Internal VRRP ?

[michael090608]

Hello, 

I have a C3750 stack which is acting as my core switch for my internal network. 
I am in the process of replacing this stack for a new stack of C9200.
As you can imagine, I need to be careful with this, as this stack is the heart of my network, it's the gateway for my whole infrastructure and also runs my whole ACL.
I recently read about VRRP, and it sounds like an interesting solution that will provide internal gateway redundancy for my network.
My questions is; can I implement VRRP by using a stack of C9200 ? If yes, How ?

Anyone has experiences with this scenario ? 

[Flavio Miranda]

Hi @michael090608 

 It supports, as you can see here:

https://www.cisco.com/c/en/us/td/docs/switches/lan/catalyst9200/software/release/16-9/configuration_guide/ip/b_169_ip_9200_cg/vrrpv3_protocol___support.html

[M02@rt37]

Hello @michael090608,

It should be good as standalone. Confirm here with you IOS version:https://cfnng.cisco.com/browse/switching/products

Stack 9200L are view as single equipement ; why VRRP ?

[MHM Cisco World]

Stack NO you cannot 
standalone Yes You can

but if you have stack why you are looking for VRRP ?
https://oswalt.dev/2011/12/redundancy-protocols-vs-stacking-pros-and-cons/

[michael090608]

I am thinking; VRRP to be able to have two internal gateways running, no matter if it is two stacks or not.
Right now, I am unable to perform maintenance services on my internal gateway because if it goes down, all my network goes down.
This is the reason why I would like to implement VRRP to be able to have two stacks of two of C9200 running as my gateway. This way if I turn down one stack, the other will keep my network up and running, this way eliminating a single point of failure.

[MHM Cisco World]

I am unable to perform maintenance services on my internal gateway because if it goes down, all my network goes down. <<- this point not clear to me can you elaborate,  SVI of VLAN must UP only the mgmt is change from one SW to other what you face exactly 

[michael090608]

Right now, I need to plan the migration of my Core switch, from a C3750 stack to a C9200 stack. 
In order to do this; my whole network will go down, because this is the gateway for my whole network.
What I am thinking is to see if I can implement VRRP on my new C9200 stack, in order to have redundancy of my internal gateway, so that if I need to perform maintenance on my Core switch, it won't disrupt operations.

[MHM Cisco World]

get your point, so we have two stack here 3750 and C9200 
if we can config HSRP between these two stack then we can within short downtime shift the traffic from one stack to other. 
it that right I will check config HSRP/VRRP inter-Stack. 
thanks 
MHM

[Joseph W. Doherty]

Ah, your reply looks to have been posted while I was composing mine.

Unclear why you believe switching to a FHRP, particularly using VRRP, will avoid network outages, or why moving to another stack will cause the whole network to go down.  Have you ever done this before?

BTW, I've done such migrations, many times, with and without having a FHRP.  A hitless change over, generally, can only be accomplished if you designed for such in the first place.  Otherwise, there's almost always an impact to the network, although the extent and the duration of the impact varies (based on different factors).

Although minimizing any network outage, from schedule maintenance, can be worthwhile, if decreasing the chances of a full or partial network outage, at anytime, is important, then I suggest you consider reviewing your design to best accomplish such (rather than just whatever about VRRP appears so worthwhile - which is not to say VRRP, or another FHRP, might be part of a new design).

[Scott Leport]

Configuring a FHRP on your network requires IP addressing changes to your VLAN SVIs and that will cause network disruption, which there is no getting away from. However, once you have a FHRP operational between the new and existing stack, you could do your migration that way.

I'd also recommend HSRP version 2 instead of VRRP, unless you've got a specific reason for VRRP. HSRP version 2 is supported:

https://www.cisco.com/c/en/us/td/docs/switches/lan/catalyst9200/software/release/17-2/configuration_guide/ip/b_172_ip_9200_cg/configuring___hsrp.html#concept_hlw_yns_3gb

[michael090608]

Hello Scott, 

I will look into your proposed solution and get back if needed. Thank you for your feedback.

[Joseph W. Doherty]

Yup, two dual stacks does increase your redundancy, however at the lost of a few features if you just had a stack of 4.

Anyway, @Flavio Miranda and M02@rt37 appear to document VRRP is supported.  Likely, HSRP is supported too, and between Cisco platforms, I would use HSRP.  I.e. I would only use VRRP with non-Cisco platforms.

Even offer better in many cases, IMO, is using GLBP, but without looking, don't know if the C9200s support that.

BTW, @MHM Cisco World notes you cannot use VRRP with a stack, assuming, I believe, you only have a single stack.  However, technically, you can still use a FHRP like VRRP on a standalone stack, but why would you?  Well, I've done it to make it easy to migrate from one stack to a replacement device, i.e. set up new whatever (often a newer stack series), and shift active gateway before I retire old stack.  It's also another way to avoid MAC(s) changing if stack master fails.