08-04-2010 02:55 AM - edited 03-04-2019 09:18 AM
Hi,
I am tying to understand this wierd behaviour when a WAN link goes down.
The below logs are from a cisco 7304 router.
Jul 19 08:31:57.704: %DUAL-5-NBRCHANGE: IP-EIGRP(0) 100: Neighbor 164.55.222.105 (POS3/0) is down: holding time expired
Jul 19 08:32:21.124: %LINEPROTO-5-UPDOWN: Line protocol on Interface POS3/0, changed state to down
Jul 19 13:34:40.276: %DUAL-5-NBRCHANGE: IP-EIGRP(0) 100: Neighbor 164.55.222.105 (POS3/0) is up: new adjacency
Jul 19 13:34:41.204: %LINEPROTO-5-UPDOWN: Line protocol on Interface POS3/0, changed state to up
If you notice when the interface actually went down, Eigrp hold timer expired before before the Line protocol shows down.
Is this behavior expected. Is there a way to fine tune the Line protocol to go down earlier and that trigger Eigrp, rather than EIGRP hold time expiring which would probably take 15 seconds for a OC3 link, hello time being 5sec.
Thanks.
08-04-2010 07:51 AM
If the goal is to cause EIGRP to notice an unreliable link faster than 15 seconds than the hello/hold timers on that link could be adjusted. This would need to occur on both sides of the link using the interface level 'ip hello-interval eigrp' and 'ip hold-time eigrp' commands. Something like the following would cause eigrp to declare the link bad after 3 seconds.
int pos x/x
ip hello-interval eigrp 100 1
ip hold-time eigrp 100 3
### 100 is the AS number
Chris
08-04-2010 09:27 AM
Hello Chris,
Thankyou for your response.
Indeed the Link loss can be identified faster by reducing the hello and Hold timers but we consider 15 sec is ok.
But i was trying to understand how much time a POS interface would actually to identify loss of framing when a link goes down and report a line protocol down on the Log.
In this case it is definitely taking more than 15 secs so thats a lot of time.
Victor
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: