02-13-2012 08:03 AM - edited 03-07-2019 04:53 AM
Within my organization, we have a TWTC fiber ring that has roughly 7 nodes comprising the ring. We have been experiencing EIGRP syslog traps on various nodes that comprise the ring sporatically. We have also reached out to TWTC, for we thought that this could potentially be due to an increase in multicast traffic since bringing on additional nodes to our ring, but are not totally certain. I wanted to see if anyone else may have experienced this type of scenario, and if there were any resolutions. Any input that you may have would be appreciated!
Regard,
Mark
02-13-2012 09:15 AM
Its just EIGRPs way of saying it lost a relationship with a neighbor.
There are lots of preliminary questions to ask first.
1. What kind of ring are you referrring to? SONET? DWDM ?
2. How is your routing done on the ring? Are all ring members peers of each other ?
3. How busy are the interfaces that touch the ring? any interface drops? bad frames?
4. How have you configured EIGRP?
02-13-2012 11:04 AM
Hi
In answer to you questions:
1. We are using a SONET fiber ring
2. Routing through the ring internally is by the use of EIGRP. All EIGRP members that comprise the ring are on the same AS#. The local IP subnet is the same for each of the nodes comprising the ring.
3. The node interfaces do get busy, however the error counters are running clean on each of the nodes.
4. Standard EIGRP configurations are applied. Just local networks being advertised out each local node on the same AS#
Hope this information helps.
02-13-2012 12:00 PM
Is there more to the error message ?
Neighbor change has about 7 things that can cause it.
listed below is an link for trouble shooting
It would be interesting to know "when' things started to happen relative to changes to the network.
02-13-2012 12:34 PM
These neighbor changes have been going on for quite some time on the ring, but they have become much more frequent since we brought on (3) additional nodes to the ring roughly a few weeks ago. Primarily we get holding time has expired messages most of the time
02-13-2012 12:54 PM
here is a generic troubleshooting path for holding time expiration.
Couple of other things, If you have multiple subnets behing the devices connected to the ring, make sure you summarize.
How is the sonet service delivered ? Ethernet? or POS ?
02-13-2012 12:56 PM
I see, the service on the ring is Ethernet.
02-13-2012 01:03 PM
D you hand off to the carrier with a routed interface or just a layer 2?
Whats the capacity of the sonet service ?
For each of your locations what are they really delivering on the transport side ?
02-13-2012 01:06 PM
We have a L2 interface that sits in the respective VLANID of 250, and then we also have a L3 SVI assigned the IP range of choice.
The capacity of SONET serivce is Gigabit Ethernet on each of the nodes entering into the ring.
02-13-2012 01:18 PM
Based on all this its possible that the issue is transport related.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide