Router Redundancy Questions

[Jkloza_2]

Looking for a recommendation.  Question about load balancing between point to point WAN links. 

I'm currently looking for a solution to provide more network redundancy over my two point to point DS3 lines
that connect my remote site.  I have 2 Cisco 3845 routers that terminate a point to point DS3 between two locations.  We are standing
up a second DS3 between the sites, and have purchased an additonal 2 routers.  I would like to either setup load balancing / sharing, or HSRP / VRRP between the routers / sites.

I would rather be able to utilize both lines, and if there is an issue with one, have traffic seemlessly transition to the other active line.  What would be the best way to accomplish this.  I'm in the process of also converting the sites from static routes to EIGRP.

Thanks, and if more information is needed, please let me know.

Jon

[Jkloza_2]

Jon,

Understood, thanks.  I'd be fine with all L3 devices receiving the default-route from my HQ site.  I still have static's that point from my HQ to the small remote site, so I shouldn't really need to redistribute anything associated with that site right? 

From what I can find on the Cisco site, it looks like all that I need to do is have a default route configured in my HQ site L3 device, then use the following command:

router eigrp 100

redistribute static

default-metric 10000 1 255 1 1500

Just a question, is this going to redistribute all static routes that are directly connected to my HQ site?  I'd rather that not happen until after my small 2nd remote site has been dropped from the network (I have a few static's pointing there, that I don't really want pushed out throughout the entire EIGRP topology).

Tyvm,

Jon

[Jon Marshall]

Jon

Yes it will redistibute all statics on the specific router you apply the command to. So if the router you apply the "redistribute static" to has other statc routes on it then you need to use a route-map to only allow the default route.

By the way when you redistribute statics into EIGRP you don't actually need the metric set which is a quirk of EIGRP because when you redistribute any other type of routes into EIGRP you do need the metric set. Won't hurt to have it there. So on the HQ router where you redistibute the statics, assuming it has other static routes -

access-list 1 permit 0.0.0.0

route-map FTR permit 10

match ip address 1

router eigrp

redistribute static route-map FTR

Jon

[Jkloza_2]

Jon,

So i've been working on this in a test environment this weekend, and I have few things that i've noticed.  The redundancy works perfectly, I've got this setup using GNS3, and when I simulate a test failure, the EIGRP topology gets updated within seconds. 

Now, let me show you what i'm seeing at my remote site L3 Switch:

show ip eigrp topology X.X.97.0/26

IP-EIGRP (AS 100): Topology entry for X.X..97.0/26

  State is Passive, Query origin flag is 1, 1 Successor(s), FD is 289280

  Routing Descriptor Blocks:

  192.168.17.2 (FastEthernet2/0), from 192.168.17.2, Send flag is 0x0

      Composite metric is (289280/286720), Route is Internal

      Vector metric:

        Minimum bandwidth is 10000 Kbit

        Total delay is 1300 microseconds

        Reliability is 255/255

        Load is 1/255

        Minimum MTU is 1500

        Hop count is 3

  192.168.15.1 (FastEthernet0/1), from 192.168.15.1, Send flag is 0x0

      Composite metric is (335360/309760), Route is Internal

      Vector metric:

        Minimum bandwidth is 10000 Kbit

        Total delay is 3100 microseconds

        Reliability is 255/255

        Load is 1/255

        Minimum MTU is 1500

        Hop count is 3

show ip cef X.X.97.0

X.X.97.0/26, version 89, epoch 0, cached adjacency 192.168.17.2

0 packets, 0 bytes

  via 192.168.17.2, FastEthernet2/0, 0 dependencies

    next hop 192.168.17.2, FastEthernet2/0

    valid cached adjacency

I see the multiple routes in the EIGRP topology, but when I look into the CEF table, I only see one cached adjacency.  I thought that if the per-destination load balancing was working, I'd see two adjacencies. 

The one thing that I did notice is that the metric is not the same for both routes, do I have to set a variance for the load balancing to work properly, or am I overthinking this?  I know this isn't a "true" test, but it gives me a warm fuzzy before implementing on my live networks.

Thanks!

Jon

[Jon Marshall]

Jon

The metric does indeed need to be the same for it to do equal cost load-balancing. You can, with EIGRP, use unequal cost load balancing using the variance command although some people have reported issues with variance Personally i have never had an issue with.

What we really should do is try and work out why the metrics are different. Perhaps you can post your GNS3  .net file and router configs. I have GNS3 on my laptop so i can take a quick look to see what is happening. 

Won't be able to to look at it until tomorrow though as it's 1:00 in the morning here in the UK.

Jon

[Jkloza_2]

Jon,

Just as I was trying to save my GNS3 configs / files, my CPU decided to lock up...  It will take me a while to re-create (work today...)  I've attached sample configs of my routers / l3 switches that I used.

Every link that I used was a Fastethernet link, so I figured the routers would calcuate the metric to be equal, but it didn't seem to be the case.

Thanks again for your help.

Jon

[Jon Marshall]

Jon

I simulated what i hope is your environment in GNS3 -

1) on DCSW1 i have created loopbacks -

loopback10  10.10.96.1 255.255.255.224

loopback11  10.10.97.1 255.255.255.0

loopback12  10.10.98.1 255.255.255.128

2) on RSSW1 i created loopback10 - 10.11.10.1 255.255.255.0

3) the DF router is not running EIGRP. It is simply used as a next-hop so i can add static routes to DCSW1 and then redistribute the default-route into EIGRP.

If you do a "sh ip route" on RSSW1 you see 2 routes for all the loopbacks and the default route.

If you do a "sh ip route" on DCSW1 you see 2 routes for the loopback advertised from RSSW1

So we need to work out where your metrics are getting changed.

Attached is a snapshot of the network layout from GNS3 + GNS3 config files.

Jon

[Jkloza_2]

Jon,

I tested this again this AM.  It worked flawlessly.  I must have had something mis-configured last night, because this AM, i see both default routes in my routing table, and I also see the redundant lines in the CEF table.

Thanks again for all of your help.

Jon

[Jon Marshall]

Jon

Thanks for letting me know. Happy to have been of help and good luck with the actual implementation.

Jon