Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. If you'd prefer to explore, try our test area to get started. And see here for current known issues.

New Member

design for redundancy of access switch

hello all,

I have the following design for a site:

 

Provider delivers channel with HSRP redundant routers, but access level in not redundant and represented by one switch.

Am I correct that there is no way no provide redundancy on access level only with access switches or there are any design proposals?!

7 REPLIES
VIP Super Bronze

Its very difficult to provide

Its very difficult to provide redundancy at the access layer as most end devices (PC, Printers, Laptop, etc..) have only one NIC and therefore can only be connected to one access switch.  In a data center environment some servers have dual NICs and therefor you can aggregate the 2 NICs together and connect them to 2 different switches.

HTH
 

Super Bronze

DisclaimerThe Author of this

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

As Reza describes, many hosts only have a single connection to network, so that single connection is always a possible point of failure.  But there's a couple of things you can do to minimize the impact of a network infrastructure device failure.

When working with small appliance type switches, you might have multiple smaller switches rather than one large switch.  For example, instead of having one 48 port switch, you might have two 24 ports switches, or six 8 port switches, etc..  If a switch fails, not all hosts lose connectivity.

You can also have additional ports, ideally enough to handle lost of any one unit of hardware,  So, for example, if you have seven 8 port switches, when you only need 48 ports, if a switch fails, you only lose 1/7 of your hosts until they can be repatched into available ports on the other switches.

If some of your hosts have multiple NICs, then there are various method to use the two NICs to avoid a single network unit failure from dropping the host.  Usually only shared servers merit that level of redundancy.

Bronze

Hello, Andrey. Let me check

Hello, Andrey. Let me check for a reference that can help your design. Is this for your existing Cisco 2960, by the way? Feel free to e-mail me (adawa@cisco.com) for additional concerns. Kind regards. 

New Member

hello all,thanks for

hello all,

thanks for suggestions.

In this case I think the best design is to put two access switches  instead of one and provide two network sockets for each work place. One of these two network sockets should be connected to one access switch, another - to another access switch. In this case, if one switch fail, customer will be able to use another working socket, which is connected to usable access switch.

Adawa, thanks, I will contact you in the case of necessity.

Super Bronze

DisclaimerThe Author of this

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

Yes, that would work, i.e. using two cable runs.

The more traditional approach would be to repatch at the switches, but sure, having the user repatch at their end would do the trick too.

Oh, and using stackable switches, as Devils Advocate proposes is great, I just want to make clear you can do  similar with two switches that are non-stackable.

New Member

Beware of naughty people!! A

Beware of naughty people!! A network loop could bring down entire network.

Lots of good ideas already

Lots of good ideas already and not much I can add.

Assuming these are end user devices for hosts like PCs/Laptops/Printers etc then have a spare switch (cold spare or 'hot type' using a stack arrangement').

Personally if you want a level of redundancy (as opposed to resiliency) then I would use 2x 48 Port 2960X switches in a stack with dual uplinks, one from each stack member.

Patch half the hosts into switch 1 and half into switch 2.

If one switch fails, repatch the hosts and HSRP will take care of the uplinks.

Thanks

 

150
Views
20
Helpful
7
Replies