Dual Sup VSS Stackwise-Virtual failover behavior

[KevinHuang8444]

If I have two Cat6K/Cat9K chassis, each with only one supervisor (dual-sup altogether), and they form a VSS/Stackwise-Virtual, what happens to the linecards when the active Sup fails?
The control plane will failover to the second sup/chassis, but will the dataplane/linecards in the first chassis continue forwarding traffic (non-disruptive)?

My understanding is the Cat6K & 9K are based on centralized forwarding - all traffic has to go through the supervisor engines (except DFC cards).
So when the original active sup in chassis 1 fails, how would the linecards in chassis 1 continue working?

TIA

[KevinHuang8444]

To be clear, I never said using VSS as access layer, or single attach devices would be a design recommendation, or best practice. I merely pointed out there are use cases.

All I really wanted was to confirm the behavior of linecards when supvervisor fails, so that we'll know what to expect.

A simple "yes, the linecards in chassis1 will also fail, won't be able to forward traffic" would've sufficed.

More a hardware architecture inquiry, but it somehow kept turning into a design options discussion.

Appreciate your feedbacks regardless.

[Reza Sharifi]

Just to answer your question with an example for a VSS setup:

All I really wanted was to confirm the behavior of linecards when supvervisor fails, so that we'll know what to expect.

A simple "yes, the linecards in chassis1 will also fail, won't be able to forward traffic" would've sufficed.

example:

VSS:

chassis-1 primary sup 

chassis-2 backup sup

If "chassis-2 back up" fails, nothing will happen, as all the line cards are being manged by "chassis-1 primary sup". All line cards in both chassis still continue forwarding traffic.

If "chassis-1 primary sup" fails, "chassis-2 backup sup" will take over the management of all line cards. All line cards in both chassis still continue forwarding traffic.

HTH

[KevinHuang8444]

Thank you, this is exactly the discussion I was hoping to have.

In Cat 6K and 9600's, the supervisors are the switching fabrics and all packets have to go through them to be forwarded, because they're based on centralized forwarding architecture. (versus Nexus 7K, where switch fabrics are the fabric cards)

What I don't understand is, if the sup fails the linecards in came chassis would lose the switching fabric, so how can they continue working?

See CiscoLive slides 69,70, 90-93 in link BB provided above...they seem to suggest half of the paths would be down. (why else would you lose 50% of the bandwidth?)

[Reza Sharifi]

See CiscoLive slides 69,70, 90-93 in link BB provided above...they seem to suggest half of the paths would be down. (why else would you lose 50% of the bandwidth?)

I am up to speed with the new VSS (StackWise Virtual) but from my days working and testing  VSS, the only time you lose 50% of the bandwidth is when one of the chassis is completely down or in the process of being rebooted. 

HTH

[KevinHuang8444]

Not questioning what you've seen & tested, but I'm curious to know how the linecards can keep forwarding packets, if they lose the switching fabric (supervisor) in the chassis...at least the CFC linecards.