• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 623
  • Last Modified:

I need assistance designing a Redundant Cisco Infrastructure

I am in the process of designing a redundant cisco infrastructure for customer.  Right now this network doesn't have any kind of hardware redundancy.  I've been tasked with implementing redundacy for the edge routers and Cisco ASAs to start.  See attached visio, this is my plan so far for redundancy.  How does it look?  Right now there is just one of each device in production, but once I come up with my plan I will be adding the redundant/failover devices.  I'm just curious if this looks like a good design that will accomplish my overall goal of redundancy.    Thanks. Redundancy Plan
0
denver218
Asked:
denver218
  • 2
  • 2
1 Solution
 
jmeggersCommented:
First off, I think you're on the right track.  However, with what you've drawn, you may need a switch on the outside of the routers to connect to the ISP unless the ISP can give you two links with two addresses from the same subnet.  

There's a bigger question, though, of how far you want to carry redundancy.  Is a single ISP really sufficient? Even if the ISP is running dual links you your location, there's probably a good chance they're running through the same media for some distance, and even if they are on separate media, they're probably right next to each other, meaning a single backhoe could take both of them out.  True redundancy means two different ISPs, with entry points on different sides of the building, with fiber going in different directions to different COs, redundant power through different supplies, etc.  (I've heard of at least one instance where a redundant setup out different ends of the building was still not really redundant when both lines wound up in the same bundle and were taken out by a backhoe.)  Of course, that's going to cost more -- more dollars, and more time and effort managing the implementation to make sure all those bases are covered.  Only your customer can determine how much redundancy they really want to pay for.
0
 
denver218Author Commented:
This customers infrastructure actually lies in a Datacenter, where the ISP actually does have dual links and they guarantee 100% up-time.  The ISP does give me two links with two WAN Addresses.  This customer has been in this datacenter for about 18 months so far without any downtime related to the ISP.  The only down-time they have had is when hardware fails, which is why I've been tasked with setting up redundancy.  I'm most concerned with hardware redundancy for the equipment I've shown above.  I understand Its hard to be 100% redundant and so does my customer.  I'm just more concerned about redundancy for the edge routers, ASAs, and Core Switches.  Thanks.
0
 
jmeggersCommented:
If the ISP links are not the problem, and if the addresses they're giving you are in the same subnet, then you should be in good shape.  It would be good to know what device they have on the outside of your routers to make sure that HSRP will failover cleanly, but it's probably not a big deal.  Typically it's a switch you control, but not in this case, so it's the only thing I see that raises a question.
0
 
denver218Author Commented:
Thanks.  I will keep going on this path then.  I appreciate your input.
0

Featured Post

Keep up with what's happening at Experts Exchange!

Sign up to receive Decoded, a new monthly digest with product updates, feature release info, continuing education opportunities, and more.

  • 2
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now