Cisco 6509-E VSS Replace Supervisor

[leonardo_liberati]

Good Morning,

we have to replace the sup720 of a vss bundle.

I've read the how-to on cisco.com web site:

http://www.cisco.com/en/US/products/ps9336/products_configuration_example09186a0080a64891.shtml

but I have some questions about the replacement:

When we perform the recovery procedure is likely to malfunction despite one of the two nodes VSS is currently up and running?

What kind of impact we can have?

May occur during recovery procedure a fault on the running machine?

In the event of a fault which these cases the recovery process and especially the estimated timing?

Note: We have a VSS with FWSM (one for each 6509) and with WISM (two for each 6509).

Thanks a lot

Leonardo

[Reza Sharifi]

Leonardo,

You need to have good size outage window to do this, as the duration for each reboot can take up to 20 minutes, specially since you have FWSM and WISM installed in each device.  Are you planning to install the new sup in the stand-by switch? If yes, it should take less time.  If you are planning to make the new sup the primary then it will require multiple boots.

Good Luck

[leonardo_liberati]

In the guide I red that only the replaced supervisor needs to be reloaded...I don't understand the outage that you mean...

[Leo Laohoo]

If you replace the primary supervisor, the secondary has to become the primary.  In VSS-world, the secondary becomes the primary supervisor by rebooting.  OK so far?

Now you put a new supervisor card in.  This, of course, becomes the secondary.  Now if you want this to become the primary, you need to boot the secondary card.  When the secondary card boots, the new card has to boot too.  Guess how long this entire process takes?  > 20 minutes. 

[Jon Marshall]

Leo

This is for my own clarification.

My understanding of VSS is if the active sup fails then there is no interrruption in traffic. The chassis with the active sup is dead but the new active doesn't need to reboot to become primary, it just takes over as active and carries on forwarding. With NSF/SSO it is almost instantanenous.

So if the primary sup has failed the system should just continue forwarding. If you then want to replace the failed sup, you simply follow the procedure outlined in the doc Leonardo linked to. The chassis with the new sup will boot up but you could leave it as secondary so why would there be an outage ?

Further question. If the chassis with the new supervisor is set to preempt the other supervisor does that mean the preempted chassis needs to reboot.

Like i say, this is for my clarification. To be honest i am surprised with all this rebooting as if it is the case you would get better redudancy with just a normal pair of 6500 switches.

Jon

[Leo Laohoo]

My understanding of VSS is if the active sup fails then there is no interrruption in traffic.

We powered down our active supervisor when we noticed that the secondary rebooted.

Further question. If the chassis with the new supervisor is set to preempt the other supervisor does that mean the preempted chassis needs to reboot.

Read somewhere AVOID using the "pre-empt".

Best Practices for VSS Priority/Preemption

6.11.5   Recommended VSS Priority/Preemption Configuration

It is recommend NOT configuring switch preemption since identical hardware setup between Active/Standby chassis is recommended with virtual router-mac address configured and that all links connected to Neighbor switch through MEC links. Preempt ensures that one particular switch will always be active in the end. However, forcing preemption transitions requires a reload and potential traffic outage.

[Jon Marshall]

Leo

Sorry, you've lost me. How does that relate to what i was asking

Edit - the bit about powering down as you've just added a bit.

Jon

[Leo Laohoo]

Sorry, you've lost me. How does that relate to what i was asking

You're too fast!

When we powered down our primary supervisor card, we had hoped that our standby card would remain active.  Y'know, like the way 2960S and 3750 would behave.

But we were mistaken to see that the standby card rebooted.  We came to the conclusion that this is probably the "norm" for the VSS cards to behave.  We have yet to determine if we've done something differently or not.  Either which case, we are booking time for a lab to get this re-checked.

I agree.  If this is going to be the norm, then going VSS method is not the way to go.  I'd rather stick to HSRP.

[Jon Marshall]

Leo

Are you talking about 2 sups in the same chassis or 2 chassis with a sup each ?

My understanding of 2 sups in the same chassis is yes, if you power down the active sup then the standby sup has to reboot because it is acting as a linecard when in standby mode so it needs to reboot to assume the role of suprevisor.

So if it is 2 sups in same chassis what you saw would make sense. If one sup per chassis then i agree, lets go back to normal pair of 6500 switches with HSRP

Jon

[Leo Laohoo]

Are you talking about 2 sups in the same chassis or 2 chassis with a sup each ?

Single-sup per chassis.

Message was edited by: Leo Laohoo  Going through my notes of the event.

[Leo Laohoo]

Freak me!

I've got my notes wrong.  Sorry Jon.

Nuts.

Yup, when we powered down the Active card the Standby went active WITHOUT a reboot.

It was the stuff we did afterwards (like upgrade the IOS) that caused both cards (and me) to go nuts. 

[Jon Marshall]

Leo

No problem, i just wanted to be sure i understood it.

Still would like to hear from Reza because he is still referring to an outage and i can't see why you would need one to be honest.

Like you say Reza seems to have done a lot with VSS so i guess we'll have to wait !

Jon

[Leo Laohoo]

Who gave me a rating for a WRONG ANSWER???

[Jon Marshall]

That would be me

And it deserved it because you confirmed what i thought to be the case but wasn't sure. I don't have any direct experience with VSS and i know you and Reza have so like i say this was for my clarification as much as the OPs.

Jon

[Leo Laohoo]

Oh ok.  Thanks for the rating Jon.

I'm getting confused becasue the VSS is in a very "critical" part of the network.  It's not like some of our switches where I can force which switch is the master or upgrade IOS at any time of the day.

FYI:  We're currently in the cross-roads.  So far, we've got a VSS pair with WiSM-1 in them (we call these two knuckle-heads our "wireless stack") and so far we are not impress.

It took us 2 hours to attempt to upgrade the IOS using the Cisco recommendation about Upgrading a VSS but to no avail.  In the end, we had to do it the old-fashion way and this method worked.

Which is good because come Monday, there's suppose to be a VSS "expert" coming to give us a talk and we're basically compiling a long list of questions and I-don't-think-so.  The list is now resembling a Chinese take-away/take-out menu.