Hey Mark, I can suggest a few tips: I suppose you can change the Bandwidth statements or to be sure, change the 'cost' on each PVC/interface and give the backup a much higher Bandwidth. You can check your routing table for the certain path to check to see if there is more than one path by issuing a 'sh ip route eigrp 123.45.6.0' Then check your costs by issuing a 'sh ip eigrp int' command. Or something like that... "Mark Z." wrote in message [EMAIL PROTECTED]">news:[EMAIL PROTECTED]... > Hi Everyone, > I know i didnt give many specs on my previous question but thats because > today was my first time out at the site. Here's what's going on there and i > believe I have the answer now for tomorrow. > > They have 2 routers in the core (A=primary, B=backup). They have 18 remote > sites, most of which are cisco and there are a few remote nu-bridges (never > heard of these). My concern is with the cisco's for now. This is how they > have it set up. Each site has a primary PVC going to core_A, and has a > "backup" PVC going to core_B. I learned that the secondary is actually not a > "backup" PVC, but it is always up, so if the primary fails then that one is > still good and info will be sent to core_B and then from that to the > destination of core_A. They are using EIGRP as the routing protocol. Now, > what the issue here is: Some of the cisco branches are using the "backup" > PVC initially, even though the primary PVC is not down, basically causing > unnecesary traffic in between the 2 cores and taking 2 hops instead of one > to core_A. I got some help tonight and was told that this was more than > likely a path selection issue and setting the metrics to make the primary > PVC the more feasable route should do the trick. Also, I was told that I > could basically set the bandwidth statements for the 2 PVC's and make the > primary's higher, therefor making it the most feasable route... > > Also, when doing a "show ip eigrp neighbors" command on one of the > trouble branches, I noticed this: Where the uptime for the neighbor > connection is, the Backup neighbor connection (pvc) would say up for like 5 > weeks, and the Primary neighbor connection (pvc) would keep bouncing every 3 > minutes or so (not the frame pvc...it's always up, but the neighbor > connection bounces). I was told that i should check out the hello timers and > make sure that they are all alike throughout the eigrp network. > > Sorry this one was so long guys, I just noticed that a couple of people > took interest. If anybody has anything to add please feel free. Themore i go > in there tomorrow with the better. I have this bad feeling that all of the > above will be set correctly on the devices and I'm gonna be sitting there > with my head in my ass wondering what's next ;) Have a good one all, and > thanks, > > Mark Z. ~ CCNP, CCDA (slowly realizing what it feels like to be a paper > CCNP...but working to change that) > FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html > Report misconduct and Nondisclosure violations to [EMAIL PROTECTED] Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=4034&t=4034 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
