, NY 10989
>845-267-4000
>
>
>
>-----Original Message-----
>From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
>Joseph Ezerski
>Sent: Thursday, July 25, 2002 7:33 PM
>To: 'Anthony Pace'; 'Colin Barber'; ccielab@groupstudy.com
>Subject: RE: MSFC2 redundancy
>
>
>Tony and Chuck:
>
>I have tried this with the 6509 and a SUP-1 MSFC1 or 2 combo and got it
>to synch. It won't specifically hurt the switch or blow it up or
>render it non functional, but it does sometime wreak havoc on the PFCs,
>which are programmed, in part, by the MSFC. In most cases, the
>hardware mismatch is a
>livable, although not Cisco supported, config. HOWEVER, as soon as you
>start enabling HA on your switch, you will see issues. Trust me, I have
>lived that nightmare. Now, all of our 6509s have like hardware in both
>slots. Just sharing the experience.
>
>-Joe
>
>TP- Best wishes on your next attempt...
>
>-----Original Message-----
>From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
>Anthony Pace
>Sent: Thursday, July 25, 2002 11:45 AM
>To: Colin Barber; 'ccielab@groupstudy.com'
>Subject: RE: MSFC2 redundancy
>
>
>I have tried this with miss-matched supervisors and they have to be
>identical for the SUPERVISOR SYCRONIZATION to occur. Identical
>including the existance or non-existatnce of NFFC. If they are not only
>one comes on line meaning no extra MSSF either. I had a buch of them
>and tried several combos. THis was my experience.
>
>Anthony PACe
>
>
>On Thu, 25 Jul 2002 08:48:06 +0100, "Colin Barber"
><Colin.Barber@telewest.co.uk> said:
> > Cisco have always stated that if using dual supervisors they must be
> > identical (sup version, PFC and MSFC). Anything else is
> > un-supported. I don't know if will work or not, I haven't tried, but
> > it wouldn't damage
>the
> > sup.
> >
> > Remember that you can purchase a redundant supervisor with MSFC for
> > half price. It is not licensed to use as a primary so you cannot
> > order one
>and
>a
> > blank chassis and then move it over (without purchasing the upgrade
>license)
> > but it does save you a large sum of money. The part numbers are
> > WS-X6K-S1A-MSFC2/2 and WS-X6K-S2-MSFC2/2.
> >
> > Colin
> >
> > -----Original Message-----
> > From: Chuck Church [mailto:cchurch@MAGNACOM.com]
> > Sent: 25 July 2002 02:01
> > To: 'ccielab@groupstudy.com'
> > Subject: OT: MSFC2 redundancy
> >
> >
> > Anyone,
> >
> > Sorry for the OT, but couldn't find a definitive answer on CCO.
> > I've got a client who purchased two 6509 switches, one with
> > redundant
>sup2's
> > and the other with redundant sup2's with MSFC2's. This was the
> > config recommended by our Cisco salesperson. Redundant MSFC2's in
> > each switch
>were
> > too pricey. Anyway, my question is can I put a sup2/MSFC2 in each
>switch's
> > slot one, with the non-MSFC2 in each slot 2? Cisco was intending
> > both MSFC2's in the same switch, but if I lose that switch totally,
> > I've lost
>all
> > inter-VLAN routing. The two switches will be tied together with
>multiple
> > (channeled) gig trunks. I was just going to go ahead and try it,
> > but
>not
>at
> > the risk of damaging a supervisor. Does anyone know if this is
>supported?
> >
> > Thanks in advance,
> >
> > Chuck Church
> > CCIE #8776, MCNE, MCSE
> > Sr. Network Engineer
> > Magnacom Technologies
> > 140 N. Rt. 303
> > Valley Cottage, NY 10989
> > 845-267-4000
This archive was generated by hypermail 2.1.4 : Sat Sep 07 2002 - 19:36:45 GMT-3